Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Apple MBPr 13 M1 cinebench R23 score (twitter.com/mnloona48_)
43 points by rezonanc on Nov 16, 2020 | hide | past | favorite | 32 comments


Apple M1 scores: 7508 multi-core and 1498 single-core

Current single-core list:

https://www.cpu-monkey.com/en/cpu_benchmark-cinebench_r23_si...

And multi-core list:

https://www.cpu-monkey.com/en/cpu_benchmark-cinebench_r23_mu...

The M1's single-core result beats Intel i9-10900KF (score 1418) and is almost at Intel I7-1165G7 (score 1504)


I wonder how this will compare to the Zen 3 mobile parts. A 4800U (Zen 2, 8 cores) apparently does 1235 single-core, 10156 multi-core @ 15W TDP.

It's not clear to me how much power the Macbook Pro draws. They specify 10W TDP for the M1 in the Air, but it seems to be higher in the Pro?


The CPU looks extremely impressive. I'm very hesitant to move away from x86 architecture until I have a great, economical CPU to use in my home server as well.

It's very unlikely, but wow it would be great if I could build a server using an M1 processor. Just the power saving alone would be phenomenal.


Aftermarket racked Mac Minis aren't a new idea in server spaces. Have you considered creating a cluster of them to fit into a rack? As far as price/performance goes the Mac Mini is about as good as you can get from Apple.


> It's very unlikely, but wow it would be great if I could build a server using an M1 processor. Just the power saving alone would be phenomenal.

Mac minis are designed with a stackable form factor, and in fact one of their selling points is to put together server farms for stuff such as running build jobs with Xcode.

If I'm not mistaken, mac minis with the M1 chip are sold starting at 700$ a pop.


Apple made server racks many years ago. This new silicon could give them a reason to re-enter that market. In the meantime, the new Mini could be a good fit for a home server.


Any gpu benchmarck yet of the M1?


"the M1 ties the Radeon RX 560 (2.6 TFLOPS), and it's just a few TFLOPS away from catching the GeForce GTX 1650 (2.9 TFLOPS)" --- https://www.tomshardware.com/news/apple-silicon-m1-graphics-...


I really want to see how these chips perform graphically in real-life games. It seems incredible how performant they are, so much so that I nearly can't believe it.


The main thing is whether this cpu will throttle under load. All intel Cpus will throttle badly in Macbook like form factors.



We know the answer to that. The result you quote (7500 MT) is MBP. Light fan, audible but not too bad. Power consumed probably around 15W, but we'll have to wait for serious measurement.

For MBA (no fan. passive cooling) same ST number, MT number is 6300. So fall off, but not that bad, large cores at maybe 80% of peak. Device gets slightly warm, not hot, under this use case. Device power maybe 8..9 W I would guess?


Per the tweets, this is a 10min test, so we'd probably see the effects of throttling at that point.


Cinebench only stresses the CPU not the GPU or the AI coprocessor. Besides I find these scores pretty dissapointing after all the hype. Yes it's faster then an i7-1165G in multicore but actually slower in single core. Slower... After all that hype.


This is a score on macOS 11.0.0, which shipped with the hardware, 11.0.1 bumps clocks a bit to 3.2GHz, which makes Apple Silicon take the ST crown by a hair.


You’re comparing a low-power CPU used in Apple’s low-end MacOS devices to a state of the art (released Q3 2020) high-end 28W TDP CPU from Intel.


I'm comparing the M1 to a cpu that would have been used in a Macbook pro 13 if it weren't for the M1 soc. The current i7 intel Macbook pro 13 inch has the 1068NG7 model. Which is in fact a better cpu then the 1065g. Which is last gens 1165g. So I think this comparison is completely valid.


If we’re comparing low-end apples to apples, the bottom-end Intel MacBook Pro (MacBookPro16,3) has an i5 part (Coffee Lake 8257U)[1] and would be a better comparison.

I don’t think I’ve seen any indication from Apple the current M1 13” is all we’re going to see in this form factor.

[1] https://en.wikipedia.org/wiki/MacBook_Pro


The price difference is only like 300 dollars so pretending that one is low end and the other high end is not really true here. They are both high end devices.

I can't buy this hypothetical stronger arm Macbook so it means nothing. Apple's marketing material said the M1 would blow everything out of the water with a HUGE margin.

Now it turns out it's actually slower then their own laptop would be with a refreshed intel chip. That's what dissapoints me.


It does blow Tiger Lake out of the water. (and btw, barely gets ahead of it on single core perf with 11.0.1 as an update)

On multicore, Tiger Lake blows well beyond its TDP range and throttles hard, ending with a 6070 score on MT for the 1165G7. (in CB R23)


It's not really slower. It's comparable, with some clear benefits. But yeah, reality is caching up.

Apple is performance-wise matching their competitors (was clear from how they positioned m1 products that this would be the case).

Now one can say that it's lower power and low end, but then we can also say that when especially AMD catches up to 5nm (within a year) and adds on-chip memory (within 2 years) they'd be ahead, with a chip with the same die size..


But it is 28 watts to the M1's 10 watts


True, in perf per watt M1 does an incredible job I'd wager. But I'm not interested in that. I simply want the highest performance in a 13 inch formfactor. Apple marketing material said the M1 would blow everything out of the water with a HUGE margin.

Now it turns out it's actually slower then their own laptop would be with a refreshed intel chip. That's what dissapoints me.


Apple's marketing material said the M1 would blow everything out of the water with a HUGE margin.

Could you provide a link to the marketing material that says the M1 would blow everything out of the water with a HUGE margin?

The M1 press release says "The World’s Best CPU Performance per Watt". That's true.

The M1 MacBook press release says "And with M1, the 13-inch MacBook Pro is up to 3x faster than the best-selling Windows laptop in its class."

The fine print: Testing conducted by Apple in October 2020 using preproduction 13-inch MacBook Pro systems with Apple M1 chip, as well as production Intel Core i7-based PC systems with Intel Iris Plus Graphics and the latest version of Windows 10 available at the time of testing. Best-selling system based on publicly available sales data over the last nine months.

Translation: those laptops you see at Best Buy that are in the $800–$1200 range? Yeah, we blows those things out the water. By a lot.


A similarly priced Zephyrus G14 Ryzen 9 4900hs scores 1246/11006 (1).

I still may buy a M1 system, but I was hoping Apple would start competing in price/performance.

[1] https://www.cpu-monkey.com/en/cpu-amd_ryzen_9_4900hs-1285


It is not a fair comparison since 4900HS is a 35w chip. A fair comparison would be 4750U, 15w chip with score of 1184/8008


I'm not comparing on the basis of TDP. I'm comparing performance/dollar.

I fully expect Apple's higher end, higher TDP chips to achieve better performance. But they will cost more money.

Like it or not the M1 is competing in a price bracket with the 4900HS.


but you will agree that - for a mobile cpu - power consumption and therefore battery life is an important aspect of the equation?


Is it though? A "Pro" laptop has, by the very name, an implication of performance.

I don't know many professionals in any industry that are constrained so heavily by power - generally when you're doing professional work, you're not far from an outlet, save some rare use cases like on-location shoots for film and television.


Absolutely, that's one of the reasons I may buy a first generation Apple product.


I too had hoped that Apple might reduce the price of the new AS lineup to promote adoption by consumers, but it doesn't seem to be going that way. Depending on how these initial models sell (well I think), we may or may not see any improvement in the price/performance ratio.


Probably doesn't make sense (yet) with chip supply constrained for a year or three.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: