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

[Jet_ski]

Summary

Nearly three months after the launch of Apple's rave-reviewed M1 Macs, Intel has fired back, not with better chips but with misleading marketing tactics of course.

 

In a slideshow shared by PCWorld this week, Intel highlighted what PCWorld described as "carefully crafted" benchmarks in an attempt to prove that laptops with the latest 11th Generation Core processors are superior to those with Apple's custom-designed M1 chip.

 

For example, Intel said that exporting a PowerPoint presentation as a PDF file is up to 2.3x faster on a Windows laptop equipped with an 11th Generation Core i7 processor and 16GB of RAM compared to completing the same task on a 13-inch MacBook Pro with the M1 chip and 16GB of RAM, with Intel noting that PowerPoint ran natively on both systems.

 

Intel also indicated that Topaz Labs' AI-based photo enlargement software Gigapixel AI performed up to 6x faster on the Core i7 system compared to the M1 MacBook Pro. Note that Topaz Labs' apps are designed to take advantage of hardware acceleration inside of Intel's processors.

 

As for gaming performance, the results were mixed, with Intel emphasizing the well-established opinion that Macs are not ideal for gaming and lack support for "countless" games like Gear Tactics, Hitman 2, and others.

 

 

Intel also conducted a "real world battery life test" and found that both the M1 MacBook Air and the Acer Swift 5 with an 11th Generation Core i7 processor achieved virtually identical 10-hour battery life when streaming Netflix with additional tabs open. Intel said both notebooks were set to 250 nits display brightness, with the MacBook Air running Safari and the Acer Swift 5 running Chrome for the test.

 

It's worth noting that Intel switched from the MacBook Pro for the performance benchmarks to the MacBook Air for the battery life test, and Intel used a different Core i7 processor SKU with a lower boost clock for the battery life tests.

 

Apple's website advertises the M1 MacBook Air as having up to 18 hours of battery life when continuously playing back 1080p content in the Apple TV app with display brightness set to 50%, and up to 15 hours of battery life when browsing 25 popular websites in Safari over Wi-Fi with display brightness set to 50%.

 

Apple columnist Jason Snell referred to Intel's benchmarks as "M1-unfriendly" in commentary shared on his website Six Colors.

 

"Inconsistent test platforms, shifting arguments, omitted data, and the not-so-faint whiff of desperation," wrote Snell. "Today's M1 processor is a low-end chip for low-end systems, so Intel only has a small window to compare itself favorably to these systems before higher-end Apple silicon Macs ship and make its job that much harder."

 

Tom's Hardware's Andrew Freedman also cautioned that all vendor-provided benchmarks should be taken with a grain of salt.

 

Apple says the M1 chip provides industry-leading performance per watt, with the latest MacBook Air outperforming a maxed-out Intel-based 16-inch MacBook Pro in Geekbench benchmarks.

 

Quotes

Quote

It's worth noting that Intel switched from the MacBook Pro for the performance benchmarks to the MacBook Air for the battery life test, and Intel used a different Core i7 processor SKU for each of these tests as well.

 

Quote

"Inconsistent test platforms, shifting arguments, omitted data, and the not-so-faint whiff of desperation," wrote Snell. "Today's M1 processor is a low-end chip for low-end systems, so Intel only has a small window to compare itself favorably to these systems before higher-end Apple silicon Macs ship and make its job that much harder."

 

My thoughts

 

It’s kinda obvious Intel went looking only for things to make themselves look better.

 

What does “2.3x faster in exporting PDFs” actually mean? How many seconds are we talking here? And what is this pdf they exported?!

 

Is the takeaway here that the Core i7 11th gen with $426 MSRP is in the same category as the M1?! That’s like the biggest complement they could have given Apple.

 

They started the benchmarks with a certain CPU then not only did they switch their own CPU, but also the laptops?!

 

For the performance benchmarks Intel used the Intel® Core™ i7-1185G7 Processor which boosts up to 4.80 GHz but then for the battery benchmarks they switch to Intel® Core™ i7-1165G7 Processor which boosts only up to 4.70 GHz.
 

 

Sources

https://www.macrumors.com/2021/02/06/intel-11th-gen-vs-apple-m1-chip-benchmarks/

 

https://ark.intel.com/content/www/us/en/ark/products/208921/intel-core-i7-1165g7-processor-12m-cache-up-to-4-70-ghz-with-ipu.html
 

https://ark.intel.com/content/www/us/en/ark/products/208664/intel-core-i7-1185g7-processor-12m-cache-up-to-4-80-ghz-with-ipu.html

[WereCatf]

Me checking Intel's latest "benchmarks" out:

[Metallus97]

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

[leadeater]

13 minutes ago, Jet_ski said:

Intel said that exporting a PowerPoint presentation as a PDF file is up to 2.3x faster on a Windows laptop equipped with an 11th Generation Core i7

And yet nobody cared, poor Intel. "But I'm good at something", still gets picked last for the baseball team lol.

 

16 minutes ago, Jet_ski said:

Intel also conducted a "real world battery life test" and found that both the M1 MacBook Air and the Acer Swift 5 with an 11th Generation Core i7 processor achieved virtually identical 10-hour battery life when streaming Netflix with additional tabs open. Intel said both notebooks were set to 250 nits display brightness, with the MacBook Air running Safari and the Acer Swift 5 running Chrome for the test.

Now this is the least believable thing of the lot. MacBooks not outright longer battery run times vs non MacBook laptops (that don't have extended battery options) just never happens. Alright Intel where are you hiding the diesel generator? Show me the hidden cables.

[Master Disaster]

Quote

We should see this moment as an opportunity to shift our focus as an industry from benchmarks to the benefits and impacts of the technology we create. The pandemic has underscored the need for technology to be purpose-built so it can meet these evolving business and consumer needs.

Bob Swan, June 2020, Intel CEO at the time.

 

Quote

"We should see this moment as an opportunity to shift our focus as an industry from benchmarks to the benefits and impacts of the technology we create"

Funny how benchmarks only don't matter when Intel cannot win them.

Quote

"The pandemic has underscored the need for technology to be purpose-built so it can meet these evolving business and consumer needs."

Like the Apple M1 chip you mean? That's certainly more "purpose built" than Intels current 14nm++++++++++ offerings.

[Ashleyyyy]

what a surprise... 

[Cyberspirit]

32 minutes ago, WereCatf said:

Me checking Intel's latest "benchmarks" out:

I'm stealing that gif, thank you.

[CTR640]

Intel never changes, eh.

[TetraSky]

It is kind of sad that they can't even win against the M1 in some of their benchmarks... Even more so in Gaming.

 

Also, this is typical of Intel, it's not the first time they've been caught screwing around with benchmarks.
PDF export... AH!

[D13H4RD]

The really funny thing is....Intel is really just acting like the spoiled, salty kid by pushing obviously skewed "benchmarks" like these. 

 

The Tiger Lake CPUs are an objectively good product for ultraportables and it plus the 10nm SuperFIN process they're on is definitely what they needed after the disappointment that was Ice Lake. However, there's no denying that Apple's new M1 chip was quite the disruptor. 

 

Mentioned it before and I think @Taf the Ghosthad a similar opinion but it's really a shame that many of Tiger Lake's strengths are basically hidden away under what's basically the marketing equivalent of a child throwing a tantrum. TGL-U is a better all-around processor than many realize, but shit like this doesn't help. 

[samcool55]

Tbh it would be more of a surprise if intel didn't do this.

Oh well, Intel will be Intel I guess.

I don't expect anything different until guys like GN or HUB will tell us that they released a CPU that's actually a decent improvement.

[flibberdipper]

1 minute ago, samcool55 said:

I don't expect anything different until guys like GN or HUB will tell us that they released a CPU that's actually a decent improvement.

Please don't say anything this outrageously funny again. I'm liable to have a heart attack laughing. We all know Intel's only trick is 14nm++++++.

[captain_to_fire]

This might be a good video idea. Test and confirm Intel’s claims@LinusTech

[WhitetailAni]

Just now, flibberdipper said:

Please don't say anything this outrageously funny again. I'm liable to have a heart attack laughing. We all know Intel's only trick is 14nm++++++.

Maybe 12th gen will be 10nm.

[Vishera]

3 hours ago, Jet_ski said:

Who cares about the performance of graphs and album organizers.

Those workloads are so light that even a refrigerator can do that:

[captain_to_fire]

6 minutes ago, flibberdipper said:

Please don't say anything this outrageously funny again. I'm liable to have a heart attack laughing. We all know Intel's only trick is 14nm++++++.

 

[D13H4RD]

6 minutes ago, ragnarok0273 said:

Maybe 12th gen will be 10nm.

It is. Alder Lake's built on the same 10nm SuperFIN node TGL is fabricated on.

[Vishera]

3 minutes ago, D13H4RD said:

It is. Alder Lake's built on the same 10nm SuperFIN node TGL is fabricated on.

And of course,new socket requiring a new motherboard.

That's a waste of good motherboards,not environmently friendly.

LGA 1700 is a huge socket for a consumer grade hardware.

[WhitetailAni]

Just now, Vishera said:

And of course,new socket requiring a new motherboard.

That's a waste of good motherboards,not environmently friendly.

LGA 1700 is a huge socket for a consumer grade hardware.

Think the mounting style will have to change?
And I wonder if they'll shrink the pins a bit so it's smaller.

[Vishera]

2 minutes ago, ragnarok0273 said:

And I wonder if they'll shrink the pins a bit so it's smaller.

Those pins are already fragile as they are now,making them smaller will make it worse

[WhitetailAni]

Just now, Vishera said:

Those pins are already fragile as they are now,making them thinner will make it worse

Didn't think of that.

I wonder why they need 500 more pins?
Does 10nm really require that many more?

[Guest]

Hell to the Naw Naw.... will be stuck in my head all day! haha. 

 

We're looking at arbitrary numbers here. Exporting a PDF means jack shit to me any ways.

[D13H4RD]

Just now, ragnarok0273 said:

Didn't think of that.

I wonder why they need 500 more pins?
Does 10nm really require that many more?

Wholly different architectural design.

 

Alder Lake is Intel's first implementation of a big.LITTLE design outside of a specialized ultraportable (Lakefield).

[Vishera]

3 minutes ago, ragnarok0273 said:

Didn't think of that.

I wonder why they need 500 more pins?
Does 10nm really require that many more?

As matter of fact a process shrink requires the same amount of pins or less,but not more of them.

Intel are probably using the headroom that the process shrink gives to them to increase performance,so it may require more pins.

[Vishera]

3 minutes ago, D13H4RD said:

Lakefield

A lake is water,a field is earth,

Is Lakefield a lake/field of mud?