Yeah, but that doesnt explain all of the millions of older MBPs still kicking.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
How long should an MBP logic board last for?
- Thread starter jmjeffrey
- Start date
- Sort by reaction score
I think we're getting away from why products are warranted in the first place. It is to guarantee against manufacturing defects which will usually show up within the warranty period. I don't often go for extended warranties because my rule of thumb is that if it lasts a year, it'll probably last 10. Usually works.
I’m sorry. An Intel CPU is not going to “desolder” itself from usage unless you, the user, are wilfully neglecting and/or abusing your machine’s upkeep (i.e., not cleaning out the fans and fins; declining to refresh thermal paste after, say, 7–8 years of constant use; overriding how the fans work and shutting them off/keeping them to lowest setting; leaving your system in direct sunlight, especially during summertime and/or in hot/humid climates; and so on).
The same neglect and abuse toward a Silicon chip will, eventually, cause not a CPU failure, but a wholesale SoC failure. That there are so few that the community know of at this time speaks to: a) how few years they have been in use on Macs; and b) how outnumbered they are against 18 years of Macs with Intel CPUs (and 12 years of PowerPC CPUs).
There is a rare case with GPU Nvidia has chip desolder itself in the past from overheating. It happens.
Let’s be clear: you wrote Intel, which suggested CPU.
I own and run an early 2008 A1261 MacBook Pro with the Nvidia GPU, the first series of which shipped with the 602 revision of the 8600M GT GPU. Mine was one of those models, but now has a “green-dot” 603 revision replacement. It is well known amongst folks who ran (or still run) with the 8600M GT that the 603 revision is stable and durable, even under high use/load/heat situations.
The 602 revision was an Nvidia design failure. No amount of keeping the system cool could get around this issue until Nvidia owned up and revised the GPU so that the “de-solder” issue was resolved (which was between GPU and substrate, not between BGA substrate underside which met the logic board).
Likewise, many of the AMD Radeon 6000 series HD GPUs from the 2011 MacBook Pros, Mac minis, and iMacs had a similar GPU design failure entirely on AMD’s behalf. Unlike Nvidia, AMD did not own up to it, did not produce revisions for the affected GPUs, and left a lot of consumers in a rotten lurch of inevitable (and eventual) failure. Apple should have been more assertive and persistent with AMD on this, but they were not.
So, once more, be clear — not for my sake, but for the same of future readers who come upon this thread.
I have a 2018 15" MBP that has had the motherboard replaced twice now. The laptop is used mainly like a desktop, and rarely moves. Both occasions were lack of power. Computer was working fine, left it in sleep mode, then, on next use the laptop was found powered off and no signs of power. In both cases, the battery health had been indicating a battery issue, but the laptop seemed to operate fine. Once the logic board and touch bar boards were replaced, the battery health showed normal. I'm still on the original battery. The first time was 2-3 months out of AppleCare in 2022, second time was last month, so I'm getting 2-3 years per logic board. Both repairs were north of $700. There was little to diagnose since the machine could not be started. Apple could only eliminate issues - no liquid, known good power adaptor.
Last edited:
Even then a chip won’t desolder itself. The melting point of solder is at least 180-200 C, well above the point where the system would shut down. The various widespread problems with chips failing in different devices has almost always been due to flaws in production, e.g., wrong solder being used.
It might work ¯\_(ツ)_/¯ and I often do the same thing. But if it dies at 366 days, or 1100 days, or 2500 days - then it’s dead. Probabilities & statistics work until they don’t.
I did electronics service for 35 years. Actual component failure was very low. And it was usually out-of-box failure. Most of my repairs were due to abuse. I've probably had only one Apple device fail on me (the display controller went bad) and that was within warranty. So I don't go for extended warranties on most stuff. Sometimes @#*& happens.
As you noted, it’s typically stuff which is out-of-box or will fail within a short time after manufacture and initial usage. Those tend to weed themselves out early. Occasionally, Apple screws up and sends out a product worthy of a recall; as to whether Apple will actually recall and repair is, unfortunately, another matter entirely.
A few components, on select models, can and do fail chronically (and, fortunately, many of those have, in the past, remained available for replacement purchase, even if not directly from Apple). For the last MacBook Pro I bought from Apple, a unibody, I chose not to buy AppleCare, which was a departure for me. On it, the keyboard failed in the first nine months, and Apple replaced it under the one-year warranty. A second keyboard failure three years after that (well after AppleCare would have been of any benefit) was one I replaced myself.
The other parts to have failed, but easily replaced, were known parts whose design left them prone to excessive wear and, eventually, failure. All failed for the first time just beyond the three-year-ownership mark: flat hard drive cables whose placement render many of them to fail within a reasonably short time (i.e., at least once in the lifetime of using that computer); and the flat wifi/BT cable on the 2011 MBP which, for reasons related to the same, also failed.
I'm more likely to get a warranty if there are lots of moving parts. I'd do a laptop (keyboard, fan, hinges, switches) over an iPhone where just the buttons move. Apple uses very good buttons and I can't recall one failing on me.
A recall, of course, is a different story. I used to get the monthly notices from the FDA for device (anything other than a drug which touches a patient) recalls and they typically ran to 10 pages or so.
Wow, that’s crazy.Power supplies sometimes do not age well. That's why I always preferred devices with a discrete PS. My oldest piece of electronics is my Heathkit digital clock. It's going to be 52 next year. Running pretty much continuously. With circuit traces a whole lot wider than they are now, older chips last virtually forever.
View attachment 2252377
They said it was either an issue to do with the display (part of the E1 Quality Program) or the logic board, but without actually looking into further it was hard to determine.
Admittedly, it was quite tricky as it was an intermittent fault where the display would go to sleep and wouldn’t wake up, or - when you powered the device on - the screen would stay black.
When I took it to Apple, they ran a diagnostics test, which came back fine, then tried to replicate the issue once. They then sent me home to reinstall the OS.
That didn’t resolve the issue, so I took it back again and they decided to replace the top case after a bit of an argument on consumer rights.
They sent it off to the Czech repair centre and the repair centre sent it back without replacing the top case 🤨 and the issue persisted.
In the process of dealing with it all now, and will update once I’m able to.
That’s what I thought - well, over a decade - and that’s what Apple say if you ask them. They actually have a script in response to the question it seems, as every time I’ve asked (three times) I’ve been told the exact same thing word for word.
Agree decades, however if you dont use the computers for an extended period of time say 18 months+ your asking for trouble with older systems.
Still have one of the ill fated and very flawed 2011 15" MPB's. I kept it in the immediate family as didnt want the headache in the event of the dGPU auto destructing. I hammered it in a professional engineering role, passed it to my daughter who mostly used it as a PlayStation. Retuned back to me as it was superficially dead, once I cleaned up the SW image It's ran flawlessly ever since.
Last it was clean installed was at the factory and it remains to be 100% stock. It's only ever shutdown when we move house and has been our media server for well over five years. Most elderly was an Early 2008 aluminium 15" MBP. Eventually junked it as the battery literally exploded (without warning), needed a 3rd set of fans, new top case and display latch. The electronics were still good to go, admittedly it was very much a "burner" from day one it ran obscenely hot 100C was the norm 🔥🔥🔥 You just learned where not to touch it LOL...
Q-6
Literally for surfing the net and emails. Nothing intensive at all. And I still glused to get the beach ball 😂
In the UK we have six years to bring a claim under what’s known as the Limitations Act 1980. It basically exists because inherent faults can sometimes take years to become apparent.
The onus is on the customer to prove the inherent fault existed when the purchase is made, which can sometimes (maybe most of the time) be tricky considering the time that has passed, but on other occasions, it’s quite easy.
Yes, you generally have more government backing for consumer protections. But I'm stuck here.
My MBP 2015 worked for 4 years before either a sensor or the SMC failed - it didn't "die" exactly but it was unusable. My MBP 2008 worked for 3 years before the GPU died.
Mac Pro '09 - 5 years. Powermac G5 - 8 years. Mac mini G4 - 1st board 6 months, 2nd board 3 days (3rd board still works to this day).
Mac Pro '09 - 5 years. Powermac G5 - 8 years. Mac mini G4 - 1st board 6 months, 2nd board 3 days (3rd board still works to this day).