Watch Apple's M1 MacBook Pro Obliterate 2020 Intel MacBook Pro in Speed Tests

[FNH15]

That’s an 8th gen Intel CPU... at the time when 11th gen is on the market. The one to be ashamed to sell that relic in 2020 is Apple, I would say ;-).

Hands-on with the Apple M1—a seriously fast x86 competitor [Updated]

Apple’s M1 proves that ARM can compete with x86 in high-end systems.

arstechnica.com

The M1 smokes the 11th gen as well. Suggest you research a bit before posting.

 

[Bauer24]

Yeah, definite *** statement. Yes, Intel *CAN* be made massively power efficient - sacrificing compute power. Intel *has* made 5W and even 1W x86 CPUs. They just can't operate at the compute power of what Apple wants.

The person you're replying to said that Intel chips can't be made power efficient in the same way as ARM. Not that you can't get low power Intel chips. Their statement is true, ARM is fundamentally different which is a huge part of the reason the power usage can go down, while the performance stays the same or improves.

Lowering the TDP while sacrificing computer power isn't improving power efficiency. That's just the same PE at a lower TDP (i.e. you're moving back down the slope, but the slope is the same flatness). Improving power efficiency means lowering TDP while maintaining or increasing compute power, which is what Apple is doing with the M1.

 

[ikramerica]

Hah! I did the "open ALL the apps!" thing, too.

On my new M1 Mac mini, I did it on the account first created, right after initial setup. On my 16" MacBook Pro, I created a new account so that at least that account would be in the same state as on the Mini. On that account, I quit all "start up for all users" apps before trying.

But I chose "all default macOS Big Sur apps in the main Applications folder - except "alternate UI" apps - Launchpad, Mission Control, Siri, Time Machine. I didn't include FCP, I didn't include anything in the "Utilities" folder.

Before "the real run", I "pre-"opened them all, went through all the "first time run" prompts each app popped up. So that when I did the mass-open-all, they would all launch to "ready to go" state. This meant adding an empty email account to Mail (I created a new iCloud account for this test,) dismissing all the "Welcome to tv!" "Welcome to Music!" "Hey, here's how to use Pages!" type prompts. I did notice that a few apps popped up a different "grab your attention" pop-up on the second launch, so had to do the third run as the "real run."

To make sure RAM caching didn't impact - once I verified that all apps launched in to "ready to use" with no obnoxious bouncing icons, I rebooted, opened QuickTime Player to screen record, opened Activity Monitor (so it would be open so I could check the RAM tab,) started the screen recording, then Cmd-O, click "Yes". Of course, Automator randomly decides when it wants to bounce-for-attention... It did for M1, didn't for Intel. ¯\_(ツ)_/¯

Thank you. That is what I figured. The original was done so unscientifically it was pointless.

 

[hladron]

Apple's first M1 Macs have defied expectations and are more powerful than anyone expected, handily beating out many other Intel Macs that Apple is continuing to sell. We've seen endless speed tests, but we thought we'd pit the M1 13-inch MacBook Pro against the model that it's replacing, the 2020 13-inch MacBook Pro with 1.4GHz quad-core Core i5 processor, Intel Iris Plus Graphics 645, and 8GB RAM.

Subscribe to the MacRumors YouTube channel for more videos.​

The prior-generation MacBook Pro was just released in May 2020, but it's already outdated and far inferior to Apple's new M1 model, as our extensive speed testing will demonstrate.

Geekbench Scores

The M1 MacBook Pro, which is the base model with 8GB storage, an M1 chip with 8-core CPU and GPU, and a 256GB SSD, earned a single-core Geekbench score of 1722 and a multi-core score of 7535.

Comparatively, our Intel MacBook Pro earned a single-core score of 871 and a multi-core score of 3786, so performance is close to double here. OpenCL scores also demonstrated a stark difference with the M1 earning a score of 19305 and the Intel chip earning a score of 6962.

SSD Speeds

There's a faster SSD in the M1 MacBook Pro and in our testing, we saw read speeds of 2800MB/s and write speeds of 2300MB/s. With the SSD in the Intel MacBook Pro, we saw read speeds of 1600MB/s and write speeds of 1100MB/s. Apple says the SSD can reach sequential read speeds of up to 3.3GB/s thanks to the new SSD controller integrated in the M1 chip.

File Transfers

When transferring a 40GB+ file, the M1 completed the task in 27 seconds while it took the Intel Mac 90 seconds. Transfer speeds started out the same, but it didn't take long for the Intel Mac to fall behind.

4K Video Export

Exporting a 10 minute 4K video from Final Cut Pro took the M1 MacBook Pro 4 minutes and 53 seconds and it took the Intel MacBook Pro 6 minutes and 47 seconds. In addition to the faster transfer speeds on the M1 Mac, the fans never came on at all, while the Intel Mac's fans were roaring.

Starting Up and Shutting Down

The M1 MacBook Pro starts up noticeably faster thanks to the new Instant Wake feature that gets it going right when you open the lid. Shutting down was also faster.

Tab Test

We opened up a dozen YouTube tabs in Safari on both Macs and the CPU load was much lower on the M1 Mac. The M1 Mac was able to play every video without issue and the fans never even kicked on, but the Intel Mac struggled and the fans were on max speed.

App Test

We opened every app in the Applications folder on both Macs, which was approximately 50 apps. The M1 excelled, while the Intel Mac lagged behind and had trouble opening everything up. It took a lot longer to open all of the apps on the Intel version, especially Final Cut Pro.

Opening up Mission Control with every single app open was seamless on the M1 Mac but the Intel Mac couldn't quite handle it and there was a lot of lag.

Tests with single apps were much closer. The M1 won out when opening up apps like Safari, Maps, Apple Music, and Final Cut Pro, but the Intel Mac wasn't too far off.

Conclusion

During our benchmarking and speed tests, the M1 MacBook Pro's fans never turned on once, so expect near silent operation for almost all tasks if you pick up one of the new MacBook Pro models. The MacBook Air has no fans at all, and the Mac mini performs similarly to the MacBook Pro.

In addition to speed, we've also been impressed with battery life. The MacBook Pro was used for an hour or two when we first got it and then most of the next day, and we never once had to plug it in to charge it even through all of the testing.

The M1 MacBook Pro beats out the 2020 Intel model, but it's also faster than the high-end 2019 16-inch MacBook Pro models in terms of CPU performance. If you're planning to buy a new Mac, at this point, it's probably worth holding out for a Mac with an M1 chip if you can. Apple is planning to update the entire lineup with Apple Silicon, a process that will take approximately two years.

Rumors suggest some of the next Macs to get M1 chips will include the iMac (there's a 24-inch model in the works) and the 16-inch MacBook Pro.

Article Link: Watch Apple's M1 MacBook Pro Obliterate 2020 Intel MacBook Pro in Speed Tests

This is exciting, I am getting a MacBook Pro soon 512GB and 16GB of Ram , I will transfer my current MacBook Pro 13” 4 ports 2018 to my wife, I am not a Pro in any sense, but not having ports on either side for when you are looking for the best side when charging, that is my only issue, not even the lack of having only two port, I would be piss if I recently got and intel model and then the M1 became available, I kind of understand some of the negative comments

 

[deputy_doofy]

M1 Mac Mini also has a fan.

That's why I said only the MacBook Air has no fan.

 

[Taz Mangus]

That's why I said only the MacBook Air has no fan.

I misread your post.

 

[ultradk]

Handbrake (Universal) after about 2-3 minutes my fan kicks on the MBP but it's crazy quiet; you have to put your head to it to notice it. Most users probably won't induce this sort of CPU load though.. But it's nice to know that for sustained performance i'd imagine the MBP would outpace the MBA over long term CPU loads like this.

Edit: That was x264 encoding above.. with x265, the fan is now audible and loud enough to hear easily but it's at least it's a pleasant woosh tone and not high pitched.. I guess this is real torture on this CPU.

Was the encoding time speedy?

 

[Mystakill]

I'll be much more interested when they start comparing it to i7 and i9 processors. i5 processors are notoriously slow for everyday usage.

 

[EugW]

I'll be much more interested when they start comparing it to i7 and i9 processors. i5 processors are notoriously slow for everyday usage.

There are already benchmarks out comparing against the higher end MacBook Pros.

BTW, personally I find the i5 processors usually fine for everyday mainstream business app type usage. For heavier stuff though they're slow.

 

[weckart]

All this does is show just how lazy Intel is and how they lack in innovation. If Apple can do this (who has less years experience in making processors than Intel) and apparently Intel can't (all they do is make faster fire-burning processors) then Intel sucks and should be ashamed of itself that Apple is whopping their butt like this.

Hardly the same, is it, though? Intel isn't so much lazy as has painted itself into a corner with having to allow for so much backward compatibility in its processors. It doesn't control the hardware or the OSes its processors will be used with and therefore cannot afford to be as ruthless as Apple in cutting off support after a few years. The enterprise sector will buy new hardware in spades but will also want to run its legacy bespoke programs on it. Apple isn't appealing to that sector any more so we get the M1 with no apologies if any of your existing software will not run on it.

The last time Intel tried to branch out with a completely new, rather than evolutionary design, the Itanium, didn't end so well for it.

 

[whyamihere]

Nice. Looking forward to the upgrade... once the 14" non-entry level pro releases.

 

[Darth Tulhu]

M1 vs i7 would have been better.

Perhaps, but I like the fact that he compared them based on price (the $999 base models).

The MBA is the single greatest value in computing right now. I really hope Apple doesn't raise prices, because they'd be worth it if they did.

 

[anticipate]

I have both a 16" 5600M based Intel MBP, and a high end 13" Apple Silicon MBP with 16GB of RAM. They are NOT the same when editing FCP video.

For one, most of the old FxPlug plugins don't work. They need to be recompiled.

Second, a nearly 6 flop GPU with dedicated 8GB of HBM definitely beats the M1 GPU. As a practical example, I am able to layer at least 3 4K videos on top of each other with LUTs and effects applied in real time on the 16" MBP. This is in "best quality" with no rendering; so real timeline performance.

I cannot do that on the M1 machine; it stutters. 8K is also very tough for the M1 Mac. It stutters, as does 6K Red Raw and so on.

(To be fair, it can handle two streams - which is impressive; the old 13" machine was useless and this machine costs $1700 vs $4000 for the 16")

An "M2" based MBP will be massively impressive; the M1 Mac barely even gets warm and the battery is at least 2X longer than the 16". But it's not at all a reasonable replacement. Yet.

My M1 Mac is going back.

 

[ani4ani]

Wow Apple truely is magical. They have managed to only consume 7 watts on RAM, SSD, WiFi, PMIC, etc. And then NPU and GPU don’t draw any power.

Why don’t you just read the article?

 

[Exdroid]

has anyone tested this for SMB transfer??

 

[LinkRS]

If we're being honest, Apple really didn't update those models, just slightly improved them. The new models carry the 10th gen Intel processors, and those 8th-gen intel models were roughly the same models from 2018. It just seems dishonest of Apple to have claimed them to be new, just as it is dishonest of Macrumors to do the same.

I agree, when comparing the "older" model of MacBook Pro, to the newer M1 model of MacBook Pro, this is 100% relevant. If (however) you are using this to compare the M1 against Intel, then this indeed dishonest. As Emanuel points out, we are comparing a 2020 designed Apple SOC against an almost two-year old Intel i5-8257U (Coffee lake or 8th gen,was launched in 2018, this specific CPU was released in Q3 2019) https://ark.intel.com/content/www/u...-8257u-processor-6m-cache-up-to-3-90-ghz.html, it is dishonest. The MacBook Pro chassis have sacrificed size for thermal performance for years now. Intel CPUs are at their "best" when they are "properly" cooled. In the case of the i5 in the "older" MacBook Pro, it is unlikely that it ever runs much faster than the base lock of 1.4 Ghz, so given equivalent architectures (which we know they are not), the M1 running at ~3.1 Ghz would be twice as fast as the Core i5. So it should surprise no-one that the new M1 MacBook Pro blows the proverbial pants off of the outgoing model. Now, if the older MacBook Pro, had the thermal headroom in the design to allow the Core i5 to turbo up to ~3.1 Ghz (max is 3.9 GHz) the comparison would be more interesting concerning the M1 vs Intel. Even still, the M1 is advertised as an 8-core CPU, while the i5 is only a 4-core, so again in multi-threaded apps would be expected to run better with more cores. Yes, the Intel CPU supports Hyperthreading, but that just means that it can execute more threads and behave as if it had 8-cores, but does not demonstrate the performance of true 8-core CPUs.

With all of that said, does it really matter that the comparison favors the Mac? I say no, it does not. You will only ever able to get the M1 in an Apple computer, and they are comparing it against the discontinued model. So, for MacBook Pro users, it is a fair comparison. It just isn't fair when comparing the Intel architecture used in the older model against the current Apple architecture, as it is biased against Intel.

Rich S.

 

[EugW]

I have both a 16" 5600M based Intel MBP, and a high end 13" Apple Silicon MBP with 16GB of RAM. They are NOT the same when editing FCP video.

For one, most of the old FxPlug plugins don't work. They need to be recompiled.

Second, a nearly 6 flop GPU with dedicated 8GB of HBM definitely beats the M1 GPU. As a practical example, I am able to layer at least 3 4K videos on top of each other with LUTs and effects applied in real time on the 16" MBP. This is in "best quality" with no rendering; so real timeline performance.

I cannot do that on the M1 machine; it stutters. 8K is also very tough for the M1 Mac. It stutters, as does 6K Red Raw and so on.

(To be fair, it can handle two streams - which is impressive; the old 13" machine was useless and this machine costs $1700 vs $4000 for the 16")

An "M2" based MBP will be massively impressive; the M1 Mac barely even gets warm and the battery is at least 2X longer than the 16". But it's not at all a reasonable replacement. Yet.

My M1 Mac is going back.

You didn't say how much RAM you have in your 16" MBP.

Judging by your experience and various reviews out there, it would seem the M1 hardware video accelerator is built for up to multiple streams of 4K video (of various formats), but cannot handle 8K. I suspect that will be addressed with the M1X (or whatever it is called).

 

[PlainBelliedSneetch]

Why don’t you just read the article?

I did read it. Anandtech has no way to measure the cpu by itself.

 

[crawfish963]

I'll be much more interested when they start comparing it to i7 and i9 processors. i5 processors are notoriously slow for everyday usage.

They have. The M1 Air beat the i9 from a 2020 16" in Cinebench in both single and multi-core.

 

[marco114]

I am sitting here typing on my new M1 Mac Mini and I can't believe how it literally smokes my 2015 i7 2.5ghz Mac book Pro. Even in Rosetta it's faster.

 

[athletejmv]

That M1 is amazing!!! Apple still years ahead of the game!

I've been running the m1 Mac mini 512Gb variant with 8Gb of RAM. Umm, this system operates on a whole other realm hardware wise. It blows my 2019 16" MBP out of the water while not even getting warm, stuttering or breaking a sweat... ugh, mind blown. I've had no issue other than the Google-Drive sync app giving an error code. I don't plan on willfully purchasing another intel chip again for the Mac lineup.

 

[bn-7bc]

The market will still be relatively strong because of Bootcamp. This is a big drawback for those that dual boot.

Windiws dis run on arm you know, so it might not be a big job for ms to do the requiered tweeks to get windowxup an running bativly on the M1 chip, wether nvidia wil port the gtx 1050 driver to thet peticular version of windows is an open question tho, most windows sw would probably just need a recomoile

 

[EdT]

Why does everyone keep comparing the M1 to the Mac Intel version of the same model? How does it stack up against Dell or HP or Asus or Lenovo? By this time most people know these processors handily beat their Intel versions released just this year but how do they compare to other manufacturers?

 

[Fthree]

Just out of curiosity how much do you think a maxed out 16 will go for in the near future? I was in the market but they were so dang high!!!

 

[athletejmv]

Your comparing the M1 chip to an 8th gen intel MacBook processor lol ok......

either way I’ll be upgrading to an M1 very soon but let’s be real now.

Actually, I own a 2019 MBP 16" and Mac mini. I have the Mac mini M1 with 8Gb RAM and 512 Gb of storage. It will be replacing it the MBP... it performs nearly identical and much more stable without stutter. You gotta see it for yourself. I would be as skeptical as you had I not experienced this new chip/SOC for myself.