iMore's reply to Linus on why macs are slower than PCs

[XenosTech]

Last time I touched a Mac Book was back in 2012-2013 when my friend was rendering stuff for class and it was uncomfortable to the touch, shit running a nintendo 64 emulator on that thing was hard as hell to do also, choppy stuttery mess.

 

8 minutes ago, 2FA said:

It does increase efficiency, but it also increases transistor density which affects heat output which as far as I'm aware does not scale linearly but I could be wrong on that, I'm not versed in thermodynamics.

If you cram more transistors in a smaller space you'll generate as much or more heat. Smaller transistors means less power required to switch them so technically they should produce less heat but if you cram more in a smaller space then you just effectively made it hotter or as hot as the bigger ones.

[Atmos]

9 hours ago, Trik'Stari said:

 

Which, again, can totally be done, but not without other, different compromises on the aforementioned noise or portability or dependency on AC rather than battery power.

is it just me or does this sentence make absolutely no f*cking sense at all. Now, im on two hours of sleep and a massive hangover, but still this makes literally no sense to me.

Wouldn't upping efficiency and lowering powerdraw of the chip as a whole while maintaining performance, which IS something that intel has done in the past, or at the very least gaining performance at the same powerdraw; how would that compromise noise? or require dependency on AC? Wouldn't it just mean it either draws less power, or exactly the same power? making no difference to noise, or infact lowering noise?

 

This "article" is FULL of broken logic like that, it reads a lot more like a blog post than anything else.

Quote

The vast majority of people don't want or need a literally water-cooled MacBook Pro. They want and need exactly the snappy feeling one in day to day use Linus mentions.

No, the vast majority of people DIDNT want that. the macbook pro used to be used by and marketed to overwhelmingly actual professionals who required it to perform well consistently on heavy workloads where ever they were. Since apple made the decision to move away from marketing to professionals and more towards marketing it as clout because it would make them more money, they have alienated their core audience for the product. Nowadays people who buy macbook pros don't require that kind of performance, but there was a time where they did, before the market for it shifted. Enthusiasts are upset because apple stopped catering to them in search of more money, because ultimately abandoning them made financial sense.

 

Quote

(Author) has been covering the personal technology industry for a decade...

and apparently has a fundamentally broken understanding of computers and computer design, and is instead willing to do anything to shill for apple as long as he stays on their gub-boy list and gets press materials and invitations to closed events. Reminds me heavily of the auto-journo industry. filled with paid shills and disingenuous yearnalists looking for easy gibs and vacation catered events. Thats all this "article" is.

[TheSLSAMG]

4 hours ago, firelighter487 said:

that still doesn't mean everyone who buys them buys them for looks. 

 

you are right. I was just saying maybe the thermals aren't too bad and you just need a custom fan curve. to be clear I would love if they went back to the 2012 design. the keyboards on those are SO much better than the new one's. 

Honestly, I'm not so sure about a fan curve being the silver bullet here. My MacBook Air 2015 has a 15W dual-core i5 and even with fan speed at max, AIDA64 will put that chip into the high 90s (and that's repasted). For bursts, the cooling solution is fine, but it's clear that Apple isn't building these things for long loads.

 

For the 15" MBPtb, this is the heatsink that is expected to do the heavy lifting for both CPU and GPU. The CPU is supposedly 45W, and I say "supposedly" because Intel has gotten screwy with their TDP ratings as of late. Coupled with the 35W Radeon Pro 560X, you're looking at that heatsink having to handle 80W+ of heat. I wouldn't mind seeing someone put MacsFanControl on it and max the fans out to see what it's capable of, but I suspect it's not going to change much since it will already run max fan speed under full load anyways.

 

It's not only Apple's fault though, since Intel wants to push the spiciest possible chips onto the market with hilariously optimistic TDPs. Can someone tell me how Intel managed to get the same TDP on the i9-8950HK as they did on the i7-7700HQ? The i9 has 2C/4T more, it has a higher base and boost clock (1GHz higher in fact) and a higher sustained clock assuming cooling and power aren't an issue (4.3GHz.) As far as I'm concerned, there's no justifiable answer beyond Intel being liars. 

 

[Ashley MLP Fangirl]

@TheSLSAMG my MacBook Pro 13" Mid-2012 will stay in the low 70's with the fans at max rpm. and stock paste. 

 

I also have a 15" Late 2011 with the radeon gpu disabled. and I mean physically disabled. i removed a resistor off the board so the GPU gets no voltage at all. a friend is currently borrowing it so I can't test it now but that thing with the fans at max rpm and stock paste got into the 80's when I tested it a few months ago, with the gpu disabled!

[Bouzoo]

41 minutes ago, TheSLSAMG said:

It's not only Apple's fault though, since Intel wants to push the spiciest possible chips onto the market with hilariously optimistic TDPs. Can someone tell me how Intel managed to get the same TDP on the i9-8950HK as they did on the i7-7700HQ? The i9 has 2C/4T more, it has a higher base and boost clock (1GHz higher in fact) and a higher sustained clock assuming cooling and power aren't an issue (4.3GHz.) As far as I'm concerned, there's no justifiable answer beyond Intel being liars. 

1. They are technically not liars, I am sure you can manipulate tests. Example: 7700K is 91W, 9900K is 95W. Yeah, not happening no matter what they say. 

2. Apple (and other OEMs) are using new parts because you gotta compete, Apple (since we're talking them hre) hasn't improved cooling in their Ultrabooks but is using much beefier CPUs. 2017 MBP that my gf has has an i5 (2C/4T), gets noticeably hot when downloading big files at high speeds when CPU needs to work a bit more. And they put an 8550U (double of C/T) inside in the 2018 model. Same cooling solution. Apple is to blame as well. 

[jagdtigger]

1 hour ago, TheSLSAMG said:

For the 15" MBPtb, this is the heatsink that is expected to do the heavy lifting for both CPU and GPU. The CPU is supposedly 45W

Now you poked up my curiosity, how high are the fins on that thing? My HP gaming notebook has  a 8750H and it can keep it at ~80°C while running p95 small FFT's and it stays at 2.4GHz....

[Trik'Stari]

2 hours ago, 2FA said:

It does increase efficiency, but it also increases transistor density which affects heat output which as far as I'm aware does not scale linearly but I could be wrong on that, I'm not versed in thermodynamics.

Not an issue if they'd slightly increase the thickness of their laptops and provide proper cooling solutions.

2 hours ago, Atmos said:

is it just me or does this sentence make absolutely no f*cking sense at all. Now, im on two hours of sleep and a massive hangover, but still this makes literally no sense to me.

Wouldn't upping efficiency and lowering powerdraw of the chip as a whole while maintaining performance, which IS something that intel has done in the past, or at the very least gaining performance at the same powerdraw; how would that compromise noise? or require dependency on AC? Wouldn't it just mean it either draws less power, or exactly the same power? making no difference to noise, or infact lowering noise?

 

This "article" is FULL of broken logic like that, it reads a lot more like a blog post than anything else.

No, the vast majority of people DIDNT want that. the macbook pro used to be used by and marketed to overwhelmingly actual professionals who required it to perform well consistently on heavy workloads where ever they were. Since apple made the decision to move away from marketing to professionals and more towards marketing it as clout because it would make them more money, they have alienated their core audience for the product. Nowadays people who buy macbook pros don't require that kind of performance, but there was a time where they did, before the market for it shifted. Enthusiasts are upset because apple stopped catering to them in search of more money, because ultimately abandoning them made financial sense.

 

and apparently has a fundamentally broken understanding of computers and computer design, and is instead willing to do anything to shill for apple as long as he stays on their gub-boy list and gets press materials and invitations to closed events. Reminds me heavily of the auto-journo industry. filled with paid shills and disingenuous yearnalists looking for easy gibs and vacation catered events. Thats all this "article" is.

I was drunk as hell at 4am and that sentence made as much sense as it does now. They basically are making the assumption that if the die gets shrunk, more cores will need to be put in (why? I have no fucking idea) and that would make it hotter.

 

Which is true, but the problem is, they wouldn't need to put more cores in. It would just be a smaller more efficient chip. Might run slightly hotter, might not, I don't know.

 

All of this by fans of a company that design a keyboard with the power button integrated in such a way that if you spill something on the keyboard, your power button doesn't work and your laptop is essentially dead.

 

Any other company? The power button is separate and still works, you can plug in a $5 USB keyboard and keep going until you can get it repaired. This is "thinking different", apparently.

[Atmos]

3 minutes ago, Trik'Stari said:

Not an issue if they'd slightly increase the thickness of their laptops and provide proper cooling solutions.

I was drunk as hell at 4am and that sentence made as much sense as it does now. They basically are making the assumption that if the die gets shrunk, more cores will need to be put in (why? I have no fucking idea) and that would make it hotter.

 

Which is true, but the problem is, they wouldn't need to put more cores in. It would just be a smaller more efficient chip. Might run slightly hotter, might not, I don't know.

 

All of this by fans of a company that design a keyboard with the power button integrated in such a way that if you spill something on the keyboard, your power button doesn't work and your laptop is essentially dead.

 

Any other company? The power button is separate and still works, you can plug in a $5 USB keyboard and keep going until you can get it repaired. This is "thinking different", apparently.

Yea i only found out about that the other night watching the rossmann drunk stream. Just insane the kind of things apple goes to, to achieve their aesthetic and cut costs, literally at the cost of functionality, time and time again. and people like the author will support them and cheer the whole time. Yay regression!

[Guest]

9 hours ago, leadeater said:

You just don't understand, they are designed to throttle so it's fine^*.

*Please ignore all MacBook Pro products before 2015 that didn't have this issue or was very minor.

My MacBook Air 2013 reached a nice 100 degrees on load and didn’t throttle. 

[BuckGup]

These threads make me hate the clickbait even more. It turns the forums into flamwars because they continue to use clickbait and titles to incite this kind of reaction

[mr moose]

8 hours ago, leadeater said:

Pointing out part of the problem that Intel CPUs produce more heat is completely valid however it's not helped by the reduction is chassis capability for cooling, compounding the issue is not exactly a smart idea. You can optimize the voltage profiles, frequency steps, fan curves as much as you like but those are not replacements for better cooling. That's a complimentary relationship, particularly important in laptops.

 

I don't think that is valid as I also don't think cooling solutions are complimentary.  They are very integral to the design, the only way thermal throttling is not the result of poor design is if Intel somehow refused to let apple read the product data sheet nor experiment with the CPU's before they went into mass production.

 

When the cooling solution is set to dissipate a maximum of 45Watts (regardless what the limitation is) out of the factory then the best you can expect from that product is base clocks in an average CPU.

[Trik'Stari]

2 hours ago, Atmos said:

Yea i only found out about that the other night watching the rossmann drunk stream. Just insane the kind of things apple goes to, to achieve their aesthetic and cut costs, literally at the cost of functionality, time and time again. and people like the author will support them and cheer the whole time. Yay regression!

Don't forget the complete and utter incompetence and nigh anti-consumer nature of the "genius" bar.

 

So genius they don't know how to repair their own fucking products properly.

[leadeater]

33 minutes ago, mr moose said:

I don't think that is valid as I also don't think cooling solutions are complimentary.  They are very integral to the design, the only way thermal throttling is not the result of poor design is if Intel somehow refused to let apple read the product data sheet nor experiment with the CPU's before they went into mass production.

 

When the cooling solution is set to dissipate a maximum of 45Watts (regardless what the limitation is) out of the factory then the best you can expect from that product is base clocks in an average CPU.

Well I was more implying that all those factors should be done, because while you can cool the chip you might want to have better control of it over the defaults to keep noise down for example. Apple is one of the few that actually attempts to do all 3, most brute force with big cooling which at least is hard to go wrong.

[SenKa]

10 hours ago, VegetableStu said:

are you using the Grammarly plugin, by chance? o_o

Yes, actually...

[Thaldor]

2 hours ago, mr moose said:

 

I don't think that is valid as I also don't think cooling solutions are complimentary.  They are very integral to the design, the only way thermal throttling is not the result of poor design is if Intel somehow refused to let apple read the product data sheet nor experiment with the CPU's before they went into mass production.

 

When the cooling solution is set to dissipate a maximum of 45Watts (regardless what the limitation is) out of the factory then the best you can expect from that product is base clocks in an average CPU.

I think Apple knows completely well that their cooling solution is complete garbage (just like other companies who try to stuff latest i7's in slim designs and those who even try to stuff i9's into laptops without extreme cooling solutions). It's just that Apple has shifted from engineering and design to design and marketing company and any complains and cries from engineering are thrown into folder "Z" and forgotten right away because it's far more important to have slim and good looking design and the most craziest numbers on the paper for marketing than compromising any of it for better performance. In general, putting those stove-CPUs to any laptop which has in any point "slim" written in it's design is just awful design.

 

My new favorite joke argument is "because it must work snappier". Like WTF, how much does i9 bring to the table compared to i7 in means of loading Facebook or opening the browser? Do we count milliseconds o even nanoseconds? I would guess latency and netspeed play far bigger role in that than the performance gain from i7 to i9. And in opening software I would guess "snappier" feel would be gained from RAIDing two SSDs than upgrading from i7 to i9 (I would argue that the moving SSD controller from the SSD itself to T2-chip made as much latency as could be gained from the upgrade of CPU). Of course unless the computer is already digging its own grave from poor memory management and 101 services running in the background.

[DrMacintosh]

14 hours ago, leadeater said:

Please ignore all MacBook Pro products before 2015 that didn't have this issue or was very minor.

That never happened. Those models “overheated” as well. 

[DrMacintosh]

6 hours ago, Atmos said:

is it just me or does this sentence make absolutely no f*cking sense at all.

Design, performance, battery life. 

 

Pick two. 

[leadeater]

4 minutes ago, DrMacintosh said:

That never happened. Those models “overheated” as well. 

They got hot, not throttled down to base or below easily. They were also designed to be hot for noise reasons. Hot isn't specifically bad unless it's having other impacts to the system.

 

3 minutes ago, DrMacintosh said:

Design, performance, battery life. 

 

Pick two. 

Older MacBook Pros had all 3.

[DrMacintosh]

6 minutes ago, leadeater said:

Older MacBook Pros had all 3.

I don’t believe that. 

[Dabombinable]

2 minutes ago, DrMacintosh said:

I don’t believe that. 

A shitload of other similarly priced laptops also have all 3.

[leadeater]

5 minutes ago, DrMacintosh said:

I don’t believe that. 

You can not believe it all you like but I've used all of them from that era, 13"/15"/17", and they were all damn excellent. They got hot, often very hot but not to the point where Final Cut would be noticeably slower. There were some issues with bad performance due to failing thermal paste but that could be fixed under Apple Care.

 

As much as I don't like Mac OS their hardware back then was only rivaled by Lenovo, which style wise looked like ass in comparison not that I minded that.

[Fnige]

19 minutes ago, DrMacintosh said:

Design

Im guessing by that you mean thin and light?

[Ashley MLP Fangirl]

23 minutes ago, DrMacintosh said:

That never happened. Those models “overheated” as well. 

only because of fan curve though. my mid-2012 with the stock paste with max rpm on the fan gets to 70 degrees and stays there. 

[Ashley MLP Fangirl]

6 minutes ago, leadeater said:

which style wise looked like ass in comparison not that I minded that.

looked like ass? still looks like ass lol. 

[Bouzoo]

13 minutes ago, firelighter487 said:

looked like ass? still looks like ass lol. 

I'd insert a Lenovo designer joke here, but they are non-existent, just like Lenovo designers.