Screen turns black when gaming?

You are 100% correct. nVidia provided bad 9600M GT cards to Apple, but Apple should have addressed this issue during their Q&A tests! The worst thing of all: this BSOD issue with the 9600M GT was first addressed TWO months ago, and still no fix or even an official word from Apple themselves!

I strongly suggest everyone who has this BSOD problem to return their MBP to Apple until they get a 9600M GT card without this issue! This is unacceptable, and it's becoming more and more clear that a firmware fix can simply not 100% fix this issue!

The only thing a firmware fix can do is
1. underclock the 9600M GT card. In other words: force lower performance than it should provide.. This is clearly unacceptable and would be a shame!
2. adjust fan control policy, which is only postpones the problem

Don't forget there are users with machines who don't have this BSOD issue at all (read this topic)! Their GPU never hits 90 degrees or higher, and they have the same fan control policy!

Also, even when I force the fans to run at max speed (6000 RPM) I still experience the BSOD. The only difference is that I can play my games most of the time for more than 1 hour before I get a BSOD. The issue is only postponed with this workaround!

I'm 100% sure this is a hardware problem with the 9600M GT card. And I have only 1 suggestion to people who experience this problem:

_*RETURN YOUR MACBOOK PRO IF YOU HAVE THIS BLACK SCREEN OF DEATH PROBLEM!
DON'T WAIT FOR A SOFTWARE UPDATE, BECAUSE THIS IS A HARDWARE PROBLEM! YOUR GRAPHICAL CARD IS FLAWED!*_

Read the investigations by The Inquirer:
http://www.theinquirer.net/inquirer/news/052/1050052/nvidia-chips-show-underfill -problems
http://www.theinquirer.net/inquirer/news/921/1049921/inquirer-confirms-apple-mac book-pros-have-nvidia-bad-bump-material

Wow, if this new update is true, and nvidia completely omitted the layer, then I guess we're up @#$# creek. I'll at least test my replacement (if the repair is ever complete.. This is taking a while...). May need to do a full return.

what's the point of replacing a computer that fails with another one that fails too?
unless they're doing something highly secret when they change the chipset, which sounds pretty dishonest to say the least…

2 reasons

1) Its possible since mine was from the very first production line (week 40 October) that NVIDIA was well aware of the issue and rushed out crap chips. Its possible they started shipping non-defective versions in the last 2 months or so.

2) Establishing a history of repairs on the model. When apple customer relations sees you've had your machine in the shop multiple times, they may take the matter more seriously or help you in getting a full refund.

Again a very interesting article by the Inquirer.

But again I do not feel comfortable with it. Sure C. said in previous articles that a mix of factors and wrong decisions lead to poor chip design on NVidia's side. But he did always have one key argument why the chips cause massive failure and why potentially all NVidia 9600M GT chips (in earlier articles even the 9400M was mentioned as potentially affected by poor material choices) are defective. In pretty much every article the tone and the key argument changed.

In the last article he claimed that it was the "bad bump" material.
Now he claims it is the underfill or the lack of a critical component which would normally lead to the chips stability. (again it is mentioned that it is a combination but the key argument is as I pointed out the underfill in the most recent article).

To quote:
" ... any strain on it will be transferred into the layers of the chip itself fairly directly. If there is too much strain, the layers of the chip peel apart and you have what is called catastrophic inter-layer delamination, and that kills the chip even deader than cracked bumps."

Again I am basically supporting the research of the INQ but it cannot be entirely appropriate to project the findings on our problem discussed in this thread.

Who actually killed his chip? If the "bad bumps" crack (as it was the alleged key problem in the previous article) the chip should be unusable.

If the underfill is not sufficient to hold up the stress then as C. wrote the layers of the chip peel apart.

I don't think this is happening with our MacBook Pro's ((yet) - yes maybe the INQ is right and in the future a high number of MacBook Pro's will die or at least will be unable to put out video signals because the chip died)

But what most people discussed here seems like a mixture of a temperature problem and something else - be it another chip issue (serial flaw or simply coincidence) or software / firmware issues.

When a BlSOD happened and after a hard reset things work (at least for 5mins in a Game) this can not be taken as evidence that the physical structure of the chip - be it bumps or underfill - has been permanently altered.

Also: If one fully believes the INQ then there is no solution. Exchanged MacBook Pro's (or new logic boards) cannot solve the problems the INQ outlines.

As I said I believe that it is a mix of temperature issues (80°C is a lot and I have also experienced temps of 90°C and up) and something else. Maybe it could still be the firmware? Or we would have to dissect one of the "good" replacement Macs and one of the "bad" ones. Yes and even put the chips under the INQ's electron microscope to see what finally could be the hardware difference between the good and bad models.

You raise a very good point.

I wonder if its always permanent damage though. Take for example layer separation. We're talking the µm range here. Assume the chip hits 90C+ and begins to stretch due to heat. All it really needs is to shift even the slightest bit to cause corruption. Once the problem causes the crash the chip will begin to cool, particularly in a reboot which will allow the chip to settle down and maybe reseat the layers properly. He does mention the layer missing allows for safer thermal stretching.

May not cause immediate permanent damage to the chip, but repeat enough times and we may have an issue.

I do agree with you though. Since we've yet to see completely destroyed chips, the article should not be taken as fact.

On the other hand, at least SOMEONE is being vocal about the problem. Unlike Apple who is not telling us anything besides "replace the logic board".

Let me ask you something else: what is the point of paying 500 - 1000 EUR extra for the 9600M GT if it doesn't work properly?

I suggest everyone to return their Macbook Pro's until you get a working one. Apple is supposed to be a 'quality product'. If your brand new macbook pro doesn't work as it should be from day 1, you should return it. When everyone with this problem returns their machine, this problem will also get a higher priority.

Just another update. I last reported that I got my replacement machine (after being contacted by an Apple representative who read my original post) and was able to game fine that evening. The next day the firmware update was released and I experienced the BSoD yet again. Since then the OS X 10.5.6 update has been released, yet I continue to get the BSoD here and there. I was actually planning to post an update earlier today, but decided I would wait to see where this goes in the coming weeks...then I started playing WoW and got the BSoD after playing for just a few minutes.

This is my first Mac. Apart from the BSoD issue, I am very happy. I have always heard great things about Apple and their products, and I'm happy with my iPod Nano and iPhone, but I am quickly losing confidence in the brand. Regardless of the cause of the problem, this all comes down to the fact that Apple is culpable and needs to publicly address the issue and start making this right. If the cause is faulty NVIDIA chips, then get it fixed. If the new unibody design somehow doesn't allow for appropriate cooling, start fixing it. I did not spend this much money to test a laptop for Apple. I expect this laptop to perform as advertised. I'm curious to know if anybody has contacted news agencies regarding the issue to see if they're interested in going public with the story? If the fact that Steve Jobs will not be attending the next Macworld Expo is news worthy, then surely this must be also?

Since this latest BSoD occurred after the update to OS X 10.5.6, I have the details that were included in the error report:

Wed Dec 17 13:18:15 2008
panic(cpu 1 caller 0x00405DEB): "NVRM: Read Error: GPU 0, PCI 0x00000200, BAR0 0xe4000000 0x60d1f000"@/SourceCache/xnu/xnu-1228.9.59/iokit/Kernel/IOLib.cpp:724
Backtrace (CPU 1), Frame : Return Address (4 potential args on stack)
0x576bfc38 : 0x12b4f3 (0x45b13c 0x576bfc6c 0x1335e4 0x0)
0x576bfc88 : 0x405deb (0x49da40 0xded280 0x576bfccc 0x576bfcbc)
0x576bfca8 : 0xbbe4f7 (0xded280 0xded280 0xdb1c88 0x0)
0x576bfce8 : 0xe75d1a (0x7a62404 0x7a66004 0x61002c 0xbfab13)
0x576bfd28 : 0xe8979a (0x7a66004 0x61002c 0x576bfd68 0xbfa9af)
0x576bfd48 : 0xcf0881 (0x7a66004 0x7a74804 0x18c3000 0x7a6f804)
0x576bfd68 : 0xd2305d (0x7a66004 0x7a74804 0x1 0x0)
0x576bfec8 : 0xca217e (0x7a66004 0x0 0x576bff18 0x576bff14)
0x576bff38 : 0xbbf0a5 (0x7a62404 0x42c6e004 0x0 0x0)
0x576bff58 : 0xbaa1f0 (0x42c6e004 0x258f96a0 0x576bff78 0x1a336f)
0x576bff78 : 0x13eed2 (0x7526000 0x42c6e004 0x1a336f 0x872f3c8)
0x576bffc8 : 0x1a017c (0x0 0x0 0x1a30b5 0x8cd28b8)
Backtrace terminated-invalid frame pointer 0
Kernel loadable modules in backtrace (with dependencies):
com.apple.nvidia.nv50hal(5.3.6)@0xe02000->0x1056fff
dependency: com.apple.NVDAResman(5.3.6)@0xba5000
com.apple.NVDAResman(5.3.6)@0xba5000->0xe01fff
dependency: com.apple.iokit.IONDRVSupport(1.7.1)@0xb97000
dependency: com.apple.iokit.IOPCIFamily(2.5)@0x5f3000
dependency: com.apple.iokit.IOGraphicsFamily(1.7.1)@0x777000

BSD process name corresponding to current thread: kernel_task

Mac OS version:
9G55

Kernel version:
Darwin Kernel Version 9.6.0: Mon Nov 24 17:37:00 PST 2008; root:xnu-1228.9.59~1/RELEASE_I386
System model name: MacBookPro5,1 (Mac-F42D86C8)

Just to post specs on my current fried MBP, in case it'll help with diagnosis:

Model Name: MacBook Pro
Model Identifier: MacBookPro5,1
Processor Name: Intel Core 2 Duo
Processor Speed: 2.53 GHz
Number Of Processors: 1
Total Number Of Cores: 2
L2 Cache: 6 MB
Memory: 4 GB
Bus Speed: 1.07 GHz
Boot ROM Version: MBP51.0074.B01
SMC Version: 1.33f8
Serial Number: W8842xxx1GN

Well, there's 52 weeks in a year so probably 48-52

Anyway, I just picked up my machine from the shop. They said they replaced the logic board and installed 10.5.6 and that should fix the problem and still claim its software. They just can't get it through their heads that 10.5.6 won't affect WINDOWS..

I'm a bit concerned though. My serial number has not changed and my Mac Address of my ethernet Jack has not either.. This makes me think they lied and actually didn't replace the logic board..

I ran a test in windows in DoD and my fans didn't increase speed at all until 100C, then they started to rev up and reduced temps to 90C. They stayed around 93C the rest of the time.. I took a screenshot of the temperature graphs for proof next time I talk to apple. 100C is unacceptable.

I also need to find out if they lied to me about the replacement.

I haven't played games long enough to confirm if the Black screen still occurs, but I'll try to give that a shot later tonight.

Played a bit longer and didn't get a BSOD. I spoke with a friend that is a genius and he mentioned that they can set a new logic board's serial with a tool so I feel better about that.

My second time playing went a bit better. Since the fans were already a bit rev'd up, it kept the machine around 85C 6000 rpm.. Things ran fine.. Will keep an eye on it and report if I have any more problems.

Message was edited by: Robert Derelanko

hi guys,

I DON'T have a MBP of the newest generation but I want to get one soon, THOUGH I think I have something interesting to contribute to this topic.

I have one of the first Macbook Airs that were shipped and as you know, they are made of the same “unibody“ enclosure like the new MBs and MBPs.
It was end of february 2008 when I first encountered problems with this notebook as a consequence of overheating. I first thought it is a hardware issue since I was not used to sth. like that from apple before. I made some research and found out that almost everyone with a MBA has those problems. Just watching a movie in iTunes let the fan turn like crazy, it gets hot that you could fry an egg on it and then suddenly turns into numbness, it takes several minutes until it's working again. That is still really annoying and also the main reason why I need a MBP (which I also had before the MBA).
So, there's a huge problem with heat and for me it seems as if it has to do with this “unibody“ thing.
Maybe I'm wrong, but like someone mentioned already here, I also think it is a problem of overheating, not the graphics. Maybe it's even both, I hope not and hope it's getting fixed soon.
It twice happened even that I had to make a total reboot forced, when this clunker didn't want to respond anymore.
Since the MBA is not made for games, I did not try to play.

Hey guys, just an update. I've managed to get my logic board replaced and so far I've had no issues of this black screen with looping audio crash while gaming in both boot camp and os x. I installed GTA 4 on boot camp and I tried to get the machine to overheat by playing with the machine on my bed. Temps reached to mid 90s(celsius) and finally, the machine just shut itself off. No looping audio, no force restart.

For the past few weeks, I've been playing Dead Space, Fallout 3, The Witcher, Crysis, GTA 4, Left 4 Dead on windows without a crash. Sure, temps did occasionally reach the low 90s but no crashes whatsoever other than that one time when I had my macbook pro on my bed while playing GTA 4 trying to get it to overheat.

IMO, I seriously think that there're a bunch of defective cards out there and I'd highly advise those of you with the problems to insist on a replacement unit or at the very very least, a replacement logic board.

honestly i don't know what to do so far… i've been able to play a lot of games without problem for a month (maybe some high peaks of temperature though), and then had a couple of BSOD on Half-Life2 EP2. also, the temp seem to never go over 90° after all the updates now, but who knows… will your brand new MBP (with piece of nvidia 9600 crap included) go BSOD again after one month of use? only time can tell… what we REALLY and DESPERATELY need is a statement from apple, with facts and solutions… otherwise this is something like a war declared on the customers and we have to do something else but be organised about it.

I've been monitoring this thread for quite a while and now I want to throw my two cents in...

I bought my Macbook Pro in October, received it right before halloween and had the black screen of death by the first week of November. I was able to play games on it for 10-20 hours completely hassle free, until one day I fired up a game and got the black screen within the first 5 minutes of playing.

After a few calls to Apple support and a forced but completely worthless trip to the Genius bar ("genius" didn't want to reproduce the problem, and it's a custom order machine so they didn't have parts to triage there) they finally sent me a box to send my laptop in for repair.

They received it 12/2 and shipped it back to me on 12/8 after changing out the logic board and battery. Keeping my fingers crossed that this would do the trick, I was able to game over the weekend another 10-20 hours without a hitch. I thought I was in the clear.

Tuesday night I got home from work and sat down to play. Within 3 minutes of starting the game up, the black screen of death returned... much to my dismay. I was running 10.5.5 at the time, so I upgraded to 10.5.6 even though I have a horrible sinking feeling that this is a rather large hardware defect. After installing 10.5.6, within another 3 minutes of starting the game up, black screen of death again.

It seems like the 9600 GPU fails over time. In my case, both failures were after 10-20 hours of flawless gaming. Now that the black screen has reappeared, every game will die within about 5 minutes of starting it up. This points to massive hardware failure.

I have another call in to the support tech I had been working with but I haven't been called back yet. I might have to resort to the 1-800 number and start the whole process over again. Even then, I have no confidence that a repair would solve this problem because either Nvidia or Apple haven't discovered what the problem even is yet. Or, they have, and they are all staying very tight lipped about it because this could lead to massive recalls or class action lawsuits.

The Inquirer has uncovered some very interesting information, but they have been met with nothing but silence as well, but at least someone out there is talking about this issue.

So thats about where I stand now, and I'm sure there are quite a few of you in my same situation. I've invested way too much money into this laptop to experience these types of failures, which seem to be from a serious engineering defect which I have a hard time believing both Nvidia and Apple missed in their QA departments.

Those of us that have purchased in California are protected by the "lemon law" or Song-Beverly Consumer Warranty Act that entitles us to extended warranty until the defect is fixed and refund after a reasonable number of attempts to correct the problem (in most cases 3 attempts). I really like the laptop, but the black screen/defective 9600 GPU hardware makes it completely unusable for me so Apple has two more strikes before I have to take more serious steps.