Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Status
The first post of this thread is a WikiPost and can be edited by anyone with the appropiate permissions. Your edits will be public.

Do you think the first benchmarks are correct?


  • Total voters
    314

consumeritis

macrumors member
Mar 9, 2015
86
43
If you look at the footnotes on Apple's M1 pages, almost all of Apple's own specific processor tests are comparing the M1 in the MacBook Pro to:

* 1.7GHz quad-core Intel Core i7–based 13-inch MacBook Pro systems

Which is the outgoing MacBook Pro 13" 2-port model with the i7 upgrade.

And the M1 in the Air to:

* 1.2GHz quad-core Intel Core i7–based MacBook Air systems

Which, again, is the outgoing highest-end MacBook Air model.

Apple's presentation was really light on specifics, and those graphs were awful, but they have been a bit more transparent with their benchmarks than many have realised.

These Cinebench scores look amazing. The M1 has awesome single-core performance. It seems like you have to go look at desktop-class chips to beat it.

For multicore, the efficiency cores are not going to be helping that much in these kind of benchmarks - for performance workloads, the M1 is effectively a four-core CPU without hyper-threading. People comparing it to a 4800U or similar (which has eight cores and sixteen threads) are being a bit unfair. Wait for an M1X with eight performance cores and come back.

And every time I have searched for a laptop using Zen 2 that has a decent high-res screen and a good keyboard, they have been thin on the ground. The Air and Pro are just a click away.

The only things holding me personally back from getting a new Air are the small performance loss against the Pro (which is dumb, because it would still be significantly faster than my current MacBook Pro, pretty sure!), and the hope that we'll get a 4-port Pro next year with an even more powerful chip.

This is a massive game-changer for the PC industry.
 
  • Like
Reactions: Sanpete

wiesson

macrumors newbie
Aug 4, 2020
10
2

I found this video of "head of tech". They have installed a lot of software (e.g. homebrew, xcode (at 3h 26m), minecraft, they also searched for pubg, but wasn't available) during their 4h+ stream. I did not watch everything, just skipped through, but it looked good so far.
 
Last edited:

leman

macrumors Core
Oct 14, 2008
19,521
19,675
Well, the 4800H and the 4800U are almost the same processor, but the U-series is clocked lower (when not boosting) and better binned. The scores are in line with other benchmarks I've looked for this CPU but that's probably a 25W laptop when boosting (which means ~35W for 15 minutes). And it will lose 1/4 of performance after the 15 minutes in most configurations. As I said earlier, the TDP is very confusing these days with all the possible boosts and configurations. So it'll be interesting to see a comparison of the performance degradation for the MBA, MBP and a 4800U-based laptop of a similiar size and weight, not only the benchmark score.

I don't necessarily disagree with you, just want to point out that these quoted scores are abnormally high when compared to trusted third-party reviews such as notebook check. For instance, Lenovo Yoga Slim 7 is known to be one of the fastest 4800U implementation out there, and yet it scores well under 9880H in Cinebench R20


It could indeed be an artifact of the first run, before the machine reaches it's sustained equilibrium, but with R23 defaulting to 10 minutes rounds... I'd just like to see some screenshots from reliable benchmarks runners
 

leman

macrumors Core
Oct 14, 2008
19,521
19,675
On single core, I wouldn't say it's among fastest. Zen 3 parts are about 10% faster, even though they are slower in geekbench.

I mean it's in basically in top 5 fastest single-core CPUs, al things considered. It's really not bad at all for an entry-level part that you can have in a $999 passively cooled laptop.
 

clangers23

macrumors 6502
Oct 27, 2016
325
447
I own an Asus G14 with a 4800HS 35W CPU, it'll comfortably outperform the i9 in my MBP 16 and ThinkPad P1. As others have posted it's all about the wattage supplied to the CPU and thermal capabilities.

On day to day use it's drawing 12-18W's. Battery life is typically 8-10 hours.
 

thrtytwo

macrumors newbie
Dec 12, 2012
24
12
I don't know if these results are good or not ¯\_(ツ)_/¯

Screen Shot 2020-11-17 at 7.19.43 pm.png


Edit: It got a pretty warm during the tests but not too hot, I don't think the fan even came on. I put my ear right up to the back of it and couldn't hear anything. I've previously owned a late 2014 15" MBP and that thing got too hot to touch and screamed like a jet engine under load.
 
Last edited:

thrtytwo

macrumors newbie
Dec 12, 2012
24
12
But how does it feel
They were offscreen benchmark tests, don't feel like anything ?

I'd love to test it on a relatively graphics intense game but the only one i can think of for mac is tomb raider or civ V and neither of them are free haha.
 
Last edited:

donth8

macrumors regular
Sep 25, 2015
106
108
I don't know if these results are good or not ¯\_(ツ)_/¯

View attachment 1670401

Edit: It got a pretty warm during the tests but not too hot, I don't think the fan even came on. I put my ear right up to the back of it and couldn't hear anything. I've previously owned a late 2014 15" MBP and that thing got too hot to touch and screamed like a jet engine under load.
This is great! Compared to the 5300M inside the 16 inch MacBook pro it is not too far away only about 13% slower:
 
  • Like
Reactions: Sanpete

thingstoponder

macrumors 6502a
Oct 23, 2014
916
1,100

I found this video of "head of tech". They have installed a lot of software (e.g. homebrew, xcode (at 3h 26m), minecraft, they also searched for pubg, but wasn't available) during their 4h+ stream. I did not watch everything, just skipped through, but it looked good so far.
It’s good news that even when maxed out GPU wise in FCP they were both barely warm to the touch on the top and they reported them both at being the same temperature.

Edit: They’re saying the bottom is definitely warm but not hot like the old ones and the Air is actually cooler.
 

Henk van Ess

macrumors demi-god
Original poster
Aug 20, 2008
314
241
Amsterdam
It’s good news that even when maxed out GPU wise in FCP they were both barely warm to the touch on the top and they reported them both at being the same temperature.

Edit: They’re saying the bottom is definitely warm but not hot like the old ones and the Air is actually cooler.
Cool! 2x
 

donth8

macrumors regular
Sep 25, 2015
106
108
This is great! Compared to the 5300M inside the 16 inch MacBook pro it is not too far away only about 13% slower:
Compared to the 5500M it is about 23% slower. The 5600M is about 70% faster but uses 50w of power. If Apple can double the gfx cores to 16 inside their next chip it would be a monster!
 

nameste

macrumors 6502
Mar 8, 2016
348
181

I found this video of "head of tech". They have installed a lot of software (e.g. homebrew, xcode (at 3h 26m), minecraft, they also searched for pubg, but wasn't available) during their 4h+ stream. I did not watch everything, just skipped through, but it looked good so far.
i watched the whole stream the mba is just amazing:Same performance with mbp without fan, battery life is same also the whole case is warmer than the mbp.I think i will send my noisy 16" back to apple.
 

thingstoponder

macrumors 6502a
Oct 23, 2014
916
1,100
Compared to the 5500M it is about 23% slower. The 5600M is about 70% faster but uses 50w of power. If Apple can double the gfx cores to 16 inside their next chip it would be a monster!
That’s good to hear. How would this compare to the 5k iMac? Would that be able to replace that GPU? Or would they need something like 3x the cores?
 

thrtytwo

macrumors newbie
Dec 12, 2012
24
12
That’s good to hear. How would this compare to the 5k iMac? Would that be able to replace that GPU? Or would they need something like 3x the cores?

Rumour is they're working on a discrete GPU for desktop class Macs, if true it would probably (looking at these results) be up to the task.
 

donth8

macrumors regular
Sep 25, 2015
106
108
That’s good to hear. How would this compare to the 5k iMac? Would that be able to replace that GPU? Or would they need something like 3x the cores?
The Radeon Pro 5700XT in the 27 inch iMac is about 100% faster, if apple puts 20 GPU cores in their desktop chips they could beat it if it scales well.
 
  • Like
Reactions: Sanpete

EntropyQ3

macrumors 6502a
Mar 20, 2009
718
824
The Radeon Pro 5700XT in the 27 inch iMac is about 100% faster, if apple puts 20 GPU cores in their desktop chips they could beat it if it scales well.
I think this is overstating things. Lets see how the M1 performs in some reasonable game benchmarks first.
Amazingly, Blizzard seem to be on board with WoW (!) so there is at least one game that people actually play that can be tested. I'm particularly interested in whether running the M1 fanless results in significant performance degradation. The MacBook Air did really well in Cinebench in that regard, and if that holds up in extended gameplay as well, that would be great!
 

Serban55

Suspended
Oct 18, 2020
2,153
4,344
Edit: It got a pretty warm during the tests but not too hot, I don't think the fan even came on. I put my ear right up to the back of it and couldn't hear anything. I've previously owned a late 2014 15" MBP and that thing got too hot to touch and screamed like a jet engine under load.
That the thing, for my Maya tests, the 16" Mbp after 2 min the chip were around at 98-99C and after the fans started to kick hard the temp went down to around 92C and the top part of the keyboard is around 42C
Here, with this M1 the temp is around 82-83C and the keys are around 35-36C so thing about the same as the human body...its insane, the Pc industry is at risk...who can challange this future? Microsoft owns win10 but they dont own arm chips ..Samsung...not, QUalcomm, doesnt have OS(and i saw how QC is running on my surface pro x), Google has android but we know how android dev are for Pro apps...and for big and scale UI displays
Again, somebody in 2-3 must challenge this...because Apple needs competition, otherwise i dont see Apple doing well after antitrust will target Apple..
 

thingstoponder

macrumors 6502a
Oct 23, 2014
916
1,100
They just finished Cinebench on the Mini and it got 7600. Pro was 7600. Air was 6600

They’re saying they haven’t heard any fan noise on the Mini, even after running cinebench. He has a 16” Pro next to him and has said it’s annoying how loud it is. I just skipped around on the last stream but did hear him say he couldn’t hear the 13” Pro either.
 
Last edited:
  • Like
Reactions: Sanpete and LFC2020
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.