Hacker News new | comments | ask | show | jobs | submit login
Apple releases software fix for MacBook Pro slowdown (sixcolors.com)
460 points by shawndumas 7 months ago | hide | past | web | favorite | 237 comments

This appears to be working as expected. Here is a graph [0] of the Intel Power Gadget while I ran Cinebench. They are not the same length because it took longer before the update and I cut out the parts before and after. Pretty striking difference IMHO and it never went below base clock after the update. I was worried about my i9 but I'm not anymore. Also if anyone is wondering it went from 780cb -> 860cb but I wouldn't put too much stock in that since I've got other apps open (same apps for both tests) and 3 external monitors attached. Some people report 1100cb score after the fix.

[0] https://www.dropbox.com/s/y30x13mi6ozhg0k/10.16.3S-Fix.png?d...

Thank you for letting us know.

Hmm my i7 8850H gets 1024cb[0] so I would expect your i9 to do better than 860cb even with some other stuff going on at the same time.


I don’t doubt it can do better, I just have a bunch running in the background and this was right after startup (after applying the patch).

But how do the temperatures differ before & after the update?

From what I've read, it's stable at ~92°C.[0]

That seems damn high to me, but then I'm most familiar with desktops and servers, where you can cool far more aggressively. So is long term ~92°C hard on CPUs?

0) https://www.cnet.com/news/apple-macbook-pro-speed-throttling...

My old MacBook Pro I had for more than 5 years is 90% of the time around 100C. Whenever I'm working or playing. It's still working good.

That's a ridiculous temperature. That is literally the thermal limit for all recent Intel CPUs - they throttle to avoid actual damage at that point. No computer, not even a thin laptop, should be hitting that temperature or anything close to it.

Wow, I've just found https://www.reddit.com/r/techsupportgore/comments/4917wn/ano... but in a different context

100C is the boiling point of water.

I hope there is no water whatsoever in your chips.

Does liquid cooling count

That's not in the chips. The liquid in a cooling loop doesn't get to the temperature of the actual die if it did it wouldn't actually be cooling the die any more. Also the temp reading are down in the die itself and there's a gradient between the die -> heatspreader -> cooling block (air or water) -> 'cooling medium' (ie air/water).

So what? What's the boiling point of water got to do with anything? Is your CPU made of water?

No but your lap is (mostly)

When people say the ‘CPU’ in this thread, they don’t mean the whole main computer unit, if that’s what you’re thinking. They mean just the little chip inside. There’s a cooling system so it isn’t actually anywhere near that hot on your lap.

True. But I've used laptops where the fan exhaust was unbearably hot. And I'm sure that was much cooler than 90°C.

So maybe someone could check exhaust temperature on a 2018 Macbook Pro with i7 under heavy load. I get that the exhaust ports are on the back, near the display hinge. But that still might be iffy, for those who cross their legs.

That I can’t really answer, I didn’t do a ton of benchmarking other than the CPU which I had seen that jagged pattern before so to see it so smooth and not fall under base clock was all I was looking for.

Does the fan seem to run more / higher rpm now?

I have not noticed that, and I did pay attention to that while running Cinebench.

Glad to hear the fix works out for you. It's rare that apple will fix something like this so fast, but it's entirely possible this just slipped past QA since apple doesn't test throttling that thoroughly it seems.

Which is strange if they're not, because their sleek and slim designs would seem to allow a very minimal margin for error on heat dissipation. you would think that would be the subject of quite a bit of testing on the hardware side of things.

Given the history of graphics issues (many of which where related to heat causing capacitors or GPUs to misbehave) I'm not at all surprised that their testing of heat-related issues is subpar to what you would expect. The fact they sell fan-less machines despite a history of heat-related issues in almost every solid-aluminimum-construction laptop they've released seems particularly negligent to me.

EDIT: If you'd like more information than just heresay, watch some of Louis Rossmann's videos on macbook heating problems (the video on tantalum capacitors is particularly interesting).

> Following extensive performance testing under numerous workloads, we’ve identified that there is a missing digital key in the firmware that impacts the thermal management system and could drive clock speeds down under heavy thermal loads on the new MacBook Pro. A bug fix is included in today’s macOS High Sierra 10.13.6 Supplemental Update and is recommended. We apologize to any customer who has experienced less than optimal performance on their new systems. Customers can expect the new 15-inch MacBook Pro to be up to 70% faster, and the 13-inch MacBook Pro with Touch Bar to be up to 2X faster, as shown in the performance results on our website.

Any idea what "digital key" could mean here?

Probably the MSR_TURBO_POWER_LIMIT value as described in https://www.reddit.com/r/macbookpro/comments/91256u/optimal_...

That analysis is incredible, how was the OP able to figure all of that out!?

I've experienced some early adopter pains with the desktop version of this CPU (x299 platform) when it came out and I used linux kernel mailing list to search for answers. I was surprised how easy it was to read the code of the related kernel modules and interact with their authors (Intel engineers).

These "magic codes" are probably just #define macros somewhere in the kernel repo, in my case it was in [1]

[1] https://github.com/torvalds/linux/tree/master/drivers/platfo...

I don't know, but apparently (1):

>> Originally posted at MacRumors by winterny >> https://forums.macrumors.com/threads/optimal-cpu-tuning-sett...

1. https://news.ycombinator.com/item?id=17602455

The Reddit post was about 20 minutes earlier.

Lots and lots of testing and lots of reading intel docs / example code on the internet. I probably ran cinebench and prime95 about 2000 times in the past few days.

Intel's HDK docs?

If you're designing hardware, this is a thing that people know. The new Xps machines have the same VMR problem. Probably he tested for a known problem and bingo! I wonder if the discoverer works for any hardware design company.

If you read that thread, what's curious is that it seems this power threshold has always been set perilously because older MacBooks seem to have the same setting for their VRMs as well.

I would expect this sort of slip from Dell or Lenovo with like 1200 skus but Apple really only has a handful.

My bet is that sensors don’t output absolute values and you need to use an offset to get the correct “reading” for your thermal profile and cooling.

Basically based on the location and type of sensor(s) and your cooling solution you will apply an offset to get the actual temperature to be used for managing the power profile and cooling of the component.

This is pretty common and the reason why many apps don’t readout the temperature of newly released components correctly.

Ofc it can be something completely different and Apple is hiding the actual cause but I don’t deal with tinfoil fashion accessories.

>Ofc it can be something completely different and Apple is hiding the actual cause but I don’t deal with tinfoil fashion accessories.

Which is a shame because the theme this year is I, Robot.

Huge performance difference - how did they not notice before?

The article made it seem like this only affected a small number of units? Their quality control probably doesn’t look at every single unit. Although it’s interesting that they didn’t find it at all, since reading HN and reddit gave me the impression it was very widespread.

Generally you can’t get a sense for how widespread an issue is from social media, because it’s a perfect medium for confirmation bias. Confirming links and stories are shared/upvoted/forwarded while non-confirming ones are ignored or dismissed (or outright attacked for some reason).

This is basically the story of the modern world as viewed through the distorting lens that is media (and especially social media). I wish people would take the two sentences you wrote and then use them as a corrective lens that undoes this distorted perspective.

Yeah it’s really hard to correct the perspective to the truth though if all you have is social media. Honestly I’ve been putting less and less trust in almost all my sources of knowledge recently.

It's so, so strange that 10-20 years ago my parents told me not to trust what I read on the internet...and now that they're all using it they have seem forgotten their own advice.

Apple always claims their issues affect a small number of units. I have seen zero people that tested the i9 MBP and did not see major throttling. The issue is clearly the result of a fundamental defect in how they were managing power and thermals.

The official explanation seems to be that the key management defect which lead to thermal management problems did not occur in the development machines used during the engineering process. This seems possible as presumably every machine used internally goes through a different provisioning process than those released into customer hands ... Thus the claimed cause of the problem seems to indicate that the flaw did not affect the internal benchmarks they ran. It would be _very_ concerning if the benchmarks they ran for engineering/marketing production were insufficient to catch this issue.

I think the official explanation seems pretty likely. If the flaw had been present in the machines used for benchmarking than either their benchmarks didn’t catch the issue (unlikely) or the benchmarking numbers were fabricated which also seems unlikely as it would invite an infeasible amount of legal risk for an organization of apples size and mindshare ...

Engineering, especially at this level, is not supposed to work like that. You certainly not do dev proto, then switch to prod both in one step and without even rechecking the tests you did on the proto. So their official explanation might be kind of right, but then it shows major deficiencies in their quality process.

Which given the track record of quality issues they have had, is possible.

> You certainly not do dev proto, then switch to prod both in one step and without even rechecking the tests you did on the proto.

Exactly. I work in software engineering for cars and even for things that are not mission critical or life-threatening, the engineering and quality process is insane because so much money is on the line. I wouldn't be surprised if macbook pros are worth more in profit, or at least revenue, than a car line and should therefore see an equal engineering process.

I can't understand how the firmware was released without the proper key, unless there was a bad merge or they dropped a pilot firmware that was testing without thermal management... maybe a clue there is a desktop in the works and the desktop firmware made it onto the macbook pros? Either way, like you said, their quality process has a major deficiency.

> It would be _very_ concerning if the benchmarks they ran for engineering/marketing production were insufficient to catch this issue.

Clarification -- I meant to say -- 'It would be _very_ concerning if the benchmarks they ran for engineering/marketing production were insufficient to catch this issue on a system configuration in which the issue manifests.'

> affect a small number of units

Same story with the keyboard that was a fundamental enough design flaw to include 2 design bandaids in as many years.

I would not trust any coverage of supposed issues with Apple hardware. There is a huge incentive for negative coverage. Remember Antenna-gate with the iPhone 4? They continued selling the iPhone 4, completely unmodified, for years after that whole debacle. That's how widespread the supposed issue was.

A more relevant example would be the Dell XPS 15 where owners and a few reviewers have reported extreme thermal throttling with i7 models.

>As mentioned in our detailed review, the XPS 15 9560 with the Core i7-7700HQ processor is prone to two types of throttling:

    Thermal throttling of the CPU or GPU (generally the CPU) when temperatures get too high

    Voltage Regulator Module (VRM) throttling caused by it getting too hot and being unable to deliver enough power

You have to wonder why a well known laptop that can't run a mainstream i7 without severe throttling is a story that couldn't gain any traction.

It isn't has bad as the Apple one though. But of course, Dell should also fix their shit.

I'd say that a similarly priced high end laptop that cannot adequately cool a four core i7 is a bigger deal, especially given that this is their second year to ship devices with the same issue.

>The service manual for the Dell XPS 15 9570 is now available and it reveals something that we wished would improve in this generation — that the XPS 15 9570 carries over the same 2-pipe heatsink assembly from the previous generation. Not just that, there is also no indication of improved thermal dissipation from the Voltage Regulator Modules (VRMs) implying that we could be in for some serious throttling issues.

The Dell XPS 15 while earning accolades for being a very capable multimedia laptop that can also game, has also earned the reputation for throttling heavily under load. Both the earlier XPS 15s suffered from both thermal and power-limit throttling. The only way to fix this was to repaste and undervolt the CPU and also apply thermal pads on the VRMs to drive away all the excess heat.

Unfortunately, it looks like Dell has chosen to retain the existing heatsink design for the XPS 15 9570. Compounding throttling fears is the availability of a Core i9-8950HK variant.


You would think that after two years with the same issue, the mainstream tech press would pick up on it if the issue is throttling and not just chasing clickbait stories.

Apple settled a fairly large lawsuit related to Antennagate and was forced to send out a bumper case to affected users so they could hold the phone the "Jobs" way and get reception rather than the way you would normally hold...literally any other phone before, during, or since Antennagate, including newer iPhone models.

They settled for $15/phone or a free case (they had already offered a free case to customers of their own accord). They kept the iPhone 4 around for 2 years after it was no longer their marquee model. Surely if Antennagate was actually a problem for a significant portion of customers they would have stopped selling it as soon as possible (aka, upon release of the 4S). But they didn't.

I think you severely underestimate the difficulty in retooling supply chains for relatively minor issues. It wasn't worth the cost of doing what you suggest when the issue could easily be fixed with a free bumper or by getting people to hold the phone in a non-ergonomic manner.

> It wasn't worth the cost of doing what you suggest when the issue could easily be fixed with a free bumper or by getting people to hold the phone in a non-ergonomic manner.

Exactly. It wasn't that big of a deal.

I mean, we haven't even gotten into how a bunch of other phones which were on sale at the time exhibited the exact same behavior.

TBH, the "Jobs" way was the normal way that most people held a phone. It was the people who smothered the whole handset in some kind of crazy death grip that had attenuation problems.

I got my bumper in the mail, but never needed or used it.

Most of the people I know who had iPhones had issues with the iphone 4 antenna, and only one or two of them are death-grippers. The problem was Apple's poor design, which Apple itself acknowledged by settling the case, offering free bumpers, and redesigning the antenna in future models.

And, as was pointed out on these very pages, immediately after they denied it was a big problem an ad for an antenna engineer appeared on their pages.

Not many manufacturers at their scale charge their prices so their QA procedure should be unorthodox as well.

Key, as in key-value pair, perhaps

I don't see how you can get more cooling with a firmware change when the fans are already going at full speed.

Does that mean the whole unit will now run hotter before it starts to throttle? How is this going to impact some of those bad solder joints MacBooks are famous for?

The issue wasn't cooling; it was how the CPU incorrectly throttle itself that caused it to overheat again too fast.

There's another thermal control unit involved and what people found was that the firmware for it was causing the CPU to drop down to 800mhz instantly when it detected the specific terminal event and when it reset that, it went back up into high-overheating load (such as the benchmark sustaining the high load) instead of staying at the right balance. In other words, it went into a loop of super low power load and then super-high load instead of the middle where the terminal balance is right for the load it is doing. Because of this low/high loop, the performance was not consistent and degraded as the overheating cause it to drop to 800mhz all the time.

Where Apple likely messed up is that the firmware was not placing the CPU at the middle to avoid going into a loop, this allows the CPU to sustain a better CPU performance consistently and for an extended period of time.

The issue was not the CPU throttling itself, it was the CPU being configured to draw north of 100W and the VRM not being able to handle it, overheating, and leading to the system board hard throttling (down to 800MHz rather than slowly stepping down) the CPU until the VRM cooled down at which point the CPU would be un-throttled and the cycle would restart.

Apparently the power configuration is the one Apple's always used but previous CPUs didn't come close to the current generation in ability to draw power, so they'd never reached the VRM's thermal limits.

By lowering the CPU's maximum TDPup to 50~55W, they avoid the VRM over-heating and while the CPU does not turbo as high as with a 100W TDP allowance it can stay there just fine.

>I don't see how you can get more cooling with a firmware change when the fans are already going at full speed

It's not about what you do when the temperature if at its peaks and the fans are going at full speed, it's about how you avoid getting there.

>How is this going to impact some of those bad solder joints MacBooks are famous for?

When did MacBooks became famous for "bad solder joints"?

> When did MacBooks became famous for "bad solder joints"?

There has been more[1] than one[2] case where GPUs in MBPs stopped working because of bad solder joints. And ifixit has a guide about how to fix your mac by putting it in the oven[3] to reflow the solder joints.

1: https://www.macupgrades.co.uk/store/product_info.php?product...

2: https://www.apple.com/support/macbookpro-videoissues/

3: https://ifixit.org/blog/6882/why-i-drilled-holes-in-my-macbo...

That was an nvidia problem. All their laptop chips were affected that were in one family. Dell replaced my motherboard four times over it.

If MacBooks are famous for it, so are all laptops with an nvidia gpu.

To be fair, most of those other laptops aren't made by a company that's famous for meticulously stressing details and restricting hardware configurations so that consumers won't need to research or worry about internals.

The PC vs Mac hardware dichotomy is best summed up as, "Apple gives you less choice so that they can exercise more quality control."

That PC vs Mac dichotomy really doesn’t carry into pre-built PCs and definitely not into pre-built laptops and other portables. Both have a responsibility for quality control.

I agree in principle, but there is still a difference here.

Dell sells way more PCs (in much more varied configurations) than Apple does. They're generally more repairable, and they're generally cheaper. They're also willing to let you max out specs and build weird configurations; configurations that arguably don't always make much sense. That range means they also sell a few objectively cruddy laptops, so you need to be careful and read reviews and do comparisons before you buy.

Apple positions itself as different from those companies. The comparison I'm making above isn't something that only self-building PC nerds understand; the same explanation is what I give to regular consumers who are trying to decide whether to go with Windows or Mac - people who will never try to build their own stuff.

"Should I buy a Windows PC or a Mac?"

"Do you want to spend a lot of time researching brands, reading reviews, and thinking about hardware specs?"

I think the point still stands that Apple is positioning itself in a different category than companies like Dell. If not, I'm gonna have some complaints about that $3K price tag.

The closest comparison is something like the Microsoft Surface - an expensive, premium device, with low repairability and configuration options, but (theoretically) increased reliability with a streamlined experience that "just works" out of the box. When the Surface line has problems, I put more blame on Microsoft than I would put on a company like Dell in an equivalent situation.

There is no correlation between how repairable a laptop is and its failure rate.

You've just made this up and now are criticising Apple/Microsoft for it.

There is a correlation between how repairable a laptop is and whether it uses a custom build process, single-body design, or if you're directly soldering chips onto a motherboard - all of which can help with reliability. Fewer moving parts and less build complexity means fewer places where things can go wrong.

I wasn't criticizing Apple or Microsoft for anything when I said their laptops weren't repairable, I was giving them the benefit of the doubt that some of their engineering decisions around repairability are made for good reasons.

I don't think that's a particularly controversial idea. I mean, no one is seriously going to argue that they are repairable. So assuming there's a good reason for that, this is the standard explanation that I and other people would give to someone complaining about repairability - that Apple optimizes for other things.

Heck, it's the explanation that Apple gives: "Our devices are more reliable when we're the only people messing with their internals."

I had one of those problematic 2011 MBPs, and reballing/reflowing/baking was only a band-aid. For many users, the issue was known to recur after implementing the fix.

Even for the eventual repair order, Apple stuck in the same logic board with the same parts. I think the main difference was that those logic boards throttled the CPU to run cooler.

That early 2011 Macbook Pro was the first laptop I ever had (by any manufacturer) that died. I was lucky I still had the MBP around by the time Apple issued the repair order. I was able to swap out my dead logic board and eventually unload the laptop.

I had one of those problem units, too. After a lot of back and forth, Apple swapped the main board. I had no subsequent problems. Yay, Applecare.

My AppleCare had lapsed the time it died. The logic board was already replaced once while it was under Applecare, and maybe 3 months after the first logic board replacement, the GPU flickering issues started recurring. At that point, I didn't have the board replaced again under AppleCare because it was clear they were going to replace the logic board with one that for all intents and purposes, had the same design (?) defect.

>There has been more[1] than one[2] case where GPUs in MBPs stopped working because of bad solder joints

More than one case in 20+ millions units sold?

I'm not sure that qualifies as "famous for".

Watch a bunch of Louis Rossman YouTube videos. He knows all about the seedy underbelly of Macbook hardware design mistakes.


A few things that I take issue with here:

1) Louis Rossman has a vested interest in drumming up attention for his videos and has, on multiple occasions, used false statements and misleading practices to do so. Just like in the initial release video of the Touch Bar MBPs where he used an unshielded USB-A to USB-C cable and then complained that WiFi wasn't working. When used with a shielded cable, the issue went away and yet he never issued a correction or retraction. The issue was with the adapter he used, not the computer. Which leads directly to...

2) The issue stated in that video that required reballing was an issue with nVidia chips and affected all laptops that used that platform, both PC and Mac. To suggest that it was an engineering issue on Apple's part is disingenuous and only further reinforces the idea that he's not making these videos to be helpful but to drive traffic to his channel and repairs to his store.

3) Just look at how he used the Linus story to get the repair done at his shop. He has a pattern and, unlike what most people think, it's not altruistic. It's all to benefit him. I don't blame him but let's not pretend that he's doing this because he cares about users or the "right to repair" movement. He cares because he can mislabel and misconstrue the positions of the people against it to benefit his business. He'd love it if Apple was forced to provide him schematics and docs. He's in it for himself.

Sure, he's got his axe to grind. I mean, you can't watch Louis and not think he's got an axe to grind -- like maybe even literally in a makeshift workbench in his kitchen with a whetstone. But even if you watch his videos with the most cynical stance, it's still very eyebrow raising. Apple has had it's share of design 'doh! That's pretty much undeniable in gory detail -- far beyond the big headlines everyone knows about.

The problem is his arguments are devoid of context or proportion. He is looking with a microscope—both metaphorically and literally—at tiny details on a circuit board which don't seem to be a problem for >99% of Apple's customers. Are these points of failure really any more common on Apple devices than their competitors?

(There is a remarkable difference in mindset for the majority of MacBook and PC laptop owners. PC laptop owners will often just ditch their machine if it stops working. It's weird to say but I see it all the time.)

Now I don't doubt that most of what Louis says is correct. But I also appreciate much of it from Apple's perspective. If everyone in the independent repair industry was at least as competent as Louis is today, Louis would have a point. But they're not, so he doesn't. Besides, Apple's intransigence creates the market which Louis takes advantage of; he shouldn't complain.

Oh and I keep getting irritated every time Louis bangs on about those damn "refurbished" iPhone screen assemblies. He should know full well that the replacement glass used on these refurbished assemblies is cheap fragile non-Gorilla junk. I know multiple people who have had their iPhones independently repaired only to have the screen break again in a matter of days or weeks, under the most innocuous of circumstances. We are being scammed, and Louis defends the scammers.

Otherwise it's a great channel.

>If everyone in the independent repair industry was at least as competent as Louis is today, Louis would have a point. But they're not, so he doesn't.

This is exactly the point that I feel like I constantly have to get across to people that support him. Louis is incredibly smart when it comes to electronics repair. It's exactly the reason that I know, for certain, that he's being disingenuous. He's too smart and too knowledgeable in most of these topics to gloss over these issues the way that he does unless he has an ulterior motive and it's clear that he does. Apple's fight against "Right to Repair" (which is misleadingly named anyways) isn't about people repairing their own devices, it never was. It's about preventing repair shops from doing those repairs and giving Apple a bad name because someone sees an iPhone where the screen fell off the front ("where the front's not supposed to fall off!").

I would be very surprised if Louis specifically endorsed non Gorilla glass replacements. It's possible that he does. I just watch his repair videos. As far as that goes: don't buy them, and don't get repairs from someone who would sell you one!

I know multiple people who have had their iPhones independently repaired only to have the screen break again in a matter of days or weeks, under the most innocuous of circumstances.

I realize this will be unpopular, but my observation about people who walk around with perennially broken glass on their phones, is that some people can't keep nice things. If your lifestyle is that rough, either use a completely enclosing polycarbonate case, or get a more robust 2nd phone. Expecting a computer made largely of glass to stand up to roughhousing contradicts common sense.

> I would be very surprised if Louis specifically endorsed non Gorilla glass replacements.

He pretty much does.

"So what they do is they take this screen ... and they replace the front glass layer. Most of the time the only thing wrong with the screen is the front glass layer is cracked. They put a new glass on it in China and then they sell it to us in the United States as a refurbished iPhone screen ... That is not a counterfeit."


I understand why Louis gets worked up about this particular issue, but he's just flat out wrong. These refurbs are dodgy as fuck. The assemblies as sold are counterfeit.

> some people can't keep nice things

Absolutely. I don't disagree, but this doesn't describe the people I spoke about. Accidents happen to all of us, even when we're particularly careful. I would never expect a glass computer to survive rough treatment but I do expect that when it's repaired with a "refurbished Apple screen" that it is made of substantially the same material as the original.

>I understand why Louis gets worked up about this particular issue, but he's just flat out wrong.

Which is exactly why Apple's distinction was the logo on the product. If it's not the exact same product, it shouldn't be able to bear their logo. Judges deciding that it doesn't matter because users will never see the logo are missing the point, imo. It only opens the door to all kinds of 3rd party parts and repairs where it's not the exact same product but you're paying for it as if it is.

Apple charges a huge premium for their machines because they supposedly integrate design and QA so that they don't ship with broken video cards like cheap commodity computers do.

Since when did Apple or anyone claim this ?

I have had a Mac since the 512K days and they have always shipped with commodity components and Apple has never had some magic pixie dust that made them fail at lower rates than if they were in a PC. Nor have they ever individually tested every machine to guarantee it works.

Apple has and will always ship with broken video cards. And broken displays. And broken CPUs etc etc.

When did MacBooks became famous for "bad solder joints"?

Have you noticed how low the bar is for "famous" these days?

Whether it's some FaceTube "star" or tech "scandal" du jour, everything is massive these days and everyone on the planet is affected. Somehow, there are no small issues anymore.

My friend and I have a hobby of linking each other to headlines of "The Internet is Outraged about <X>" where the article body embeds 10 random tweets.

How would ten randomly picked tweets have anything to do with X ?

Hey, now. Don't go robbing the masses of their Two Minutes Hate.

Please don't do this here.

Wow. I'm pretty sure this is the first time in 8 years I've commented anything that's been flagged. I'm not entirely sure what "this" is.

Is it saying "don't go robbing the masses"? Is it agreeing with the parent's comment by means of a literary reference that I thought was an approachable and analogous representation of the phenomenon the parent was describing? Seems whoever flagged this read it in a wildly different way than I intended it. Would it not be "this" if I had worded the same point differently--e.g., "You know, you're right, parent. These events totally remind me of 1984's Two Minutes Hate."?

Genuinely curious what "this" is that I shouldn't do, dang, because I've spent 8 years intentionally avoiding doing it.

It's simple: the comment was unsubstantive and snarky. Bad combo, especially on flame-prone topics. We're looking for thoughtful comments here, as you know.

The issue isn't just the quality of a comment in its own right—it's the quality of what it leads to. These things compound.

Thanks for being a fine HN commenter and user!

there were a whole series of macbook pros from around 2010-2012 which had the nvidia discrete GPUs, BGA chips, come unsoldered from the motherboard on their own under normal usage.

(edit: nvidia, not amd/ati radeon)

Discrete GPU on my former MBP 2010 would regularly crash. Would need to force it on the integrated, slower though less power hungry Intel GPU.

Supposedly the whole Nvidia debacle is the reason why Apple hardware doesn't sport Nvidia anymore. Though you can get them to work sometimes with iGPU, its more safe to keep up with AMD.

I'm not sure about FreeSync vs Gsync though, and I suppose Moonlight doesn't work on AMD.

Early 2008 MacBook Pro as well.

Note that bad solder joints only existed for 1-2 years and were solved long ago. AFAIK they were caused by NVidia (and MS) using untested lead-free solder balls without underfill.

This earlier HN thread has all the details: https://news.ycombinator.com/item?id=17602146

There are no additional details in that thread.

My guess is a sensor requires a signed connection, much like how HDCP works. Seems a bit over-engineered if Apple is building protected sensors, but would also love to hear from someone in the know.

My guess would be that they shipped with an updated firmware version to the one programmed in the factory. It should have upgraded the first time the device is powered on but because of a missing security signature or maybe key to verify a signature, the firmware update wasn't applied. This is still just a guess but signed firmware is becoming much more common.

"key" as in dictionary key, not cryptographic key

I would bet my money on the Apple T2 SoC: https://en.wikipedia.org/wiki/Apple-designed_processors#Appl...

Yikes, pretty bad showing by Apple here. They don't have the leeway that they did when they had Steve Jobs... they have tremendous brand loyalty, but will need to be more careful in the future.

They had the same issues before Steve Jobs, 99% of people seriously do not care or even knew about this.

What's more important is fixing and moving on to the next problem.

Technologies are so complex nowadays that it is becoming virtually impossible to produce anything without issues.

They had the issues with Steve Jobs but they also had Steve Jobs to save their ass.

And yet, we've survived more "gates" without Jobs in the past few years as well.

Batterygate recently, or bendgate?

I always felt "bendgate" was a non-issue. Just a bunch of hipsters who were too dumb to understand that you don't put a phone in the back pocket of your skinny jeans.

It's a slab of glass. What kind of person thinks that's a good idea?

What kind of company makes a device commonly put in pockets out of a material that can't stand that environment? ;)

At the same time it is a phone ... who thinks to put it in their pocket?

> they have tremendous brand loyalty, but will need to be more careful in the future.

Does that really matter though? I was just discussing this with someone over the weekend... his thesis was that people buy Apple because of branding, which might be true as far as the majority of customers are concerned... But for people like me (powerusers), MacBooks are simply the best possible choice. Everything else is either adware (Windows 10), or not usable (Linux laptops lack deiver support (last I heard) You can’t install software on Chromebooks, Windows 7 doesn’t have a Unix subsystem). Not to mention that Apple hardware is beyond superior, and the software mostly just works (things like wi-fi, backups, sharing, ...)

It's pretty trivial to disable all of the telemetry in windows 10. I'm not saying microsoft should get a pass for such nonsense but it's a little over the top to act like windows 10 is DOA botnet.

Whether it's trivial is irrelevant. It should never need to be done. Ads and tracking in the default OS is a scummy move by Microsoft.

Ubuntu did it with Amazon. Mozilla did it with Pocket. I'm sure there are endless other examples.

Dismissing windows 10 as merely "adware" is inflammatory to the point of meaninglessness. It's really the kind of rhetoric I would expect on /g/ not here. The OP has also been rightly called out by others for his broad dismissal of linux.

I agree that Windows 10 should not be dismissed as adware. But just because Ubuntu and Mozilla did it doesn't make it right. It's something that absolutely shouldn't happen in the first place. Microsoft does not need to default to invasive tracking and ads in the default UX, and there's no question that it's an anti-consumer move in the first place.

You're right, it's not exactly adware. Adware is software that's supported by ads. Windows 10 users pay for it, and also get the ads (for free!), so technically it's worse than adware.

I've had no driver problems running linux on my laptop, just sayin.

I don't have driver problems if I actively fix them, but I haven't had a linux laptop just work basically ever.

Currently, my XPS 15 with Xubuntu doesn't function after coming back from suspend and I haven't bothered to dig into why. I don't mind, but it's absolutely not a problem I experienced with windows on the same hardware or any mac I've ever owned.

Same, since 2009. Even got better battery than on windows (and by far since windows 10)

But it's true that you better buy a laptop that you know has good support. eg: almost any dell, almost any asus, any thinkpad.

Another comment today mentioned the low quality of asus laptops. I have never owned one, but buying an apple laptop is supposed to relieve all these doubts about good brands, years and models.

Show me a MacBook as robust as a T or W series Thinkpad. They play in the same price/performance class, except that the former has a fancy touchpad and the latter a trackpoint. And it is unlikely to break, and the non-touch versions come with a matte screen that actually works as long as you don't have the sun in your back. And even then, cleaning the screen from dust and increasing the contrast/gamma does wonders.

I've currently got no driver issues per se. But even in 2018 I don't have hibernation out of the box, battery life and wifi are both worse than Windows on the same machine, and the fans are on all the time.

I mean, I still love it and these aren't life or death issues, it's just something fundamental about the nature of Linux in an ecosystem of proprietary hardware specs and commercial drivers. It will always be a fact of life. Just like Linux package management, because the software is open source, will always be better than on Windows or a Mac. That's a fact of life that can't really change because of the nature of _their_ ecosystems.

No harm in admitting the differences, but good for you if everything works out of the box!

Wasn't a Linux laptop with hibernation, sound, and wireless working at the same time one of the UN's Millenium Goals?

That was the original draft, but they chickened out and just decided to halve poverty by 2015 instead.

Err, what do I want to hibernate? S3 is pretty energy efficient and you'd likely want security patches anyway before the battery runs out on LPDDR3/LPDDR4 ram.

Everything else is either adware (Windows 10)

I am running Windows 10 Pro on a Surface. I haven’t seen a single ad, anywhere. If that’s your only reason let me set your mind at rest. Surface is a better Mac than a Mac at this point. That’s why I upgraded my old MBP to one rather than a new MBP.

The Surface is probably a great Surface, and you might prefer it to a Mac, but a Surface is a terrible Mac, for much the same reason a Mac is a terrible Surface.

As I say, I upgraded my MacBook Pro - for that is what MBP means - to a Surface, so I am quite familiar with both thanks.

But by all means ignore my advice and pay through the nose for a device that is blatantly not fit for purpose, I'll be over here enjoying the superior computing experience that Steve Jobs would have wanted me to have!

I would hardly call the Surface family cheap.

What price do you put on a nice keyboard with a bit of travel, and a processor that can hold its clock speed?

Windows now tells me my computer is at risk because I refuse to give all my data to OneDrive. Their rationale is I'm at risk for cryptolockers, but the ulterior motive is clear.

I had to change a registry key to stop it from installing Facebook and Candy Crush on its own.

I bought a PC that came with Windows 10 pre-installed and the start menu had a bunch of ads for video games, photoshop and even a little picture of Trump in there https://i.imgur.com/WMLndK3.jpg

Putting aside the stub applications that are promoting other products (e.g., "Get Office"), the Trump photo is just the News app showing a live tile view of the top story for the moment.

Yes, the promotional app stubs are advertisements, but they're easily removed and don't re-appear. In my opinion, removing the tiles you don't want is closer to an act of operating system "customization" than the user experience with ads on the web.


> "Poweruser"? No, I feel the more accurate term would be power-consumer.

Possibly. I guess there are many different kinds of powerusers. The interpretation I had in mind is that I use the computer (1) a lot, (2) not just for internet/media consumption, and (3) I use it to program so I want/need easy accessibility to many software packages (which is a problem on Windows, IME).

Just because I can solve all of the problems/configuration and use all the low-level backup/networking tools, it doesn't mean that I want to (it's just not worth my time).

If you use computers "a lot", the percentage of time required to learning a select few Linux userspace administrative tools, as well as the time dedicated to configuration and maintenance, is less than nothing compared to time spent on general everyday usage. The common perception that Linux is a timesink is false (not to mention most things are very much plug-and-play nowadays, e.g. networking via DHCP). And the benefits in doing so are huge, in that you now control your needs using a variety of independent tools, as opposed to the far more locked-in, proprietary solutions that Apple offers. I simply disagree with your claim that Macbooks are the best option for so-called power-users when on multiple aspects from hardware to software they're not. Macbooks/The Apple ecosystem is A choice, not THE choice.

> Possibly. I guess there are many different kinds of powerusers.

It's simply that "programmer who doesn't like to configure things in detail until everything they use fits them like a glove and doesn't know much about hardware" and "power user" are different and orthogonal things.

Apple hardware is good-ish. For proof you should look at the equivalent laptops from Dell etc and notice that they are cheaper, but not by much. Beyond superior? Nope. They still have better screens (colors not number of pixels!) and faster storage, but that's about it. And they fucked up the keyboards lately.

The one thing that makes Apple crap still superior is the HW/OS integration. I switched from Linux to OS X because of that, not the actual hardware.

Back to the power something, I do custom embedded Linux-es for a living. I still would like my base OS to "just work". <Cough> systemd... can't trust Linux any more.

Really? What's the best Linux distro/toolset to use in order to get anything to show up in a usable way in HDPI? I don't think you can buy new monitors anymore that are at low enough a resolution to support modern Linux.

Err, you must be looking in the wrong aisle/store. You can certainly buy new, good IPS/MVA panels (as a finsihed, desktop-ready unit) with more classic dpi. E.g. the classic format 19" 1280*1024: iiyama PROLITE P1905S-B2 available for 150~200$ new. To be fair, this is probably one of the, if not the lowest DPI/resolution desktop screen they offer, but still. it's at about 86 DPI, they also sell a 32" 1080p screen running at 68 DPI. I do like it a bit higher between about 100~110 DPI, and above 120 DPI requires high DPI technology. So, yes, please consider reasonable hardware. Don't buy HighDPI if you don't want to handle HighDPI with legacy software.

...Fedora? Or anything recently modern?

I literally run FC28 on 4K monitors and at 1.25 scale on my 1440p T480 and I don't even think about it.

Wayland pretty much just works and has for at least a year.

What country are you living where you can't have access to such "old" monitors? An imaginary one?

Two classic form factors are still readily available: 1920x1200 or 1920x1080 at 24", and 2560x1440 at 27".

All the ultrawide 21:9 screens are "LoDPI" as well, and at 1440p or 1600p they're great for coding.

Happy to see Dave2D at the forefront of this. He gives solid advice and puts out crispy 4K content on YouTube.

Link to his channel: https://www.youtube.com/channel/UCVYamHliCI9rw1tHR1xbkfw

I feel bad for the other famous YouTuber who published a video basically saying that everyone are overreacting.

For every internet action there's an equal and opposite overreaction.

The 3rd Law of Emotion.

The 3rd Law of Emoticon


Wait who?

Dave (the guy who Apple mentioned working with to find this issue) already tweeted that the performance is "soo much better". Source: https://twitter.com/Dave2Dtv/status/1021817313555755009

Has he tried setting it on his bare legs during the test?

Why would anyone put a high end laptop on their legs while doing a prolonged intensive task such as video rendering??

Louis Rossmann on YouTube has a take on this "you're using it wrong" culture of Apple products I find interesting:


I find Louis hard to watch if I am honest. Yes I know Apple don't make the best products but let's be honest they are not actually that bad. I know he doesn't like Apple, he hates all the soldered components making things hard to impossible to repair, etc. He loves his ThinkPad's which is cool (I love them also) but I just got bored of his constant Apple hating.

> his constant Apple hating

I guess that's normal for someone fixing Apple stuff as his day job and thus always only seeing apples bad side. Especially with seemingly Apple not wanting such repairs to happen.

And the whole Apple threatening to take down his YouTube channel thing.

Watch any content provider enough and you'll probably get tired of them IMO

All computers get really hot when doing stuff like video rendering, not really apple specific

Actually i only sweat with my thinkpad, but that might come from it having 1) slightly lower TDP (30W instead of the 50 I read above), and 2) a thermal design that does not result in any part being too hot to touch. Even in ambient temperatures that are no longer healthy for a resting human.

Isn't part of the definition of a laptop that it should be able to be used comfortably resting on the user's lap?

The versatility of a laptop doesn’t mean that it should be expected to violate the laws of physics, or only use chips with maximum TDP such that it is always comfortable on your lap.

Having a chip which can max out thermal load while plugged in at your desk, with the expected thermal results, and then throttle down when you go mobile, again with the expected thermal results, is actually the best of both worlds.

And with one click the user can actually choose which power mode they want to operate in.

A laptop that throttles back to avoid getting hot when you need the CPU/GPU operating at peak performance would be much less useful, particularly for professional use.

Violate the laws of physics? Is that the term for limiting the amount of power you dump into the chassis? There's an unlimited amount of air to soak up the bulk of the heat.

That's why they're explicitly not referred to as laptops but notebook computers. It's the reason Apple calls them MacBooks only and not laptops.

You have to be kidding.

What you're referring to is a branding tactic. It has nothing to do with some subconscious direction on how to use the device. It's not only reasonable, but expected, that customers will use the device in their laps... no matter what the naming convention. Come on.

Not kidding at all. The entire industry made a huge hard stop and turn on the terminology around 2001/2001 after Dell was sued because a man claimed that his "laptop" burned his thighs/penis when the CPU/battery heated up and the fans didn't turn on. The courts ruled that "laptop" implied that the machine was safe to use in that manner and the guy was awarded damages. After that, it became exclusively "notebook" or "portable" computer industry-wide. I haven't really put any thought into it recently but I think you'd be hard-pressed to find any manufacturer referring to their machines as "laptops" now.

Edit: It was a Dell Latitude that burned the man's penis: http://ftp.compal.com/Download/NB/NCL50/User-Service%20manua...

>I haven't really put any thought into it recently but I think you'd be hard-pressed to find any manufacturer referring to their machines as "laptops" now.

Oh, really? Let's see.

Dell: https://www.dell.com/en-us/shop/dell-laptops/sc/laptops

HP: https://store.hp.com/us/en/cat/laptops

ASUS: https://www.asus.com/us/Laptops/

Lenovo: https://www.lenovo.com/us/en/laptops/c/LAPTOPS

Toshiba: http://laptops.toshiba.com/computers-tablets/home/

Acer: https://us-store.acer.com/laptops

Samsung: https://www.samsung.com/us/computing/windows-laptops/

I will stop there, because it's obvious that 'the entire industry' did not make a 'huge hard stop and turn on the terminology'. Stop being a fool.

I'm not being a fool. The lawsuit must have worked out in their favor or they appealed. I worked for 2 separate computer manufacturers during that time where we were explicitly told not to refer to any of the portable computers as "laptop" computers in any of our outgoing communications or marketing. I had associates at other manufacturers that were told the same thing.

If that's not the case anymore, then that's just because I haven't bothered to keep up with the industry as I no longer work in comms for PC manufacturers. As I said, I haven't put any thought into it recently. I'm neither being a fool nor am I lying, as you seem to be implying. Times change and I prefaced my statement to admit that the info might be outdated.

I can't even believe you put all that effort into collecting those links. That literally does not change the fact that there was an industry-wide aversion to using the term "laptop".

Laptop is the same thing as notebook computer and Macbook is just a brand name while the actual device is laptop.

Well you can put it on your bare legs if you really want to and during general use doing so is fine but rendering video is going to make it get hot and very uncomfortable.

Place a W-series Thinpad on your bare legs on a warm summer day (think you can just still survive/think at all without A/C). You will want to get it away as you can't sweat well under it, and it's dumping some heat into your blood. That's it, despite some discomfort you won't mind it. And apparently the make them with more TDP than apple seems to be able to handle, but instead of an i9 they appear to jump directly to a Xeon. Those start at over $3500 in base configuration though....

If you look carefully, Apple (and probably other manufacturers) call their portable computers "notebooks" and not "laptops" because using a computer on your lap is bad both thermally and ergonomically.

Laptop and notebook is the same thing...

... except that one conspicuously avoids the assertion that one should use it on one's lap.

You never need to compile things sitting in a coffee shop? I do it at least two three times a week.

So you put it on your bare legs? That's a good do-it-yourself vasectomy alternative.

Sure but I put it on a table so the fans can work properly.

is compiling as taxing as video rendering though?

Sure, if the compiler is able to leverage all cores in an intensive task

Once, the place I was staying in was so cold that I put my laptop onto a CPU intensive task and stuck it in between the covers as a sort of hi-tech hot water bottle.

Shouldn't there be performance testing for common tasks (rendering video) like what Dave2D did? For a company like Apple that has a lot of resources, wouldn't it make sense to have that department that does just that, performance testing on specific apps. The testers can just focus on Apple software for ex (iMovie, Final Cut) and they could have caught this "bug" earlier.

There is and they should. But bugs tend to find a way anyway.

Yeah, but... let's be honest. This kind of integration flaw is a very... well, let's just say it's a very "Hewlett Packard" kind of bug, and the kind of thing we've come to not expect from "Apple Computer".

They're a small upcoming company, they're doing their best, let's cut them some slack.

Nice, I was going to buy the lowest spec CPU version of the 2018 yes or yes so seeing that the thermal issues are getting improved just affirms my decision. I think the 2018 might be the new 2015. Apple MacBooks are like classic wines, some years are better than others, but you'll pay through the nose for every vintage anyway ;)

My 2008 alu worked without a hitch for 10 years until a bad third party battery started to give problems together with the keyboard a month ago and I decided not to try to fix it anymore: that wasn't a bad investment.

That's my hope as well, as soon as I saw the iFixit report on the silicone cones on the keys I pulled the trigger on the 15" i9 32GB RAM 1TB SSD and I love it. This throttling issue worried me a lot but after the fix I am seeing it not fall below clock [0]. I came from a late-2013 which was a great laptop (though I know the 2015 was the sweet spot, I was going to upgrade to that if the 2018 didn't address my top issues but it did so I went for the 2018).

[0] https://news.ycombinator.com/item?id=17603760

That's sweet. I wasn't sure this fix would be able to tame the i9.

I’m glad it was just a software bug. It still seems like Apple’s quality control has gone way done over the last few years. My hypothesis is that there has been a change in culture as the company gets bigger with less emphasis on rigorous testing of the code?

It would be interesting to get some perspective from a longtime Apple employee on this.

It still seems like Apple’s quality control has gone way done over the last few years.

I doubt it. They've had problems for years... bulging batteries, bad solder, bendy iPhones, etc.

But, is the quality really going down, or is Apple just an easy target? I tend to think it's more of the later. I've had Lenovos and Dells for work and they've not been perfect.

That’s a good point, for me it’s more that I perceive that the quality of Apple quality control used to be very high, and now it’s coming down to earth. It’s this shift that has people upset I think not the fact that Apple products have more issues than those of other vendors.

A friend had one of those white G3 iBooks. It got replaced about five times, and I think they eventually gave him a successor G4 one. Apple's never been perfect.

My mom had one too. They put the GPU right under the wristrest, so that you'd crack the solder just by typing. Eventually they added a plastic collar around it but that only moved the stress to the motherboard.

I doubt QA percentage failures are going down, but rather volume is going up. A failure rate of 0.1% of 100,000 is 100, but for 100,000,000, then it's 100,000, which would increase the publicity of failures significantly.

The alternative is this issue came up, and they had a trade off to make with heat + noise vs cooler + quieter, and out of the gate they bet on cooler and quieter; with the backlash, they’re realizing it was the wrong bet, and fixing a “bug.” I have to imagine that basic stress testing would have caught this somewhere.

That’s true, but it seems like this issue doesn’t affect all MacBooks? And looking at the reddit thread someone posted here it does look like the behavior wasn’t actually throttling properly.

Great to see Apple address this so quickly. The throttling issues didn't effect me and my workload (iOS and Android development) but it clearly caused bit issues for those doing video editing.

I have installed the update and so far so good. I have only had this 2018 model for a little over a day but I am loving the keyboard changes and TrueTone display and performance for me is superb.

Thanks Apple, now back to keyboard and Touch Bar.

They did (quietly) roll out improvements[1] to the keyboard, though time will tell if they actually help enough...

1: https://www.theverge.com/circuitbreaker/2018/7/19/17590100/a...

They need to improve the key travel. If I wanted to type on a flat board, I'd get a Yoga Book.


I for one am much happier with the newer keyboard and now the old one feels weirdly mushy and gross.

I reckon keyboards are a matter of taste ultimately.

Yes, I like the 2017 keyboard. I'm not ecstatic over it but definitely usable for me. I also like the touchbar. Maybe I'm just weird.

TouchBar is quite amazing once you know the gestures, too. Tapping and holding the volume/brightness key and sliding it for super-precise volume or brightness control is an amazing feature.

> I reckon keyboards are a matter of taste ultimately.

People can debate about the feel of the keyboard but its fact that the new keyboards are louder compared to the 2015 keyboard. The noise gets annoying when you are working late night.

There is lots of people happily typing on Cherry MX Blue switches, notorious for being loud. I know some using IBM Model M which is even louder.

Even the noise is a subjective taste, ultimately.

Anecdotally, the 2018 key travel feels better than the 2017. But I can't stand the touch bar even after a year with it.

Does this impact all macbook pros with the touch bar, or just the 2018 models?

Only the recently released ones were affected.

OK, thanks for the answer instead of the downvote :P

Possible details on what went wrong: https://www.reddit.com/r/macbookpro/comments/91256u/optimal_...

TL;DR The CPU is stuck in a cycle of switching between minimum clock frequency and turbo speed.

I didn’t follow my own advice and I bought a new top spec i9 MacBook Pro. I’ve an iMac pro with a persistent issue that causes kernel panics. Apple is gathering information about the issue and I’ve had a few calls with an engineer. So far it isn’t fixed.

When I got my i9 mbp I noticed the fan was quite active and it was fast, but not significantly better than my 2014 mbp quad core. I just updated with the fix this morning and so far I notice way less fan usage and it did seem to be a bit snappier. I will run some benchmarks later and see where it lands. If it isn’t working I’m about ready to abandon platform. Sucks because my entire music studio is Mac based.

It says the fix is for High Sierra. For everyone running Mojave on the new MBP, will there be a fix for that as well, or was it previously patched in Mojave already?

As Mojave is in beta right now, I don't think Apple feels obligated to rush out a Supplemental update like they did for their production OS. Also not sure how they would have already patched in Mojave if they hadn't had it fixed in HS when the hardware was released.

I expect the next Mojave beta release to resolve this, but don't expect the next release to come faster due to this bug.

works ok for me after the update. Cinebench 1020 and CPU clock speed stays at 3ghz under load. Ran it several times and it's the same each time. 99fps for the GPU in cb opengl test, though that drops a couple of fps each time you run the test.

Temps still high, up to 95% during cb but I guess that's expected with this type of laptop body under extreme load and can be somewhat mitigated with external fans.

I was going to return this but will keep it now.

so still slower than older 2.6GHz i7

Other reason for thermal slowdown: https://news.ycombinator.com/item?id=16689070


> Yeah, and if this will make the Macbook run so hot it will bend...

That's a big "if". They're not removing the thermal controls, just making them work correctly.

Well, VW etc didn't remove emissions control either. They just changed them such that they work better in "benchmarks" at other costs.

Now these laptops work better in benchmark, but at what costs? Longetivity of the laptop or battery? Keyboard melting away?

Obviously it's not going to bend. That was a snarky comment. But you get my point.

> But you get my point.

I don't, really. They had a bug that caused the chips to throttle down way more than necessary. It's like VW having a car that won't go more than 20mph on a hot day because of some bad software - there's no reason that fixing the software will mean VW cars suddenly can't handle going normal 65mph speeds.

It sound like the thermals might be fine, but the interpretation of the metrics were faulty, in such a way that it made the controllers panic and think they needed to throttle the CPU and run the fans hard.

I understand your point but disagree. This fix should provide higher performance at the same temperature. A new risk isn’t being introduced. It sounds counter-intuitive that software can improve hardware performance with no tradeoffs and I’m not sure how to best get past that other than running your own tests and reading others’.

Well that's quite nice

If they are on the up and up and that really is the problem.

I hope it is though. I was getting really depressed at the idea of switching to something else.

Congrats to Apple for handling this correctly: not suing the guy. The bad news is this escaped your QA process. That never would have happened 10 years ago.

> That never would have happened 10 years ago

The Macbook Pro from 2008 literally had a thermal issue. It made some units not boot at all. And it required a logic board replace, not just a software fix.


"If the fans run at high RPMs, but the computer does not boot, there is probably an issue with the thermal sensors.

Late 2008/Early 2009 models feature a thermal sensor on the heat sink. First, ensure that the thermal sensor on the heat sink is plugged into the logic board. If it is, try replacing the heat sink. If this does not fix the problem, the logic board most likely needs to be replaced."

Yeah, 10 years ago we were just frying mainboards with the NVIDIA chips and insufficient cooling: https://www.google.com/amp/s/gizmodo.com/5061605/apple-confi...

Is this really a software bug or are they now going to have heat related failures by voiding Intel's specs?

Just running an already hot laptop hotter can't be good.

Even with the fix, I'm afraid you're still shilling out a lot of money for top hardware that can't be used to its maximum extent. Because of the thin form factor limitations, insufficient thermal design and possibly even something as silly as not using the best thermal paste.

And the system as a whole will still run quite hot for high CPU+GPU loads, so who knows how that affects the electronics and battery life longer term.

Please, Apple, just make it 1mm thicker already.

People bought them knowing that they were making tradeoffs in performance, but the real problem was that the computers were not even hitting their advertised specs.

They're still very good when you consider everything across the board, including the battery life, ecosystem and support, and apparently the best in some areas like SSD performance:

[0] https://www.macrumors.com/2018/07/13/2018-macbook-pro-fastes...

[1] https://9to5mac.com/2016/11/01/2016-macbook-pro-ssd/

I had a serious slowdown issue on my 2010 MacBook Pro that took me a few years to figure out. It started after I spilled tea on it. The system appeared to function normally, but was very slow. When I ran top as root, it showed that kernel process was always using one core (>100% out of 200). I tried reinstalling the OS, changing options and upgrading, to no avail.

Apparently, the OS does this to prevent thermal damage. It perceives that the CPU is overheating, so it sends one core into an idle loop to prevent it from doing real work.

Finally, I took it to an Apple repair shop whose diagnostics determined the CPU thermal sensor was not working. That didn’t immediately help. After much web searching, I found that a fix for this it was to completely disable the power management profile by deleting it. Now the machine works perfectly normally in terms of everything - heat, fans and performance. The machine seems to work identically to its original behavior, so I am puzzled at what function the power management profile performs.

"could drive clock speeds down under heavy thermal loads"

So now it just don't give a shit and fires away at the highest clock, without a care in the world about temperatures?

The clock speed should go down under "heavy thermal load", that's not a bug...

The CPU itself had plenty of thermal headroom, but was overdrawing the power system - which then thermally throttled itself way more aggressively and in a less-controlled manner. The update changes the curve to ensure that the CPU doesn't draw more power than the power system can handle for extended periods. The CPU will still downclock itself as needed if it experiences its own thermal overload.

Are you saying Intel CPUs now can adapt to power loss, instead of just not working? I doubt that very much. The CPU got too hot(easily reaching 100C) so there was definetly not "plenty of headroom", in fact, there was no headroom at all.

It's not a powerloss but voltage goes down. Modern VRMs can also signal their power capabilies and reserves.

If the CPU draws more current than the VRM can handle, this is usually okay within bounds for a very short time, after than you'll get dropping voltage when the VRM starts to self regulate and the CPU will downclock in response.

I doubt if this patch will decrease the lifespan of MacBook. Because for silicon chips, 10 degree Celsius rise in operating temperature will roughly decrease life span by half. If Apple relax the temperature constraint by 20 degree Celsius (e.g. from 50 degree to 70 degree), it is expected that the lifespan of some component will decrease by 75%.

Except macbooks almost never see 50, and rarely 70 degrees, its 80-100 all day when actually using cpu.

Oh my god. That number is unimaginable. Even if my full power desktop i7 can't reach this number. Chips will die very quickly in this temperature. I think Apple is having a real trouble.

Apple used an inadequately cooled VRM that's incapable of sustaining the power required to run the CPU at full speed. See the link below. https://www.reddit.com/r/macbookpro/comments/91256u/optimal_...

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact