Q: 2011 MacBook Pro and Discrete Graphics Card

Richard Liu wrote:

 

What is the big deal with gMux version 1.9.24?  My hardware is original and that is the reported gMux version.

I don't recall claiming that it was a big deal.  LA_slacker posted post repair system info for his GPU and asked whether anybody whose machine still had the original motherboard in it could spot any differences.  I merely offered the information that most, but not all my post repair system info was the same as his and left it up to him to decide whether it was a "big deal" or not.  He apparently didn't think it was, and I certainly didn't intend it to be.

Thanks...

I have not been following this this thread closely so wondered if I missed something important.

Judging by your last sentence,

Hope I was clear enough this time.

I think you were replying to me.  Yes, thank you.

Regarding point 5:  Of course, people rarely open up their machines just to check whether the manufacturing quality lives up to their high standards before beginning to use it, so I doubt that many will have reapplied the thermal paste as a preventive measure, especially since removing the heat sink just to see how the thermal paste was applied commits one to reapplying new thermal paste.

After reading about thermal paste in this thread, I have the impression that it's not the way Apple is applying the thermal paste that is the problem.  For if, as you say, the Apple technicians are applying "too much" thermal paste according to some Apple guideline (e.g., "better too much than too little"), and if that causes problems -- seeping, according to one poster in the thread, does not always cause problems --, I would expect practically every Apple product that contains a CPU and a heat skink to be affected.

Did you happen to notice what kind of thermal paste Apple is using?  What kind did you use?

<Link Edited By Host>

Apple replaced my logic board 3 times because of that failed GPU, and finally they just ended up giving me a brand new 15 inch MacBook Pro retina!  Thank you Apple!

I notice that you reported that on April 21.  How many times was it repaired under the repair extension program?

Twice under the repair program, and on the 3rd time I took it in the manager hooked me up.

So, once before, twice under?  Regardless of the exact number of times, I think the willingness to exchange "irreparable" (in this sense) machines for new ones is encouraging and should convince those who hesitate for any reason that they have nothing to lose by having their afflicted machines repaired.

Although I am convinced that Apple is sincere about solving this problem, I am equally convinced that they've handled the whole thing very poorly.  Compare how Steve Jobs confronted the so-called Antenna-Gate Affair head-on with an information campaign, and how four years after some of our machines were produced we still have no explanation of the causes of the video issues from Apple.

Well, the thread you posted is quite controversy and at some points I must agree.

Speaking in heat transfer, if you can make contact with heat sink and chip diode without any imperfections, I would say not to use thermal compound at all, only thing I would be concerned about that is a corrosion between two different metals, that I am not familiar to.

So, as I understand this whole debate about thermal compound, the idea of using it is to fill the air gap between heat sink and chip diode.

Speaking in terms of too much or too little I would say that you need to use right amount, and that is not hard to achieve.

I don't know what kind of thermal compound apple uses, but some technicians are claiming that is one of cheapest on the market, that gets dry very quick and that means that is not transferring heat so good any more. And if you combine it with a fact that there is too  much of it and it is spread around chip diode it actually isolates heat inside the chip and not transferring it.

I use Keratherm KP 12 and tend to reapply it once per year.

if you can make contact with heat sink and chip diode without any imperfections, I would say not to use thermal compound at all, only thing I would be concerned about that is a corrosion between two different metals

Perfect contact would be too expensive, if not impossible, to achieve.  If the heat spreader of the chip and the heat sink were in perfect contact with one another, there would be no air in between, and hence no oxidation (corrosion).

in terms of too much or too little I would say that you need to use right amount, and that is not hard to achieve

I think the problem is the definition of "right amount".  Ideally, you would want the thermal paste to cover the heat spreader of the chip completely from edge to edge, thereby ensuring perfect thermal conduction between it and the heat sink.  Because the paste spreads when the heat sink is pressed onto the chip, the question is, is thermal paste seeping over the edge of the heat spreader a bad thing?  If it is not, then the right amount is a thin layer that just covers the heat spreader before the heat sink is applied.  If it is, the right amount is more difficult to calculate, being the amount that it will spread exactly to the edge of the heat spreader but no further when the heat sink is pressed down on it.  From the discussion in the thread I conclude that, for those who have some experience with thermal paste, a bit of "seeping" is OK, perhaps even preferable to none at all, but that messing with thermal paste isn't something for the inexperienced.

Over and above that, though, I fail to see the need to redo what Apple did.  Although some postings in this thread relate that resoldering or replacing the GPU fixed the video problems, I do not recall any saying that solely reapplying less or another kind of thermal paste solved any.  So any benefit of preventively reapplying thermal paste is, for the time being, theoretical and must be balanced against the risk of causing a problem by applying a suboptimal amount or brand.  At any rate, I assume that it would immediately invalidate any warranty, even the 90-day warranty of repairs performed under the repair extension program.

Just to inform that I went for a Logic Board replacement on saturday. They found that it was exactly the trouble that is under the repair program and it took 3 hours. At least, didn't made a 400km trip for nothing.

I had a good service.

They told me they were replacing the LB with a fresh new model that shouldn't bust.

Thanks for your help!

I posted previously that I am having a bad lag due to bad blocks on the HDD, today came the SSD, installed and boom everything is now okey and hoping that my replaced logic board will last long.

Hi I'm new in here but have read a lot of the posts. I got some questions.

I've had this computer since May 2011 (Early MacBook Pro 15-inch) I've had many of the same problems as described here, and it went to that point where I couldn't boot the computer. Last month the logic board was replaced with a new. However efter testing the graphics with Unigine's Heaven I get these results:

Is the computer simply not capable of handle the test or is the gpu chip breaking again?

Should I go to the AASP again and get it tested?

(I work with applications for graphics but haven't noticed any glitches while using them like I did before the replacement, but maybe it will show up again?)

- Joachim

A couple of months ago my Late 2011 15" MBP started behaving strangely, exhibiting the symptoms described in this thread and, last week died completely. Having identified it as a GPU issue, searched around and discovered this was a known issue and found out about the replacement programme, contacted Apple who checked serial no etc and booked me in to the Liverpool Apple Service Centre for today. The machine failed its tests and the Genius advised 5-7days for repair as the part is a new design logic board for this model and is on special order. Surprised they hadn't got one ready for me since I'd already waited a week! The Genius phoned round and suggested I could go instead to Manchester where they had 3 boards! and made an appointment for this afternoon. The Genius at Manchester seemed to know about the case and immediately booked in for replapcement logic board but again advised 5-7 days. When I explained I'd driven all the way to Liverpool then to Manchester as they had the parts, he said they'd try and do it in 2-3 days. Both guys were very helpful but does it normally take this long? i thought they would do it whilst I waited!

Unigine's Heaven is not a diagnostic program; it simply places extraordinary demands on the GPU.

Should I go to the AASP again and get it tested?

I think that is an excellent idea.  After all, the new motherboard is still under warranty, and the diagnostic programs that will be run are supposed to test, not tax, the GPU.  Be sure to take along the photos that you posted, just in case your machine is too shy to fail the diagnostic tests.

Please keep us advised.  Good luck!

Like you, I began to experience the video problems just around the time that the repair extension program was announced, and Dr. Google ranked this thread near the top of his (her?) list of retrieved documents when I began gathering information about the symptoms.  We are fortunate.  The first entry in this thread is from February 2013, and I am convinced that other reports of these video issues predate it.  So some people have been -- excuse my language -- urinating against the wind for a long time before Apple finally did something about it that seems, at least from feedback about the program in this thread, moderately effective.  Five to seven days is nothing compared to the years that other have waited.

Please post back here after you get your repaired machine back.

when I took mine, they ordered the part and had it installed in 2 days. they originally estimated 3-4 but finished up quickly. i would guess each shop just has their own workload, estimates, etc. Perhaps the places you've gone are simply very busy? or maybe they're like Mr. Scott on Star Trek. estimate double and appear as a miracle worker when you're done in half the time