Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
MacBook Pro 16-inch M1 Max Stockfish Benchmarks
- Thread starter Appletoni
- Start date
- Sort by reaction score
Looks like no one shows M1 Max CPU results, maybe they are to bad:
Performance Metrics:
openbenchmarking.org
Performance Metrics:
Stockfish Benchmark - OpenBenchmarking.org
DennisdeWit, I am very interested as well. Just start from the original position, select in the menu Engine/Start Infinite Analysis and let it run for 5 minutes. Then take a screenshot. That would be very much appreciated.
note: it would help if you do not run anything else at the same as to not split the power of the cpu.
My M1 is not great for Stockfish, very curious to know the difference with the M1X.
note: it would help if you do not run anything else at the same as to not split the power of the cpu.
My M1 is not great for Stockfish, very curious to know the difference with the M1X.
If you can tell me how to run Stockfish Benchmarks, I am more than happy to share the results with you. Even if the results are bad (despite my glasses, I don't like chess.)
Does this help you?
View attachment 1881984
To measure and to compare the engine speed, you can use a free chess software or Fritz 17 or ChessBase 16 or BanksiaGUI or ChessOK Aquarium 2021 and then click the analysis button at the beginning of a game (starting position). After 1 minute is over you click the button again to stop the analysis and take the kn/s and run the test again with other engines on different hardware.
You only need to decide at the beginning if you want to compare tests with one core or more cores or maybe something like 2 vs 4 cores.
You can also run only the Stockfish.exe and the benchmark or do something like: go nodes 100000
You will never see exactly the same kn/s but that is today not important to compare.
Stockfish:
kn/s: 1234 (only the first (here the 1xxx) is important to compare). On another hardware it could be 2000 (2xxx).
kn/s: 12345 (only the first + second (here 12xxx) is important to compare).
kn/s: 123456 (only the first + second + third (here 123xxx is important to compare)
The last 3 xxx are always different, even on exactly the same hardware.
If you get a result, please send it to Ipman Chess:
Hardware speed:
Ipman Chess
Measure speed:
Build Tester v.1.4.7.0
Build Tester
Build Tester,download,chesschess.massimilianogoi.com
Compare benchmarks with your hardware:
acepoint's home
Engine Benchmarks on Apple’s M1 | acepoint's home
Benchmarks with more than one thread on an Apple M1 | acepoint's home
Apple M1 Max chip problems:
Stockfish:
Stockfish - Open Source Chess Engine
Stockfish is a powerful and open source chess engine.Stockfish Testing Framework
FishCooking - Google Groups
GitHub - official-stockfish/Stockfish: UCI chess engine
UCI chess engine. Contribute to official-stockfish/Stockfish development by creating an account on GitHub.github.com
See also Stockfish Discord developer channel https://discord.com/invite/nv8gDtt
The most people won‘t provide benchmarks when they already know that the results will be bad.
-Stockfish = extreme slow. Other cpus are between four and 14 times faster.
-SD (UHS-II) = slow and very old technology. SD Express (UHS-III) is at least twice as fast.
-LC0 = extreme slow. Other gpus are much faster.
-Wifi 6 = slow and old. Wifi 6e is much faster.
-Integer math = extreme slow.
-Bluetooth 5.0 = slow and old. Bluetooth 5.3 is much better.
-HDMI 2.0 = only 60hz and very old technology. HDMI 2.1 is much better with 120hz.
-No 5G and that’s very bad.
-Media Engines are available, but Integer math and other engines are still missed.
Computing News, Laptop Reviews, Windows, macOS, and More | Digital Trends
News, reviews, and discussion about desktop PCs, laptops, Macs, GPUs, and everything else in the world of computing, including in-depth reviews and buying guides.
www.digitaltrends.com
-40 CPU cores and much faster cores
-128 GB DDR5 RAM and twice as fast
-16 TB or 2x 8 TB SSD and twice as fast
If I understand, this is one of those gaming threads where people complain that the MBP sucks at gaming with stuff not written for Macs compared to the machines it is written for? Yes, we know. Chess games and tools are no different than from other games that way.
The most likely reason people haven't been doing this test is that it's the first they've heard of it, so they have no idea what it's for or what the result would be apart from what you've said, and they don't see the point. But insulting them might help.
The most likely reason people haven't been doing this test is that it's the first they've heard of it, so they have no idea what it's for or what the result would be apart from what you've said, and they don't see the point. But insulting them might help.
Does this only apply to benchmarks Where Apple-silicon does poorly? People seem to care alot about apple vs- oranges benches like Geekbench which they most probably also have no idea what it is for...
It applies to tests they've never heard of that have no apparent purpose that concerns them. Geekbench is very well known and established as a benchmark. People here generally know what it's for.
The thread is here:
Apple M1 CPU & GPU speed is very disappointing
Benchmark: Stockfish (chess) speed M1 CPU = 13000 kn/s i7 3930k overclocked (from 2011 = 10 years old) = 13000 kn/s Others = 40000 kn/s - 80000 kn/s Desktop CPUs = 230000 kn/s and much stronger
forums.macrumors.com
It was locked, so no more replies are allowed, but it was not deleted, so is still readable.
and they said there was nothing more to optimize because M1 sucks...
forums.macrumors.com
Stockfish patch to improve performance on M1 by up to 80%.
Optimization work has been ongoing recently and the result is this PR, which some report to provide up to 80% performance increase (one person reports 32%, the other 80-85%. The results from the latter person presented below). I got some numbers from M1 that are comparable with other numbers at...
forums.macrumors.com
Mea culpa. But everything said has been in that thread.The thread is here:
Apple M1 CPU & GPU speed is very disappointing
Benchmark: Stockfish (chess) speed M1 CPU = 13000 kn/s i7 3930k overclocked (from 2011 = 10 years old) = 13000 kn/s Others = 40000 kn/s - 80000 kn/s Desktop CPUs = 230000 kn/s and much stronger
It was locked, so no more replies are allowed, but it was not deleted, so is still readable.
The chess developers went back actully did pretty much all optimizations suggested in that threads and guess what.. performance on M1 was still 2-3 times below AMDs 7nm laptop CPUs.. (they could only squeeze like 20% more perf. by doing M1 optimizations.. The conslusion really wes that M1 simply lacks potential to compete with modern 8-core SMT x86 7nm competitors.The thread is here:
Apple M1 CPU & GPU speed is very disappointing
Benchmark: Stockfish (chess) speed M1 CPU = 13000 kn/s i7 3930k overclocked (from 2011 = 10 years old) = 13000 kn/s Others = 40000 kn/s - 80000 kn/s Desktop CPUs = 230000 kn/s and much stronger
It was locked, so no more replies are allowed, but it was not deleted, so is still readable.
The sad fact still remains... M1 (and M1 Max) pretty much sucks potenital-performance wise compared to ryzen 7 and 9:s.
Don't know what that means, and so far have no reason to care. The M1 and M1 Pro/Max do very well at what they're designed to do.
No, I'm telling you what I actually said. Obviously Stockfish isn't everything chess-related, not even close. Obviously the M1 and M1 Pro/Max aren't designed or limited in excellence for only single-threaded work. Hope that helps.
Still have no good reason to care about Stockfish, and it appears you can't give one.
Bench LC0 then, the second best chessengine... M1, Pro and MAX Still sucks (even more on the GPU part of things), (even so if you do CPU inference back-ends to LC0)...
I am reasonably sure you are one of those persons who would not care about ANY benchmark, if it does not confirm/validate your bias and special brand-love / loyalty.
Out of curiosity which benchmark against an alder-lake or 5980H, or an RTX 3080 on the GPU-side, do you really care about?
Last edited:
I have no idea what numbers you are looking at but M1 Pro/Max scores 16080837 nodes/sec (according to this test #9) in patched stockfish 4.1 while the Ryzen 9 4900H scores 14999806 (http://ipmanchess.yolasite.com/amd--intel-chess-bench-stockfish.php) - that's 8% faster.
There were no benchmarks of high-end Zen3 mobile CPUs I could find, but we know that the 5900HX is around 10% faster vs. the 4900H. So it should perform similar to the M1 Pro in the end. Of course, M1 Pro will still use somewhere between 30% and 50% less power.
And of course, you have to consider that stockfish devs had years to optimize for x86 while the current ARM performance is literally based on a single patch that a single developer wrote in a couple of days — and if I understand it correctly, this developer does not even have access to native hardware for testing!
Last edited:
yeah, they suck except their numbers are better than Ryan 9 4900H, after only a couple weeks (vs years) optimizing, with more optimizations already identified but not yet implemented.
You’re correct other than that.
What most people talk about is actual real world performance. I don’t buy my machine to run benchmarks. I buy it to do actual work. Can you share actual workload tests and comparisons? Or do you only buy a computer to run chess benchmarks?
I am reasonably sure you ignore half of the stuff people have said in all of these threads. Especially the stuff mentioned by actual chip designers and developers that go against your narrative.
I like benchmarks. Since I don't play games on my computer (much less games where the machine does the thinking for me), gaming benchmarks are only a curiosity, though I've sometimes done them for people who have a more direct interest in them. People who grew up on Earth and understand insults aren't the best way to ask for help.
For popular computer games, the RTX 3080 generally smokes Apple chips, mainly because it's the kind of hardware the games are written for. I made that plain in a general way in my first post in this thread. It's such a long thread, you probably missed it.
Laptops with chips like the 3080 are also useful as space heaters and white noise generators, whereas this one generally remains at most merely warm under load, is too quiet to be useful, and doesn't run twice as well plugged in as on battery.