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

[leadeater]

2 minutes ago, Bouzoo said:

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

1990 Thinkpad

 

2018 Thinkpad

[Bouzoo]

2 minutes ago, leadeater said:

1990 Thinkpad

 

2018 Thinkpad

 

Yeah, I can see where the designers went for internships.

[Drak3]

37 minutes ago, leadeater said:

not throttled down to base

If they're not going below base, they're not throttling.

[Dylanc1500]

10 minutes ago, leadeater said:

1990 Thinkpad

 

2018 Thinkpad

 

I'm going to be honest, that is probably my favorite aspect of some of lenovos products.

 

Now they just need to bring back that damn 3.5" floppy drive and internal power supply...

[Guest]

1 hour ago, Jack_of_all_Trades said:

 

That does not sounds very healthy for the chip ....

Worked fine for 3 years till I sold and upgraded it. 

[SenKa]

3 hours ago, VegetableStu said:

yeah there're problems with using grammarly on this forum (maybe also others using the invision engine). the suggestion so far has been to turn off grammarly when commenting on this forum ._.

Yep. Had the same issue on Cooler Master's forum. I just deleted it, as I don't use it anyways...

[justpoet]

The reality here is that the general population are not hardware enthusiasts.  The general population want thin and light, even if we'd rather have full power as enthusiasts.  Apple goes for that market, and that market generally likes them.

 

Apple does actually design quite well most of the time.  Are there exceptions, like the silly keyboard and the OLD GPU that wouldn't stay stuck on because the graphics partner wouldn't provide the chip in the package format they had designed for?  Of course.

 

But, the reality of today is that, while the typical PC will tune fans and the like to make you deaf to keep temps down a bit longer and keep the system less throttled, Apple keeps the fans quieter longer, optimizing for noise and battery life over raw power.  If I'm sitting in the living room working on the laptop around others, I appreciate the quieter fan curves, lighter weight, and longer battery life.  If I'm at my desk on my own, or gaming, I use SMC Fan Control and crank them up ahead of time, then never get to throttle temps (i7 in mine, not the impossible to cool in thin laptop i9).  I know to do that because I'm an enthusiast.

 

Sure, we could say it would be nice to have a "performance" or "silent" switch someplace to adjust that for end users…but even in Windows that has similar ways to manage power profiles, people don't actually use the feature…so why bother?

[Loote]

I see no problem with that video, the title is what it is, but Linus explains the problems with Macs which you can't deny. iMore might be right that there are products with similar problems, but they shouldn't aim to be better than the worst, but to be among the best. Ultrabooks with i9s are an upgrade over i7s for a very small amount of users, most producers put that kind of CPUs in bigger form factor devices, where they can be properly cooled and utilised. As unbelievable as it sounds, I believe Apple should make their performance oriented devices bigger, because you can't fool physics at this point and it'd be much easier way to get top performance from macOS than virtualisation on tower PC/big thick laptop.

[Atmos]

8 hours ago, DrMacintosh said:

Design, performance, battery life. 

 

Pick two. 

I wont lie, im about half a bottle of vodka in right now and it took me a solid like 5 minutes to grasp what ye were trying to say.

100% this though seems to be apples reasoning.

I dont know though off the top of my head if the macbooks even get that solid of battery life compared to competitors, because they certainly appear to have sacrificed about everything they can to have that clout worthy design. I can't deny though, apple products feel high end when you handle them at least lol. i could go one for literally paragraphs at this point lol but imma stop myself here; damn vodka lmao

-edit holy shit agreeing your post was really hard lmao

[Morgan MLGman]

15 hours 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.

Intel are being liars with their TDP ratings, however that doesn't make MacBooks throttle. Poor cooling design does.

What I'm getting at - it's not Intel's fault in the slightest, even if they are being assholes with their TDP numbers.

Apple has skilled, educated and experienced engineers designing those laptops and there's no way that they just slapped the cooler on and called it a day without any testing. They knew how their cooling solution performed before they started selling those MBP's. They just did not care.

[mr moose]

39 minutes ago, Morgan MLGman said:

Intel are being liars with their TDP ratings, however that doesn't make MacBooks throttle. Poor cooling design does.

What I'm getting at - it's not Intel's fault in the slightest, even if they are being assholes with their TDP numbers.

 

Why do people still tout this?  There is absolutely nothing erroneous with Intel's TDP figures.  There has not been a single case where Intel's chips have required more cooling to maintain Tj than the spec called for.  

[Morgan MLGman]

9 minutes ago, mr moose said:

Why do people still tout this?  There is absolutely nothing erroneous with Intel's TDP figures.  There has not been a single case where Intel's chips have required more cooling to maintain Tj than the spec called for.  

They're designed to be misleading and are measured differently to what AMD does with their TDP ratings.

9900K - 8C/16T (3.6/5.0GHz) has 95W TDP, 6700K - 4C/8T (4.0/4.2GHz) has 91W TDP. They're the same architecture... How is that rating accurate? Also, the 7800X which is a 6C/12T chip also based on the same architecture (it has a different cache architecture but that doesn't cause a drastic TDP increase) is rated for a TDP of 140W.

AMD's Threadripper 1920X has TDP rating of 180W and yet at stock settings it draws only 10W of actual power more than the 95W 9900K Note that it has four physical cores more...

[mr moose]

Just now, Morgan MLGman said:

They're designed to be misleading and are measured differently to what AMD does with their TDP ratings.\

 

What?  how are they misleading?  and why does it have to be thee same as AMD's? 

Just now, Morgan MLGman said:

9900K - 8C/16T (3.6/5.0GHz) has 95W TDP, 6700K - 4C/8T (4.0/4.2GHz) has 91W TDP. They're the same architecture... How is that rating accurate? Also, the 7800X which is a 6C/12T chip also based on the same architecture is rated for a TDP of 140W.

AMD's Threadripper 1920X has TDP rating of 180W and yet at stock settings it draws only 10W of actual power more than the 95W 9900K

 

You can't compare Intel's TDP to AMD's that not how it works.   They use different criteria and have different guarantees on performance.

 

Here's a reasonable write up, and it;s even written from the perspective of someone who doesn't like it:

 

https://www.anandtech.com/show/13544/why-intel-processors-draw-more-power-than-expected-tdp-turbo

 

[Morgan MLGman]

22 minutes ago, mr moose said:

What?  how are they misleading?  and why does it have to be thee same as AMD's?

To quote anandtech:

Quote

By Intel’s own definitions, the TDP is an indicator of the cooling performance required for a processor to maintain its base frequency. In this case, if a user can only cool 95W, they can expect to realistically get only 3.6 GHz on a shiny new Core i9-9900K. That magic TDP value does not take into account any turbo values, even if the all-core turbo (such as 4.7 GHz in this case) is way above that 95W rating.

So Intel can pretty much add more cores to their CPUs and simultaneously lower their base clock to keep the same TDP value as on previous gen despite the fact that those chips can draw a lot of power, even twice as much as previous 91W generations (based on the same architechture), don't you think it's made to mislead people who don't know the specifics of how TDP is measured? And that it even varies from vendor to vendor?

I prefer the way AMD gets their TDP figures because they're a bit more "realistic", considering that I'm pretty sure no 9900K owner runs it at its base clock of 3,6GHz and that fact makes the TDP value useless because you can't even realistically look at this number to buy a good-enough CPU cooler... If you bought a 95W cooling solution to use with a 9900K, you'd have a pretty bad time


To conclude what I mean - Intel's TDP rating is technically not wrong, meaning that it's possible that the 9900K running ONLY at its base clock can be considered a 95W chip. That doesn't change the fact that AMD's own 1800X with the same core & thread count and the same TDP rating draws 50W less of power when both run at stock settings. Both chips have the same base clock speed.
 

22 minutes ago, mr moose said:

You can't compare Intel's TDP to AMD's that not how it works.   They use different criteria and have different guarantees on performance.

I know that, but I can compare their TDP ratings to the actual power draw of those CPUs. And to be fair, you can't even compare Intel's TDP to Intel's TDP either.

7820X - same architecture (with a different cache structure), same core count and the same base clock has a 140W TDP
9900K - same architecture, the same base clock & core count has a TDP of 95W

Makes sense, right?

[RejZoR]

35 minutes ago, Morgan MLGman said:

They're designed to be misleading and are measured differently to what AMD does with their TDP ratings.

9900K - 8C/16T (3.6/5.0GHz) has 95W TDP, 6700K - 4C/8T (4.0/4.2GHz) has 91W TDP. They're the same architecture... How is that rating accurate? Also, the 7800X which is a 6C/12T chip also based on the same architecture (it has a different cache architecture but that doesn't cause a drastic TDP increase) is rated for a TDP of 140W.

AMD's Threadripper 1920X has TDP rating of 180W and yet at stock settings it draws only 10W of actual power more than the 95W 9900K Note that it has four physical cores more...

Iirc Intel measures TDP at base clock, AMD measures TDP at Turbo clocks.

[Morgan MLGman]

1 minute ago, RejZoR said:

Iirc Intel measures TDP at base clock

Yes, I put that in a quote in my second post.

[mr moose]

6 minutes ago, Morgan MLGman said:

To quote anandtech:

So Intel can pretty much add more cores to their CPUs and simultaneously lower their base clock to keep the same TDP value as on previous gen despite the fact that those chips can draw a lot of power, even twice as much as previous 91W generations (based on the same architechture), don't you think it's made to mislead people who don't know the specifics of how TDP is measured? And that it even varies from vendor to vendor?

Nope, I think it's carefully measured so when OEM and system designers use pl2 or insufficient cooling solutions they can't blame Intel.  It's pretty clear what they do and why they do it, how can something so clearly explained be misleading?

6 minutes ago, Morgan MLGman said:

 

I prefer the way AMD gets their TDP figures because they're a bit more "realistic", considering that I'm pretty sure no 9900K owner runs it at its base clock of 3,6GHz and that fact makes the TDP value useless because you can't even realistically look at this number to buy a good-enough CPU cooler... If you bought a 95W cooling solution to use with a 9900K, you'd have a pretty bad time

You are welcome to like anything better, but preferring the way AMD do things doesn't make Intel liars or misleading.   Also I don't know any coolers that are sold with watt ratings,  probably a very good reason for that.

6 minutes ago, Morgan MLGman said:

 

To conclude what I mean - Intel's TDP rating is technically not wrong, meaning that it's possible that the 9900K running ONLY at its base clock can be considered a 95W chip. That doesn't change the fact that AMD's own 1800X with the same core & thread count and the same TDP rating draws 50W less of power when both run at stock settings. Both chips have the same base clock speed.
 

It's not even technically not wrong, it's just not wrong full stop.  AMD express theirs differently.

6 minutes ago, Morgan MLGman said:

 

I know that, but I can compare their TDP ratings to the actual power draw of those CPUs. And to be fair, you can't even compare Intel's TDP to Intel's TDP either.

 

No you can't compare Intel's TDP spec to total power draw at some clock range above base.  In fact you can't even use Intel's TDP to calculate power draw beyond base clock.  Why would you? That's not what it's for.

6 minutes ago, Morgan MLGman said:

7820X - same architecture (with a different cache structure), same core count and the same base clock has a 140W TDP
9900K - same architecture, the same base clock & core count has a TDP of 95W

Makes sense, right?

Yes actually it does,  I don't even need to know specifically why they are different, All I need to know is that to maintain the base clock under almost full load on all cores I need a bigger cooler for thee 7820X.  Beyond that you are subject to all the variables of silicon lottery, motherboard VRM, cooling solutions flaws, PSY limitations, enclosure design etc. Given I am not designing a cooling solution for it I can buy one of the shelf and 99% sure it will be fine (like most are).

[leadeater]

22 minutes ago, Morgan MLGman said:

7820X - same architecture (with a different cache structure), same core count and the same base clock has a 140W TDP
9900K - same architecture, the same base clock & core count has a TDP of 95W

One is Skylake and the other is Coffee Lake, a bit of power efficiency was gained between those (not much). The larger quad channel memory controller and larger PCIe controller also uses more power, quad channel IMC would likely be the larger increase of all of these.

[Dylanc1500]

13 minutes ago, leadeater said:

One is Skylake and the other is Coffee Lake, a bit of power efficiency was gained between those (not much). The larger quad channel memory controller and larger PCIe controller also uses more power, quad channel IMC would likely be the larger increase of all of these.

Well, the 7820x also has two AVX-512 FMA units, the 9900K does not.

[leadeater]

1 minute ago, Dylanc1500 said:

Well, the 7820x also has two AVX-512 FMA units, the 9900K does not.

Forgot about that, should add far more towards peak power but it's another thing sucking power at base.

[Morgan MLGman]

27 minutes ago, mr moose said:

Nope, I think it's carefully measured so when OEM and system designers use pl2 or insufficient cooling solutions they can't blame Intel.  It's pretty clear what they do and why they do it, how can something so clearly explained be misleading?

Honestly, if it was for OEM's only, it wouldn't even need to be public information but is for a reason...

27 minutes ago, mr moose said:

Also I don't know any coolers that are sold with watt ratings,  probably a very good reason for that.

Entire Be Quiet!'s CPU cooler lineup has TDP ratings written on the boxes...

It used to be more common for CPU cooler manufacturers to use that metric on their coolers.

23 minutes ago, leadeater said:

One is Skylake and the other is Coffee Lake, a bit of power efficiency was gained between those (not much). The larger quad channel memory controller and large PCIe controller also uses more power, quad channel IMC would likely be the larger increase of all of these.

Yeah, you're probably right on that, I didn't take into account other perks of X299 over Z390

But it doesn't explain why does the 7800X (6-core with 100MHz lower boost clock and less cache) have the same TDP as the 7820X I mentioned

 

[Stefan Payne]

21 hours ago, firelighter487 said:

no it isn't. I didn't buy my MacBook for looks. 

You don't but most do.

Might ask Louis Rossmann about his usual clients.

There is only a small amount of people that really need an Apple Product.

Most of Louis Clients are either talked into it because some professor told them to or for prestige.

 

And what most people want/need is a device that can run a webbrowser and some kind of office program. And that's it.

The rest for looks or prestige.

In normie circles a MACBOOK is more better than a Thinkpad.

[mr moose]

7 minutes ago, Morgan MLGman said:

Honestly, if it was for OEM's only, it wouldn't even need to be public information but is for a reason...

Of course the spec is public,  its definitely public for a reason.   You know nearly every single chip has the Tj specs publicly listed?  It's hardly thee nefarious reason people seem to make out.

Quote

Entire Be Quiet!'s CPU cooler lineup has TDP ratings written on the boxes...

 

I wonder if they ever sold one that was 45W

 

 

[Morgan MLGman]

1 minute ago, mr moose said:

Of course the spec is public,  it definitely public for a reason.   You know nearly every single chip has the Tj specs public ally listed.  It's hardly thee nefarious reason people seem to make out.

I wonder if they ever sold one that was 45W

Intel actually used to depict TDP pretty much the same way as AMD does. They changed it not for OEMs, but for the fact that it looks better to customers that don't really know what that rating is - this is my problem with their TDP and why I said it's misleading. It's not misleading to me or you, because we actually have any kind of knowledge about the topic, that doesn't mean that it's not misleading at all.
I can give you several examples of politicians using precisely the same tactic to mislead people about data that they find only partially good for their agenda. They're not wrong, and you can't call them out for being wrong, but they also didn't give you the bigger picture on purpose that would change the way you interpret that data.

Fun fact: Intel actually tried to do TDP ratings a better way not so long ago (in the times of mobile Ivy Bridge & Haswell Y-Series CPUs), they created something called SDP (Scenario Design Power) which according to Intel represented "the average power consumption of a processor using a certain mix of benchmark programs to simulate "real-world" scenarios". But they're Intel so they ditched it.

Intel's definition of SDP was:

Quote

Intel's description of Scenario Design Power (SDP): "SDP is an additional thermal reference point meant to represent thermally relevant device usage in real-world environmental scenarios. It balances performance and power requirements across system workloads to represent real-world power usage. "

And IMO it's a more realistic metric. Or useful at all, unlike current TDP
 

[RejZoR]

Intel kinda is lying. What CPU, unless stupendously thermally limited runs at base clock? None. They all turbo boost the clocks, meaning not a single Intel CPU is running at specified TDP. Where AMD's, when under load they actually consume as much as it's stated and that is for the turbo clock. This info goes out the window if you use the ALL Turbo feature in BIOS which locks all cores to max turbo multiplier. But that's not stock experience so no one can be blamed for it.