AMD Dual Core Timing Issue AKA Mouse Bug

[cite] Firewalk:Will test as i have triple boot xp. can anyone recommend any benchmarking proggies for xp and vista. to test i shall revert settings and run benchmarks, then change settings and run benchmarks to compare.. strangely my keyboard seems more responsive.. Firewalk

I don't think thats strange everything in my system seems more responsive with the slower settings
Not sure about best benchmarking software, sisoft sandra, 3dmark?

this strange mouse movement could be memory related too you know. have you noticed any performance lag lately?

lanxxx, gotta say this has made my system feel smoother more reponsive.

one again thank you for the idea

Firwalk

P.S need something else to fix now, quickly running out of bugs real life looming (need a crash)..

[cite] WinLinMac01:this strange mouse movement could be memory related too you know. have you noticed any performance lag lately?

The fact that slowing down the hypertransport link makes a huge difference I would think suggests very strongly that it could be memory related.

Question is why are some people having this problem and not others what is the difference in those systems?

@firewalk, I was just wondering, what are your system specifications?

IMHO, the verdict is still out on this one because it can take hours or days until this slowdown occurs, from my experience. i'm taking extra care not to restart my system so i can see if it is indeed memory related or not. i am also trying to run performance-intensive apps to speed up the results.

we've got varying setups, from what i can see. firewalk's specs as he told me are:
"asus 690 chipset sb600 i think m2a-vm hdmi,CPU: amd x2 4800 (no overclocking),RAM: 4gb ddr2 800mhz, Video Card: ati X1650 ddr3 256 dual dvi."

he and i have nothing in common, system-wise. i have an NF4 chipset and 1GB DDR2 400MHz, not to mention a different GPU. so that sounds a little fishy to me. Ianxxx, what were your specs?

I too am having this issue, I was referred to this thread after I posted my own. Anyway I seem to recall having a problem where my Tiger install would lock up periodically, I was overclocked at the time (stable in XP) so I put it back to stock and it fixed my issues. Now after reading the replies here it makes sense that turning the HT link down or loosening up the RAM timings might in fact fix this issue. I am currently overclocked right now which hasn't made it any worse than before (different CPU than I had with Tiger). I am going to loosen the timings and I will let you know if it makes any difference.

AMD Opteron 165 (Dual Core)
DFI Ultra-D nForce4 Ultra
G.Skill 2gb RAM
eVGA Geforce 6800GS (QE/CI)
Buffalo WiFi (Broadcom chipset)
WD 120gb SATA (Leopard)
WD 250gb SATA (Time Machine)
WD 250gb SATA (Windows XP)
WD 500gb SAT

nforce 4 ultra
3800 + socket 939
2 gig DDR 400
So it looks like me and you have similar systems but firewalks doesn't fit.!

I think we should have all posted under this heading, is it me or does this make anyone else laugh.

Slow jerky mouse movement with HD munching

lanxxx, my system never fits i built it myself.. still running smooth..mouse not gone mad.

@Ian: Hahahaha. Part of me wants you to make a new thread specifically just to have that new heading, but part of me knows we have waaaay too many threads already for this topic (hence my manic gathering of them today).

[cite] agrafuese:Hahahaha. Part of me wants you to make a new thread specifically just to have that new heading, but part of me knows we have waaaay too many threads already for this topic (hence my manic gathering of them today).

could HD stand for something else?

[cite] Ianxxx:[cite] Firewalk:lanxxx, my system never fits i built it myself..

Me too lol

Well waddya know? ME TOO! Hey, this is the common element to bring us all together: we built our own hacks. So the answer must be to buy a Dell, right?

[cite] Ianxxx:[cite] agrafuese:Hahahaha. Part of me wants you to make a new thread specifically just to have that new heading, but part of me knows we have waaaay too many threads already for this topic (hence my manic gathering of them today).

could HD stand for something else?

HotDog...definitely HotDog. HotDog munching.

[cite] agrafuese:[cite] Ianxxx:[cite] Firewalk:lanxxx, my system never fits i built it myself..

Me too lol

Well waddya know? ME TOO! Hey, this is the common element to bring us all together: we built our own hacks. So the answer must be to buy a Dell, right?

I would rather sell my soul to satan and become a hot dog muncher.

aaahahahahahah. oh man. i am actually loling. i needed that.

Wow, this derailed fast LOL I think I was under some influence when I wrote that

[cite] StealthMode:Wow, this derailed fast LOL I think I was under some influence when I wrote that

Where you watching porn at the time?

Have any of the suggestions made any difference to your errr problem?

well, i guess the next logical question is: do you notice the slowdowns after said influence?
or, there's always the classic half-baked question: ever tried using Leopard...ON WEEEEEED?

[cite] agrafuese:well, i guess the next logical question is: do you notice the slowdowns after said influence? or, there's always the classic half-baked question: ever tried using Leopard...ON WEEEEEED?

It's leopard shouldn't it be catnip?

who said catnip, you got any go on sell us some catnip..(shake shake shiver shiver).

i guess that would explain the erratic mouse movements

well, bad news for me guys. i just tried stressing my system by opening all of the apps i have at once. at the same time, i was browsing with firefox (checking my email) when it suddenly crashed. i also noticed that the last app i clicked on to open (which was illustrator) didn't open. so i went to restart firefox, and i got this message: "The application "Firefox" cannot be launched. -10810" i've actually had this error before on random occasions. whenever it happens, it does this for all of my apps. so, for example, i tried to open illustrator again, and it gave me the same error. i then closed all of my apps and tried to open them again, and it gave me the same error for every single one. i tried to open a new finder window to test browsing some folders, but it crashed finder all together and all i could see was my dock. at this point i had to log out and log back in (using keyboard shortcuts, since i no longer had a menu bar). well, that sucked. and worse, now i have to start over my long-term stability test for these memory settings since i basically had to re-initiate a new leopard session by logging out/in. at this point, i am fantasizing about going back to tiger. it's not a good feeling. baaarrrgggghhh! where's the damn catnip now!?!

same thing here, ahd alot of apps open and eyetv same crash 1080 on all apps, restarted and now ok again. seems like its not fixed to me. no mouse jumping to speak of the odd speed up but nothing in comparison to the default hypertransport settings. will try different hypertransport settings to see if i get any better results..

Firewalk

what we really need to know is if anyone has had this problem on a non amd system. Is it purely amd related or is it just a bug in leopard that will be fixed in the near future by apple?

well, i actually looked up the 10810 thing on google, and it's rare to say the least. apple dev's seem a bit clueless about it, so i'm guessing it's not common on "real" macs. there was a post on the apple forums that said this could have something to do with LaunchServices database needing to be rebuilt. the guy gave a link to an app that would do this, but i'm a little hesitant to try it out. i don't want to screw my install up anymore than it may already be. regardless, here's the post if you've got the cajones to try it yourselves (i have a feeling it wouldn't help though): discussions.apple.com/thread.jspa?threadID=1201327&tstart=60 (scroll down to MarkDouma's post for the app)

well I've run the app and nothing seems screwed but will let you know in about four hours time.

I've been reading through this thread and I have both these issues along with some other which I will not mention yet.

Firewalker: Same problem here after running for a while mouse goes mad like its on d***s jumps to corners and so on

Agrafuese: "The application "Safari" cannot be launched. -10810" error (mine is with Saf, not FF)

It's hard for me to believe these issues are latency related. I think when having latency probs we should experience a lot more issues like crashes, slowdowns, freezes even. And these issues even occur when our systems run idle.
I'll try relaxing them anyway. I'm slightly overclocked and timings are also slightly tightend. Mind you, for those playing games in windows and loosening the timings by a lot you will loose some FPS. That could be a prob if you need all the FPS you can get.
I believe my HT is still at the default of 5x so I will try the 3x but I think it's more of a placebo effect lads.

Anyway, thx for all the info you guys put here and hopefully we can sort this issue out soonish.

Laters

[cite] agrafuese:well, i actually looked up the 10810 thing on google, and it's rare to say the least. apple dev's seem a bit clueless about it, so i'm guessing it's not common on "real" macs. there was a post on the apple forums that said this could have something to do with LaunchServices database needing to be rebuilt. the guy gave…

This is worth someone else trying it, it certainly doesn't seem to brake anything which is the main thing. I'm not sure its working as I said earlier I don't get the funny mouse movement thing any more since changing memory. But my computer has been on since about 3 am thats over 7 hours ago, nothing has crashed so far.

[cite] farcaller:Can someone please tell me how could hardware problem (timings) be magically fixed by just re-spawning a bunch of processes?..

It couldn't

But that's like asking how could cpus=1 fix a networking problem.

Alright lads, a little update.

I've ran my PC with HT x3 this morning and got even worse problems then before.

I'll post a little diary on what happend:

- Safari Crashes
- When trying to restart Saf it shows 10810 error
- Menu icons have dissapeared except for Quickeys (3rd part app) and Spotlight
- Cannot start a single app anymore
- When closing apps that are running I loose a bit of the screen content, meaning, Dock quits, then Menubar closes when trying to quit from the menu.
- Can't even reboot anymore and nothing is displayed on the screen except for tha wallpaper.
- Need to RESET PC to get things fixed.

Voila, there you have it. Before this Saf would just crash or I would get the mouse behaiour but nothing liek this. Coincidence or cause of HT adjustment, I have no idea. I do know though that the mouse (or all else for that mater) wasnt any smoother by changing HT setting.

I have now lowered my timings too so I'm running HTx3 with 3-4-4-8 timings (default timings for my RAM)
I'll keep you guys updated!

Btw for testing I suggest you open iTunes, Saf, Mail and another app like Photoshop and leave it there for a few hours. Then come back to the PC and start to work with it.

Good luck!

And now Time machine doesnt work anymore too. Could not find backup volume error. It has been a while since I backed up so can't tell what happend since then. Nothing major though.

Tried Erasing, reselecting, other SMBIOS & IONetworking, all to no result.

I'm another one suffering from these issues. I think it's possibly connected to having multiple cores. I found this thread on Insanelymac about X2 users, not sure if its still valid for leopard.

http://forum.insanelymac.com/index.php?showtopic=7405&a… ↗

It talks about the use of kernel flags such as idlehalt=0, cpuidle=0, platform=X86PC, fn=3. Also here's a quote from the thread:- "I think the key it is in FSB…OSX loves to run in high FSB."

Edit: Firefox is also crashing with the 'divide by zero' error. Not sure if thats connected at all?

Well i've found out what the -10810 error is. It happens when you have runaway spawn processes. Apparently its often caused by having SMB sharing on, i've turned that off for testing now. Fingers crossed, not sure if this is related to the mouse issue.

Do any of you guys have SMB sharing turned on?

@voyn1x: no, i don't have SMB turned on.

@farcaller: it seems i recall the same keyboard issues.

@snow: next time instead of force-restarting, do apple+shift+Q and it will allow you to log out. then you can log back in with GUI working again.

ok here's a little report for now: i've had my system up for 10 hours now without reboot. i have ianxxx's suggested memory settings: cas latency @ 3, cycle time @ 18, row refresh @ 18, HT @ 3x. things are still running fine, however, i haven't tried stressing the system out again because i don't want it to freak out and give me the 10810 error once more (thus forcing me to reboot and start my long-term testing all over again). admittedly, most of these 10 hours were spent with the system being idle because i was asleep, so only about 2-3 hours of that time was me actually using the system. so, nothing intensive was going on. just a little bit of itunes, some web browsing, and some entourage email checking.

@ianxxx: not to be a pest, but for me a 7-hour stint still isn't enough. i've had this problem happen after many more hours than that. in fact, to me it seems like it may not even be an issue of time at all, but rather just the right combination of apps and usage. for example, i've had this issue happen after using Linotype Font Explorer (it's a great app for font organization, btw, if anyone is interested. google it.). this app, i believe, it highly RAM and CPU intensive, and i did notice once or twice that my system started acting up after it had been up for an hour or so. also, as i mentioned in a previous post, i used to run Tomato Torrent (not anymore though, because it sucks in general) and that also seemed to cause the problem, especially when i'd leave it running with torrents over night. i'll run Font Explorer now for a while and see if i can get the system to act up again.

now, here's a noob-ish question: what is this talk about "spawns"? can someone explain that to me? if it is what i think it is, then it makes sense because essentially Tomato torrent has many simultaneous processes going on (connecting to many peers, gathering data, etc.)...and Font Explorer also has this same thing going on because it has to process many different fonts (gathering info on the font, displaying the name of the font with the actual typeface, etc.).

ok that's all for now. sorry for the TLDR guys, hope you actually read all of this

I stopped reading at the @ sign bro, then I went back to reading when I saw my nick mentioned

Thx for the suggestion on logging out.

So to add to this, I havent got any probs so far running my RAM at 3-4-4-8 and HTx3. Before this I was running my RAM at 2,5-3-3-7 and HTx5. And this time, with relaxing my RAM timings Leo does feel a bit more responsive overall. When this turns out well I think I will bump the HT back up to x5 cause I have a feeling it might only be the RAM timings.
Let's hope it stays this way. I'm willing to cope with the lower timings if it fixes the whole crash/mouse issues.

Thx all for the input and fingers crossed!

Just a little update: I've got music playing in iTunes, a large file transfer going on in Transmit, iChat open talking to someone, Safari with multiple windows, and Firefox just sitting idle (wow, it's not crashing!). I also just opened Font Explorer up and activated an additional 60 fonts (up to about 350 fonts all together now). If this stays stable for a while, I'll add a little more stress by opening up Photoshop and Illustrator. At that point, it will be a pretty typical working scenario for me, and therefore a good test for the problem to arise. Here's one question I have though to derail this whole memory issue:

I've never bought RAM for a "real" Mac before. So how does it work? Can you just go on newegg.com and buy some Kingston RAM or Corsair, or some other fairly "standard" RAM, or do you HAVE to buy it from an Apple store? The reason I ask is because what would happen if I stuck my RAM in a real Mac? Would these problems be happening? Furthermore, can you change memory timings on a real Mac? It seems to me that if these things aren't necessary for a real Mac that there could in fact be some kind of tweaks done to Zeph's release so that we don't have to worry about these memory timing settings. Also, this kind of safeguard MUST have been implemented in my old Tubgirl version of Tiger 10.4.10 because wouldn't it also need to have specific memory timings if this weren't the case? Don't get me wrong (Ianxxx, lol), I'm not trying to kill your theories, but I am still skeptical because it seems really ridiculous for every AMD user to have to tinker around with his system. And what about Intel PC users? Are they worrying about this? Or does Kalyway take care of it somehow? I really think we need to push this in Zeph's direction and see what he has to say. We have different systems! There are enough of us here now that this can't be small potatoes. There must be a blanket solution for us all. Anyone with me on this?

damn, where did everyone go? anyone else have any news on this?
i was running for a solid 21 hours without rebooting. no crashes, no slowdowns. then, suddenly, safari started crashing out of nowhere. the first two times it crashed, i could start it back up, but then after the third, it gave me the dreaded -10810 error on all of my apps and nothing would start anymore. if anyone cares to look through it, here is my console log:

3/17/08 11:20:18 PM ReportCrash[796] Formulating crash report for process Safari[721] 
3/17/08 11:20:19 PM ReportCrash[796] Saved crashreport to /Users/agrafuese/Library/Logs/CrashReporter/Safari_2008-03-17-232013_AMD-Mak.crash using uid: 501 gid: 20, euid: 501 egid: 20 
3/17/08 11:20:19 PM com.apple.launchd[95] ([0x0-0x99099].com.apple.Safari[721]) Exited abnormally: Floating point exception 
3/17/08 11:20:20 PM kernel CODE SIGNING: cs_invalid_page: p=798[ATSServer] clearing CS_VALID 
3/17/08 11:20:36 PM /usr/sbin/ocspd[800] starting 
3/17/08 11:24:19 PM ReportCrash[802] Formulating crash report for process Safari[799] 
3/17/08 11:24:20 PM ReportCrash[802] Saved crashreport to /Users/agrafuese/Library/Logs/CrashReporter/Safari_2008-03-17-232417_AMD-Mak.crash using uid: 501 gid: 20, euid: 501 egid: 20 
3/17/08 11:24:20 PM com.apple.launchd[95] ([0x0-0xad0ad].com.apple.Safari[799]) Exited abnormally: Floating point exception 
3/17/08 11:24:21 PM kernel CODE SIGNING: cs_invalid_page: p=804[ATSServer] clearing CS_VALID 
3/17/08 11:26:58 PM /usr/sbin/ocspd[809] starting 
3/17/08 11:28:39 PM ReportCrash[813] Formulating crash report for process launchd[95] 
3/17/08 11:28:39 PM ReportCrash[813] Saved crashreport to /Library/Logs/CrashReporter/launchd_2008-03-17-232839_AMD-Mak.crash using uid: 0 gid: 0, euid: 0 egid: 0 
3/17/08 11:28:39 PM com.apple.launchd[1] (com.apple.launchd.peruser.501[95]) Exited abnormally: Floating point exception 
3/17/08 11:28:39 PM kernel CODE SIGNING: cs_invalid_page: p=814[mdworker] clearing CS_VALID 
3/17/08 11:28:42 PM ReportCrash[813] Formulating crash report for process Safari[807] 
3/17/08 11:28:43 PM ReportCrash[813] Saved crashreport to /Library/Logs/CrashReporter/Safari_2008-03-17-232839_AMD-Mak.crash using uid: 0 gid: 0, euid: 0 egid: 0 
3/17/08 11:28:53 PM /System/Library/CoreServices/Dock.app/Contents/MacOS/Dock[107] spawn_via_launchd() failed, errno=22 label=[0x0-0xb20b2].com.apple.Safari path=/Applications/Safari.app/Contents/MacOS/Safari flags=1 

the good news is, it seems the slowdowns have disappeared for now. if i could choose, i suppose i much prefer this -10810 error to the slowdowns, but the truth of the matter is i have to reboot my system either way. i still haven't run that app that Ianxxx tried from the link i posted. maybe i should take the leap now :\

Argh bad news, I could say it's just Saf being a dick and that real macs also have this issue but since the error is rare I think we can still conclude it's AMD related (or at least OSX86)

I need to do some more testing to post updates on this matter m8.

I'll keep you updated.

After a lot of searching, i have discovered that Apple ensures that all memory modules meet their timing requirements by not POSTing if incompatible memory is installed in the system and that it is very important that Mac users should only use memory that's specifically made for the Mac.

Because all Macs configure their memory timings based on the SPD that Apple specifies, all memory is configured to have the same timings regardless of what the modules are capable of - in other words, all Mac memory performs the same!

This would explain why Ianxxx suggestion of playing with the ram timings seems to work.

@Agrafuse: Open up terminal and type 'top'. This gives the terminal equivalent of Activity Monitor. Keep that open and when you start to get the -10810 error have a look to see how many processes are running, you should be running 55-80 on average. If the amount has sky rocketed you should be able to see the offending process.

I'm gonna throw something new in to the mix, my broadband got disconnected last night problems with provider, so I did a test, put my memory overclocked etc.
Its still running this morning, I left it on when I came out of the house and will be back to see whats happening around dinner, but I'm begining to think that this is in someway network related.
And that the memory is just like the cpus=1 which helps but is not the cure.
for info I am using a realtek 8169 lan card with the apple kext supplied on 10.5.2 rev 1 dvd.
what configs are others using?

Actualy I was thinking the same thing yesterday. Remember the crash issues with the net on Zephs rev1 release (first one)
Mind you, I'm not using the Nforce LAN with fixed forcedeth or whatever but I'm using the MArvel Yukon (other LAN port on my ASUS A8N SLI)

And btw, by overclocked do you mean higher speed and/or tighter timings?

Cause I'm still running my ram overclocked at 230 but with lower timings 3-4-4-8 (default)
Only the timings seem to affect the issue for me so far (I'm not jumping yet)

So for me I went from HTx5 FSB 230 RAM 2,5-3-3-7 (which I have always used for such a small OC) to HTx4 FSB 230 RAM 3-4-4-8

Hope that helps with info

my ram is now set at htx3 fsb 450 3448

The more I think about it the more i think this is a network bug. I ran my system succesfully for over 8 hours no problems and "no internet".
So the question is "us with the bug, what do we have or do different to those without.
So question to anyone who doesn't have it, did you change anything network related since you installed leopard? Different IONetworkfamily.kext? Etc?
For those who do have it, are you like me no changes to network since installation. Also any thing else anyone can think of, I'm thinking this must be something really simple and really subtle as some don't and some do and there doesn't seem any apparent link, what are we all missing?

Yes! I woke up this morning and began to believe it was network related too! Here's why: I can almost 100% guarantee that this problem will occur when I've left my computer on overnight with torrents running (this time I was using BitRocket, before it was Tomato Torrent). I have a feeling it is also upstream-related because I only had one torrent downloading. It was only around 100MB, and it finished just before I went to bed. So performance was still good when it finished and started to seed. I have an auto-stop when ratio hits 2:0, so after I seeded 200MB, there could be no more peers connecting. So this problem must happen fast, with only a few peers connecting over a short period of time. I should also say that I have Little Snitch running, and I use it in demo mode (so it stops after 3 hours into the night). I have no idea if this has anything to do with the problem at all, but it's just some more info to throw at you all. Also, one last thing I've noticed is ALWAYS associated with this problem is this console message that repeatedly occurs throughout the night:

3/18/08 8:41:32 AM mDNSResponder[17] mDNSPlatformRawTime went backwards by 4 ticks; setting correction factor to 206737245 

The number of "ticks" differs, and it seems that the higher the number, the worse my mouse movements and GUI will perform. I've had it go up to 15 or 20 in the past. Also, the more frequently this message occurs, the worse performance is too. For instance, right now the mouse and GUI are slow but not as bad as usual, and I noticed that the tick numbers are all around 3-4 and the message only occurred a total of 5 times at random instances throughout the night/morning.

...now, for the cherry on top...I found this little line in there as well:

3/18/08 9:11:38 AM kernel skgeosx ERR: ouputPacket() zero segments 

skgeosx is the network driver I am using.

I am using the same network setup as Snow. Marvell Yukon 88E8001. I too have the nLan adapter, but I haven't been using it because I had problems with it in the past. I am running 10.5.2 rev1 right now, so I don't know if it would work better now with this release. Maybe I'll give it another try. What network adapter/driver are you guys using?

[cite] agrafuese:Yes! I woke up this morning and began to believe it was network related too! Here's why: I can almost 100% guarantee that this problem will occur when I've left my computer on overnight with torrents running (this time I was using BitRocket, before it was Tomato Torrent). I have a feeling it is also upstream-related…

I am using a realtek 8169, so thats different, just out of interest are you using a router? If so what have you set mtu to in osx and on router?

yeah, i'm running the classic WRT54G router. MTU is set to auto (1500) in the router and auto (1500) in osx as well. i've always had these settings in osx since the days of tiger 10.4.8. what does your realtek adapter use for a driver? does it use its own third-party driver?

yet another update:

i decided to install another fresh copy of 10.5.2 rev1 on a spare partition just now, but this time i am using nLan for ethernet (forcedeth) instead of my marvell (skgeosx), which I have now disabled completely in BIOS. everything else is totally pure on this installation except for the video kexts i need to modify for my video card to run. i also updated leopard (including the new safari 3.1), and patched everything with marvin's. for now, i've decided not to install AppleSMBIOS-27 or the fixed IONetworkingFamily, and i am leaving the mach_kernel at the default 9.2.0. if i get problems with this setup, i will replace the AppleSMBIOS and then try the fixed IONetworkingFamily. as a test, i will be downloading a 3GB torrent with bitrocket and little snitch running. may leave it to seed over night as well. i'll report back with my results when i find out something new.

[cite] agrafuese:yeah, i'm running the classic WRT54G router. MTU is set to auto (1500) in the router and auto (1500) in osx as well. i've always had these settings in osx since the days of tiger 10.4.8. what does your realtek adapter use for a driver? does it use its own third-party driver?

My lan card uses apples driver, so I haven't changed IONetworking, I'm not sure which AppleSMBIOS I'm using can't remember wether I changed it on this installation or not.
I have a wag54gs router just updated the firmware on it "you never know".

Hmm, I still have a Realtek 8139 PCI network adapter laying around that I actualy used for JAS 10.4.8 and Tubgirls 10.4.9 (10) in the past. Maybe I should try the card again. I haven't had any issues so far though so I'll wait till they start to occur again.
Im' still at HTx4 FSB230 3-4-4-8 here.
I remember the issue with the nForce LAN was that the OS crashed when using iTunes radio stations or Front Row movie trailers or cd previews. I don't even know which skge kext we're running anymore lol, is it a modded one from Zeph or?
Thing is though, Lanxxx is using a totaly different network solution and still has the same issue.
For now we can just update this thread with our latest experiences.
Did all of us (with the issues) change the SMBIOS kext? I wonder. My Time Machine broke too and I can't figure out what happend. I did change the SMBIOS cause I had issues with Photoshop but, I've tried the old ones again and Time Machine is still broken. Photoshop doesn't crash anymore though, for now.

Laters

Snow, I just thought of something. Maybe when you changed your SMBIOS, it affected the way your MAC address is displayed to Time Machine, hence the MAC address error it's giving you? Perhaps when you changed it back, you didn't reload the kext cache? We're a little off topic on this now, I guess.

Oh, I checked the skgeosx driver that we have from Zeph and it appears to be the same one that was initially released by the guy who made it. Forcedeth, on the other hand, was modded by Zeph, afaik. And I don't know if I even have the latest one. I am using 10.5.2 rev1, but has he done anything new to that kext since then? Anyone know? It doesn't say so in his rev2 thread. It worked alright for me, but it didn't display or connect to network computers properly, so I had to dump it and go back to skgeosx for my Marvell instead. Bah. So much for that test.

Hmm, I always use Kext Helper lately which seems to update the cache.

Btw , Agra, YOUR BLOODY INBOX IS FULL ON INSANELYMAC! lol
Guess I wasn't the one to clean out my nox after all.

Anyway, is there any way to completely reset or even reinstall Time Machine?
Ah so it does work, no crashes but it doesnt work well on the local network. I don't think he updated it since rev. 1 m8.

Was about to send you the cpui.txt until I got the inbox is full error from you so I will paste the link here bud:

send space.com/file/jhq2sg

I've edited the file so it only displays the one file that needed to be patched. This is after I've done Safari 3.1 and latest Airport and Security updates.

Good luck son!

Hiya all
well hope in a way I'm wrong about this, cos it will piss me off a little if it turns out to be right.
So far uptime just over 5 hours.
I have to go out in a few mo's and won't be back till tomoz so can't test any longer till then, so maybe someone would like to try this if you haven't already.
I only made two changes
1 I updated the firmware for my router. linksys wag54gs
2 configured my lan card to half duplex 10baseT <-really hope this isn't the answer. Having gone out and purchased a lan card for £7

Anyway so far no crashes no weird mouse, just running fine. opened about a dozen apps I'm not using just to stress pc a little more, still fine.

now 6 hours no faults

I don't have a router nor did I configured the lan card.

So far I haven't ran into any issues after I've relaxed my RAM timings. Only thing that's been broken now is Time Machine.
I just need to keep working until issues occur (hopefully this wont happen)

I do suggest trying one thing at a time. I changed the applesmbios recently and that could be the cause of Time Machine not working anymore but I don't seem to have the Photoshop crash.
So for now I'll just focus on the ram timings.

Uptime here about 4 hours

[cite] Snow:I don't have a router nor did I configured the lan card. So far I haven't ran into any issues after I've relaxed my RAM timings. Only thing that's been broken now is Time Machine. I just need to keep working until issues occur (hopefully this wont happen) I do suggest trying one thing at a time. I changed the applesmbios…

Did you use the fix for time machine?

Yeah bud, no good.

In time machine prefs I get: Backup volume could not be found

My thread about this issue is here: osx.kb ot.de/comments.php?DiscussionID=775&page=1#Item_8

Last night I gave up on the nLan adapter after I discovered that it doesn't handle local network shares properly. So I went back to my original Leopard installation with skgeosx/Marvell lan adapter. System has been up 14 hours now and slow jerky mouse is back. This time it's less intense, but I'm leaving my system on to see if it gets worse over time. Some notes: Last night when I left the system on, I forgot that Little Snitch was off, and I was running Transmission. Coincidentally, Transmission crashed only two hours after I fell asleep, so I didn't get a good test out of that. However, I re-opened it, and turned Little Snitch back on. After about 2-3 hours, the mouse jerking is happening. This is also what I was doing while it happened:

Safari 3.1 - I was aggressively browsing and downloading about 30 MP3 files ranging from 2MB-16MB in size.
Quicktime - I was sampling each of the MP3s I was downloading.
iChat - talking to one friend (about the MP3s, hehehe).
Transmission - as I already mentioned, downloading a 3GB file from approx. 20 peers and seeding at the same time to others.

As soon as I noticed the mouse jerking, I stopped Little Snitch to see if it would help. It either didn't help or the difference was very slight.

Now, as I check my console again, what do I see? Oh! The f**king mDNSresponder tick messages and a host of cs_invalid_page messages again! This has to be somehow related, and the only things that come up on google about these messages are related to hackintoshes or VERY rare "real" mac problems not related to our issues at all. Here's a text file of the console messages. Go through them if you'd like, and let me know if you get anything at all similar to this. Most importantly, I want to know what "lssave" is, and if it has anything to do with Little Snitch, because I am getting that particular cs_invalid_page message A LOT (as you'll see).

<a href="media fire.com/?gyaimyb06ks">media fire.com/?gyaimyb06ks</a>

Ok! I just googled the "lssave" thing and found out that it has to do with the same LaunchServices thing I was talking about a couple days ago. Coincidentally, I found an apple discussion link that once again pointed to that LaunchServices database rebuilder that I gave the link to. I will now try it out and report back with results (if any).

EDIT: Ok, tried it out, and no luck on clearing these "lssave" cs_invalid_page messages. I don't think it will help me out in any other way, but I'll report back if it does somehow.

I know this might sound like a very basic problem, but what is the severity level of this issue? Is this mouse movement noticed in all of your computers everyone? Or am I the only one not facing problems with this release, lol. =)

I would like to hear. Thanks everyone.

I had a whole big post typed out but i accidentally hit command+left arrow and lost it

Long story short, I'm in the same boat as the rest of you (computer is a x2 3800+ OC'd to 2.5Ghz via FSB increase running zephyroth 10.5.2 rev1) and have the seemingly random 10810 errors. I get random reboots too while I'm away from my computer, so it's never on long enough to see the jerky mouse all the time.

I tried running everything I have at once to run it out of threads, but it didn't work. I'm going to try relaxing the FSB and see if that helps tomorrow.

what mobo/chipset do you have? and what kind of ram do you have? do you have any constantly running apps in the background (like little snitch)? at this point i've given up on finding a solution, and i just want to know what common link divides those of us who are having these issues from those who are not. there has to be an answer.

for me, a sure-fire way to get the system to start acting strange was to run torrents with little snitch turned on. so i turned off little snitch and started downloading a torrent, and so far i've been running for 10 hours without an issue. additionally, i've only seen ONE mDNSresponder "tick" error since. normally i would have seen about 10 by now. that's all i've got to say though i'm sure this problem will return in some way or another.

here's a little something to try out, posted by bhast2:

osx.kb ot.de/comments.php?DiscussionID=676&page=1#Comment_10072

the theory is that cutting down on system logs will free up performance a bit. try it for yourselves, but give it some time guys. post back with your results. after i do it, i'm gonna try for a 24 hour run with some torrent downloads and little snitch.

that would be amazing if cutting the system log works, crossing my fingers. =)

[cite] agrafuese:here's a little something to try out, posted by bhast2: osx.kb ot.de/comments.php?DiscussionID=676&page=1#Comment_10072 the theory is that cutting down on system logs will free up performance a bit. try it for yourselves, but give it some time guys. post back with your results. after i do it, i'm gonna try for a 24…

As i can see this doesn't work at all still get the errors

I have an NF4 DFI Lanparty Ultra-D with G.SKILL 2GB (2 x 1GB) 184-Pin DDR SDRAM DDR 500 (PC 4000), and a radeon x1900xt. I had little snitch installed at one point then I decided I didn't need it and uninstalled it. I haven't ran a torrent for a few weeks. Nothing starts at login.

How is everyone getting on with this? Still having problems?

Still no issues here bud. Same RAM settings (HTx4 / 3-4-4-8)

If it turns out to be true then you made my system 100% stable, ya silly tit

[cite] Snow:Still no issues here bud. Same RAM settings (HTx4 / 3-4-4-8) If it turns out to be true then you made my system 100% stable, ya silly tit

Gee thanks

Woops spoke too soon bud, PS just crashed again. Hmm, this didnt happend in a while though. I'll just post the console log here:

20/03/08 20 maart 20:34:04 Adobe Photoshop CS3[339] *** _NSAutoreleaseNoPool: Object 0x1df78820 of class NSThread autoreleased with no pool in place - just leaking
Stack: (0x925aa12f 0x924b6ec2 0x924bc954 0x1e32c6 0x52e71a 0x61e9e9 0x219bee 0x219d31 0x3272 0x3199)
20/03/08 20 maart 20:34:07 [0x0-0x4a04a].com.adobe.Photoshop[339] objc[339]: Class EpicController is implemented in both /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_personalization.framework/adobe_personalization and /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_registration.framework/adobe_registration. Using implementation from /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_registration.framework/adobe_registration.
20/03/08 20 maart 20:34:07 [0x0-0x4a04a].com.adobe.Photoshop[339] objc[339]: Class EpicJSWindowExternalHandler is implemented in both /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_personalization.framework/adobe_personalization and /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_registration.framework/adobe_registration. Using implementation from /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_registration.framework/adobe_registration.
20/03/08 20 maart 20:34:07 [0x0-0x4a04a].com.adobe.Photoshop[339] objc[339]: Class EpicWizardController is implemented in both /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_personalization.framework/adobe_personalization and /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_registration.framework/adobe_registration. Using implementation from /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_registration.framework/adobe_registration.
20/03/08 20 maart 20:34:07 [0x0-0x4a04a].com.adobe.Photoshop[339] objc[339]: Class EpicPanel is implemented in both /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_personalization.framework/adobe_personalization and /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_registration.framework/adobe_registration. Using implementation from /Applications/Adobe Photoshop CS3/Adobe Photoshop CS3.app/Contents/Frameworks/adobe_registration.framework/adobe_registration.
20/03/08 20 maart 20:35:33 kernel CODE SIGNING: cs_invalid_page: p=361[lssave] clearing CS_VALID
20/03/08 20 maart 20:40:43 kernel CODE SIGNING: cs_invalid_page: p=364[mdworker] clearing CS_VALID
20/03/08 20 maart 20:45:53 kernel CODE SIGNING: cs_invalid_page: p=369[mdworker] clearing CS_VALID
20/03/08 20 maart 20:46:07 mDNSResponder[23] mDNSPlatformRawTime: last_mach_absolute_time EC45A25300000EAB
20/03/08 20 maart 20:46:07 mDNSResponder[23] mDNSPlatformRawTime: this_mach_absolute_time EC3520C400000EAB
20/03/08 20 maart 20:48:04 Adobe Crash Reporter[373] Process Id:
20/03/08 20 maart 20:48:04 Adobe Crash Reporter[373] 339
20/03/08 20 maart 20:48:04 [0x0-0x4a04a].com.adobe.Photoshop[339] 2008-03-20 20:48:04.270 Adobe Crash Reporter[373:10b] Process Id:
20/03/08 20 maart 20:48:04 [0x0-0x4a04a].com.adobe.Photoshop[339] 2008-03-20 20:48:04.271 Adobe Crash Reporter[373:10b] 339
20/03/08 20 maart 20:48:13 ReportCrash[374] Formulating crash report for process Adobe Photoshop CS3[339]
20/03/08 20 maart 20:48:19 ReportCrash[374] Saved crashreport to /Users/Snow/Library/Logs/CrashReporter/Adobe Photoshop CS3_2008-03-20-204805_Moeder.crash using uid: 501 gid: 20, euid: 501 egid: 20
20/03/08 20 maart 20:48:19 com.apple.launchd[119] ([0x0-0x4a04a].com.adobe.Photoshop[339]) Exited abnormally: Bus error
20/03/08 20 maart 20:48:20 kernel CODE SIGNING: cs_invalid_page: p=376[ATSServer] clearing CS_VALID
20/03/08 20 maart 20:48:37 kernel CODE SIGNING: cs_invalid_page: p=380[lssave] clearing CS_VALID
20/03/08 20 maart 20:48:38 kernel CODE SIGNING: cs_invalid_page: p=381[Console] clearing CS_VALID

I booted PS back up and bam, after a few mins another crash.

Log:

20/03/08 20 maart 21:05:31 Adobe Crash Reporter[400] Process Id:
20/03/08 20 maart 21:05:31 Adobe Crash Reporter[400] 388
20/03/08 20 maart 21:05:31 [0x0-0x53053].com.adobe.Photoshop[388] 2008-03-20 21:05:31.407 Adobe Crash Reporter[400:10b] Process Id:
20/03/08 20 maart 21:05:31 [0x0-0x53053].com.adobe.Photoshop[388] 2008-03-20 21:05:31.409 Adobe Crash Reporter[400:10b] 388
20/03/08 20 maart 21:05:35 Adobe Crash Reporter[400] (null)
20/03/08 20 maart 21:05:35 [0x0-0x53053].com.adobe.Photoshop[388] reproStepsString:
20/03/08 20 maart 21:05:35 [0x0-0x53053].com.adobe.Photoshop[388] 2008-03-20 21:05:35.530 Adobe Crash Reporter[400:10b] (null)
20/03/08 20 maart 21:05:41 com.apple.launchd[119] ([0x0-0x53053].com.adobe.Photoshop[388]) Exited: Terminated
20/03/08 20 maart 21:05:49 Adobe Crash Reporter[373] (null)
20/03/08 20 maart 21:05:49 [0x0-0x4a04a].com.adobe.Photoshop reproStepsString:
20/03/08 20 maart 21:05:49 [0x0-0x4a04a].com.adobe.Photoshop 2008-03-20 21:05:49.353 Adobe Crash Reporter[373:10b] (null)

[cite] Snow:I booted PS back up and bam, after a few mins another crash. Log: 20/03/08 20 maart 21:05:31 Adobe Crash Reporter[400] Process Id: 20/03/08 20 maart 21:05:31 Adobe Crash Reporter[400] 388 20/03/08 20 maart 21:05:31 [0x0-0x53053].com.adobe.Photoshop[388] 2008-03-20 21:05:31.407 Adobe Crash Reporter[400:10b] Process Id:…

Oh well, so we know memory times improves things but doesn't cure. I am 100% its network related and even more specifically internet.
Snow how are you connecting to the internet? router, modem etc?

Don't be too sure of that yet Ianxxx. I thought the same for a while, but...

Last night, I turned Little Snitch off (to rule it out) and I was running Transmission to download a couple of torrents. I added an XVid torrent and started downloading it to my desktop. I looked over at my Console and suddenly, I noticed that "quicklook" (the process that creates thumbnails for files) was trying to create a thumbnail for the XVid movie. The thumbnailing process failed (probably because the file wasn't done downloading yet), and it was rapidly spitting out this same error message over and over again:

3/20/08 2:06:44 AM quicklookd[1193] [QL ERROR] 'Creating thumbnail' timed out for '<QLThumbnailRequest /Users/agrafuese/Desktop/xxxyyyzzz.avi>'
3/20/08 2:07:35 AM kernel CODE SIGNING: cs_invalid_page: p=1206[lssave] clearing CS_VALID  

It repeated this about a thousand times (not exaggerating), and the "p=" part of that second line grew higher and higher in number.

Now, I know why this happened: it is because Finder tries to create thumbnails for any file that is in an "active" window. Since the desktop is technically a constantly "active" window, it wouldn't stop trying to create the thumbnail as it failed each time. So, because this error was repeating and Finder was trying soooo hard to make it happen, my system bogged down A LOT, and lo and behold: The return of the jerky mouse! I eventually stopped downloading the file, moved it to a folder that wouldn't be in an "active" window, and the errors in Console stopped. However, the jerky mouse remained. It appears that when Leopard gets stressed, it doesn't recover after the stressful incident has stopped. It doesn't "refresh" itself in any way, apparently.

Now, this thumbnail scenario is very rare, and I know it is not THE ONLY reason that a jerky mouse will occur. BUT, I believe the answer lies in this following statement:

The mouse, GUI, and performance problems will occur when the system is under large amounts of stress of any kind. Period. There is no ONE solution, and it is not always related to the internet or networking. It could be anything: an intensive process, a power-hungry app, too much file-processing, repetitive errors (in my case), and on and on and on.

So what do we do about it? Well, if I am correct, there is nothing that can be done. We don't know if Zeph can do anything, because this may be a genuine Apple bug straight from the developers' desks. All we can do is keep Zeph notified and hope he chooses to help us out in any way he can.

Well guys, that's my 2 cents. I could be wrong, and I hope I am, but I have a feeling I'm not. Good luck to us all.

I went back to non-overclock setup, and so far I haven't gotten any 10810 errors, nor system reboots, but after coming back to my computer after a while of being idle (an hour or two), i've come back to a crashed photoshop, and just now, a crashed firefox. This is an improvement, but still not what I'd like. If I stay stable like this for a few more days I'll install logitech mouse drivers again and see if they run fine.

The mouse, GUI, and performance problems will occur when the system is under large amounts of stress of any kind.

I agree with this totally agrafuese. Thing is now for me these faults always happen after 3 hours use, I left my system on but no internet connected and it didn't happen.
To me it sounds like some kind of memory leak, whatever it is it obviously mounts up over time and then causes the instability.
I would love for someone else to try using there system for over 3 hours with no internet just to see if they get the same results!

[cite] Ianxxx:To me it sounds like some kind of memory leak

Yes! I was waiting for someone to say this because I wasn't sure if that was the correct term to use! (I know nothing about advanced crap like that, haha) Also, back in the days of Tiger, I recall Netkas (I think) dealing with some setting inside the kernel that had to do with processor timings. I wonder if I can go that far back on his site and find what he was talking about? This is exactly the kind of thing that needs to be investigated.

Question Ianxxx, what are you doing with your system while it is unhooked from internet? Any apps, anything intensive? And when it IS connected to the internet, what are you doing? And what if you simply connect it to the internet and walk away? Would it still slow you down? After the slowdowns, are you seeing anything excessive in Console? Anything repeating? Any "CODE SIGNING" messages? Sorry for so many questions, but they're all important

If I leave the system with just firefox running connected to net but do nothing. After 3 hours as soon as I start to use apps it won't be long before I get a crash "usually firefox" programmes that upload as well as download seem to do this best.
Not connected to the net seems to work fine after 3 hours photoshop coding videos etc etc.
Just need someone else to test this out, if I am right at least it narrows things down a little.
A few folks have mentioned java " I think thats a possibility" except this problem seems to be amd only.

At the moment I have one error in console that persists

com.apple.launchd[1] we crashed at instruction: 0xffff0315 (sent by PID 0)

The error is there at startup

What is this anyone know, or more importantly how to get rid of it?

@lanxxx: I'm using a modem, Ethernet conn, proxy.

I think you guys are on the right track here. Memory leaks could well be the prob. If we look back to our windows days we had similar probs like this. Games in particular can be affected by this.

My scenario yesterday when I had the crash was not very intensive. I was viewing a PS tutorial in iTunes while performing the excersice in PS CS3. PS then crashed while not doing anything particular so I can't simulate the behaviour. No specific error msg, just the log thingy. Then when trying to run PS again it crashes again after a very short period of time.

I still don't know if I have the seperate PS crash issue (due to not proper installation or non crack usage) or if this is related to the mouse bug and other crash probs.
Thing is, PS still crashes, but less and I don't have the mouse bug anymore.

Now, Lanxxx, I'm willing to do the network test but what exactly do you want me to do. unplug the ethernet cable from my pc and let the pc run for a few hours? How long? what apps to run? how much stress do you want me to put on the system? Gimme some spec details fo test scenario.

We should also look into what kexts we are running, those with and without issues cause it might just be a simple kext replacement for all I know. What I'll do know is completely get rid of the d-forcedeth kext. It's still on my disk just in case I ever need the nLAN but I remember some having issues with the net even though they weren't using the nLAN and I think Zeph or someone else recommended removing the forcedeth kext completely.

Sorry if this has been thought or said already, or if it's wrong, but I didn't have the time to read any posts in the thread!

I remember exactly this Issue! However, this only occured with an early 10.5.1 version for me.

Now it never happened again.

I thought that it was somehow related to nForce LAN, because when I switched to Wi-Fi and disabled nF4-LAN in BIOS, it never happened again.

[cite] Ianxxx:How many people who are suffering from this also have random crashes from firefox, for me I'm pretty sure the issues are related? Also this bug used to come on if I downloaded/uploaded a lot. Are these the things other people recognise?

in getting random crashes from firefox

and then some regular pc crashing:

Once = computer stayed on all night, next morning all applications failed to launch (gave me an error), had to reboot.

Everyother time = left PC on all night (without azureus) system log says (?:?am) We crashed at 0x0000…forgot the numbers)

with azureus i think its a Java issue, i dont know, ill leave it on again tonight and get a log

[cite] Snow:@lanxxx: I'm using a modem, Ethernet conn, proxy. I think you guys are on the right track here. Memory leaks could well be the prob. If we look back to our windows days we had similar probs like this. Games in particular can be affected by this. My scenario yesterday when I had the crash was not very intensive. I was…

Hiya Snow
yep disconnect lan would be good and run for over 4 hours 3-4 hours seems to be the crucial amount of time. Put the system under as much load as you can be bothered, if you don't get any problems after 4 hours I think we have narrowed down our culprit. I'll be away and won't be back till sunday night, but would really appreciate you doing this test. If we can narrow it down to internet related I think that would be very useful.

K, I'll try this today bud. Do I get payed though?

Next post will be with results!

[cite] Snow:K, I'll try this today bud. Do I get payed though? Next post will be with results!

Only payement is that you'll know for yourself if this is net related.
And possibly an appreciative following. "you could end up famous HONESTLY"

With eddie11c's my mouse had a problem once where they would be a 2 second delay. It resolved after I reset the safari browser. I don't think this equates with the problem though.

Okay I don't have time to fully test this until tomorrow but for anyone wanting to try a possible solution.
If your memory isn't running in accordance with its spd. Then try it, if your motherboard doesnt have the option of run by spd use cpuz in windows and get the info from there.
As I said this is looking promising so far but can't continue my tests till tomorrow.

[cite] Ianxxx: Hiya Snow yep disconnect lan would be good and run for over 4 hours 3-4 hours seems to be the crucial amount of time. Put the system under as much load as you can be bothered, if you don't get any problems after 4 hours I think we have narrowed down our culprit. I'll be away and won't be back till sunday night, but…

Hmm, yeah, curious to see what you guys come up with on this one. Right now, I am trying to think of a way that I can disconnect from the internet, but at the same time invoke some kind of repeat Console error message, such as the one I described with the thumbnail creation error above. It in itself wasn't an internet problem of course, but it happened because I was in the middle of downloading a video that needed a thumbnail created...sooo...kind of a catch 22, haha. I'll try to think of something though. There's a certain Office 2008 problem I am considering. Hmm...could Microsoft help us for once?

k guys. I unplugged Ethernet cable but ran Mail and Safari, Photoshop, iTunes showing a tutorial vid and Garageband playing a few tracks.

No probs but I do have to say that my system wasn't crashing that often anyway. And I've only had probs with PS so far. Saf has crashed before with the same error as Agra but that happend very rarely.
So I'm not too excited yet but so far it's stable.

Errors displayed while testing

21/03/08 21 maart 18:33:11 kernel CODE SIGNING: cs_invalid_page: p=395[lssave] clearing CS_VALID
21/03/08 21 maart 18:34:04 kernel CODE SIGNING: cs_invalid_page: p=398 clearing CS_VALID
21/03/08 21 maart 18:38:17 kernel CODE SIGNING: cs_invalid_page: p=403[mdworker] clearing CS_VALID
21/03/08 21 maart 18:43:27 kernel CODE SIGNING: cs_invalid_page: p=404[mdworker] clearing CS_VALID
21/03/08 21 maart 19:07:34 kernel CODE SIGNING: cs_invalid_page: p=407[mdworker] clearing CS_VALID
21/03/08 21 maart 19:46:50 kernel CODE SIGNING: cs_invalid_page: p=412[mdworker] clearing CS_VALID
21/03/08 21 maart 19:52:35 kernel CODE SIGNING: cs_invalid_page: p=413[mdworker] clearing CS_VALID
21/03/08 21 maart 19:59:14 kernel CODE SIGNING: cs_invalid_page: p=414[mdworker] clearing CS_VALID
21/03/08 21 maart 20:09:15 kernel CODE SIGNING: cs_invalid_page: p=417[mdworker] clearing CS_VALID
21/03/08 21 maart 20:35:18 kernel CODE SIGNING: cs_invalid_page: p=422[mdworker] clearing CS_VALID
21/03/08 21 maart 20:53:15 kernel CODE SIGNING: cs_invalid_page: p=423[mdworker] clearing CS_VALID
21/03/08 21 maart 20:59:27 kernel CODE SIGNING: cs_invalid_page: p=424[mdworker] clearing CS_VALID
21/03/08 21 maart 21:06:12 kernel CODE SIGNING: cs_invalid_page: p=427[mdworker] clearing CS_VALID
21/03/08 21 maart 21:18:40 kernel CODE SIGNING: cs_invalid_page: p=428[mdworker] clearing CS_VALID
21/03/08 21 maart 21:26:14 kernel CODE SIGNING: cs_invalid_page: p=433[lssave] clearing CS_VALID
21/03/08 21 maart 21:26:43 kernel CODE SIGNING: cs_invalid_page: p=435[mdworker] clearing CS_VALID

+ 3 hours so far but I'll let it run till I'm off.

Before I try the thing I mentioned previously, I do want to say that I'm up 14 hours here. Internet is on, Little Snitch is on, and I have every single app on my dock running at once, including CS3 apps, Office 08 apps, Linotype Font Explorer, iTunes, and too many more to mention - about 15 in all, plus others that I've been opening and then closing (small stuff). On the flip side, I am not running torrents this time. My console is rock solid stable, aside from the usual CODE SIGNING messages (like Snow just posted) and the occasional mDNSresponder messages which seem to have lessened since I upgraded to Netkas' new AppleSMBIOS. Aside from this new AppleSMBIOS, nothing else has changed on my system aside from the new security update and Safari 3.1. I'm wondering if any of these changes have helped solve the -10810 errors? Though, I have to say, I think I was getting those errors after I tried Ianxxx's memory timing settings (which I have since changed back to default). Also, Firefox seems rock solid, although I am only doing light browsing with it. All of my aggressive browsing is being done in Safari (which I am now in love with since 3.1 came out). Oh, and I have been doing some heavy FTP downloading too throughout this period. No uploading, though. Maybe the next test will be uploading some files, as it seems like that could be the reason torrents cause the slowdowns when I run them.

You find it freaky only because you ain't go it The rest of us find it a pain in the arrrrsseeee!

On a side note, I just want to say a big W-T-F?!?!? I've been running now for 18 hours straight and still no problems! Is it torrents? This is the only thing I haven't done yet. And Firefox is behaving! What the hell is up with this shit?

i was telling winlinmac, I come up with a crazy theory.
I was reading about mac memory specs. The importance of spd was stressed a lot in what I was reading.
As real macs don't use bios, but efi.
Is it possible that osx somewhere is reading the spd and using that and the bios is set differently?
I've set my memory to run with spd settings which is a bit of an arse for me as my memory is 2T memory. But so be it.
I've noticed in the past that if I open enough applications that although I might not get apps crashing and mouse jumping, I can tell that the mouse movement isn't quite as smooth.
With memory running at spd this doesn't seem to be the case, so testing will continue.
And thanks for testing my other idea snow.

Ian, what is spd? Can you explain it for a memory noob such as myself? Or link us to the thing you were reading? Btw, here is my latest report:

I'm coming up on 20 hours straight now. I'm still jerky-mouse-free! Still FireFox crash-free. No -10810s, no crashes anywhere, actually. I had all of my apps open for a very long time, but I've closed most of them except for a couple CS3 apps, Linotype, and iTunes. I'm constantly opening and closing Safari and FireFox. Both are behaving really well. I was even running YouTube videos in them, and that's about as taxing as you can get with a web browser, short of maybe Flash games or something? I still haven't run torrents. So now I'm starting to think that my Leopard's weakness is torrents and repetitive system-logging, because otherwise I've done just about everything I can think of to make my system freak out. And as I said, I think the 10810s haven't come back because I am back to default memory timings. I'm not doing any overclocking either. Full default in BIOS. So, take that how you want, but this is my report. I'm going to leave my system on as long as it will let me and see what happens. So far it's perfect. If I could have it be this way forever, I would no longer have a single complaint, and I could stop dreaming about Tiger

I can't link to the stuff I was reading but basically I think I googled mac ddr memory or something similar.
SPD is the firmware chip on your memory that tells bios etc the exact speeds to run it at. I bet if you boot up in to windows and run cpuz you will see that the memory settings your running at are the same as the ones set by spd. I'm pretty dam certain that will be the case.

So wait, how would I know what the spd settings ARE? Unfortunately, I no longer have Windows on this machine (took out the hard drive it was on). I can tell you this: I have my memory settings set to "auto" in BIOS. Does that mean that it will set them to whatever spd tells them to be?

Yep I would guess so. Reads info from chip.
And as I said I'm begining to think that this is the important thing rather than the actual speed.

But you and I have the same RAM don't we? Kingston DDR 400 Dual-Channel? I may be wrong about the brand though. It's been a while since I looked at the chips. I have two 512MB sticks. When was the last time you had it on auto settings in BIOS?

Today is the only time i have had it on auto or in case of my board use spd.
Although we both have kingston DDR 400 mine are 1 gig modules and quite frankly there are loads of manufacturers that the actual chips could have been made by, I think samsung is a favourite of kingston but certainly not the only one they use. You could only say we have the same memory if we compared numbers printed on actual chips.

I have set the memory slower than the spd setting but it never occured to me before that it might have something to do with firmware, I maybe totally barking up the wrong tree, I don't even know if osx does read memory firmware or if this is done purely by efi. But hell this seems to be the best "stablest" yet so who knows.

Man, you know a ton about RAM, I'll give you that! All I know is: makes shit run, and the more you have, the better the shit will perform, haha. Good to know, though. Thing is, I've only ever changed my memory settings ONCE and that was when you told me to do it. Since then, I've changed back...soooo...I dunno...:\

[cite] agrafuese:Man, you know a ton about RAM, I'll give you that! All I know is: makes shit run, and the more you have, the better the shit will perform, haha. Good to know, though. Thing is, I've only ever changed my memory settings ONCE and that was when you told me to do it. Since then, I've changed back...soooo...I dunno...:\

Hmmm, strange. Mind you I must say I still have HT link at 4x rather than 5x

This is confusing I think other than through trial and error as we are doing at the moment it would take someone like netkas or who ever wrote the efi patches for hacks to tell us the full story, but anyway so far so good my system is certainly behaving better than it did.
As for knowing about memory, you could say hardware is my obsession lol. I've been building PCs for about 13 years.

Hmmm that made me think - I wonder what would happen if I ran Leo without EFI? The last time I did this was with 10.5.1 rev1 and I can't recall whether or not I had any performance problems. If I can clear a spare drive, maybe I can give it a try with a standard MBR Leo 10.5.2 rev1 install. I think I'd have to take out my 8800GT, though, since it needs the graphics update to run properly (and updates are out of the question without EFI, right?). I can always swap it out for my old card, I guess. Does this sound like it would do any good as a test for our problems, or just a waste of time?

I wouldn't bother I ran 10.5.1 with efi and can't remember having this issue.
Though i must admit I ran tiger without efi and its the stablest thing I've ever used. Just used to piss me off cos reboot wouldn't work and there seems to be no chance of updating it to 10.4.11 everyone whos doing anything is doing it with leopard.

I must say though for all the shit we've been having we're running something on hardware it wasn't meant to be run on and its still better than vista. Always good to keep perspective.

I don't think my system is cured. No crashes but that roughness of the mouse movement is back and I've only been online one and a half hours. Boll*cks

You're definitely right on perspective.
So, I'm on hour 21-22 now and I just got done running some really stressful processes in Photoshop and Illustrator as a test. I took a look at my Console after a while, and to my dismay, I saw my first mDNSresponder "ticks" message. Ugh! I know now that it's all going down from here. This is the thing that I never want to see because as soon as it happens, the mouse gets rough. It's the same way you describe it Ian - not exactly jerky yet, but rough - though, if I leave it this way long enough it will eventually get jerky. So can you do me a favor? Can you look in your Console, make sure "All Messages" is clicked over on the left, and then do a a string search for "ticks" and see if anything comes up? I really hope we can at least find a common ground with that message if nothing else. That's my last hope of finding any explanation, though I don't even know what the "ticks" message means in terms of what's actually causing it other than system overload.

EDIT: I was thinking of building an Intel also, once I get some more cash. Thinking about using a GA-P35-DS3L for the board. What about you?

[cite] agrafuese:You're definitely right on perspective. So, I'm on hour 21-22 now and I just got done running some really stressful processes in Photoshop and Illustrator as a test. I took a look at my Console after a while, and to my dismay, I saw my first mDNSresponder "ticks" message. Ugh! I know now that it's all going down from…

Similar but I was thinking of going for a cheaper version GA-31-S3G
its a lot cheaper but also uses the ICH7 controller I think that it will probably be more compatible with vanilla, but I could be wrong. Its only downside that I can see is max 4 gig ram. But hell I aint gonna buy more than that anyway probably wont use more than 2. On top of that I've been thinking about processors and quite frankly I think Im gonna use a e2160 on 1333 bus that would give me 3 gig processor for £44. Im not sure I can justify spending a hundred quid for 3 mb of cache.

As for console 32 clearing cs_valid messages but nothing else

Just a quickie: 28 hours uptime, still good. Only ran one tiny torrent and then I closed Transmission immediately. Didn't seed. I thought I was doomed when I saw a couple of "tick" messages, but they stopped happening all together, and my mouse is still fine. No crashes, except for one Safari crash that probably happened because of the site I was on. It started right back up, though. So no 10810s, nothing else to speak of. Either I got some stability from Netkas' AppleSMBIOS or the security update, or this is just one loooonnnggg 28-hour stream of good luck. Whatever it is, it's nice, and I won't complain. I'll also avoid torrents like the plague

Do you want me to push the RED button now Agra? It lasted long enough don't you think?

good news though bro, I'm happy for you cause I know you were about to change pets.

Take care

"change pets" ... i like that one, it's almost better than "change girlfriends" ... almost

please, no red button. i'm on 33 hours! this might be a record for me.

Glad to see your up and running agrafuese 33 hours cool.
I seem to have found a solution for myself too.
I now have my processor overclocked, the timings on my memory are a lot faster than they should be, seem to be able to alter anything and system remains stable. Been uploading downloading torrents & encoding videos, running as many apps as I can watching two movies at the same time. Every thing is still smooth as it should be.
The answer for me was my memory is now running at 398 mhz as apposed to 400mhz.
Weird but it seems to work 100% so far.

Red bull anyone? Go to you local store and pick up a few energy drinks to consume, lol. =)

[cite] agrafuese:"change pets" ... i like that one, it's almost better than "change girlfriends" ... almost please, no red button. i'm on 33 hours! this might be a record for me.

Haha...more like we need to buy red bull for our computers No more slowdown issues right? Or maybe it would be too fast then

Well I've been up for days now...I THINK. I am faced with a very strange issue now. I left my computer on since Saturday morning, left for the weekend and came back now (it is now Monday afternoon where I am). Before I left, I had nothing running except for Transmission downloading 4 large torrents. When I woke my display up just now, I was presented with the Login window. So I thought my computer had restarted itself. I logged in and went to Console to see if I had any error messages. The odd thing is, there was no report of any kind of rebooting! So at first, I thought that Leopard had some kind of built-in auto-logout security feature if there was no user activity for two days. But then, I opened up Transmission and found that my torrents had stopped. So why would Leopard log me out if torrents were still running? Does anyone know anything about this? When I look at my Console messages over the past couple of days, there are only two messages worth noting:

3/22/08 4:34:16 PM bootlog[53] BOOT_TIME: 1206228844 0 
3/23/08 4:34:10 PM com.apple.launchd[1] (com.apple.dashboard.advisory.fetch[222]) Exited with exit code: 1 

Another strange thing is, I cleared my console before I left on Saturday, and there is nothing before those two lines of code. Also, even if I were logged out by Leopard, wouldn't Transmission still be sitting there when I log back in? Am I wrong about that? I am so confused right now :\

That is very strange, but is this mouse-related? Looks like an entirely different issue.

[cite] agrafuese:Haha...more like we need to buy red bull for our computers No more slowdown issues right? Or maybe it would be too fast then Well I've been up for days now...I THINK. I am faced with a very strange issue now. I left my computer on since Saturday morning, left for the weekend and came back now (it is now Monday…

Yeah, you're right, sorry. I kind of went off on a tangent there. I should post this in another thread. I've grown too comfortable blabbing on in this one, hehe.

didnt read all 9 sites of this thread - weird i dontknow if the mouse problem is still active...
anyway i had the same prob and i enter following kernel flag to my boot.plist:

<key>Kernel Flags</key> 
<string>idlehalt=0</string>

thats all
never had any bad mouse issue again

______________________________________________

is it possible that only amd dual cores have this problem?
would be an explanation for that, coz win xp needs the amd dual core optimitzer for running fine.
and coz there is certainly no mac verison of the optimizer the second core should or should not interact in idle.
hmm sounds logically, anyway the kernel flag works for me

This can work, definitely. I can't believe it works, lol. =) - Cheers indeed.

[cite] onAir:hey didnt read all 9 sites of this thread - weird i dontknow if the mouse problem is still active... anyway i had the same prob and i enter following kernel flag to my boot.plist: <key>Kernel Flags</key> <string>idlehalt=0</string> thats all never had any bad mouse issue again

what exactly is that string meant to do (aside from it fixing your mouse) ?

im think its preventing the second core for stopping in idle mode

hehe sorry im not sure about the exact function, but it works for me and i thought i have to let u all know about that
if u find more infos about that i would really appreciate sharing them with us
although im not as interested in it as i would google around for it hrhr

Hmm interesting, this is what the osx86 wiki says:

"[use it]..."if you experience stutter like (shutter) syndrome; usualy on dual core processors"

Who knows...worth a try, I guess...

A general site that talked about unix kernels said that this flag is also found in other kernels, and is used to disable energy saving features...?

Gonna give this a go too, i had 24hrs problem free rebooted and then it came back again

Unrelated, but a follow-up to that strange issue that I posted an hour ago here. I found out that my log wasn't showing anything before because it had "turned over". Anyway, I looked into my archived system logs and guess what I found as the culprit for the mystery:

Mar 22 16:30:38 AMD-Mak com.apple.launchd[1]: We crashed at instruction: 0xffff0315 (sent by PID 0)

In case you don't know, I've been trying to link this error message with the jerky mouse movement and the -10810 errors. So just to re-iterate, I was running torrents when this happened. This is also what I am doing when I get the mouse problems. I'm a little unsure about the 10810s though, however, I haven't seen them since the time I was experimenting with my memory timings. So I guess I'm not 100% cool with Leopard anymore. I thought running it for over 33 hours meant I was in the clear, but apparently not. There's also nothing I can do to prevent this except stop running torrents, and well, why should I have to do that? Bullshit.

[cite] agrafuese:Unrelated, but a follow-up to that strange issue that I posted an hour ago here. I found out that my log wasn't showing anything before because it had "turned over". Anyway, I looked into my archived system logs and guess what I found as the culprit for the mystery: Mar 22 16:30:38 AMD-Mak com.apple.launchd[1]: We…

I've found with transmission the higher the download speed the quicker this problem will come on.
also i have that error

Ugh. That makes sense since I was downloading 4 torrents at once. It's gotta be partially torrent related too because I've downloaded many many large files via web browser (FF and Saf) and it never happens when my speed goes high. For example, I've downloaded Ubuntu at my max ISP speed (700kB/s) as a test, and it was fine. Did this numerous times actually. This is really discouraging. I don't think it's limited to Transmission either because I had weird issues with BitRocket and TomatoTorrent in the past. I find it hard to believe that thousands of real mac users and/or the hundreds of Intel hackintosh users out there have problems with torrents like we do. What else could this be?

I did a google search and did see that a couple of real mac users seem to be having the problem but wether it was a different problem with same symptoms who knows.
For the moment I'm trying this

<key>Kernel Flags</key>
<string>idlehalt=0</string>

and as the memory thing seems to make some difference this makes sense.
Anyway let you know more tomorrow keeping my fingers crossed.

I'll give it a try too right now. I'll re-start the torrents after I reboot. That'll be the ultimate test for me to see if the system stays up despite torrents downloading.

disabling energy saving features of the cpu sounds pretty clear...
i bet if i try without that but disable the second core the mouse problem will be gone...
but i wont try coz i dont have the time to test that, more have another problem

btw...anyone knows how to fix the boot with inserted dvd only problem?
since i copied around my boot partitions (leo only for main and for tests) its occuring all the time
think the mbr or efi is damaged in any way... actually messing up with that at the moment,
would be cool if someone knows an open thread here for that

yeah, look around. i don't know your exact situation, but there are many many many threads about booting issues. i'll be happy to join in on another thread that's more on topic

Well so far after three and a half hours "downloading torrents" this is looking very promising, except my cpu temp is about the same as running prime95. Oh well suppose you can't have everything.

Ian, how are you monitoring CPU temp in Leo? I couldn't find a good app back when I was looking for one. Any suggestions?

I have been running with idlehalt=0 and torrents for 5.5 hours now and it's looking good for me too! i wonder if we have any decline in performance though? i don't notice anything, but i haven't run any benchmarks yet...let's keep updating here and hopefully we've (finally) found our fix!

Seems a lot has been discussed since yesterday. I will also try this idlehalt setting. I've heard of it before but didn't think it would do any good anymore with the new kernels etc...

I need a new install too but I don't know which of the releases to try, Zeph's latest or Leo4all latest, hmmm.

Okay system on overnight no crashes no funny mouse so this is looking really good, final test system overclocked -> that also is looking very good.
agrafuese my temperature monitor is hardware I have a thermometer wired to the underside of my thermalright xp120 heatsink, and believe me when I say this is running hotter than it has ever reached even running benchmarking software, so I would be careful anyone who doesn't have a temperature monitor.
I would say also definately no decrease in performance, possibly a slight increase.
So the only downside is temperature which I can live with I live in a cold house lol.

i*ve tried the same and for me it's getting even more worse than before. Mouse starts to jump around after an hour uptime +/- a few moments.

While typing this text Firefox crashed at least 2 times =(

I don't know about the temperature since i don't have a thermometer for my CPU, nontheless i'm pretty sure it can't be high since my CPU-fan runs slow (speed of my fan is controlled by CPU-temp).

Apps running right now:
NetnewsWire
Firefox
Thunderbird

Transmission crashed while i was sleeping so i had to restart my machine around an hour ago.

Update:
My machine crashed completely this time. The loading circle appeared and that's it … another restart.

I really hope someone is able to find a solution.

Cheers,
Tom

Just to update the temperature thing my processor usually runs around 35degrees maxing out at around 50 under full load.
With this in place its nearer 60degrees

So I'm guessing this is the final straw for me I'm going to invest in an Intel processor something I haven't done for nearly ten years

Was just about to aks you on the temp numbers. From 50 to 60 under load. What's idle like? If idle also ups by 10 then it's not worth it but if it's only on load I can live with it.

Sad to hear we're going to loose another AMD user who contributed a lot to this board, hmm, can't say that for myself.

But on the other hand, I do encourage you to jump the Intel bandwagon, so good luck!

I don't have a temp monitor like you do but I'll try the idlehalt tweak anyway. Thx for the info lad.

Ahhh shit! I was worried about that. I don't have a CPU thermometer either and so I have no clue what's going on in there. I wonder if I can quickly restart my computer and look at my temps in BIOS really fast before the temp goes down? Anyone know if that works? Upside is, I too have been stable with torrents for over 18 hours now! I don't overclock anymore, so maybe my temps will be lower than yours Ian?

@subsonic: It's really strange that you have more problems with this setting than before. Are you doing any overclocking or memory tweaks in BIOS? And what are your specs?

Your BIOS would be a great place to start. Try that first.

[cite] agrafuese:Ahhh shit! I was worried about that. I don't have a CPU thermometer either and so I have no clue what's going on in there. I wonder if I can quickly restart my computer and look at my temps in BIOS really fast before the temp goes down? Anyone know if that works? Upside is, I too have been stable with torrents for…

I had this issue (mouse jumping and reboots after hours of bittorrent and heavy download overnight) a while ago (10.5.1) and after reinstalling Leopard many times, replacing kext and hardware components like usb2 and network i found It could probably be related to APIC implementation in BIOS. What led me to this was the fact that my crash happens overnight or after waking up from the screensaver. So I tought about ACPI problems. All of this only happened when running both CPUS. With cpus=1 all worked well. Searching around I found that some users had issues even in windows with heavy overclock and dual core AMDs related to APIC Advanced Programmable Interrupt Controller that in dual core processors could led to some stability issues. The solution for me was to select the 1.1 version in Bios and not the 1.4 (default). Also I set the IRQ manually. I have a NF4 DFI Ultra-D motherboard. Hope this helps.

@ agrafuesinski

Open up case and remove the heatsink from the CPU while it's running. Then read temp on the side of the heatsink that makes contact with the CPU. It's hard to explain how the process works but it really works so try it, now!

@Snow: Huh? Wha? I don't know what you mean. Don't I need a thermometer for that? I don't have one...shoot me a message over at insanely mac

So here's my report for now: After 22 hours straight of running torrents, I restarted my system and went into BIOS to read my temperatures. At first, CPU was 51. I let it sit in BIOS for about a half hour and it went down to 44. When I was in Leopard prior to booting into BIOS, I'd say I was only putting about a "normal" amount of load on the CPU. Just running the torrents and doing some web browsing/emailing. A few hours earlier, I was trying to watch a hi-res quicktime video online (of the Fallout 3 trailer, hehe) and it was putting A LOT of load on my system - to the point of causing spinning beachballs and short freezes, etc. I can only imagine what my temps were like at that point, but I don't know for sure. Probably in the upper 50s at least. Anyway, I'm just going to leave the idlehalt=0 flag on for the time being. It works fine, and the system is pretty damn stable. Plus, I don't usually put too much stress on the CPU on a daily basis, and I don't overclock, so I should be okay. If I can get ahold of a hardware thermometer, I'll be able to keep my eye on it a little better, but that won't be for a while. If I have anymore news for you guys, I'll be sure to post, but for now I'm gonna close this case until something new arises. Good luck everyone!

Well been out and bought myself a cheepo Intel setup, haven't put it together yet. But no matter what can't see me leaving you lot completely, might even have to keep an AMD somewhere in the house to give me an excuse lol.
Haven't put the new gear together yet "tomorrows project".

Leo4All is in your hands.

[cite] Ianxxx:Well been out and bought myself a cheepo Intel setup, haven't put it together yet. But no matter what can't see me leaving you lot completely, might even have to keep an AMD somewhere in the house to give me an excuse lol. Haven't put the new gear together yet "tomorrows project".

[cite] WinLinMac01:Leo4All is in your hands. [cite] Ianxxx:Well been out and bought myself a cheepo Intel setup, haven't put it together yet. But no matter what can't see me leaving you lot completely, might even have to keep an AMD somewhere in the house to give me an excuse lol. Haven't put the new gear together yet "tomorrows project".

Still trying to download it lol.
But to be honest not sure wether to use it or I have a brazil patched 10.5 or kalyway. I think its either gonna be Eddies or Brazil. Might end up trying both.

havent ever thought about temperature problems with my x2 4600+ running on 2,8 GHz with a big scythe mugen cooling fan on it.
by the matter of fact that my cpu temps are about 55 degrees on vista64 running crysis or prime95 and no errors on osx with idlehalt=0 (no errors since 10.5.2 rev.2! and yea i fixed the boot problem by simply re-installing pc-efi v8 from zeph's boot-dvd) i couldnt warn u all about that risk. i didnt think so far and im sorry about that. furthermore i didnt know what idlehalt=0 does exactly do.

the thing is that i always had some weird problems with osx86 and i already decided that rev.2 is the last try for mac on my pc. now, very lucky , everything is running flawlessly (crossing fingers to keep that). despite that my next system upgrade will be the change to intel e8400 cpu on an x48 board and ddr3 ram for homogeneous 1333 MHz bus speed and much upgrade and overclock potential. is a good way to monster speed for 600€ i think (if i had more i would buy a mac pro of course... i love mac)

maybe this will be the next challenge for osx86 as no more errors would occur on my amd system hehe. otherwise i still i have my girlfriend's macbook hihi

so sorry again for that risky hint and keep ur eyes on the cpu temps.
hope not to need more visits to these sites except by interest in the scene and wanting to share experience.

just a last update on the temperature thing before I dismantle my system.
temp running around 45 degrees no overclocking but I do have a thermalright xp-120 so I'm guessing the temps would be hell of a lot higher for some people, to give you a guide I usually have my processor runnging @ just under 2800mhz and it never gets hotter than around 45. i'm thinking someone with a standard amd cooler will definatly be up in the 60's.

@agrafuese

Hey =) Regarding your question, no i don't do any overclocking or better said i can't because my Bios won't let me. My Bios is pretty static, i can't alter memory timings or anything related to my CPU. That's one bad thing about preconfigured HP/Compaq-machines =(

Anyways, here are my specs:
-----------------------------------
Modellname: Mac
Modell-Identifizierung: EP164AA-ABE SR1759ES ES610
Prozessortyp: AMD Athlon(tm) 64 X2 Dual Core Processor 4200+
Prozessorgeschwindigkeit: 2.2 GHz
Gesamtzahl der Kerne: 1
L2-Cache: 512 KB
CPU-Funktionen: FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH MMX FXSR SSE SSE2 HTT SSE3
Speicher: 1 GB
Busgeschwindigkeit: 200 MHz
Boot-ROM-Version: Hack.88Z.int.0sh.061193037 (Phoenix Technologies, LTD)

My mainboard is an MSI Amethyst M, a pretty cheap piece of hardware

1GB DDR 400 Ram
---------------------------------------
Hm, when i look at the number of my cores it states "1". Weird, shouldn't it say "2"? Is it because i cloned my old HDD to a new 500GB SATA HDD with the following bootloader options (might sound stupid but i can't say it often enough that i'm a ubernoob with Leopard and the whole OSX stuff *hehe*):

dd if=./guid/boot1h of=/dev/rdiskXsY bs=512 count=1

dd if=./guid/boot0 of=/dev/diskX bs=400 count=1

Please let me know if you need anything else. Thanks and cheers,
Tom

Usually the number of cores reported in System Profiler doesn't mean anything, unless you have "cpus=1" in your boot.plist. If you run Activity Monitor and see two CPU graphs moving, then you have two cores running. Although, my System Profiler is reporting both cores and I'm running the latest Netkas version of AppleSMBIOS. Maybe that could be why you're only seeing one? Anyway, that shouldn't be any reason for your system to act up. But actually, you may want to try that "cpus=1" flag instead of the "idlehalt=0". I've heard it has the same effect, but it probably reduces performance on your machine, since you'd only be running one core. It's worth a try. Certainly wont heat your CPU up like idlehalt does.

[cite] agrafuese:Usually the number of cores reported in System Profiler doesn't mean anything, unless you have "cpus=1" in your boot.plist. If you run Activity Monitor and see two CPU graphs moving, then you have two cores running. Although, my System Profiler is reporting both cores and I'm running the latest Netkas version of…

How you getting on Agrafuese?
I'm well and truly converted to the darkside and £144 poorer.

Well, I don't smell any smoke yet, sooooo...I guess I'm fine for now. Ask me in a couple of days, haha Seriously though, the temps are high, but the system has never been this stable. I was thinking...I believe the kernel I'm running (Netkas 9.2.0 w/ shutdown fix) has a speedstep function in it. I am wondering what my system would be like if I went back to the version before he implemented the speedstep. My theory is, if idlehalt deals with energy saving and turning it off helps me, maybe using a kernel without speedstep would help me just the same, and perhaps even without the heating up? I don't know if I'm on to anything, but I'll give it a try.

It sounds like I should be asking YOU how you're doing with the Intel...any persuasive words for us?

well although I think things got a little worse after I had installed that same kernel, but before then firefox would randomly crash it didn't at all after the idlehalt thing.

I don't want to blow smoke in anyones face but I tell you if you can afford it I'd definatly go intel. I just wish apple had partnered up with amd.

Well, there seems to be a lot of talk about it in rumor circles, but for me it'll be quicker to save money and buy an Intel than it will be to wait for Apple to cozy up with AMD. I'm still tortured by the fact that we've got these issues in Leopard, but not in the old Tiger releases. I really wish I could understand how that would be possible, but I have neither the knowledge nor the time to spend investigating it. Maybe we should be harassing Netkas instead of Zeph. Anyway, I'm doing some tests right now, so I'll be back with reports.

ok, small update:

apparently, there is a newer non-speedstep kernel by Netkas that i never knew about. i'm trying it out right now with some torrents and i'll let everyone know how it works after some hours of monitoring. i checked the idle temperatures on it, and it's waaaay lower than my previous kernel with "idlehalt=0". it's a difference of about 12 degrees. so that voids my theory that idlehalt is the same thing as speedstep when it comes to temperatures. definitely not so we'll see what happens.

@mrburns05: not exactly sure. i think rev1 uses the original 9.2.0 non-speedstep kernel that netkas first released. i don't know which one rev2 uses...possibly speedstep kernel with shutdown fix. the kernel i am using right now is neither of the two. and while i'm here, how about an update:

the newer netkas 9.2.0 non-speedstep kernel didn't help me. i've been running torrents for about 5 hours now and the jerky mouse is back. so it looks like idlehalt=0 is the only solution for me...although, the +12 degrees on my CPU that it brings is definitely not fun or exciting. so once again, case closed for me with no "real" fix. just a "hot" fix

okay now i left azureus on all night downloading a 6gb file, conections limited @ 250/300 peers, no crash!

System: using whatever comes on 10.5.2 rev1 disk

only things ive changed:

Disabled Indexing, turned off time machine, turned off screensaver, and in bios set cpu fan from 90% to full speed ( i believe at idle im around 32-33 Celcius)

I am over clocking (64 X2 3800+"2Ghz" OC'ed @ 2.2Ghz) DDR400 running at 440mhz

Does anybody actually have a full working leopard install on AMD, ie without the bugs listed here. My system randomly crashes using torrents or just heavy workload. i have tried all suggestions here and on insanelymac to no avail , although the cpu=1 is the closest to a fix for me (takes about 8hrs to crash). Is there any point continuing on the AMD platform for Leopard or should i just get and Intel setup. It would be a shame to go Intel given all the hard work done by Zephyroth and others, but for me i cant live with random crashing and the mad mouse problems.

Firewalk

idlehalt=0
totally fixed my system apart from temperature issues, but I decided to go intel anyway, its just easier!

Hi lanxxx, I tried idlehalt=0 but my system would reset itself after 30, 50 mins. mouse would freeze then the busy cursor then mouse would dissapear then i would get the spinning logo on desktop and after a minute or so system would reset. so far ive had best results with the cpus=1 command (still results in reset but atleast i can use it for 4-5 hours). Do you know if this problem is purely AMD related as i dont want to buy all intel to hit the same wall again..

AMD OS X has been fun though..

Firewalk

[cite] Firewalk:Hi lanxxx, I tried idlehalt=0 but my system would reset itself after 30, 50 mins. mouse would freeze then the busy cursor then mouse would dissapear then i would get the spinning logo on desktop and after a minute or so system would reset. so far ive had best results with the cpus=1 command (still results in reset but…

As far as I know its only amd, I wonder if your system was overheating with idlehalt=0 sounds to me like that might be the case. I now have a gigabyte ga-p35-ds3l and I can tell you everything is spot on the only "negative is two channel audio" other wise, everything seems perfect, updated using software update etc etc.

If you think your system might be overheating then I found this temperature monitor

version tracker.com/dyn/moreinfo/macosx/19994

Hi lanxxx, Thanks for the link, although it only shows temps for my hard drives..with the cpu=1 command my system has been running since 9am this morning uk time. so far my uptime is 5.3 hrs with no probs..still gonna look at intel boards and cpus though..

Yeah, that app doesn't work for AMDs...tried it some time ago.

So Ianxxx, you went with the DS3L? I thought you were gonna go with that other cheapo board? That's good news though, because that's the board I want. Glad to hear another account of the DS3L working great. IMO, 2-channel audio is fine. It's all I use. Besides, I've never had anything more than that with my AMD anyway!

Oh, but AMD users may still have a fighting chance to have a stable system! Did anyone see that JaS is back on the scene? Check it out: http://forum.insanelymac.com/index.php?showtopic=92524 ↗

I was all smiles when I read the news, but he wont be making an AMD release for quite some time it seems. Still, it's something to look forward to!

[cite] Firewalk:Hi lanxxx, Thanks for the link, although it only shows temps for my hard drives..with the cpu=1 command my system has been running since 9am this morning uk time. so far my uptime is 5.3 hrs with no probs..still gonna look at intel boards and cpus though..

Are you OC'ing at all? If so, try setting everything in your BIOS to default and then try idlehalt=0 again. In my case, I have no more headroom for any kind of tinkering with OC because if I did it, my system would definitely fry. At this point, I'm at about 52 degrees at idle and I'm worried enough about the ambient temperature going up in my house. :\ I may even have to look into more cooling - if I don't get an Intel system first, that is. Oh, and also try not using forcedeth.kext, just to rule it out. Then let us know how you're getting along.

Hi agrafuese, i dont overclock my system so bios is set to defaults, cooling shouldnt be a problem in the antec case i have there are two big 3 speed exhaust fans right next to the cpu. i cant check cpu temp as i havent found a proggie for it, but if i open system and touch the heatsink when running its cold to touch (i know its stupid sticking your fingers in live hardware, so i dont recommend anybody doing this). my record uptime has been 8.5 hours (not good for me).

i am at a loss , will keep researching and testing and let you know how i get on

Firewalk

P.S upon checking i dont have forcedepth.kext installed in extension folder,(isnt forcedepth for nvidia based gfx and network) if so my gfx are ati and network is gigabit.

ah yeah, forcedeth is for nforce lan. very strange...i haven't a clue as to why you're experiencing those issues. we should all just have a "switch-to-intel" party. anyone in? i'll bring the chips and salsa

I don't want to stress much about this, but I had the same issues as you with torrent and amule and now I'm 3 days stable 24/7. I founded that the APIC setting in BIOS was the matter. Plus I set the IRQ not being assigned automatically. I think this is a problem related to 2 core AMDs. Nobody told me to do this, but after month of trying everything possible this was what kept me from buyng another system. I also used the idlehalt=0 fn=3 parameter.

so are you still using idlehalt=0 now? and what does fn=3 do?
what did you do with your APIC setting? on, off, ?

[cite] agrafuese:Yeah, that app doesn't work for AMDs...tried it some time ago. So Ianxxx, you went with the DS3L? I thought you were gonna go with that other cheapo board? That's good news though, because that's the board I want. Glad to hear another account of the DS3L working great. IMO, 2-channel audio is fine. It's all I use.…

When I went shopping the place I use only had DS3Ls in stock so I thought what the hell. It seems certain components are very in demand. E2180s for isntance. I bought an E2160 unfortunatly it will only clock to 2700 but hell it's running at 25degrees 1.375v so I can live with that. If you do get one of these boards the only downsides I can see are, reports of max bus speed 450mhz (but I think thats high enough as far as I'm concerned) and it seems a little fussy with memory bought 800mhz 444-12 memory it runs at 555-15 but again it will do this at 900mhz default voltage no problem. I used kalyways disc to install and his 10.5.2 update, then just updated everything by software update. Results seem almost perfect, even X11 working. Xbench scores:

Good stuff man. My #1 priority for a mobo is that it is compatible with Leopard to the point of feeling like a real mac. I had it this way on my current system with Tiger, and I intend to get it back in Leopard with an Intel system. I honestly could care less about overclocking. I figure if I really cared that much about speeding things up, I'd rather just put another $100-$200 into a better processor or more RAM. Overclocking is fun and interesting, but I really don't enjoy it as much as others might. It's a bit tedious for me, and I'm constantly paranoid that I'm going to do something wrong. I just want a stable system. So the DS3L sounds like a perfect board as far as that is concerned. As far as a processor goes, I'm a little torn between getting a Q6600 or an E6850. I'm still not clear on whether or not the quad-core would even do anything for me in Leopard, and the E6850 is stock at 3.0ghz with a higher FSB. It's only a $20 difference between the two, so I'm a little on the fence there. Ianxxx, do you know if quad-core would be beneficial at all? Keep in mind OC isn't a concern for me...

[cite] agrafuese:Good stuff man. My #1 priority for a mobo is that it is compatible with Leopard to the point of feeling like a real mac. I had it this way on my current system with Tiger, and I intend to get it back in Leopard with an Intel system. I honestly could care less about overclocking. I figure if I really cared that much…

As far as I understand it quads come into there own when doing things like converting video files etc. It's a tough one because you have to take into account each piece of software, can it use quad core? If I where you and I was going to spend that type of money I'd consider an E8400 you can get an oem for around £120 or a boxed version for £140, 3Ghz 6mb cache 45 nano tech.as far as I know there is no problem with OS X. I know you said you're not interested in overclocking but reports are that this processor will easily overclock to 4Ghz I think that says something about the processor overclocked or not.

I have an intel system as well (GA-P35-DS4-E6850-4gb-7600gt) look my xbench, maybe this can help you.
I have leo installed in a software raid.

[cite] roisoft:Hi! I have an intel system as well (E6850-4gb-7600gt) look my xbench, maybe this can help you. I have leo installed in a software raid.

roisoft what make of hard drives are you using, I was just wondering wether to set up software raid or buy a better hard drive?

2 x 400 WD 16mb cache and 2x 320 Samsung 16mb , same performance +- (both)

@roisoft, your bench scores are absolutely sexy. is your DS4 board working 100% with full-features? anything not working? are you overclocking? is your leopard vanilla?

@Ianxxx, thanks for the advice. I'll check that CPU out.

roisoft how come your graphics scores are so high? You easily outscore my xfx7900gs?
Also could you tell me what kext you use to recognise all 4 sata ports?

yeah my 7900 gt/ko @ 580/1680 gets

Quartz Graphics Test 174.14
Line 132.96 8.85 Klines/sec [50% alpha]
Rectangle 180.65 53.93 Krects/sec [50% alpha]
Circle 148.21 12.08 Kcircles/sec [50% alpha]
Bezier 139.63 3.52 Kbeziers/sec [50% alpha]
Text 572.67 35.82 Kchars/sec
OpenGL Graphics Test 108.79
Spinning Squares 108.79 138.00 frames/sec

E6850=3.4ghz( 24/7) +400mhz OC, 3.8 (stable), 3.95 my top score, but i don´t want to force too much and i don´t need it because the performance is awesome. All the P35-DS series are perfect for OS X, workig 100%, ethernet, 6+2 Sata2, E-sata works, Sleep/Reboot works, Shutdown 50%, 100% stable. Digital Out/In works perfect, and if you want 5.1 you need set up an agregate device in midi preferences and that´s all.

I´ve installed from retail dvd adding dsmos.kext, efi-strings for audio and Gfx, and deleting AppleIntelCpupowermanagement.kext .. that´s all you need.

I´ve enabled the QuartzGL and disable the Beamsync for do this open a terminal and type

sudo defaults write /Library/Preferences/com.apple.windowserver QuartzGLEnabled -boolean YES sudo defaults write /Library/Preferences/com.apple.windowserver Compositor -dict deferredUpdates 0

LogOut-IN

So the answer to the AMD problem is to buy a new intel board and CPU. Joy...

I´m not saying this, I ´m very happy with my 3 amd systems, i love them, I don´t have this kind of problems with them...Did you try another install method? I always use a retail dvd from a spare partition.

I'm very happy with my AMD as well

Works flawlessly since 10.5.1 Rev.2 Beta (Out of the Box for my system), which I still use - (updated with eddie11c's updater to 10.5.2)

Me too, but i just cant seem to get leopard to stop crashing. i have sorted the mad mouse movement bug and tons of others mainly driver related problems i'd been having. but it seems me and others just cant get a stable system ie random crashing, torrents,cant leave on overnight as it crashes. the reason for my above post was, it seems looking at this thread that alot of people have just gone out and bought intel to overcome there problems. please dont feel that i am unaprecitive of all the hard work the devs have done and all the support recieved. i do enjoy the 4-5 hours i can use leopard before a crash its just a tad frustrating..

Firewalk

what do you have for ram in your system?

[cite] Firewalk:Me too, but i just cant seem to get leopard to stop crashing. i have sorted the mad mouse movement bug and tons of others mainly driver related problems i'd been having. but it seems me and others just cant get a stable system ie random crashing, torrents,cant leave on overnight as it crashes. the reason for my above…

Hi bhast2, I have 4x1 gig ddr2 800mhz unbuffered.. not sure of the brand but its heatsinked.

Loving those black window decorations how did you change those? i cant find any UI mods for leopard with shapeshifter not compatible.

[cite] roisoft:E6850=3.4ghz( 24/7) +400mhz OC, 3.8 (stable), 3.95 my top score, but i don´t want to force too much and i don´t need it because the performance is awesome. All the P35-DS series are perfect for OS X, workig 100%, ethernet, 6+2 Sata2, E-sata works, Sleep/Reboot works, Shutdown 50%, 100% stable. Digital Out/In works…

LogOut-IN

i had problems with my system crashing all the time
i had 4GB of ram and when i pulled one of them out it quit crashing
just an idea but pull one of your sticks out and see what happens

[cite] Firewalk:Hi bhast2, I have 4x1 gig ddr2 800mhz unbuffered.. not sure of the brand but its heatsinked.

@roisoft and puttabong:

i couldn't help but look up your signatures over at insanelymac and i happened to notice something veeerryyy interesting:

puttabong: RAM - 1 GB Kingston HyperX DDR
roisoft: RAM - 2GB Kingston HyperX PC3200

okay, it might be a little coincidence, but could it be the reason? roisoft, what are your other 2 AMDs using for RAM?

and finally the big question:

ARE YOU GUYS RUNNING TORRENTS?!?!?! (and if so, for how long?)

sorry, I'll calm down now.

after pulling that 1 ram stick out
i am running torrents now just download about 10GB worth of stuff and no problems what so ever
i was burning a movie and viewing the web and had photoshop open and my system didn't crash at all
no sign of crashing no lag no nothen
I HAVE NOT HAD A SYSTEM CRASH IN 36 HOURS OR ERRORS! i even got ilife working great

2 x A8N32-Sli dlx with 2 gb Kingston HyperX 3200(both) 4200/4600 x2
1 x GA-K8N-Sli with 2GB Kingston Value Ram. 4000+

I use Azureus for Mac & utorrent via VMware Fusion (the best option for me) 5/6 dowloads simultaneuosly , I´m using wifi adapters 2 x Asus WL-138G v2 (broadcom chipset) working perfect like airport and appleyukon.kext from tiger for my A8N32-sli , i don´t use forcedeth. How long? all the time i need.

I always install from retail dvd patched with marvin, restored into a spare partition.

Hi bhast, thanks for your advice. i have now removed 2 gig of ram from the system will test and let you know. although i have tried this before (but i did have different problems at the time..

Firewalk

i hope it works because when you get your system running right you will never work with windows or linux or whatever again

[cite] Firewalk:Hi bhast, thanks for your advice. i have now removed 2 gig of ram from the system will test and let you know. although i have tried this before (but i did have different problems at the time.. Firewalk

@roisoft: so you don't use zephyroth's releases? what method do you use to restore the DVD into spare partition? just normal Diskutility? And is there a method you use to make the installer bootable? and what version of the Leopard install DVD do you use?

also, thanks for the info on the wireless adapters. i may check them out. i've always hated the idea of getting a new lan adapter when i already have two built into my mobo (marvell and nLan), but if it's a wireless adapter, i wouldn't mind.

I have an external disk (usb/firewire), with 1 working leo, 1 tiger and 1 partition for installers. I like to see all the installers (zephyroth, iatkos, kalyway, eddie11c, XxX, tubgirl...).

I use the retail dvd 10.5.0, apply the efi bootloader for that partition and copy mach_kernel (toh), dsmos, AppleNforce..and delete Appleintelcpu,,,finally patch with marvin the partition. Don´t forget once the install finishes you need repeat the steps with the new install (copy kernel,dsmos…patch with marvin).

thanks for the info, i'll give it a try when i have some more free time. do you find that using the retail DVD has any advantages over Zeph's DVDs, or is it just for your own peace of mind?

riosoft!

how the heck is your graphics/OpenGL scores in xbench so high?

i have a 7900GT/KO that scores far lower than your 7600

my clock speeds are even factory oc'ed @ 580/1680

there is something magical about the 7600. it's favored by OSX, apparently. i can tell you that scores with my 8800GT are nowhere NEAR the ones roisoft is getting. actually, OSX hates my card :\

Real mac users use the retail dvd, so i like to be more close possible to real mac you know ?...don´t get me wrong, zeph´s releases are great and i think his work is awesome, admirable and amd users must be very happy and thankful.. [is this the word? sorry for my english ]

I have a 7950gt as well and 7600 have better results... the relation drivers/card are perfect.

hey roisoft, i'm going to send you a PM over at insanelymac.

hmm...i guess that explains it then ...i thought maybe i had a problem to fix, hehe. thanks for explaining.

[cite] emaxxx:Hi guys. I had this issue (mouse jumping and reboots after hours of bittorrent and heavy download overnight) a while ago (10.5.1) and after reinstalling Leopard many times, replacing kext and hardware components like usb2 and network i found It could probably be related to APIC implementation in BIOS. What led me to this…

This solved my issues, too!

[cite] ghostdex:This solved my issues, too!

I actually looked into my BIOS for the APIC setting, and mine was grayed out for some reason. It also looked like I would only have the choice between "on" or "off" ...so I don't think it would have let me pick a version number, like emaxxx said. So no go for me.

On my board its called MPS Version control i have a DFI Ultra-ii-m2 Nforce4-Ultra

My system finally seems stable 48 hrs uptime no bugs, used transmission for torrents still fine, stress tested to hell and back, all seems ok. for me the answer was to remove 2 gig of ram from my 4 gig system now working like a charm..thanks to the devs and all the people on the forums for there help..

Firewalk

its something with 4 gigs of ram that kills the system

[cite] Firewalk:My system finally seems stable 48 hrs uptime no bugs, used transmission for torrents still fine, stress tested to hell and back, all seems ok. for me the answer was to remove 2 gig of ram from my 4 gig system now working like a charm..thanks to the devs and all the people on the forums for there help.. Firewalk

Hi bhast2, do you think 3 gig would be ok..if so in what banks..

Firewalk

I have 4gb ram and i had this same problem. I had to turn OFF "Memory Hole Remapping" or Remap Memory Hole some BIOS are different. I found this option in my DRAM setting menu where i set CAS/RAS/TRCD /etc..

[cite] bhast2:its something with 4 gigs of ram that kills the system [cite] Firewalk:My system finally seems stable 48 hrs uptime no bugs, used transmission for torrents still fine, stress tested to hell and back, all seems ok. for me the answer was to remove 2 gig of ram from my 4 gig system now working like a charm..thanks to the…

3 gigs are fine i have mine in Dimm1 Dimm 2 Dimm 4

[cite] Firewalk:Hi bhast2, do you think 3 gig would be ok..if so in what banks.. Firewalk

Just a thought:

Whom of you guys with the choppy mouse is using Firefox as their main browser? I had those problems too, i stopped using Firefox anx anything releated to it (i.e. Thunderbird) and i had no crashes for 2 days (47 hours uptime). Could it be related to Firefox?

That's what I was thinking originally but someone said they don't use it just Safari and still had the problem.

Just thought I'd chime in and say I made the move over to Intel
I needed to get serious, and working with an AMD felt like playing with a little toy. I'm not sure what my experience will be like with this new setup, as I've only just put it all together, but I installed from the Retail DVD, replaced a few kexts, and it's running beautifully so far. Just to give a little perspective:

AMD 3800+ (2.0ghz stock)
1GB 400mhz RAM
Nvidia 8800GT 512MB
---------------------------
xBench score = 116 (at best, with no disk test)

Intel 6750 (2.6ghz stock)
2GB 800mhz RAM
Nvidia 6200TC 64MB (old school)
---------------------------
xBench score = 230 avg. (no disk test)

I'll be testing out my new system with the 8800GT later, but will probably keep the 6200 in it just because I still want to use my AMD for gaming. Also, I may be swapping out the E6750 for an E8400 in a few days. Either way, I really can't complain about a 100% performance increase.

[cite] agrafuese:Just thought I'd chime in and say I made the move over to Intel I needed to get serious, and working with an AMD felt like playing with a little toy. I'm not sure what my experience will be like with this new setup, as I've only just put it all together, but I installed from the Retail DVD, replaced a few kexts, and…

Welcome to the new club lol.
I wouldn't mind seeing your xbench scores, what mobo did you buy?

Hey Ian, no prob bud. I'll send you a bunch of different benches at insanelymac after I am finished doing some tests today - don't want to mix too much Intel talk over here, hehe. I'll be trying out AHCI vs IDE, time machine tests, may possibly swap in the 8800GT, and also might try out the 9.2.2 hacked kernel. At the moment, I am using the retail 9.2.2 kernel and it works really well! I get shutdown/restart, but I haven't tried sleep yet. See you over at insanelymac.

EDIT: Wow! Sleep works, and wake works with power button-tap! So I have a 100% working retail/vanilla system.

Same here, i'm going to say bye bye to AMD soon. I'm tired of having a not perfectly running OS since i need to work with it. It's not just some fun. I know, some ppl might say "then use another, stable OS". Anyways, my decision is to use Leopard so i do whatever is necessary to get it stable (except buying a real and overpriced Mac).

So i've ordered the same hardware lanxxx got:
Gigabyte GA-EP35-DS3R
Intel Core 2 Quad Q6600 4 x 2,4GHz
and 4 Gig Kingston HyperX 2 GB Kit DDR2-1066

I'll give it a try with 4 gigs, i'm sure it'll work =) At least i hope so ^

yeah 4 gigs will work, i think. i only have 2, but one of the guides i read for DS3* boards said there's a way to get it to work. just can't remember right now i'm with you on the importance of a stable system. i have a feeling that a good number of people on these forums either don't run anything demanding on their systems, or they just don't use their systems enough to notice the issues we've been having. but, hell, i'll keep running my AMD as a hackintosh with Tubgirl 10.4.10 ... but that's about all it's good for in the hackintosh realm. :\

[cite] agrafuese:Hey Ian, no prob bud. I'll send you a bunch of different benches at insanelymac after I am finished doing some tests today - don't want to mix too much Intel talk over here, hehe. I'll be trying out AHCI vs IDE, time machine tests, may possibly swap in the 8800GT, and also might try out the 9.2.2 hacked kernel. At the…

When I use vanilla kernel shutdown only works now and again what release etc. are you using? I can't find a hacked 9.2.2 kernel, or do you mean the one thats floating round this forum?
also subsonic if you get board with ICH9R then I don't think you have a problem if you get board with 9ICH then you need to use the new driver for it.

I'm going to put my 2 cents out there, I have been looking at this problem for this past 2 weeks and I've carefully monitored the work load on my pc. What I found was I don't have the performance bog monster until I play about 2 hours of warcraft 3. After that you can clearly see the Warcraft 3 menus moving up and down sparaticly, at that moment everything goes glitchy and mouse movements crazy that exact moment when warcraft 3 for me hits the crazy menu bars movement is when the bogging starts. That's all I really know from now but I can maintain a controlled environment, until I get into warcraft 3, warcraft 3 is a guarantee catalyst to the problem. It's the only application that can cause the mystery system bog monster . Hope that helps somehow.

My Specs;

Asus A8N-SLI Deluxe
2GB Patriot Dual Channel 3-3-3-8 Ram
AMD Opteron 180
8800GT

I'm all for someone brilliant fixing this problem!

Is there anyway yet known to prolong the strange problem clearing memory erasing cache something?

It seems that memory intensive apps can cause issues, as prolonged use of Photoshop CS3 causes this bug too. Adjusting the memory/performance settings within PS itself can stave off the bug for a while longer, but eventually it catches up. It's the only app that causes this for me which is a great shame as I use PS quite a lot.

I am running Leo4all v3 10.5.2

I also had this issue, but when i switched from PCI WiFi card (D-Link DWL-G510 Rev c) to the USB WiFi stick(D-Link DWL-G122) , the problem solved.

Xbench score = 155 (Without Disk Test)

Xbench Score = 122 (With Disk Test)

My sys config::

AMD Athlon FX-60 @ 2.6 GHz
Asus A8N -SLI Deluxe
2 GB Mushkin DDR500 Ram @ 400 MHz
eVGA 8800GT SC 512 MB (GFX-String with NVinstaller.41 and patched Div ID Files, QE / CI Enabled)
Asus SATA DVD-RW
Seagate 250 GB SATA II HDD

One more update :
For people who having Logitech Mouse, there is Logitech driver and control utility for Mac OS x on Apple's site.

I found serious improvement using my Logitech USB keyboard/mouse.

www.apple.com/downloads/macosx/drivers/logitechcontrolcenter.html

check out

I only notice the mouse stutter bug when I am transferring large files, across HD's for instance. Otherwise everything works great. Is it something to do with the ata/sata drivers?

[cite] seren:On my board its called MPS Version control i have a DFI Ultra-ii-m2 Nforce4-Ultra

Same on mine. I got a DFI Lanparty Nforce 4 SLI-DR.
Just switched it to 1.1
Will report back on what happens.

one more suggestion, in motherboard BIOS, try changing your ACPI -APIC settings. Try S3.

Avoid S1 and S3 together.

[cite] snyder9:[cite] seren:On my board its called MPS Version control i have a DFI Ultra-ii-m2 Nforce4-Ultra

Same on mine. I got a DFI Lanparty Nforce 4 SLI-DR.
Just switched it to 1.1
Will report back on what happens.
System running for 30+ hours now. One Firefox crash (probably website related).
No mouse movement bug though as yet.

Unfortunately about 5-6 hours later, the mouse movement bug reappeared along with some weird behavior in Safari and segmentation errors repeatedly crashing Adium.
I changed the ACPI settings to S3 but this didn't seem to help.
Installed the 9.2.2 kernel, and though the system felt a little smoother - the mouse bug was still persistent.

Would this bug show up if I install leo4allv3?

[cite] snyder9:Unfortunately about 5-6 hours later, the mouse movement bug reappeared along with some weird behavior in Safari and segmentation errors repeatedly crashing Adium. I changed the ACPI settings to S3 but this didn't seem to help. Installed the 9.2.2 kernel, and though the system felt a little smoother - the mouse bug was…

I am using Leo4all v3, and i have no problems, so i guess you can give it a shot.

I've reverted to using CPUS=1 until a proper fix has been found

Stability is more important to me than performance. Nothing worse than an app quitting and losing work!

Voyn1x, if you haven't already, try Leo4Allv3. It might solve the problem.

However, I use zephy's release and it seems to work fine, even after about 10 hours of uptime.

You seem to be one of the lucky ones Puttabong. My machine runs great, but as soon as I start loading files into Photoshop, run a few filters and layer styles, the bug appears and a reboot is necessary. I'm toying with installing Leo4All until my Intel kit arrives.

@Puttabong, yeah i've thought about it - i'm hoping it might of fixed itself after updating to 10.5.3. I haven't got around to trying it again without CPUS=1 yet.

I'll report back when I have.

[cite] rohitkpatil:[cite] snyder9:Unfortunately about 5-6 hours later, the mouse movement bug reappeared along with some weird behavior in Safari and segmentation errors repeatedly crashing Adium. I changed the ACPI settings to S3 but this didn't seem to help. Installed the 9.2.2 kernel, and though the system felt a little smoother -…

I am using Leo4all v3, and i have no problems, so i guess you can give it a shot.

Can you tell me what all you have changed in the BIOS (if anything) to solve the problem?
Memory Timings, APIC, etc.

The stuttering problems usually for me occurs after watching 720p content which puts a lot of strain on my 4200+X2 with VLC.

Also, does it matter that I have 512mb x 4 running in dual channel in my system? I read earlier in this thread that someone solved the problem running 3gb instead or 4gb. Is this the same sort of problem?

[cite] snyder9:[cite] rohitkpatil:[cite] snyder9:Unfortunately about 5-6 hours later, the mouse movement bug reappeared along with some weird behavior in Safari and segmentation errors repeatedly crashing Adium. I changed the ACPI settings to S3 but this didn't seem to help. Installed the 9.2.2 kernel, and though the system felt a…

I am using Leo4all v3, and i have no problems, so i guess you can give it a shot.

Can you tell me what all you have changed in the BIOS (if anything) to solve the problem?
Memory Timings, APIC, etc.

The stuttering problems usually for me occurs after watching 720p content which puts a lot of strain on my 4200+X2 with VLC.

Also, does it matter that I have 512mb x 4 running in dual channel in my system? I read earlier in this thread that someone solved the problem running 3gb instead or 4gb. Is this the same sort of problem?

Your memory configuration should make no difference the 4gb problem is caused by 32 bit limitations. As for solving the mouse movement bug as far as I know the only cure is still the idle halt in com.apple.boot.plist "see earlier in this thread". If you use it keep a close eye on your processor temperature.

I thought IDLEHALT=0 had cured it for me initially, but after working for a period of time in photoshop the jumpy mouse came back unfortunately

It only seems to delay the problem.

Have you tested the keyboard mouse? Go in System Preferences -> Universal Access -> Mouse and switch on the keyboard mouse. Now you should be able to control the pointer with the numpad keys.

Have you a function 'USB Legacy support' enabled in BIOS? Some computers have problems with mouse and/or keyboard if it's enabled.

[cite] Voyn1x:I thought IDLEHALT=0 had cured it for me initially, but after working for a period of time in photoshop the jumpy mouse came back unfortunately It only seems to delay the problem.

Exactly the same problem here mate.

Hm, the only problems with mouse I ever had were artefacts caused by a wrong gfx driver. On 5 machines, all using Logitech PS/2 mouse/keyboard combos. But USB works too.

I read a long time ago that AMD stuttering problems could be solved to set the CPU multiplicator to a .5 value if the BIOS supports this. so 10.5 insted of 11. But this was maily ment to Dual-Core CPU problems.

I'm going to try running with the cpus=1 flag and see how that works out. From what I gather it might well be attributed to a timing issue between the two cores, so rather than use idlehalt flag I'll see how this pans out.

cpus=1 is very stable for me.
It's not ideal though if I you can't use both cores. I can barely play 720p with one core only.

As for the multiplicator, if I change it to 10.5 what else do I have to adjust as my system does not POST on this setting.
FSB is at 200 and LDT/FSB freq ratio is on Auto. I have a 4200+ X2.

It is perhaps depending on the mouse resoulution? My pointer speed is set at the third marker from the left, but I need only 2 x 2 cm to span the 1600 x 1200 pixel screen size. Don't know which resolution my mouse has, maybe 800 or so. It's an optical mouse.

[cite] naquaada:Have you tested the keyboard mouse? Go in System Preferences -> Universal Access -> Mouse and switch on the keyboard mouse. Now you should be able to control the pointer with the numpad keys. Have you a function 'USB Legacy support' enabled in BIOS? Some computers have problems with mouse and/or keyboard if it's enabled.

I do have legacy support enabled in the bios. Surely if it was that then cpus=1 shouldn't make a difference but it does. With this kernel flag my system is solid as a rock, unfortunately losing 25 points off my xbench score in the process

@ Rogue, i am inclined to agree it must be a timing issue - seems everyone who has experienced this problem are all running x2 processors.

I have an Opteron 185, it's a also a Dual-Core cpu, it's working perfectly, no mouse problems. I even use it overclocked from 2,6 Ghz to 2,86 GHz, still no problems.

Do you have the mouse problem already in the installation DVD or in safe mode (boot with -x). Maybe it's a gfx driver problem. I also have deleted all GeForce, NVDA* and other ATI gfx kexts except my RadeonX1000 kexts.

My machine has now been up for well over 12 hours without any sign of the bug. Last night I was working in Photoshop without a hitch and after leaving the machine on over night it's still running along quite happily. I'm wondering if certain apps that use SMP are causing the problem - more specifically third party apps. I think it's safe to assume that this is not a 'mouse bug' per se, but a timing issue between the two cores making the system skip and stutter and the mouse pointer to freak out. I'll keep on testing and report back in later.

Could you tell me your bus frequency and the multiplicator settings? Mine is actually 220 MHz, Multiplicator 13x, Vcore 1.3125 V. I'll test the settings you have when you post them. But I presume you both thave an Athlon 64 X2 processor which have 2x 512K Level-2-Cache. The Opteron series and the FX-60 (Socket 939) have 2x 1MB L2-Cache. I phoned with the AMD support and they said they're replacing the missing 512K with 200MHz more bus frequency.

Was a great game yesterday, Germany-Portugal. I'm no big soccer fan, but it was good for me: I got a 7 Euro Paypal-Coupon from ebay 'cause we're in the semi-final

Voyn1x, we have the same CPU, hence I don't believe the mouse stuttering problem is related to Dual Core. Actually I can remember that I had this problem a long time ago, when I was starting with OSx86. I think it must be due to (driver?) incompatibilities because apparently I didn't experience this problem for months, since my machine is configured fine.

Tell me what I shall do to test whether I really don't have the problem - which tasks shall I run?
I use a lot of resources of my machine and most of the time got it running for several hours.

@naquadda,

I'll be able to give my bios settings tomorrow, i'm away from my hack until then

@Puttabong,

Try running some heavy filters in photoshop, or do something cpu intensive.

The problem doesn't normally appear until at least 10hrs+ have passed. You'll notice it most if you try to drag a window around, you should see it jump about. Leave Firefox open and see if it quits with a 'divide by zero' error at all, it normally quits every couple of hours for me.

Eventually you should start to get the 10810 error which is only fixable by logging out and back in again. I believe this is all related to the mouse bug since all of these issues are solved if i you boot with cpus=1.

to add, when using cpus=1 my system was up for 2 weeks with no reboots or errors of any kind.

Firefox 2 quits several times this is true. Preview exits if I use the 'Undo' function too often (Alt-Z/Option-Z).

I just tested something with CPU load: A Logic 8 project with 69-tracks, 5.1 surround realtime mixing with fullscreen video and playing a DivX movie (widescreen but full screen width) with Quicktime simultaneously. I could move the windows around VERY fast and smooth, no problems at all. Each CPU needed about 70-80%.

Yes. Leo4All+ 10.5.3 AMD Update. But I also had no problems with Zephyroth 10.5.1 RC2. I now will install a new 10.5.1 RC2 plus a full patching to 10.5.3 with the combined Apple Software update using Zephyroth's Updater. I must check why Xcode crashes during installation.

Only to compare: I'm using a Socket 939 CPU with nForce 4 SLI chipset. Are you guys using AM2 and a nForce 5xx chipset?

Ahaa. Have you checked your CPU and bus speed now in System Profiler?

I can't right now as I'm at work, but why do you ask?

Only for comparison. Maybe I can try the same settings, but I presume your total CPU frquency (bus*multiplier) is higher than I can set mine.

Wow, Zephyroth's Updater found 132 cpuids in the combined 10.5.3 update. It's 549 MB in size.

iirc the system profiler does not correctly report the Bus Speed correctly. It states mine as being 200Mhz when in fact it should be 2Ghz (2 x 1000Mhz).

edit: Maybe I am wrong about the Bus Speed reading - I'm probably getting confused with the HT speed. :\

No, the update from 10.5.1 -> 10.5.3 was not working. I have always errors on loginwindow, atsserver and mds.

22 hours of up time using CPUS=1 boot flag and my machine is still rock solid. I've spent some time in Photoshop creating artwork and applying filters with no visible sign of performance degradation.

Any Leo4All users experiencing this bug or is it just those of us using Zeph?

For now I can live with one core @3Ghz (though not ideal) until either A) I install Leo4All or B) my Intel kit arrives.

Can you reduce the CPU clock speed in the BIOS? As I o'clocked my system to 3.06 GHz (on 10.5.2) it crashed, couldn't boot anymore. On 10.5.3 I can't have to stay beneath 3 GHz. The actual Macs also only have 3.06 Ghz max, could the high cpu speed be the problem?

@naquaada: I'll take a look at my bios settings in due course, but I first want to test this machine thoroughly before rebooting. So far it has been up for over 24 hours and I've hammered it without so much as a hitch. It's definitely a Dual Core problem as far as I am concerned.

It has to be, before I upgraded my A64 3200+ to my Opteron 165 x2 I didn't have this problem.

I've been speaking with JaS about this and he's convinced that this is a TSC timing issue. He's kindly going to find out about getting a patch to fix the timing issue. I'll keep you all posted.

http://developer.amd.com/assets/TSC_Dual-Core_Utility.pdf ↗

That's really good news! Rogue, after reading that pdf you linked, this would also explain those strange time correction entries in the console when running both cores. eg.

3/18/08 8:41:32 AM mDNSResponder[17] mDNSPlatformRawTime went backwards by 4 ticks; setting correction factor to 206737245

Vony1x: You also have a Socket AM2 board I presume. Could this be a problem, maybe because of the DDR2 RAM? Puttabong and me have Socket 939 and it's working fine here. Have you disabled Cool'n'Quiet in the BIOS? I have it enabled.

nope socket 939... Cool'n'Quiet not sure if i have an option for that, i'll have a look

after a bit of research (ie. Google), it seems there's no way to disable Cool'n'Quiet on my mobo

@ Rogue, do you have an option for CnQ in yours, If so is it enabled?

What kind of motherboard do you have?

EDIT: Found it in your account infos. Which version is it? there are three (giga byte.us/Search/Search_List.aspx?Keyword=GA-K8NF-9&SearchType=Spec) different.

EDIT2: Here is the manual for the version GA-K8NF-9-RH (asia.giga byte.com/FileList/Manual/motherboard_manual_k8nf-9-rh_e.pdf). I presume the BIOS versions are rather similar.

EDIT3: I Checked the BIOS setting and the complete manual, not a word about Cool'n'Quiet. This is crazy.

Have you tried to disable all unneccessary devices? Today this could be Floppy controller, LAN Boot ROM (is your onboard LAN working?), Parallel Port, Modem Port, Gameport, MIDI Port. If you use a PS/2 mouse and keyboard, I would set all three Legacy USB options to 'disabled'. In the 'MB Intelligent Tewaker' settings must be an option 'CPU Spread Spectrum', disable it.

[cite] Voyn1x:after a bit of research (ie. Google), it seems there's no way to disable Cool'n'Quiet on my mobo @ Rogue, do you have an option for CnQ in yours, If so is it enabled?

C'n'Q is located in the Cell menu on my motherboard. Mine is enabled because I don't want my system running at full CPU when in an idle state or merely just chatting and surfing.

mines the Rev 1, onboard lan is disabled... In theory most of the options you've suggested shouldn't make a difference if they're enabled or not since the machine is fine running cpus=1. I'll give disabling the spread spectrum a go to see if that helps at all though.

Voyn1x, on your mobo under M.I.T. you have an option named 'K8 CPU Clock Ratio', if that is set to Auto then C'n'Q is enabled, if you've changed it then it's disabled.

With idlehalt=0 and cpus=1 I am still having the mouse stuttering bug, more so when Transmission is checking data on an existing torrent. Large file transfers seem to be smoother though.. Still, the mouse movement still stutters. This is in Leo4Allv3 10.5.3.

3 days 12 hours uptime with cpus=1.

Stable as a rock for me. Using Leo4allv3 10.5.3.

can't understand it, it works great on my Opteron. Have you tried boot option -legacy? This are my com.apple.boot.plist settings:

<plist version="1.0">
<dict>
	<key>Kernel</key>
	<string>mach_kernel</string>
	<key>Kernel Flags</key>
	<string>-legacy</string>
	<key>Timeout</key>
	<string>3</string>
</dict>
</plist>

I deleted ALF.kext on my system, also all other gfx drivers which I don't need for my own card (Radeon X1000 series).

[cite] StealthMode:With idlehalt=0 and cpus=1 I am still having the mouse stuttering bug, more so when Transmission is checking data on an existing torrent. Large file transfers seem to be smoother though.. Still, the mouse movement still stutters. This is in Leo4Allv3 10.5.3.

I think you've got a different issue there, could be something to with your hard disk/network drivers. Do you have any other symptoms? Apps quitting etc..?

I get the 1080 error in Zephs after awhile, iChat crahses a lot. Leo4Allv3 seems pretty stable app wise, but the mouse bug is worse I think.
I have a Buffalo WLI-12 wireless nic (broadcom chipset), I did not install the forcedeath kext for the nforce LAN, and I have the marvel LAN installed. (DFI Ultra-D mobo) The bug never shows up till torrents are being checked or I am transferring large files. I'm confused.

Right, this morning(7hrs ago) I disabled the spread spectrum in my bios and booted with the -legacy flag.

To test I opened Firefox, every CS3 application and started playing a 4.3gb wmv-hd file.

I've just got home and all apps had quit on me. I had to logout and back in again because I was getting the -10810 error. Oh well back to the drawing board…

Days logs are below:

23/06/2008 10:15:46 com.apple.launchd[147] ([0x0-0x19019].org.mozilla.firefox[208]) Exited with exit code: 3 
23/06/2008 10:17:42 com.apple.launchd[147] ([0x0-0x1b01b].com.adobe.illustrator[212]) Exited abnormally: Floating point exception 
23/06/2008 10:17:51 com.apple.launchd[147] (com.apple.ReportCrash[995]) Exited abnormally: Floating point exception 
23/06/2008 10:17:56 com.apple.launchd[147] ([0x0-0x24024].com.apple.quicktimeplayer[399]) Exited abnormally: Floating point exception 
23/06/2008 10:19:01 com.apple.launchd[147] ([0x0-0x1d01d].com.adobe.InDesign[214]) Exited: Terminated 
23/06/2008 10:19:30 com.apple.launchd[147] ([0x0-0x1a01a].com.adobe.dreamweaver-9.0[211]) Exited: Terminated 
23/06/2008 10:20:09 com.apple.launchd[1] (com.apple.UserNotificationCenter[999]) Exited abnormally: Floating point exception 
23/06/2008 10:25:31 Adobe Crash Reporter[1057] Process Id:  
23/06/2008 10:25:31 Adobe Crash Reporter[1057] 213 
23/06/2008 10:25:31 [0x0-0x1c01c].com.adobe.Photoshop[213] 2008-06-23 10:25:31.523 Adobe Crash Reporter[1057:10b] Process Id:  
23/06/2008 10:25:31 [0x0-0x1c01c].com.adobe.Photoshop[213] 2008-06-23 10:25:31.525 Adobe Crash Reporter[1057:10b] 213 
23/06/2008 10:25:41 com.apple.launchd[147] ([0x0-0x1c01c].com.adobe.Photoshop[213]) Exited abnormally: Bus error 
23/06/2008 10:26:27 com.apple.launchd[1] (com.apple.UserNotificationCenter[1060]) Exited abnormally: Floating point exception 
23/06/2008 10:33:28 com.apple.launchd[147] ([0x0-0xa00a].com.apple.systemuiserver[162]) Exited abnormally: Floating point exception 
23/06/2008 10:46:33 com.apple.launchd[147] ([0x0-0x39039].com.apple.systemuiserver[1075]) Exited abnormally: Floating point exception 
23/06/2008 10:54:49 com.apple.launchd[147] ([0x0-0x3a03a].com.apple.systemuiserver[1081]) Exited abnormally: Floating point exception 
23/06/2008 11:41:17 com.apple.launchd[147] ([0x0-0x3c03c].com.apple.systemuiserver[1088]) Exited abnormally: Floating point exception 
23/06/2008 12:17:59 com.apple.launchd[147] ([0x0-0x41041].com.apple.systemuiserver[1114]) Exited abnormally: Floating point exception 
23/06/2008 12:57:43 com.apple.launchd[147] ([0x0-0x43043].com.apple.systemuiserver[1130]) Exited abnormally: Floating point exception 
23/06/2008 13:00:53 com.apple.launchd[147] ([0x0-0x48048].com.apple.systemuiserver[1151]) Exited abnormally: Floating point exception 
23/06/2008 13:01:30 com.apple.launchd[147] ([0x0-0x4a04a].com.apple.systemuiserver[1157]) Exited abnormally: Floating point exception 
23/06/2008 17:49:15 com.apple.launchd[147] ([0x0-0x4b04b].com.apple.systemuiserver[1159]) Exited abnormally: Floating point exception 
23/06/2008 17:53:10 com.apple.launchd[1] (com.apple.launchd.peruser.501[147]) Exited abnormally: Floating point exception 
23/06/2008 17:53:13 com.apple.loginwindow[53] LSOpenApplication returned error -10810, path=/System/Library/CoreServices/SystemUIServer.app 
23/06/2008 18:28:48 UserEventAgent[156] CalAlarmAgent :: WARNING: could not launch iCal Helper 
23/06/2008 18:38:44 Dock[161] [QL ERROR] Can't get server port for <QLClient com.apple.quicklook>: (ipc/send) invalid destination port 
23/06/2008 18:38:44 com.apple.loginwindow[53] LSOpenApplication returned error -10810, path=/System/Library/CoreServices/Dock.app 

Have you tested a second time? I'm not sure that these erros are because of the BIOS setting and the -legacy flag. Man, I'm working with Logic and it works! Only Firefox seems to crash the system sometimes.

No these are the errors I get normally when running both cores - basically there was no change after making the amends to the com.apple.boot.plist and bios. I really think changing my mobo is the only solution for the time being (or sticking with cpus=1).

Even after two hours of uptime with the -legacy flag I got the dreaded -10810 error. Back to cpus=1 ...

I was having the mouse bug and random crashes particularly with transmission but also Quicktime and Safari and also apparently I was getting random restarts as well. I tried a few different things but the kernel flags notsc=1 and nohpet=1 seem to be doing the trick. My system has been stable now for 2 days with only 1 safari crash. Before I was lucky to get 3 hours out of it. The cpus=1 did the trick as well, but if I can be stable and dual-core, I'll take it.

Specs:
Athlon 64 X2 6000+
3GB DDR2 667
Geforce 7900 GS
Apple USB Keyboard
Logitech MX400 mouse
OS X 10.5.3 (Leo4Allv3 updated)

It is my understanding that notsc flag keeps the kernel from checking the time stamp counters on the cores of the cpu. AMD time stamp counters have a tendency to drift apart from one another, so it can cause strange crashes. I did the nohpet flag because I wasn't sure if I had hpet.

uptime just passed 62 hours, stable as a rock w/ dual core enabled. I think we may have a winner.

Amazing News!
I can't wait for Voyn1x and Rogue to confirm that it works - I hope.

Thanks ClawHammer.

I can't believe it, light at the end of the tunnel at last! Must say I've never heard of these flags until now.

Clawhammer, If this works you're a star...

* Adds kernel flags and restarts hack *

I think it fixed my mouse lag when transferring large files across hd's and such. I only used the notsc flag so far. Does this mean I might actually be able to start using Leo full time?!

Edit: maybe not... Might have to add the other kernel flag. Maybe I am having other issues…
Well, I just realized that my music stutters too when I have the mouse stutter bug (while transferring large files).. Both kernel flags are added, am I having issues with the ATA kext?

I couldnt read all the stuff written, its too much, but im having the same problem. My hardware:

GA-K8NXP-SLI
A64 3800+ X2 2ghz @ 2.6ghz (260fsb)
4gb DDR 333mhz @ 420mhz

bla bla…

anyway, i opened all the aplications i had and nothing happened, i do think its something related to the net. Anyway ill wait to see your results with the kernel flag and then ill try !

I added four flags to my com.apple.boot.plist and left the machine running over night with several apps running plus a download. In this time frame there have been no 'unexpected quits', -10810 errors or stuttering/lag. I'll keep you all posted.

<key>Kernel Flags</key>
<string>apicpmtimer=1 notsc=1 nohpet=1 cpus=2</string>

Do you always use the flag cpus=2 ? I don't use it. Is your system gettin slower by disabling HPET ?

I don't always use the cpus=2 flag, this is just a test. HPET is just another timing method, and no, it's not got any slower.

I presume it's for HyPEThreading. But the option notsc ? I have no idea for this.

HPET stands for High Precision Event Timer. TSC is Time Stamp Counter.

I just noticed that when i was hash-checking my torrents the problem appeared, but when i finished, its normal again, i really dont know WHEN it happens to me. But yesterday happened while i was using excel ...lol (of course, torrents in background and other apps).

I want any answer, did it worked with the new flags? Thanks

edit: what i mean by "WHEN it happens to me" i mean when it happens and stays like that…

[cite] pmcnano:I just noticed that when i was hash-checking my torrents the problem appeared, but when i finished, its normal again, i really dont know WHEN it happens to me. But yesterday happened while i was using excel ...lol (of course, torrents in background and other apps). I want any answer, did it worked with the new flags?…

I found that it always happened around the 3 hour mark no matter what I was doing

I had been without the issue a while, whithout doing anything else, lol…meh..i want it to happen again to check some stuff on console, etc

Well, transmission crashed again, so i reboot with the flags posted (hpet tsc) and it got worse! LOL, they crashed like in a little while and i couldnt open anything again! D:

I have found, sadly, that my system is still not perfectly stable. My mouse movement problem is fixed with the kernel flags, and I needed no restarts for 3 days, but then I got the 10810 error again. It is getting really annoying. Itunes also gave me a division by 0 crash. I know it has to be some problem with multiple cores, but what?

After a few days of testing I cannot maintain a stable dual core system with any of the aforementioned flags. Looks like we're going to have to wait for either a patch or switch to Intel. For now I have reverted to using cpus=1 in my config.

well mine crashes right now like in 3 days or something, dont know whats the breaking point, but i wont put cpus=1 lol...i dont want to lose my core =)!

I only have the problem that Firefox crashes rather often. Version 2 and 3.

@pmcnano: Unfortunately it appears to be an AMD dual core issue that needs to be resolved at base code level. For stability, the only solution is to use cpus=1 :-\

I think it is a board/ram/cpu issue. Puttabongs and my Dual-Core system are working without problems. My system has maybe a speciality: I installed it on a single-core computer with the same board and restored the image on my dual-core machine. Even real big applications like Final Cut and Logic are working without problems. BTW: My new M-Audio ProFire Lightbridge 34-channel audio interface is fantastic!

The only progs which are crashing sometimes are Firefox, Toast and the Quicktime Player, but I have the same problems on my single core machine. I presume these programs fail sometimes because the DivX-Encoder is faulty installed. The DivX Player and the DivX converter are also crashing. Firefox, Toast and Quicktime are all using the DivX-plugin. Bad is that I can't repair DivX even after removing all files, so I have to reinstall the whole system I'll do it again on my single-core machine and copy the finished image to my other machines. I have 4 computers with 3 different boards, but all are using variants of the nForce 4 chipset and using a Radeon X1600, so porting the OS is easy - just restore it with Disk Utility from an image

Thought i'd chime in with my results. I have found it certainly a lot more stable, going at least 15hrs without issues - where previously it would only manage 3hrs tops before an app would quit. I also haven't experienced the jumpy mouse since adding the flags, but have still received the -10810 error a couple of times after prolonged uptime.

Also Rogue kindly pointed out that Firefox contains 2 cpuids, so I promptly marvinised the bastards. This seemed to have solved the random crashing of FF. I had overlooked this, I've run Marvins in the past but since Mozilla is constantly releasing updates (especially for FF3) I think the cpuids had been rewritten somewhere along the line.

Oh well, new mobo here I come, maybe an intel chip while i'm at it!

aww i want a new computer too …well…im trying atm this kernel flags, i will report:

<key>Kernel Flags</key>
	<string>apicpmtimer=1 notsc=1 nohpet=1 cpus=2 maxmem=3072</string>

Lets see what happens…atm its working fine..wthout a problem

@All: which distro's are you using?

Zeph 10.5.2 Rev 1 upgraded to 10.5.3 using ASU.

Also using an EFI gfx string, MacNub's AppleSMBIOS27, Modbin 9.3.0 kernel and 9.3.0 system.kext

naquaada, I might try Leo4ALL V3 ppf patched to 10.5.3 to see how it goes, anyway, with the kernel flags i had been using for like 12 hours, i didnt had any problem, lets see how it goes

leo4all with ppf went wrong i cant get it boot, i get kernel panic before anything, i cant even boot in single user mode.

I have been reading, theeres a post in insanelymac about the divide by zero crashes on Tiger, it was because the HTT recognition in the kernel, they had to mod some hex parameters to get it working and suppostly it worked flawlesly, what do you think? It has something to do with the kernel recognizing 2 cpu's instead of 2 cores, or something like that.

http://forum.insanelymac.com/index.php?showtopic=24907&… ↗

What do you think? hope it helps!

This was in August 2006, so maybe 10.4.7 or 10.4.8 = 8.4.1 or 8.8.1 kernel. And only two posts... that won't work here, the kernel is totally different now.

What shows System Profiler on your system? It's 1 cpus and 1 cores for me.

Same here, it was just a though

Anyway, i dont know..lol…I still get some crashes, especially ff and transmission, im not using any kernel flags atm...i just reinstalled and updated. We'll see

If Transmission and Firefox fails, could the network driver be a problem? What are you using? I'm actually using skge.kext because I had problems with the old forcedeth driver. The forcedeth-v driver is working without problems.

Well but its the divide by zero error.

anyway im using a Realtek 8169 card.

So, nvidia drivers are working fine now? The last thing i knew is that it made system's crash

bump?..im using cpus=1 atm =(! no crashes or anything, still mouse movement thing when hdd has heavy loads..

It seems like I just experienced the Mouse Movement Bug while playing around with Photoshop for a few hours...

Ouch.

Finally!

Photoshop does seem to cause the cores to freak out.

Im selling my motherboard/cpu/ram...(well trying to sell them) if i have success i will buy c2d !
It would be first time with intel in like 8 years.

With cpus=1 expose seems slow etc...bah!

[cite] pmcnano:Im selling my motherboard/cpu/ram...(well trying to sell them) if i have success i will buy c2d ! It would be first time with intel in like 8 years. With cpus=1 expose seems slow etc...bah!

Thats what I did and it was the first time in twelve years lol

Lol the problem is…selling my current stuff, i have to "scam" someone lol…XD if i sell it to someone that knows...ill get crap for it =(! and its a great board! =(!

I just want to add that,
i'm using intel cpu with 680i chipset,
everythings run fine, just the mouse will lag slightly (a very slight one, but noticiable) sometimes ...
but it's annoying!

I've tried just about everything mentioned on this thread with no success. Once my CPU undergoes significant load, the various problems start surfacing.

But for now, I can live with using one core. I can manage to play 720p with Mplayer OSX 1.0 RC2 with only minor instances of lagging. The lag was a concern for me using a single core but I realized I wasn't using the RC2 version of the player.

I just added the cpus=1 flag to my boot.plist, it's time for me as well...
The only application that seems to be able to cause the mouse movement bug is Photoshop CS3.

I use it a lot at the moment. Rebooting now.

I never had problems, like strange mouse movements, mouse jumpung or things like that. But what i noticed after installing OS-X onto 2 AMD machines, one Intel C2D desktop and Intel C2D laptop machine is that somehow on my amd machine mouse movements are somehow very unnatural comparing to windows or linux. Its very annoying. On intel machines its just great but only on AMD machine very unnatural and really not comfort using mouse as in windows. I tested with 3 different mice, the same. I am used to this almost, after 5-6 months, but what i wanted to say is this kind of movement i really didnt notice on my other Intel machine.

The more AMD users recognise this is problem the quicker we can find a solution. I know Jas might be looking into it, I wonder if the great Zephyroth has any input/suggestions?

To summarise this for future reference; it is not a 'mouse bug' per se, but a dual core timing issue.
Symptoms include:
• Mouse lag - early signs are drag and drop inaccuracies, trail skipping
• GUI lag - Finder window movement becomes choppy and acceleration degrades
• Random restarts - caused by the 'divide by zero' error resulting in kernel panic
• -10810 error - applications will not start until a restart or log out/in
• increased CPU fan noise and temperatures - due to dual core synchronisation errors
• general erratic behaviour - system feels sluggish, unresponsive at times

Remedy (Temporary)
• Boot using the 'cpus=1' kernel flag - this can be done at the beginning of each boot at the Darwin boot prompt, or permanently by editing the com.apple.boot.plist which can be found in: /Library/Preferences/SystemConfiguration/

I would like to ask, this happened in all versions of leopard?

Does it happens on tiger?

Is there any kernel that detects the 2 cores (in system profiler).

I do think it could have something to do with the HTT issue that happened in tiger.

It also must be a problem depending on the hardware. My Dual Core Opteron 185 (2 x 2.86 GHz overclocked) works absolutely perfect. Firefox quits sometimes, but otherwise I have no problems. I'm running Leo4All V3 + 10.5.3 update by ASU. The board is nForce 4 SLI, Socket 939 with 4 GB DDR RAM 200 Mhz. Puttabongs Dual Core machine also runs fine as far as I know.

Maybe it's a software problem, especially Photoshop? I'm using Logic and have no problems. Or the gfx card? I presume you all have nVidia, I have an ATI.

[cite] naquaada:Puttabongs Dual Core machine also runs fine as far as I know.

Read a few posts back, Puttabong's hack recently became susceptible to the problem. This issue has been around (according to JaS) since the Tiger days, and it is not hardware related it is a timing issue at software level. You're just one of the lucky ones.

Yeah i am one of those lucky people too cause i don't have problems with this

and have a AMD Athlon 64 x2 5000+ and not one problem at all

To the people who say they have no problems, how long on average do you leave your hack on without rebooting.

Many of these issues don't appear until long period of uptime has passed. Also if any apps quit on you during this time with a divide by zero error then you are suffering from the dual core timing issue, albit mildly.

Ya, I've never had a problem, but I don't leave my any of my machines running 24-7. My son probably does - but he's using a single core Turion64 socket 754 CPU. The longest any of my dual (quad) cores run is 15-20 hours at a stretch (well, except when, ummm, using Transmission a few times when they had to be up for a couple of days while Transmission, ummm, finished... what it was doing...)

the only time i reboot or shut down my computer is when i have to for an update

so my computer is always running it could be weeks before i do a restart

I just switched back to two cores, because a single core is just too slow for most applications I use.

i think it is a hit and miss for people because non of my apps crash and I hardly ever turn off photoshop it is always running in the background

Do anyone knows if this happened in tiger? I dont remember, in fact when i upgraded to dual core i changed to leopard so i dont know. I want to attempt to install tiger and update to 10.4.11 but i remember i couldnt before, Can i use the software updater? (Zephyroth's). I know there are kernels, but i dont know about updates.

@bhast2, yes, im with 1 core and its slow... meh

i've start to fiddle arround with some bios settings... maybe there are some options to solve that kind of problem!

changed:
disable: CnQ
memory CL timings raised from CL5 to CL6
Disabled some ACPI settings
Disabled Enhanced halt state
Disabled Advanced Virtual Machine
Disabled USB Legacy Mode
Disabled PCI Busmaster

leopard is working fine atm since 2 hours no crashes no mousejumps, i will report if the issue happens again!

after 5 hours intensive use of macOS with the settings above still no mousejumps or crashes!
i made some hardcore tests to run many heavy applications at the same time e.g. eyeTV, photoshop, reason, toast, adium, UT2003, firefox (about 50tabs lol) and adium(about 10chattabwindows) without any issues!
i will check it out till today evening, if the problems doesnt come back i will change the bios settings back to normal and lemme see what happens!

As far as this issue is concerned I'm one of the lucky ones. I have has sporadic problems with -10xxx errors and the mouse movement. My CPU is a X2 Brisbane 5000+ Black Edition (Unlocked Multi) Stock 2.6GHZ Running at 3.2GHZ with 4gb ddr2 800 @ 886. I'm running a new install of Zephs Rev2 with 10.5.4 (ASU) and modbin kernel 9.4.0. My current uptime is over 2 days not a single issue.

For what its worth i use this to stress test rather than running a bunch of apps, but using this plus a bunch of processes should stress nicely. sof t32.com/Download/Free/CPUTest/4-199513-1.html . The method I use for CPU Test is 32 Instances small and 10 repetitions . YMMV .

PS

2 Days up; and running Transmission (seeding many GB) + the usual's, Firefox will Crash sometimes over night but not resulting in -10xxxx errors anymore with this install.

My computer is online for about 3 or 4 days. Transmission crashed last night, division by zero. My installation has maybe a speciality: It was installed on a single-core system, then ported on a dual-core computer. Have you guys with the stuttering problems tested Onyx? I made some additional settings in my system, maybe this will to the trick.

If it's a mouse problem (like the thread was called first), we should compare the mouse connections. What is you mouse connection and configuration? Here's mine:

Mouse: Wireless Mouse of the Logitech Cordless Internet Navigator Keyboard/Mouse combination
Connection: PS/2
Type: Optical
dpi: 800 (not sure)
Speed setting: 3 (third bar from the left)
Space needed for desktop: 2x2 cm (0,78x0,78 inch) for 1600x1200 pixel
Underground: plain table (!)

I'm actually running seren's test program, it finished over 920 tests sucessfully. EDIT: And I get every time that damn authentifcication error!

nah, its not the mouse…its something with the kernel...or software based error

If it's an software/kernel error which appears at mathematical problems, then Photoshop could be affected because it's using a lot of these functions. So also rendering programs must have this problem, resulting by crashes or incorrect graphics. Has anyone tried this? If there's no other idea, post the mouse configurations for comparison.

BTW: Photoshop and all other Adobe apps can't be processed by the Xslimmer, the blacklist claims various errors. Maybe this is a problem too. Has anyone tried to patch the Adobe Suite's installer packages?

Seren's test program absolved actually over 2380 tests successfully on my machine.

Here are my test results: 3471 of 3471 tests successfully, 49 to go. After this it hang in an enless loop in Selftest 7, but that issue was described in the readme file, it also could happen on real macs. Here's a screenshot (naquaada.na.fun pic.de/naq/osx86/forum/infinitemac/naquaada/pix/CPUTest01.jpg), here's the log file (naquaada.na.fun pic.de/naq/osx86/forum/infinitemac/naquaada/docs/CPUTest01.txt).

These applications were running in background: Activity Monitor, Adress Book, Cyberduck, Dockainer, Firefox, Mail, System Preferences, Transmission (paused), VLC Player.

Info: I always have the Activity Monitor running in the Dock (starting at login), showing cpu history. Maybe another thing to test.

My system was installed originally on a single core processor. A couple of months later I went and replaced it with a dual core X2 so i'm not that dissimilar from you.

The mouse is not the cause since it works fine running on one core. It's just a side effect like the other issues.

Why the posibility that is a HTT problem is discarded (i asked that before). IF you notice, at least in tiger it appears (in leopard i cant seem to find it) but it detects HTT and detects 2 cpu's with 1 core each one when its the other way arround.

whatever... my bios settings fixed that issues for me, the tiger is up and running since 10hours or more without crashes or mouse-hickups!
ok firefox crashed one time, but it wasnt a -10xxxx error
before that i was only able to use the tiger for at least 1 hour (intensive app use) after that it was unusable (mousejumps firefox adium crashes etc.) only a reboot fixed that for a small time... very anoying!

well i'll keep u guys informed

NOOOOOOOOOOOOOOOO! =(

Fire, you are using tigeR? damn

By the way, what are you using for cpus? I'm using an Opteron 185, it's a Dual Core Server CPU with 1 MB Level-2-Cache per cpu. The most Athlon 64 X2 cpus have only 512K Level-2-Cache.

I still believe it's a Photoshop / Adobe problem, all users reporting this problem are Photoshop users. Adobe installs so much shit you don't see and everything is not patched for AMD. The software isn't unproblematic, XSlimmmer won't patch it because it doesn't work without the PPC part. The speed difference between x86 and ppc mode is rather high on AMD, don't know how's it on Intel cpus. This could be another reason.

I have a 3800 X2 cpu, and i dont have any adobe software.

What have you guys for power supplies? My PSU has only 480W. Seems not very much, but the combined power (3,3+5V) is 384W! Various PSUs have a lot higer maximum watts, (550, 650 and so on) but much less combined power. My 550W RaptoXX had only 145W, and it was dead after two years using my 3700+ and 4000+.

Is anyone able to record the screen? I want to see how the mouse movement bug looks like.

[cite] pmcnano:NOOOOOOOOOOOOOOOO! =( Fire, you are using tigeR? damn

sorry i meant of course Leopard

My specs:
Motherboard: Asus A8N-E (Bios 1013)
CPU: Athlon X2 3800+ (2Ghz)
Memory: 3Gb (2x1Gb, 2x512Mb)

I've used in BIOS auto-overclocking profile which sad - '10% overclock'.
So my CPU was 2200Mhz
BUS Speed 220Mhz
RAM 440Mhz
and HT Link 1100 Mhz

When i installed Leo4All v3, i started experiencing mentioned above problems, so i switched to one core and updated my system to 10.5.4 and kernel 9.40 without problems.
Then i've manually overclocked my system, so i've set
Bus Speed to 240 Mhz
Multiplier to X10
HT Link to X4 (960 Mhz)
and RAM running on a divider 240/12

Then i removed 'cpus=1' flag, and system was up for abt 13 Hours, Transmission Downloading, Photoshop, stress tests and so on. I haven't noticed any problems. So i think for me it was HT Link: if it is above 1000, then system becomes unstable.

P.S. In 'Hardware Overview' next to 'Total number of cores' it shows 1? But in 'Activity monitor' i've two cpu graphs?
Is the second core recognized?

It is recognized, System profiler doesn't show the infos correct. My system is also overclocked, from 2.6 to 2.86 Ghz.

[cite] taikn: P.S. In 'Hardware Overview' next to 'Total number of cores' it shows 1? But in 'Activity monitor' i've two cpu graphs? Is the second core recognized?

thats normal, if the activity monitor shows two cpu's, the system uses both cores

so after 20hrs hardcore test, leopard runs without any of the anoying issues!
no mousejumps no crashes!
try to raise your memory CAS settings to 6 or higher, if i lower them back to normal (e.g. CAS 5) i got back the mousejumps and crashes!

I'm not sure how to check the HT speed, but I've set the mulitiplier to 5x, so with my bus frequency of 220 MHz it should be 1100 MHz. I'll see what happens.

Is anyone able to record the movement bugs?

Well i can tell you, when i start posting here i had less errors with both cores and i had it overclocked.

I have a 3800+ X2 2ghz OC @ 2.6ghz with a 333mhz ram @ 420mhz.

now that i took off the overclock, i have my problems with dual core, so yes, overclocking reduces our problems.

I could try increasing memory timing but i dont know anything about it to be honest.

You have 333 MHz RAM? It is labeld 333 or does your board not like it? I had on my Asus boards various problems with different RAM types, I have always bundles with 3200 speed, but they only work in correct combinations. Are you using Dual-Core or Single-Core RAM?

I guess you refer to dual channel, yea im using 2x2gb dual channel ddr 333mhz ECC memory.

With the overclock on my system the ram is setup to 420mhz, and i dont have that much issues compared with non overclocked rig.

Hm, I can't overclock my RAM, computer won't start then. But my RAM is 400 MHz without overclocking.

if you want to OC RAM sometimes its possible to higher your ram voltage!
some mobos have special settings for that

Yes, I have settings for that. But I don't try it, my system runs fine like it is now. Always say: NTARS

naquaada, had you tried decreasing the HT?...if i have it in 5x it wont start! 4x or lower works fine.

Anyway, atm im trying to update tiger to 10.4.11 ! IF i success ill try and see how it works there

I have overclocked my computer to 2,86, so the bus freuency is 220 MHz * 13 for the multiplicator. HT frequency is 1100 MHz, 220 MHz (bus frequency) x5 (manual setting). RAM is not overclocked and the cpu core voltage is reduced to about 1.325 Volts.

With these settings my system is absolutely stable, but i never had any problems yet.

OH , ok so you underclock memory and overclock cpu, something like this :S ? :

underclock ram to 360mhz

overclock cpu to 2.6 and the ram is ""overclocked"" to 400mhz...

yes? lol...anyway...im still trying to install 10.4.11 to amd, its amazing that theres anything about it!

no, RAM is standard @ 400 MHz

why do you try to install Tiger, does Leopard not work?

Well, we have the mouse bug, and the -10080 errors or w/e. But i want to at least have the latest tiger version working.

With 10.4.9 i have no issues with mouse, not even when stressing the hard drive or w/e.

Success! 10.4.11 runs fine…and i dont have the dual core problems (ATM) Ill let it powered and see how it goes. Only the system profiler crash.

Some observations after I removed my Kingston DDR400 kit and replaced it with a OCZ one.
Earlier I was running a OCZ kit (2 x 512mb) and the Kingston (2 x 512mb) in dual channel.
Now I'm running four OCZ modules in dual channel (2 x 512mb and 2 x 1gb) and set them at CL3 (CL2 is default).

Didn't use any flags at boot. System is running fine after 28 hours. No Mouse bug, no 10080 errors so far. Will report back once if I get them later (hopefully not).

Interesting! I have 4x Samsung RAM in my board, two bundles with 2x1 GB. I also want to remember: Check your boards if the RAM modules are placed in the correct sockets. Today's mainboards are very sensible in that. I have these bundles with two modules each:

2 GB Kingston HyperXX 2GB 400 MHz
2 GB Corsair XMS 4000 500 MHz (running @ 400 MHz)
1 GB GEIL GLX1GB4400DC 550 Mz (running @ 400 MHz)
2 GB Samsung 2 GB 400 MHz (bought in 2005)
2 GB Samsung 2 GB 400 MHz (bought in 2008)
2 GB Samsung 2 GB 333 MHz (bought in 2008, should be 400 Mhz but only works in 333)
1 GB No-Name RAM 400 MHz

I can only combine the GEIL and the Corsair RAM modules and the two Samsung 400 MHz modules, no other combinations work. Has anyone of the guys who have the mouse problem tried to use Single-Channel instead of Dual-Channel RAM?

I've had this for a while and couldn't find any info before. I'm happy to know that it's not an exotic problem on my machine, though.

I was noticing two errors I get in the console quite often:
> mDNSResponder[16]: mDNSPlatformRawTime: last_mach_absolute_time 409FDBDC0000333D
> mDNSResponder[16]: mDNSPlatformRawTime: this_mach_absolute_time 40987CA10000333D

Checking the source I gathered that somehow the timers of the cores get out of sync and in whichever core the mDNSResponder thread is executing it gets a different time. In some cases this seems to crash the gettimeofday function which probably relies on those timers. Sometimes the crash only affects applications that can easily be restarted but sometimes a system process crashes which either gets restarted (resulting in the lag - i usually notice crashrepoter hogging cpu when that happens) or result in errors like the -10810 one.

In my case it doesn't happen that often but rather regularly, about once a day. I don't think Photoshop is a trigger for it in my case. In "light" cases I notice that Firefox and Transmission are the most affected and sometimes I get "bad" cases when almost all apps begin to crash. Usually, if I just let it sit for a while the situation calms down again.

I tried to tweak my memory timings with no success.

Unfortunately, the mouse bug has returned - though this one took much longer than usual (some 32 hrs.)

cpus=1, here I come.

@ Adrian: Have you activated time syncronisation by internet?

@naquaada
I had it disabled but activated it recently. Though if I'm not mistaken that didn't change anything...

I will help out with this in any way I can (test my machine, etc). I have problems with this all the time.

I used to get a ton of restarts. Then I upgraded my intellimouse software and that went away. Now I get a ton of program crashing, especially in waves like Adrian mentioned, and I also suffer from the occasional 10810 (every two days or so). I'm NForce570.

well i can report that theres no dual core issues in tiger, no crashes no anything. The only thing i had some random kernel panics, im trying to figure out why, i think it was a lan card, i just deactive it, ill report later. Also, i cant find 10.4.11 kernel, what pisses me off mostly is that i had it! but i deleted it and now its nowhere! lol

=)

I got the first time ever this -10810 bug. The only program could have caused it must be the VLC player. I listenend to an mp3 and suddenly the sound stopped. After this the VLC player quits, but without an 'VLC Player quits unexpectedly' message. After this I couldn't start any program getting the -18010 error. I pressed Ctrl+F12 to get the reboot menu, but the screen went grey for a short time and the desktop was back again. It seemed to be like a fast logout/login procedure, without entering a password. After this all programs worked fine again, no problems.

For guys who have problems with Firefox 3: Try Flock (fl ock.com/beta/download/). The version 2 beta is based on Firefox 3, it even accepts the plugins. Firefox crashed rather often, but I don't think it's because of the Dual-Core bug. Flock seems to be faster and more stable.

Firefox contains a couple of cpuids. If you patch it with Marvins the crashes stop

Ok, i unfortunately have to join you all here…
My system : Benq p52 laptop with Turion 64X2 TL-52, 667Mhz Ram, Mobility Radeon X1600, SB Chipset.

System works perfectly for about 2-3 hours. After that problems keep showing up.
I experience no restarts or crashes though but :

- Sound is becoming more and more bugged ( flashing sound and noises )
- GUI works strange - slower acceleration, mouse movement are slowed down a little.
- Maybe...not really sure - cooling works little harder.

I guess that`s all of my problems ( fortunately ).
I have added cpus=1 flag - will se what happens now.

Edited :::

I wonder…because i`ve noticed one thing - that`s not my first Leo install, but that`s the first time i have this problem. Earlier i used to leave my pc on for almost whole day, play few hours using Fusion - and it worked like a charm i guess...the only difference was that i installed new PowerManagement.bundle then - and i did not this time cause it didnt bring my bat. indi. anyway. Could it be related in any way ? I have no idea but i`ll install it now as well.

I finally decided to get some Intel hardware.

cpus=1 doesn't cut it for me. I have to close every application to run 720p smoothly and I have a lot of 1080p content (which barely runs obviously). Don't want to go back to windows so this was the only option with no working solution for this bug.

Here's what I ordered

Arctic Cooling Freezer 7 Pro LGA775 2500RPM 45CFM
G.SKILL F2-8500CL5D-4GBPK PC2-8500 4GB 2X2GB DDR2-1066 CL5-5-5-15 240PIN Dual Channel Memory Kit
DFI LANParty Dk P35-T2RS LGA775 P35 ATX DDR2 1PCI-E16 3PCI SATA2 RAID HD Sound GBLAN Motherboard
Intel Core 2 Quad Q6600 Quad Core Processor LGA775 Kentsfield 2.40GHZ 1066FSB 8MB

There`s one of possible solutions that works on my laptop - do not install iTunes with Leopard. Use Lawless relase without it. I don`t have this problem anymore, but first time when i installed iTunes - problem came back, CPU was heating like hell etc. just after installing iTunes, i didn`t even have to run it. So - for me, maybe for someone else as well - no itunes = no problem. I hope i won`t install another program like that.

no iTunes? this is crazy
For me no solution cause I got an iPhone

But what did the trick for me was ...finally... to "idlehalt=0"

Since then I have pretty stable system, no Thunderbird problems,
Systems running now for 2 Days without any problems or mouse bugs

Lawless 10.5.4 / Stage XNU / Disabed QnC / and Idlehalt=0

Sly55 can You explain what idlehalt=0 does ?

Anyway, iTunes 8 is working for me at last, no heating after installing it and newest QuickTime, but i grabbed it from this forum ( someone posted pre-patched version ) so - maybe asu does`nt work very well on my os ?

idlehalt=0 prevents your cpu from stopping when it is idle. This prevents the timers from drifting from each other.

I'm working now with idlehalt=0 and have no problems so far. But I still get mach_time messages in the logs. But the timestamps do not differ very much.

Important for everyone

Guys we have found out some important information regarding the timing issue.

The P-state and C-state changes can affect the TSC and can result in TSC drift amongst the processor cores. The drift occurs only when the operating system uses TSC as the time keeping source. TSC drift can occur on K8 AMD multi-processor platforms and single-processor dual-core platforms as they do not provide frequency independent…

http://developer.amd.com/documentation/articles/Pages/1… ↗

Mercurysquad is now aware of this and is aiming to incorporate the Linux fix into the voodoo kernel.

I have a question for everyone suffering from this 'bug'. Is your motherboard based on the K8 chipset? I know mine is...

This is now fixed as of Voodoo beta 2

One question though: would people prefer to have it as a kext or in-kernel as it is now?

Am I reading right? The 'Mouse movement bug' has finally been put to an end? Wonderful!

Thank you XNU Dev Labs.

Yes, you read it right. I've been testing a pre beta 2 kernel with the relevant fix and I have to say that my system has never been so stable. No mDNSResponder errors, no divide by zero, no -10810 errors -- nothing!

Good job XNU-Dev!

it`s not yet released right? is it possible to get this beta ? and what is the release date cause 8th was delayed

According to mercurysquad the official beta 2 should be released before Monday, "hopefully".

Does the 9.5 kernel work with 10.5.3 too? I don't have the mouse mevement bug, but sometimes dividion by zero and maybe two times the -10810 error since I own my dual-core cpu which is since May. Which packages do I need?

i think it may work, at least it works with 10.5.4. anyway i wonder why some bugs appear on some cpus but do not show up on others...i`ve never seen division by zero problem or any errors. i`ve got only mouse-bug and sound stuttering

The 'mouse bug' and audio stutter are both symptoms of TSC drift which has now been resolved.

Not to throw this into the wrong discussion but with the voodoo_beta kernel, my original TSC delta was -235690 and the new delta with CPU1 is 36.

Is there any way to narrow the gap down closer to 0?

@mercurysquadThis is now fixed as of Voodoo beta 2 One question though: would people prefer to have it as a kext or in-kernel as it is now?

This is fantastic news, huge thanks to the XNU team for solving this at last! :

Is there any benefit to having this fix as a kext? Would it not be simpler if it stays in-kernel?

R0GUE just identified a bug where sleep/resume resets the workaround. So we'll probably package this up as a .kext which AMD users can dump in their Extensions folder and it will take care of everything - rather than modifying the kernel's sleep/resume functions. People who use other kernels can use the kext too.

This also means beta 2 will be further delayed. Please be patient

not workin 4 me

I have been using the VooDoo_beta2 for a while and it has not fixed it for me. I still get mDNS responder errors too. I am running a 3.5ghz C2D with 4gb of 750mhz ddr2 on an ASUS P5N-t Deluxe mobo. I have also not yet tried the HT freq. x3 yet but will now. just letting you know that the kernel does not seem to be a total fix for everyone. cheers all~

This fix incorporated into the kernel is only aimed at K8 socket AMD owners. If you are experiencing problems you should fill out the beta test report and post it on Google Code.

ah sorry bouut that, is it known to happen to lga775 mobos?

It seems that installing Paragon NTFS gives me the mouse bug. Using the cpus=1' kernel flag didn't fix it for me.
But I logged into Windows and using Macdrive deleted the Application from Applications. It still gives me read and write of NTFS but fixed the problem.

I had a similar problem on my machine

After using standby (accidentally as I had left the PC for a while and had forgot to change power management settings) the whole OS was very laggy, System monitor showed high CPU usage.

I think the OS crashed when trying to resume from standby and after a reboot then I had the high CPU usage problem. No matter how many times I rebooted, nothing fixed it.

I only have a single core Athlon 64 though. Also, booting with -x option would make it run smooth but video drivers didn't load so couldn't run anything fancy.

Never worked it out and eventually reinstalled

I can't seem to find a fix for my mouse bug.
Booting cpus=1 doesn't do anything.
It seems to happen randomly when I boot up. The screen goes black and the monitor says no signal, then it goes light blue, then the desktop comes up very slowly and then I can't move my mouse. If I restart I'll get the same thing.
Tried booting with the fsb= boot argument but no go.
I've even tried iDeneb 1.4 same thing.

I need help!!!

Well im new to the hackintosh world so please if anyone can help me do.
i still havent installed the program. well what im trying to say is i havent even picked out one ie. (IDeneb, Kalyway, or a iPC) so if anyone could also help me with this that be great, so my rig is as follow,

M3A79-T Deluxe
Phenom x4 9550
XFX GeForce 9800 GT - 512MB GDDR3
Seagate Barracuda 7200.10 250GB Hard Drive

my optical drive is a cd-rom but do really need a dvd-rom
note* the mobo is a crossfire x ready chipset

Max .C.

@BiosquirrelWell im new to the hackintosh world so please if anyone can help me do. i still havent installed the program. well what im trying to say is i havent even picked out one ie. (IDeneb, Kalyway, or a iPC) so if anyone could also help me with this that be great, so my rig is as follow, M3A79-T Deluxe Phenom x4 9550 XFX GeForce 9800 GT -…

This isn't exactly the thread to post that.

I've had the most luck with ideneb. Make sure you get the most resent iDeneb 1.4.
You need to search the forums for the right kexts (drivers) for your hardware. You can also check out the insanelymac.com wiki and forums.

Yes you need a dvd drive os x to to big to be on cd. Unless you but it on a USB key and boot from that.