Q: 2011 MacBook Pro and Discrete Graphics Card

Perhaps 10 or 15% of all MBP's with replaced logic boards or "re-balling" and still working is enough for you to deem a given solution a success.

So you're an electronics engineer, eh?  As you know every industry has its hacks, also-rans, and guru's.  Besides that, the world is filled with mediocrity.  So I would have to determine which of these you are before I could consider your opinion helpful.  Not knowing anything about you personally, I'm left with the laws of probabilities.

But if you believe that 10 - 15 or even 20% success rate is enough to deem a solution a succees, then I already know more about you than I care to.

I was speaking you plural.  Not just Paigoomein.  I was using his particular story because it sounds like so many others I've read regarding logic board, soldering, GPU card causing the freezes and various corrupted screens.

To the best of my knowledge, this thread is about 1 problem only as the title clearly indicates.  That's the only problem (and supposed solution) that I'm talking about anyway.

You'd be surprised what can go wrong when soldering these boards.

Especially BGA's.

CindyBruce wrote:

You guys speak almost as if Apple assembly line workers have never soldered before the early 2011 MBP's.

I just don't get it.

You do realise its not soldere by " Apple assembly line workers" but by Chinese manufacturers?

They are not soldered manually but fully automated reflowed?

It is most likely bad soldering of the BGA.

I'm sorry, is English your first language? Just a question, don't mean to offend or disrespect you. There isn't only one solution supposed solution, there are multiple. You now combined two separate solutions, replacing the logic board and re-balling, into one success percentage... Re-balling has been way more successful than replacing the logic board; I've maybe read about one instance in which it did not work.

Yes, I understand that.  I was attempting to be brief.  The robotics are programmed by software engineers and the hoppers are filled by workers, etc.

I'm not saying the logic board assembly is superior or inferior and may indeed be an indirect contributor.  But if that were clearly the problem, then Apple, wishing not to spend any more than they have to on this fiasco, most likely would insist on superior assembled logic boards as the appropriate solution, and step up the QA to ensure this was happening.

However, with the amount of failures recurring after the replacements, I'm left with one of two possibilities:

1.  The exact same inferior materials and/or assembly was repeated for the replacement logic boards.

2.  This ain't the problem.  (though indirectly could be a contributor of sorts.)

My apparently brilliant mind has deduced that #2 is the most logical and makes the most sense.  Especially after 2+ years of this ordeal.

That's not to say it couldn't be #1, but again, the law of probabilities have to kick in somewhere and I see no reason to defy it this time around.

To the best of my knowledge, Apple unofficially suggests replacing the logic board while some of the rabbit hole chasers thinking they have an even better grasp of the situaiton have conocted another "solution" called re-balling.  Yes, two variations of a solution for 1 problem.  I think I get it.  And maybe I need to work on my engrish a bit.

If you ever have a chance to see a high-speed computer manufacturing line, you should take it.  Some parts are "pick-and-place", and others are "tape-and-reel".  Although I'm a hardware engineer, I think the software guys who program the robots to perform these operations are geniuses.  Wear hearing protection - the pneumatics sound like machine guns.

Reading through this thread, it's clear that many posters believe that reflowing doesn't work, and the only real solution is to reball.  "psspecialist" - I hope I got that correct - has been recommended by many posters, and he has verified that he uses a "lead-free" process.  Can anyone speculate why the Apple vendor's lead-free process does not work, and psspecialist's process does?

I find it very interesting that this conversation about lead vs lead-free solder is going on and at the same time while doing unrelated research I find this lecture by Dr. John McDougall:

http://www.youtube.com/watch?v=81xnvgOlHaY&feature=em-subs_digest-vrecs

In this lecture, the Dr. explains what he feels is the source of Steve Jobs cancer - working with lead solder and inhaling the fumes of the lead solder.  The Dr. predicts that Steve Job's cancer started to grow in his pancreas when Steve was as young as 24 years of age.  I found the entire lecture very interesting and a good watch, but for those of you that want to cut to the chase, this is stated at approximately the 30 minute mark.

I'm not claiming to be a Dr., an engineered, or taking a stance one way or another on any argument that has erupted on a support forum - rather just sharing what I feel to be an interesting correlated topic.

Lead free solder itself doesn't cause the issue. As an engineer you must know that there are many different kinds of lead free sold as well. From the other engineers that I've talked to about "bumpgate," a similar issue that occurs to Nvidia GPU's in the millions around 2008 is that they simply did not understand the behavior of the material as well as leaded solder when they designed these boards.

These links make for a interesting read:

http://www.theinquirer.net/inquirer/news/1004378/why-nvidia-chips-defective

http://www.theinquirer.net/inquirer/news/1013947/why-nvidia-duff-chips-shoddy-en gineering

http://www.theinquirer.net/inquirer/news/1036374/what-nvidia

These links explain what happened to the 2008 MBP's with NVidia GPU's as well.

My mbp has been working for the past three weeks without issue using gfxCardStatus v.2.2.1, smcFanControl and only the Integrated GPU (see my previous post for fix). Until such a time as a recall is announced or reball is warranted I'm now sticking with it and sod the Discrete GPU.

I can use Adobe CS, play 3d games on low settings and watch video with no problem.

Just check before putting to sleep that Energy Saver prefs are on 'Better battery life' and gfx is still on 'Intergrated Only'.

I've been following this thread for the last few months and my inbox was full, I've kept a few relevant posts for reference but now it seems the forum has descended into a petty slanging match; with immature trolling, mostly unproductive comments wasting my time and probably to Apple's delight.

Apologies to anyone with constructive and pertinent information.

Elvis has left the building.

Thank you. Do you know if Window 8.1 also only uses the discrete graphics?

Lest anyone forget, I reported the other day about my encountering and enduring what seemed to be the exact same problem, even to the point where the mbp would not boot at all (no login prompt). Even after many, many variations of reboots. I applied a fix of sorts and my early 2011 mbp is as new ever since and I did it all within Apple's policies, procedures, and/or standards.  In my case anyway, the problem was not what Apple claimed it to be, nor was the solution.  Even clicking on the "Auto Switching" GPU box in Energy Saver works every time, where previously clicking it meant instant freeze and reboot.  And no I did not download any gpux or any other software tool to disable either GPU, etc.

Remembering what I speculated on that worst of all days and knowing what I know now with a proper solution (staying entirely within Apple's guidelines), I'm quite confident where the problem is not and approximately where it is.

But I've only "fixed" it once and hopefully won't ever have to do it again.  But as I said before, even if it died tonight, that single experience convinced me where the problem is not.  In my book it's just not possible.  Could inferior solder joints be causing this or other problems?  Of course.  But since my solution had nothing to do with this specific area, and my mbp has worked beautifully (as well as or perhaps even better than ever) ever since, I see no reason to jump on the bandwagon of those who believe otherwise when they still have crippled or bricked mbp's.

When you think about it, all it takes is for one very senior well-respected engineer to misdiagnosis a problem like this one for all heck to break loose.  If that's the case here, it certainly wouldn't be the first time.

For all the nay sayers, just keep asking yourself, why has this very serious problem been outstanding for so long?  The length of time for a problem this significant is simply incredible.  I can't think of a single reason unless Apple doesn't really know for sure.

Well, let's hope you have it and we soon all get the benefit.

Strange. I posted and it disappeared. And now it's back. Nevermind.

Well if Apple never responds to you, you can always just sell your fix services to people with the problem.