Intel caught fudging benchmarks in M1 vs Core i7 11th Gen comparison

[D13H4RD]

1 hour ago, AndreiArgeanu said:

Can someone genuinely explain to me why would anyone care about Office 365 performance? Like cool, it can export PowerPoint to PDF and sort stuff in excel a little faster but can you actually tell the difference when using it?

It's a longtime metric used to gauge typical office productivity performance and has usually been where Intel does fine in. 

 

I'd also argue that it's largely pointless as many would also deem a Core i3 machine from 2011 to also be sufficient for Microsoft Office. 

[saltycaramel]

So, now that we have both sides of the story, is it worse

- to post (as the underdog) a generic-but-honest performance-per-watt graph like Apple did in November (making Linus lose his temper for some reason, or better for some clickbaity reason)

- to post (as the incumbent) super specific cherry picked figures that probably mean next to nothing

 

?

[leadeater]

3 hours ago, saltycaramel said:

So, now that we have both sides of the story, is it worse

- to post (as the underdog) a generic-but-honest performance-per-watt graph like Apple did in November (making Linus lose his temper for some reason, or better for some clickbaity reason)

- to post (as the incumbent) super specific cherry picked figures that probably mean next to nothing

 

?

Two companies with garbage marketing doesn't create one company with good marketing 

 

Two things can be shit, do we really need to compete for who's is worse? In keeping with this day and age both get a participation award and everyone is a winner.

[Video Beagle]

1 hour ago, saltycaramel said:

So, now that we have both sides of the story, is it worse

- to post (as the underdog) a generic-but-honest performance-per-watt graph like Apple did in November (making Linus lose his temper for some reason, or better for some clickbaity reason)

- to post (as the incumbent) super specific cherry picked figures that probably mean next to nothing

 

?

 

I think 2nd....

 

I don't understand this as a move. Intel had to know they'd get called out for their benchmarks being...dodgy..if not dishonest.....why make the story be about that? do the comparisons against their last generation to tout their improvements...sure, "new media" like LTT will want to do "How does intel compare to the M1!" videos, but that's not a story that mainstream are going to care about. They're going to look at PC World and other "top level" tech media to get the info they'll break down for their headlines. So instead of ..frobes let's say, reporting "Intel's new chip is 100times faster than last year's model" they'll see "intel fudged benchmarks"....just feels like a weird self score.

[RedRound2]

9 hours ago, leadeater said:

So does Final Cut for Intel CPUs, that's what makes it so good compared to software intended to run on a generalized non standardized set of hardware so focusing in on a single set of hardware acceleration features not actually found on higher end workstations is pointless.

What? Final Cut runs a lot faster in M1 compared to last gen Intel Macbooks, so I dont get your point. What I said was that comparing final cut on Mac and premiere pro in Windows is the kind of tests that would introduce a variability in software optimizations, similar to what Intel did in one of the tests. That's not really a comparison you use to potray the superiority of the chip and it's intel trying to be disingenuous at best

Quote

It's not translation or emulation, if you run Linux natively or by way of a hypervisor then what is being said is the performance is significantly worse and has basically nothing to do with drivers or the like at all, especially when running from within a hypervisor (I don't know I haven't personally checked but what was being said isn't what you are thinking). The performance difference will 100% be software optimization related, M1 is a new and very different hardware architecture and it is extremely unlikely any Linux software packages have been optimized for it in any way. This situation can of course be remedied by those that choose to do said optimization.

Can you give me a source for this? Becasue all the articles on Linux running on M1 talks about how the non standard components of the Mac an the lack of any proper drives the reason why it isnt as usable in the current state

Quote

Also saying MacOS and Apple software are better optimized for the hardware they run isn't a negative or a criticism, it's more praise than anything else. It's just context that does need to be considered when talking about other hardware, operating systems and software.

 

7 hours ago, D13H4RD said:

So are we just going to throw away just how much effort Apple's software engineers put into making sure a lot of the M1's potential actually translate to strong real-world performance? One of Appls's key pièce de résistance is the really tight integration between hardware and software, now being much more so thanks to in-house silicon. You can keep parroting the "oh, Apple's fast because they're brute forcing it" phrase, but it doesn't mean Apple's software work is no-less important. The fact that a whole new architecture with a different instruction set hasn't been a right mess and even runs emulation without much issue is nothing short of remarkable.

Can you re-read what I wrote before. I never disregarded Apple's software optimizations. That's the only reason Macs especially with the new M1 chips are very snappy and "iPad like" to use. Their first party softwares does a great job taking advatage of the all the hardware and software APIs that accelerate performance. 

 

Circling back, what I did say was that brute benchmarks like running geekbench or some PDF exporter is measure of raw performance (unless it makes use of accelerator in one and doesnt in the other) that has nothing to do with Apple's optimization. The guy I replied to attributed all significant performance gains to just "software optimizations" which isnt true at all. 

[D13H4RD]

34 minutes ago, RedRound2 said:

Circling back, what I did say was that brute benchmarks like running geekbench or some PDF exporter is measure of raw performance (unless it makes use of accelerator in one and doesnt in the other) that has nothing to do with Apple's optimization. The guy I replied to attributed all significant performance gains to just "software optimizations" which isnt true at all. 

Maybe you came to a different conclusion, but my takeaway from his comment wasn't that all of Apple's gains came from their tight software:hardware integration but that the M1 Macs had a distinct advantage of their own because they are running in an environment that's optimized with the chip in mind, and that people arguing against Intel picking suites that are optimized for their system shouldn't discount that sheer fact alone.

 

It's not that all of Apple's gains came from brilliant optimization, but a good chunk of it does, and that is something that isn't lost on anyone. This is exactly how Apple manages to extract the best performance out of the hardware they have, even more so when the core hardware is their own doing. The chip is fast but a fast chip only really begins to truly shine when it is mixed with software that knows how to properly utilize it, which is what I think is the point he's conveying.

[leadeater]

2 hours ago, RedRound2 said:

What? Final Cut runs a lot faster in M1 compared to last gen Intel Macbooks, so I dont get your point.

The point was Apple specifically optimized Final Cut for Intel CPUs and it is this optimization that makes it so efficient at the task, this optimization Apple did is what allows it to do so well. This has also been done for Apple's M1 SoC. You said optimization has nothing to do with it, however it very much does. Adobe has brought in similar support for Intel iGPU acceleration in to Premier and it shows significant performance gains.

 

But you were also criticizing Premier while discounting the specific advantage Apple and Final Cut has, this is a logical contradiction to your complaint about Intel using Topaz Labs. The issue is neither that it's not fair to use either Topaz Labs or Final Cut but more rather the limited sample size, if you only compare against one thing and pick the best case possible for a particular hardware product then it's a worthless comparison for general hardware performance comparison. That's why with independent benchmarks you see Blender, V-Ray, Premier etc etc so there is not a sample size of one.

 

2 hours ago, RedRound2 said:

What I said was that comparing final cut on Mac and premiere pro in Windows is the kind of tests that would introduce a variability in software optimizations, similar to what Intel did in one of the tests. That's not really a comparison you use to potray the superiority of the chip and it's intel trying to be disingenuous at best

That isn't what Intel has done here though, Intel choose software and ran them on both. I don't know what relevance this has to Intel being disingenuous, which they are being, but this is not it. I see zero comparisons to Final Cut in their presentation and data.

 

You complained that Intel choose software that was optimized for Intel CPUs, while that is true so is Final Cut for both Intel and M1. The problem is independent reviewers have already tested Final Cut Intel vs M1 so there isn't any new ground there for Intel to tread and it's not a comparison they come out on top either.

 

It however is completely fair to choose software that is best optimized for the hardware otherwise it would not be fair to compare Final Cut to Premier at all and last I checked that is a completely fair comparison to do. Final Cut isn't always better and can't match everything Premier can do and the shortcomings of Final Cut don't take away anything from it's good points. It's the purpose of the comparison that matters.

 

A Swiss Army Knife may well be highly versatile but if all you need is a screw driver then you're better off with a screw driver, certainly don't hope your one has one either because that would make you look stupid ( *cough* The Verge *cough*).

 

2 hours ago, RedRound2 said:

Circling back, what I did say was that brute benchmarks like running geekbench or some PDF exporter is measure of raw performance (unless it makes use of accelerator in one and doesnt in the other) that has nothing to do with Apple's optimization. The guy I replied to attributed all significant performance gains to just "software optimizations" which isnt true at all. 

I saw no such thing being done, nowhere was all of it attributed to software optimization.

 

But before you jump back to "there are no optimizations" even just comparing Spec2006 Linux vs Windows there is around a 5% performance difference, I certainly think there is room for Apple to optimize Mac OS very well for the M1 that there would be performance advantage to running the same software under Mac OS compared to Windows.

 

2 hours ago, RedRound2 said:

Can you give me a source for this? Becasue all the articles on Linux running on M1 talks about how the non standard components of the Mac an the lack of any proper drives the reason why it isnt as usable in the current state

The only driver issue for the M1 is graphic drivers, if you want to run CPU tests you can do so just fine. However like I said doing so without updating the Linux Kernel and software for the M1 would be a waste of time and highly uninformative. The Linux Kernel had a lot of updates to properly support Zen and subsequent generations of Zen, each new Intel architecture also gets patched by Intel too to properly support it. Running performance tests without doing this isn't a simple matter of just saying it's showing the most basic raw performance of the SoC either, by not properly building in the correct support for the architecture performance could be unfairly impacted and not at all represent the performance of it.

 

Remember I said I haven't looked in to performance comparisons at all, my comment was explaining to you your misunderstanding of what was said. However there are tutorials of functioning and usable Linux installs for the M1. Those are more targeted to running it natively, you can run it far easier as a VM but you'll get endless criticism if you try and give benchmark data doing that regardless of how little it may or may not impact the performance, kind of fair enough though. You can't know the impact without comparing to native either.

[Taf the Ghost]

9 hours ago, Arika S said:

ITT: People who for some reason STILL think that first party benchmarks can be trusted. Every single manufacturer cherry picks things that make them look the best in their press releases.

There's cherrypicking, then there's Recent Intel. Though I'm not quite sure it tops Peak Performance Nvidia slides. Everything is 2X improved or better. Even the badness of the charts.

[leadeater]

7 minutes ago, Taf the Ghost said:

Though I'm not quite sure it tops Peak Performance Nvidia slides. Everything is 2X improved or better. Even the badness of the charts.

2x the leather jackets over the last generation, SLI jacket belt costs extra.

[Jet_ski]

21 minutes ago, leadeater said:

2x the leather jackets over the last generation, SLI jacket belt costs extra.

Many have talked about the kitchen and the jacket. But who noticed Jensen’s bust in the RTX announcement video?! The ego.

[D13H4RD]

1 hour ago, Taf the Ghost said:

There's cherrypicking, then there's Recent Intel. Though I'm not quite sure it tops Peak Performance Nvidia slides. Everything is 2X improved or better. Even the badness of the charts.

2X the spatulas rendered with RTX on, enhanced with DLSS. 

[WereCat]

40 minutes ago, Jet_ski said:

Many have talked about the kitchen and the jacket. But who noticed Jensen’s bust in the RTX announcement video?! The ego.

I had no idea that he got busted

[Aereldor]

On 2/7/2021 at 4:53 PM, Metallus97 said:

HEHEHEHE wellll you just cant compare em.

But for the normal user the M1 busts some serious a. 

I thik that this machine is way way better than any current normal laptop on da planet 

Except hardcore gaming

You can actually game on the M1 with crossover and it does surprisingly well. There are people running Witcher 3, 1080 30fps high settings

[Spindel]

8 hours ago, AndreiArgeanu said:

Can someone genuinely explain to me why would anyone care about Office 365 performance? Like cool, it can export PowerPoint to PDF and sort stuff in excel a little faster but can you actually tell the difference when using it?

I care about excel performance. Just using some formulas on large datasets can take minutes to calculate when one value changes. Might not sound like much but the longer the wait the more disruption in your workflow. 
 

And before people comment that I shouldn’t use excel for large data sets: it is the tool I have and it does the job. 
 

The most extreme one is my monster file that went from 7 min to complete on my old iMac from 2013 (same on my windows laptop from 2019, but that one is unresponsive in the process) to 2 min under rosetta on my M1 mini. 

[RedRound2]

5 hours ago, D13H4RD said:

Maybe you came to a different conclusion, but my takeaway from his comment wasn't that all of Apple's gains came from their tight software:hardware integration but that the M1 Macs had a distinct advantage of their own because they are running in an environment that's optimized with the chip in mind, and that people arguing against Intel picking suites that are optimized for their system shouldn't discount that sheer fact alone.

 

It's not that all of Apple's gains came from brilliant optimization, but a good chunk of it does, and that is something that isn't lost on anyone. This is exactly how Apple manages to extract the best performance out of the hardware they have, even more so when the core hardware is their own doing. The chip is fast but a fast chip only really begins to truly shine when it is mixed with software that knows how to properly utilize it, which is what I think is the point he's conveying.

I still dont get how any suite of benchmarks that Intel ran on the M1 were blessed by the software optimization in macOS. A purely mathematical calculation cannot be further increased in speed because of the operating system. Where you could argue about OS optimization is like how some Youtubers do a real world speed test on phone, booting the phone, opening up diffferent applications, closing them, keeping in memory, etc where an iPhone with measily 4/6GB RAM ends up being better that 8/12GB RAM android phone

 

I dont see how running a PDF extractor, or calculating data in excel, encrypting (both of which have dedicated accelerators) has anything to do with macOS optimizations

[RedRound2]

4 hours ago, leadeater said:

The point was Apple specifically optimized Final Cut for Intel CPUs and it is this optimization that makes it so efficient at the task, this optimization Apple did is what allows it to do so well. This has also been done for Apple's M1 SoC. You said optimization has nothing to do with it, however it very much does. Adobe has brought in similar support for Intel iGPU acceleration in to Premier and it shows significant performance gains.

 

But you were also criticizing Premier while discounting the specific advantage Apple and Final Cut has, this is a logical contradiction to your complaint about Intel using Topaz Labs. The issue is neither that it's not fair to use either Topaz Labs or Final Cut but more rather the limited sample size, if you only compare against one thing and pick the best case possible for a particular hardware product then it's a worthless comparison for general hardware performance comparison. That's why with independent benchmarks you see Blender, V-Ray, Premier etc etc so there is not a sample size of one.

 

That isn't what Intel has done here though, Intel choose software and ran them on both. I don't know what relevance this has to Intel being disingenuous, which they are being, but this is not it. I see zero comparisons to Final Cut in their presentation and data.

 

You complained that Intel choose software that was optimized for Intel CPUs, while that is true so is Final Cut for both Intel and M1. The problem is independent reviewers have already tested Final Cut Intel vs M1 so there isn't any new ground there for Intel to tread and it's not a comparison they come out on top either.

 

It however is completely fair to choose software that is best optimized for the hardware otherwise it would not be fair to compare Final Cut to Premier at all and last I checked that is a completely fair comparison to do. Final Cut isn't always better and can't match everything Premier can do and the shortcomings of Final Cut don't take away anything from it's good points. It's the purpose of the comparison that matters.

 

A Swiss Army Knife may well be highly versatile but if all you need is a screw driver then you're better off with a screw driver, certainly don't hope your one has one either because that would make you look stupid ( *cough* The Verge *cough*).

Did you not read my earlier comment before? I said that comparing premiere pro and final cut is an example of a bad comparison, where the former beats the latter just due to it's better optimizations. Drawing parallels to why the Topaz Labs AI optimization is bad example since it makes use of dedicated harware accelerator on Intel's chip, but doesnt make use of the Neural Engine on the M1. That is Intel being disingenuous at best. 

 

Ideally to have showcase their product, they should have chosen benchmarks that forces the chips to brute force the solution and see the raw speed. Anyway, even if any of these benchmarks have any sort of truth to it, its quite pathetic that Intel's latest and greatest just beats Apple's lowest end, designed to be fanless chips

4 hours ago, leadeater said:

I saw no such thing being done, nowhere was all of it attributed to software optimization.

 

But before you jump back to "there are no optimizations" even just comparing Spec2006 Linux vs Windows there is around a 5% performance difference, I certainly think there is room for Apple to optimize Mac OS very well for the M1 that there would be performance advantage to running the same software under Mac OS compared to Windows.

 

The only driver issue for the M1 is graphic drivers, if you want to run CPU tests you can do so just fine. However like I said doing so without updating the Linux Kernel and software for the M1 would be a waste of time and highly uninformative. The Linux Kernel had a lot of updates to properly support Zen and subsequent generations of Zen, each new Intel architecture also gets patched by Intel too to properly support it. Running performance tests without doing this isn't a simple matter of just saying it's showing the most basic raw performance of the SoC either, by not properly building in the correct support for the architecture performance could be unfairly impacted and not at all represent the performance of it.

Precisely why I said that the lack of development and proper kernal support impacts performance, while the original comment I replied to claimed that Apple's CPU was only fast because macOS had magical optimization (which in reality is basic harware support and nothing groundbreaking that would lead it to export PDF faster)

 

[D13H4RD]

1 hour ago, RedRound2 said:

I still dont get how any suite of benchmarks that Intel ran on the M1 were blessed by the software optimization in macOS. A purely mathematical calculation cannot be further increased in speed because of the operating system. Where you could argue about OS optimization is like how some Youtubers do a real world speed test on phone, booting the phone, opening up diffferent applications, closing them, keeping in memory, etc where an iPhone with measily 4/6GB RAM ends up being better that 8/12GB RAM android phone

It's not just macOS but also the individual programs. Both the M1 and TGL-U processors feature a host of accelerators that benefit certain applications especially if they are written to take advantage of them. A lot of the M1's impressive performance comes down to its acceleration hardware alongside applications coded to take good advantage of it.

[Metallus97]

4 hours ago, Aereldor said:

You can actually game on the M1 with crossover and it does surprisingly well. There are people running Witcher 3, 1080 30fps high settings

whuuuuut? U got a link for this?

Thats duble emulation lol 

[Spindel]

45 minutes ago, Metallus97 said:

whuuuuut? U got a link for this?

Thats duble emulation lol 

https://www.techarp.com/games/apple-m1-witcher-3-gaming/?amp=1
 

[leadeater]

7 hours ago, RedRound2 said:

Precisely why I said that the lack of development and proper kernal support impacts performance, while the original comment I replied to claimed that Apple's CPU was only fast because macOS had magical optimization (which in reality is basic harware support and nothing groundbreaking that would lead it to export PDF faster)

No I still disagree that was being said. You've made similar comments yet because it's come from you it's fine, however if someone else says similar things then you get very defensive and run down a rabbit hole and argue about something that actually wasn't said at all.

 

Just because someone points out that there is an optimization advantage using an OS that is not at all saying that it's only good due to that, in a highly completive market where products are very close in performance getting a 5% gain (re: Spec2006) could be the difference between being faster or slower.

 

And it really is not as simple as saying they are just running math and it cannot be different, if that were actually true and that simple the Spec2006 difference between Windows and Linux would be 0% but it's not. Memory management and process management along with other specific software optimization for each architecture do make a difference and there can be differences between OS's even by the very same software developers because the OS's are just simply different. 

[RedRound2]

12 hours ago, D13H4RD said:

It's not just macOS but also the individual programs. Both the M1 and TGL-U processors feature a host of accelerators that benefit certain applications especially if they are written to take advantage of them. A lot of the M1's impressive performance comes down to its acceleration hardware alongside applications coded to take good advantage of it.

Both Intel and M1 have dedicated accelerators for certain applications. And I agree that is an important metric when choosing platform. That was never the issue. But when Intel decides to use a software that specifically took advatage of their accelerator on their processor and did not make use of the NPU in the M1, that's where the issue arises (they could've either ommitted it, or found a softare that made use of both). And then going on screaming about how they achieved some insane performance is being disingenous

9 hours ago, leadeater said:

No I still disagree that was being said. You've made similar comments yet because it's come from you it's fine, however if someone else says similar things then you get very defensive and run down a rabbit hole and argue about something that actually wasn't said at all.

Can you give some context on things you talk about. You dont even read my comment properly and then make up some BS I never stated. Precisely like your last reply on Final Cut when I wrote about how it was similar and a bad example

 

It's as simple as this, he said all of M1's perforamance come from mostly from software optmizations - which is not the case. Sure software optmizations play a role in certain cases like comapring FInal cut and premiere performance across same or different platforms. The tests that I could make out from Intel's charts were pure brute force kind of tests (except they kept changing laptops as per their convinience and forgot to mention that one of them made use of Intel's dedicated accelerator, while the other one didn't - even though it was available)

9 hours ago, leadeater said:

Just because someone points out that there is an optimization advantage using an OS that is not at all saying that it's only good due to that, in a highly completive market where products are very close in performance getting a 5% gain (re: Spec2006) could be the difference between being faster or slower.

And we dont really see a 5% difference in those benchmarks, do we? Those 5-10% could be due to OS overhead, but that clearly doesnt seem to be the case

9 hours ago, leadeater said:

And it really is not as simple as saying they are just running math and it cannot be different, if that were actually true and that simple the Spec2006 difference between Windows and Linux would be 0% but it's not. Memory management and process management along with other specific software optimization for each architecture do make a difference and there can be differences between OS's even by the very same software developers because the OS's are just simply different. 

Same as above

[leadeater]

2 minutes ago, RedRound2 said:

It's as simple as this, he said all of M1's perforamance come from mostly from software optmizations

Well first before anything else, show me where in his words, then we can continue.

[leadeater]

25 minutes ago, RedRound2 said:

Can you give some context on things you talk about. You dont even read my comment properly and then make up some BS I never stated. Precisely like your last reply on Final Cut when I wrote about how it was similar and a bad example

Like I said in my reply but maybe you just didn't understand, it's not similar. Intel used the EXACT same software on both systems, the difference were only the OS and the hardware. That isn't anything like Final Cut vs Premier example you gave nor to your statement of "Similar to what Intel did in one of their tests".

 

As I mentioned in that post Final Cut vs Final Cut both on Mac OS (obviously) would be a much better test but it's already been done and Intel does not have a performance advantage with that software so clearly they are not going to use it. Why would a company trying to show off superior aspects of their product choose a test where they are behind, that's illogical from their standpoint. This isn't supposed to be objective performance comparisons, that's not what this marketing nonsense is for, anyone thinking or expecting it to do so is gravely mistaken.

[RedRound2]

4 hours ago, leadeater said:

Well first before anything else, show me where in his words, then we can continue.

This whole paragraph below goes into great lengths on how software optimizations somehow made things faster on M1 machine which is far from true

On 2/7/2021 at 9:39 PM, tim0901 said:

For example, MacRumors is quick to point out that, yes, Topaz Lab's apps are designed to take advantage of Intel's hardware acceleration suite. But they neglect to mention, as the original PC World article pointed out:

Quote

This is the same magic that’s made Apples phones shine for so long. 

Which is completely true. This is exactly how Apple is squeezing so much performance out of their processors (M1 included) in the first place - that tight integration between software and hardware goes a long way to boosting Apple's position here. Intel's chips - running plain old windows - don't get to benefit from that level of integration, which does hurt their position when making direct comparisons. As those who have been playing around with the M1 have found out, performance is lost when running other operating systems on M1 Macs - far beyond the level expected from being run through a hypervisor. It's a bit disingenuous to criticise Intel for choosing one benchmark where they have a hardware-level advantage, but to then neglect to mention when Apple has that same advantage for pretty much all of the other tests conducted. If you're going to criticise Intel, you have to at least acknowledge Apple's position as well.

OS optimizations plays into RAM management, OS snappiness, launching apps, boot up etc. It plays very little into brute force application that use math to do tasks. And both Windows and macOS are matured operating system to not have some random heavy background windows/macOS process have more priority over the task the active window is doing

4 hours ago, leadeater said:

Like I said in my reply but maybe you just didn't understand, it's not similar. Intel used the EXACT same software on both systems, the difference were only the OS and the hardware. That isn't anything like Final Cut vs Premier example you gave nor to your statement of "Similar to what Intel did in one of their tests".

 

As I mentioned in that post Final Cut vs Final Cut both on Mac OS (obviously) would be a much better test but it's already been done and Intel does not have a performance advantage with that software so clearly they are not going to use it. Why would a company trying to show off superior aspects of their product choose a test where they are behind, that's illogical from their standpoint. This isn't supposed to be objective performance comparisons, that's not what this marketing nonsense is for, anyone thinking or expecting it to do so is gravely mistaken.

Do you have some issue reading? I have never stated that they should've used Final Cut for their comparisons. I dont not understand why you keep making up these things in your head

 

What I said was the Topaz lab comparison was particularly unfair. It used Intel's dedicated accelerator while it didnt make use of M1 NPU. There is a block of code that makes use of the extra hardware to run faster on Intel's side and that's unfair. This is similar to comparing Final Cut on Intel's accelerated by Intel's quicksync vs premiere pre quicksync patch (I'm not sure why it's faster on M1, so I'm not going to use M1 for the sake of drawing parallels). That was the comparison I made, to point out why it was a bad comparison. I don't know how to be more clear to you

 

What they should've done was either take a software that made of use of both the accelerators, or disable Intel's accelerator and test it. Which obviously Intel doesn't want to do, hence them being disingenous

[leadeater]

1 hour ago, RedRound2 said:

Do you have some issue reading? I have never stated that they should've used Final Cut for their comparisons. I dont not understand why you keep making up these things in your head

I'm not saying you did, your example was straight up bad and incorrect because of what you said not the example itself. Everything I said is just pure explainer after that. So I'll make it simple, your Final Cut sv Premier comment was wrong and was not like what Intel did. Intel wasn't swapping software in the performance tests, at no point did Intel compare X software on Mac OS to Y software on Windows. It was always X vs X.

 

All I did after saying your statement was wrong is explain how and why a Final Cut test would be better and why Intel didn't use it. Because you're complaining that Intel choose software that unfairly utilized their hardware features and not Apple's, well of course they did that was the objective, it's marketing.

 

 

1 hour ago, RedRound2 said:

This is similar to comparing Final Cut on Intel's accelerated by Intel's quicksync vs premiere pre quicksync patch (I'm not sure why it's faster on M1, so I'm not going to use M1 for the sake of drawing parallels). That was the comparison I made, to point out why it was a bad comparison. I don't know how to be more clear to you

Not it is not, you added on to that saying Intel did similar to this which they did not.

 

On 2/8/2021 at 10:41 PM, RedRound2 said:

What? Final Cut runs a lot faster in M1 compared to last gen Intel Macbooks, so I dont get your point. What I said was that comparing final cut on Mac and premiere pro in Windows is the kind of tests that would introduce a variability in software optimizations, similar to what Intel did in one of the tests.

 

There is only a single test where the software was changed, in the battery life test as using Chrome on Mac OS would definitely be unfair and would not be generating the same power load as Chrome on Windows, that's why Safari was used instead, to make it more fair not less.

 

You specifically pointed towards a criticism that did not actually happen. I'm all for criticizing a company for something they did wrong, I'm however not when it's something they didn't actually do. There's plenty of BS marketing here, but there isn't any funny business with changing software in "like for like" comparisons other than the battery test. Intel could potentially be butting up against laws if they did do that, marketing is often the art of truthful lying.