Apple to replace Intel and move to ARM - *** Confirmed as "Apple Silicon" ***

https://browser.geekbench.com/v5/cpu/4731213

Rosetta Geekbench:
Single: 1331
Multi: 5888

Native:
Single: 1700
Multi: 7500

The drop is about 22% (inline with A12Z results as well, they also dropped about 25%).

Usually emulated performance at synthetic benchmarks is the highest, and it is lower in custom complex workloads. So this 22% drop is a best case scenario, so headlines which say "M1 is faster at running x86 through Rosetta than Intel chips at native" are not accurate, these may be that fast, but not these benchmarks don't conclude that.
 
ChrisD. said:
Code: 
Processor Information
Name    VirtualApple
Topology    1 Processor, 8 Cores
Identifier    GenuineIntel Family 6 Model 44 Stepping 0
Base Frequency    2.40 GHz

Are they all reported like that on Geekbench? I thought they actually report as an M1 with 3.2Ghz base frequency?

Edit, I see, that's how it's reported when using Rosetta.
Click to expand...

Yeah, it shows as Intel for compatibility reasons. I'm guessing Apple has paid Intel for this :D
 
As expected, Geekbench results were hugely biased towards apple/arm

Using Cinebench R23, here are the scores:

M1 (Single thread/Multi thread) - 990/4530
Ryzen 5 3600X - 1300/9500
Ryzen 7 4800HS - 1230/10600

Not exactly apples/oranges comparison, but I honestly saw some apple sheep who were claiming that this M1 chip was going to be more powerful than ryzen desktop chips lmao. Maybe they got ahead of themselves after seeing some leaked geekbench scores
 
ZP9 said:
As expected, Geekbench results were hugely biased towards apple/arm

Using Cinebench R23, here are the scores:

M1 (Single thread/Multi thread) - 990/4530
Ryzen 5 3600X - 1300/9500
Ryzen 7 4800HS - 1230/10600
Click to expand...

You're quoting A12Z results, not M1.

M1 gets 1500/7500 in Cinebench, as reported (here and everywhere) many times.

https://images.anandtech.com/graphs/graph16252/119145.png

https://images.anandtech.com/graphs/graph16252/119372.png


Anandtech review is up, so stop spreading nonsense.

https://www.anandtech.com/show/16252/mac-mini-apple-m1-tested/2

ZP9 said:
Not exactly apples/oranges comparison, but I honestly saw some apple sheep who were claiming that this M1 chip was going to be more powerful than ryzen desktop chips lmao. Maybe they got ahead of themselves after seeing some leaked geekbench scores
Click to expand...

Against throwing insults, as you did in the past, and it's pretty clear that it's YOU who got ahead of himself.
 
Last edited by a moderator:
HACO said:
Anandtech review is up, so stop spreading nonsense.

https://www.anandtech.com/show/16252/mac-mini-apple-m1-tested/2
Click to expand...

The conclusion summary details this too:

The performance of the new M1 in this “maximum performance” design with a small fan is outstandingly good. The M1 undisputedly outperforms the core performance of everything Intel has to offer, and battles it with AMD’s new Zen3, winning some, losing some. And in the mobile space in particular, there doesn’t seem to be an equivalent in either ST or MT performance – at least within the same power budgets.

What’s really important for the general public and Apple’s success is the fact that the performance of the M1 doesn’t feel any different than if you were using a very high-end Intel or AMD CPU. Apple achieving this in-house with their own design is a paradigm shift, and in the future will allow them to achieve a certain level of software-hardware vertical integration that just hasn’t been seen before and isn’t achieved yet by anybody else.
Click to expand...

The fact that this is a 'low power' chip and is hitting some of these performance margins is absolutely awe inspiring and I can't wait to see what the next three years of the x86 to ARM transition means for Apple when we see the likes of iMac/Mac Pro dedicated SoCs and compare them to the latest and greatest of Intel/AMD. People seem to love the fact that AMD/Intel are just about winning on some benchmarks compared to a teeny, tiny low-powered gen 1 CPU haha.
 
You have to think that Apple are sandbagging a little bit here. Well, not quite, but they've put in a chip that does exactly what it needs to do without thinking about a billion cores and mega clock speed.
 
HACO said:
You're quoting A12Z results, not M1.

M1 gets 1500/7500 in Cinebench, as reported (here and everywhere) many times.

https://images.anandtech.com/graphs/graph16252/119145.png

https://images.anandtech.com/graphs/graph16252/119372.png


Anandtech review is up, so stop spreading nonsense.

https://www.anandtech.com/show/16252/mac-mini-apple-m1-tested/2



Against throwing insults, as you did in the past, and it's pretty clear that it's YOU who got ahead of himself.
Click to expand...

Yeah sorry my mistake!

Either way if you take a look at this, https://wccftech.com/intel-and-amd-...apples-m1-in-cinebench-r23-benchmark-results/

They are still behind the likes of intel and AMD who are running on 10nm/7nm chips

The fact is that when it comes to low watts, thats where ARM is best and x86 conversely is generally quite poor at low watts, so its quite concerning when the M1 cant be beating these low powered x86 chips (at least AMD) and what apple arm does best? Because I highly doubt it going to outperform the bigger more core/thread chips in desktops
 
Last edited by a moderator:
ChrisD. said:
You have to think that Apple are sandbagging a little bit here. Well, not quite, but they've put in a chip that does exactly what it needs to do without thinking about a billion cores and mega clock speed.
Click to expand...

I dont think so, they are just playing to ARM's strengths in that they perform well at low power. When it comes to high power I doubt they stand much of a chance at all because arm does not perform anywhere near as well at higher wattage
 
Throrik said:
The conclusion summary details this too:



The fact that this is a 'low power' chip and is hitting some of these performance margins is absolutely awe inspiring and I can't wait to see what the next three years of the x86 to ARM transition means for Apple when we see the likes of iMac/Mac Pro dedicated SoCs and compare them to the latest and greatest of Intel/AMD. People seem to love the fact that AMD/Intel are just about winning on some benchmarks compared to a teeny, tiny low-powered gen 1 CPU haha.
Click to expand...

imac pros/mac pros wont do anywhere near as well as these lower power chips because of how arm is
 
ZP9 said:
Yeah sorry my mistake!

Either way if you take a look at this, https://wccftech.com/intel-and-amd-...apples-m1-in-cinebench-r23-benchmark-results/

They are still behind the likes of intel and AMD who are running on 10nm/7nm chips

The fact is that when it comes to low watts, thats where ARM is best and x86 conversely is generally quite poor at low watts, so its quite concerning when the M1 cant be beating these low powered x86 chips (at least AMD) and what apple arm does best? Because I highly doubt it going to outperform the bigger more core/thread chips in desktops
Click to expand...

You're missing the point, those Intel/AMD chips still have significantly higher TDPs than M1, just because they're laptop 8 cores doesn't mean they're the same. Some of those Intel chips can boost to 20w per core in single-core workloads, Apple peaks at about 6w per core, as Anandtech showed.

ZP9 said:
I dont think so, they are just playing to ARM's strengths in that they perform well at low power. When it comes to high power I doubt they stand much of a chance at all because arm does not perform anywhere near as well at higher wattage
Click to expand...

There's no ARM ISA strength there, it's simply a matter of microarchitecture improvements. 5 years ago no ARM chip had any performance per watt advantage over Intel, 7 years ago it was behind. ARM ISA didn't change all that much, microarchitectures did.

ZP9 said:
imac pros/mac pros wont do anywhere near as well as these lower power chips because of how arm is
Click to expand...

Dude seriously, you have no idea what you're talking about. Stop embarrassing yourself (or trolling) as you've done time and time again here, lol.
 
HACO said:
You're missing the point, those Intel/AMD chips still have significantly higher TDPs than M1, just because they're laptop 8 cores doesn't mean they're the same. Some of those Intel chips can boost to 20w per core in single-core workloads, Apple peaks at about 6w per core, as Anandtech showed.



There's no ARM ISA strength there, it's simply a matter of microarchitecture improvements. 5 years ago no ARM chip had any performance per watt advantage over Intel, 7 years ago it was behind. ARM ISA didn't change all that much, microarchitectures did.



Dude seriously, you have no idea what you're talking about. Stop embarrassing yourself (or trolling) as you've done time and time again here, lol.
Click to expand...

10w vs 15w is significantly higher?

I know exactly what im talking about and so do most people who are interested in this sorta stuff. x86 will never be as good as arm in low power environments and vice versa. So if apple are struggling to beat low power x86 chips currently (mainly amd ones) then they wont have much of a chance at much higher tdp's which is where x86 shines, and arm is weak at.
 
ZP9 said:
10w vs 15w is significantly higher?

I know exactly what im talking about and so do most people who are interested in this sorta stuff. x86 will never be as good as arm in low power environments and vice versa. So if apple are struggling to beat low power x86 chips currently (mainly amd ones) then they wont have much of a chance at much higher tdp's which is where x86 shines, and arm is weak at.
Click to expand...

There's no 10 or 15w, you need to understand how TDP figures works (clearly you don't).

And your comment about ISA (bold sentence), well just frankly, is just uninformed. So let's move on. I suggest just reading up that Anandtech article, it will help clear things up for you.
 
ZP9 said:
I suggest you read this article https://wccftech.com/intel-and-amd-...apples-m1-in-cinebench-r23-benchmark-results/
Click to expand...

That's not an article - it's just speculation, Wccftech has never been a reliable source

WCCFTech said:
The problem is that ARM architecture doesn't clock very high and has high leakage at high power levels (this is also conversely true, x86 is not very good at low power scenarios). This means that the x86 market's bread and butter segments are safe from Apple's ARM influence - unless companies like NUVIA have something to say about that too.
Click to expand...

With no source to back this claim up
 
I'd really like Steve Burke to get his hands on one, although I don't think that's going to happen as he/GN has always been about gaming PCs.
 
Wccftech said:
On the other hand, the Apple M1 is easily thrashed by an Intel 4-core variant on 10nm: the Intel Core i7 1185G7 which is roughly 1 node behind TSMC's 5nm process.
Click to expand...

Thrashed? It's 40 points, or 2.7 % faster. Hardly a thrashing. And at the cost of how much power? They forgot that part.

Also:

Wccftech said:
But of course, the single-core benchmarks are purely cosmetic as most applications use multithreading.
Click to expand...

No, they really don't. :D I've just reminded myself why I tend to avoid that site with the dribble in that article.
 
You must log in or register to reply here.
Back
Top Bottom