Fixing Apple's Engineering in an Hour [video] | Hacker News Reader

Fixing Apple's Engineering in an Hour [video]

youtube.com

41 pointsdevxpy31 comments

To be fair isn’t this probably the case with just about every laptop? Kit it out with some crazy totally impractical cooling setup and you can squeeze out a bit more performance. Even after all the craziness they edged out enough performer to detect it on a stress test but probably not enough to materially impact user experience.

This video sort of misses the point that real-world design is balancing multiple factors, not just over-optimizing one factor at the cost of others.

People don’t buy a MacBook Air because their looking for the fastest processor performance processor. That’s not the point of the product or the market it’s made for.

Hmm, fair point, but I do expect the CPU and the fan to at least be to be connected!

The machine quickly jumps to 100 degrees on load, which is something I do care about, from a usability perspective. I have to regularly turn on the aircon to make the keyboard on my MacBook Pro even usable -- something I did not have to do on a ThinkPad T440p.

And then there's the chance of damaging internal components because of the sustained high heat over a long time.

I think the intention of the video, initially, was to "squeeze out a bit more performance.", but it ended up demonstrating some arguably obvious oversights in the thermal solution of a rather expensive machine.

I came here expecting an in depth view of Apple's Software Engineering practices... I am disappointed.

That would be Louis, not Linus.

Consumer hardware engineering these days is all about getting away as close to the edge as possible. And that'd be fine, if there was an option for better margins at a higher price. But there isn't...

Can't have that MBA beating the base spec MBP... -_-

Watch the whole video. They managed to keep the laptop ice cold with a custom heatsink, but even then only managed to eek out a 12% performance improvement. As it turns out, Apple actually knows what they're doing.

I think you misunderstood that. They hit the power limit, which just means after deciding the thermal limit Apple saw no reason to give it a CPU power supply that can go much beyond that. (Which might be either a hardware or a software limitation.)

From what I understand it's the CPU power management/frequency profile that is limiting it. Even when satisfying the thermal requirements they were still getting sub-par performance because presumably the CPU was being limited in software anyway.

A combination of good cooling + a tweaked power management profile to accommodate the better thermals should give a substantial performance increase.

> presumably the CPU was being limited in software anyway

Would this be in the OS or the bios (or whatever they call it on a Mac)?

If it's the former, I wonder if the test would have been more interesting if the MBA was running a different OS.

Firmware.

Doing it within the operating system is suicidal as a malfunction of the OS (whether accidental or due to malware) would force the machine to operate beyond its thermal limits and potentially damage itself.

The thermal management is done in userland. This is probably the same.

- https://twitter.com/tomcoates/status/1255170545366097922

This is another misunderstanding/misconnection. "kernel_task" is used to report "missing" CPU time so the CPU usage accounting remains consistent even when the CPU is throttled. That does not mean either the kernel or userland have any input to the thermal throttling functions.

There's also multiple levels of protection and throttling. At the lowest level, various ICs on the board (e.g. power regulators) will turn themselves off if they overheat. Somewhere above that, firmware will definitely exert /some/ control, to cover odd situations like the machine hanging while booting. Above that, the kernel or userspace can add another layer — but like each of the layers before, that layer can only lower the limits further.

Having thermal management in userland (and nowhere else) would be legal suicide. Imagine the lawsuits from people accidentally burning their legs...

Thanks for this.

I believe you are correct, although I did think that kernel_task was there to help cool the components down, and not simply as a devide to report throttled CPU usage consistently with unthrottled.

That said, I didn't mean to say that thermal throttling only happened in userland but rather that it did happen there, even though it might also happen elsewhere.

Within the context of this thread, this is relevant since in this case the CPU was not being thermally throttled as it was running cool, but something was still throttling the amount of power going to it.

I was merely suggesting that Nextgrid's affirmation that it must have been the firmware throttling the CPU is not necessarily true. Given that there is a software component to thermal management (or at least I thought so), it's only logical to assume that there may be a software component of whatever other throttling was occurring without it necessarily being suicidal as suggested.

Firmware, good luck overclocking a macbook

Apple ran that experiment for you. It’s called a 13” MBP.

> A combination of good cooling + a tweaked power management profile to accommodate the better thermals should give a substantial performance increase.

Give it a try and report back.

edit: Wow, HN has changed. There was a time when experimentation was encouraged here.

I don't think that's terribly useful. Regardless of how much processing power you can extract from the CPU, there is the parallel question of CPU lifetime.

You can see in the video how hot the CPU runs. If I remember correctly, a 20°C temperature increase halves the lifetime of electronics. "Normal" laptops tend to stick around 60°C-70°C. Apple apparently decided that halving or even quartering the CPU lifetime compared to that is fine.

That said, most MBA CPUs probably spend their CPU lifetime closer to idle. It'd be interesting to get temperature numbers for that. Still, it's worse than it could - and IMHO should - be.

> I don't think that's terribly useful.

The entire overclocking community would disagree with you.

The entire overclocking community also decided lifetime is secondary to performance, yet I still feel rather displeased with Apple deciding to make lifetime secondary to saving a few cents on a better cooling solution :D

If the CPUs on your macbook pros are dying, you're doing something very wrong. If they're not, then you're creating an issue where none exists.

> Give it a try and report back.

Modifying the thermal & power management profile is very difficult as it would require reverse-engineering the firmware.

I guess a temporary hack would be to mess with the current sensing circuitry to make the machine believe the CPU is drawing less current that it actually is and encourage it to give it more power.

They dropped the temperature under load by a good 10 degrees C just by milling a few mm off the heatsink so it actually made contact with the CPU... sure, that didn't increase the performance drastically, but it seems like a massive oversight on Apple's part

I have, but the fact the heatsink didn't event mount properly is kind of annoying... I agree, they know what they're doing, but I also wish they'd have designed better cooling.

12% is a pretty decent improvement.

The problem is that the 13" MBP is over twice as fast in many benchmarks, but people think "if they had just put a better fan/heatsink on the air" they would be the same speed.

I wonder how much faster the 13" Pro would be if they did a similar setup.

In this[0] iFixit teardown (of last year's 13-inch, they don't have one for this year yet, but I assume the inside layout hasn't changed much), the cooling setup is a lot more traditional with a heatpipe from the CPU to the fan. I don't think it'd get a lot of benefit from the watercooling that LTT did (at least, without Apple's power profile taking over and throttling).

The only reason I can think of that Apple set up the MacBook Air like that was because it was supposed to be fanless and at the last second they couldn't figure out a way to eek out performance, so they added a fan where it would fit. That doesn't sound much like Apple's engineering though, so I don't know why they did it.

0: https://www.ifixit.com/Teardown/MacBook+Pro+13-Inch+Two+Thun...

Yeah, I was kinda disappointed by the end. They really didn't gain that much performance.

Because Apple limited the CPU to only be able to use x amount of Watts, because otherwise the cooling/the power consumption would be outside of their "product design goals".

But they probably still advertise "latest Intel CPU with xyz GHz and blah blah amount of cores.". But that would be like buying a Ferrari with advertised "an engine capable of gazillion horsepowers", but having it limited to only be able to go up to 2000 RPM...

> But they probably still advertise "latest Intel CPU with xyz GHz and blah blah amount of cores."

From what I have seen of LTT, their biggest criticism of Apple is their tendency to emphasize the turbo boost clock speed when the CPU is nearly always being throttled (presumably below the base clock speed) due to the machine being unable to dissipate the heat.

If you want to see an example of Apple advertising the turbo boost speed in very large print, check out the "Overview" page on the 13" MacBook Pro.

I found it unreasonable that ipad keyboard is only $300. simultaneously, Windows is very bad OS: bitlocker, updates, userland, no solution for game anti cheats so every game has to run a special program to create anti cheat program, visual studio is terrible.

yet, in the long run I think the OS itself is beast!

Reply on news.ycombinator.com