Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
ROM Revision: 113-B2250F-212
EFI Driver Version: 01.00.212

For me and my 2.8 Extreme. I have experienced freeze ups from cold boot and random black screens as well as wake up problems. I am using Leopard.
 
I agree. Both my iMacs are 212 and one freezes, the other doesn't.

OK, this is getting confusing. So many people claim it's a software issue, but since both your iMacs are running the same software doesn't it imply that it's a hardware fault? Have you tried returning the one that freezes for a new one?
 
OK, this is getting confusing. So many people claim it's a software issue, but since both your iMacs are running the same software doesn't it imply that it's a hardware fault? Have you tried returning the one that freezes for a new one?

I might do that next week. I am giving Apple the benefit of the doubt when they said a fix was in the works for the end of the month.
 
I might do that next week. I am giving Apple the benefit of the doubt when they said a fix was in the works for the end of the month.

I think I can trust Apple to keep its words about the fix at the end of the month. If it doesn't arrive I really don't know what I'm going to do. I desperately need a new computer but there aren't any other Macs that fits my needs.

If it's a hardware issue can it really be fixed merely by a software update?
 
I think I can trust Apple to keep its words about the fix at the end of the month. If it doesn't arrive I really don't know what I'm going to do. I desperately need a new computer but there aren't any other Macs that fits my needs.

If it's a hardware issue can it really be fixed merely by a software update?

We shall see. Some hardware problems can be "worked around" with software.
 
Here is mine, and I have not had any freezing issues.

ATI Radeon HD 2600 Pro:
Chipset Model: ATI,RadeonHD2600
Type: Display
Bus: PCIe
PCIe Lane Width: x16
VRAM (Total): 256 MB
Vendor: ATI (0x1002)
Device ID: 0x9583
Revision ID: 0x0000
ROM Revision: 113-B2250F-212
EFI Driver Version: 01.00.212
 
In Tiger I froze only while playing WoW and now in Leopard I just random freeze everywhere. It froze literaly 1 minute into a clean install of Leopard, as it was freezing like hell when I updgraded from Tiger. WoW now in Leopard is super unstable I cannot even think about full screen as this will complely freeze it for sure. Anything video intesive just seems to go straight to hell. Photobooth will most certainly freeze it and completly crash the graphics and I know its only the graphics because I will be on Skype with friends and the screen will go black but I can still hear them and they can hear me. I hope that they come with a fix today for this it's really starting to piss me off bad.
 
OSX not only requires a fast video chipset and memory, but also one that can process complex compositing instructions in parallel pipelines using various APIs. OSX is just as demanding as some advanced videogames in terms of compositing.
If any part of the compositing hardware is below spec or faulty, you get random problems with display output.
In a very real sense, you're better off with a lowend videcard that makes no attempt to render CoreImage/CoreVideo/CoreAnimation, than one that "can do it" but crashes in the process.

Most of the reports seem to be associated with iMacs that have the 2600 Pro chipset (and 24" display), which is in all likelihood, exactly the same chipset as used in the 2400, except that it is running at higher frequencies and has more pipelines/paths etc... enabled.

ATi might be able to "make" 2400 chipsets from 2600 chipsets, by simply flashing the ROM with 2400 settings, using samples that either cannot run at 2600 specs, or alternatively by flashing-down "good" 2600s in order to meet a current demand for the 2400.
They cannot however, reliably "make" 2600s out of 2400s.

This is what I think might be happening with some iMacs.

Just a guess based on anecdotal reports from iMac owners, and my own experience with failures due to overclocking and flashing videocards in PC gaming machines.
The symptoms are virtually the same for both scenarios.

One thing I would try is to reduce the desktop resolution to one of the lower settings to reduce the load on the GPU; if the lockups and display corruption are reduced or eliminated, I would suspect that the GPU is simply working beyond its capabilities at the higher native resolution.
Reducing the clock frequencies and disabling pipelines via third party utility (if available for OSX) would be another possibility that would lead to a similar conclusion.
 
One thing I would try is to reduce the desktop resolution to one of the lower settings to reduce the load on the GPU; if the lockups and display corruption are reduced or eliminated, I would suspect that the GPU is simply working beyond its capabilities at the higher native resolution.
Reducing the clock frequencies and disabling pipelines via third party utility (if available for OSX) would be another possibility that would lead to a similar conclusion.

So did this method work for you? Or are you one of those people who didn't get freezing iMacs?
 
One thing I would try is to reduce the desktop resolution to one of the lower settings to reduce the load on the GPU; if the lockups and display corruption are reduced or eliminated, I would suspect that the GPU is simply working beyond its capabilities at the higher native resolution.
Reducing the clock frequencies and disabling pipelines via third party utility (if available for OSX) would be another possibility that would lead to a similar conclusion.

If that were the case, things would not lock up literally ten seconds after a cold boot which they do.
 
If that were the case, things would not lock up literally ten seconds after a cold boot which they do.

I've tested all the iMacs at our local Apple Store and some of them freeze when doing something graphically intensive (scrolling on the Dock etc), but I have yet to see one that locks up from doing nothing.
 
If that were the case, things would not lock up literally ten seconds after a cold boot which they do.

My assertion does not assume that the "freezing" is caused by overheating, which is what you are implying here.
ie: "If the temps are normal (10 seconds after boot up), the GPU should not crash"
 
My assertion does not assume that the "freezing" is caused by overheating, which is what you are implying here.
ie: "If the temps are normal (10 seconds after boot up), the GPU should not crash"

I can't see anything that would tax the GPU in any way on a cold boot. No apps running. No animations or other graphical eye-candy. I do think its a GPU bug but I don't think its due to any stress or limitation of the GPU working beyond its capabilities.
 
The problem iMac I've seen first-hand, just seemed unstable.
Almost like it was dramatically overclocked and always on the verge of crashing.

It's not at all like the typical driver issues I've seen that can be replicated, such as launching a specific game or application always causing the same problems.
 
The problem iMac I've seen first-hand, just seemed unstable.
Almost like it was dramatically overclocked and always on the verge of crashing.

It's not at all like the typical driver issues I've seen that can be replicated, such as launching a specific game or application always causing the same problems.

Mine is similar but the pattern is incredibly random. Warcraft can be going at full settings top res for 3 hours with no crashes but it can die on opening up Mail or even just idling.
 
I can't see anything that would tax the GPU in any way on a cold boot. No apps running. No animations or other graphical eye-candy. I do think its a GPU bug but I don't think its due to any stress or limitation of the GPU working beyond its capabilities.

Simply drawing the composited OSX desktop constitutes a faulty video chipset deal-breaker.
ie: the chipset "chooses" to draw CoreImage, CoreAnimation etc..., even if the associated registers and pipelines are faulty.
If these instructions are executed incorrectly, what is likely to happen?
 
Simply drawing the composited OSX desktop constitutes a faulty video chipset deal-breaker.
ie: the chipset "chooses" to draw CoreImage, CoreAnimation etc..., even if the associated registers and pipelines are faulty.
If these instructions are executed incorrectly, what is likely to happen?

I see you point. Given that the crashing occurs almost exclusively in OSX, the finger would point to the OSX drivers rather than the card firmware.
 
I will be pleasantly surprised and happy for those affected, if Apple releases a display driver update that transforms these iMacs into the reliable workhorses they should be.
This is assuming the driver "update" doesn't simply cripple every iMac it's installed on.
 
I will be pleasantly surprised and happy for those affected, if Apple releases a display driver update that transforms these iMacs into the reliable workhorses they should be.
This is assuming the driver "update" doesn't simply cripple every iMac it's installed on.

If that's the case, and I have a non-freezing iMac to compare it to, its going straight back to Apple.
 
The biggest problem with the "software driver is the problem not the hardware" theory, is that identical machines are functioning perfectly with the "faulty display driver".

If it were the OSX display driver software, the problem would almost certainly affect all iMacs.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.