3 ms·

Well, firstly, it isn't. There are higher Geekbench 6 CPU scores, even forgetting the ones that appear to be fake or at least broken. But secondly, that would

by jchw 2y ago

Well, firstly, it isn't. There are higher Geekbench 6 CPU scores, even forgetting the ones that appear to be fake or at least broken. But secondly, that would absolutely not indicate that it is the "fastest single-core performer in consumer computing". That would indicate that it is the highest scoring Geekbench 6 CPU in consumer computing. Whether or not that's actually a good proxy for the former statement is up to taste, but in my opinion it's not. It gives you a rough idea of where the performance stands, but what you really need to be able to compare CPUs is a healthy mix of synthetic benchmarks and real-world workloads. Things like the time it takes to compile some software, scores in video game benchmarks, running different kinds of computations, time to render videos in Premiere or scenes in Blender, etc. In practice though, it's hard to make a good Apples-to-Intels performance comparison, since it will wind up crossing both OS boundaries and CPU architecture boundaries, which adds a lot of variables. At least real world tests will give an idea of what it would be like day-to-day even if it doesn't necessarily reveal truisms about which CPU is the absolute best design. Of course it's reasonable to use Geekbench numbers to get an idea of where a processor stands, especially relative to similar processors, but making a strong claim like this based off of Geekbench numbers is pretty silly, all things considered. Still... these results are truly quite excellent. It would suffice to say that if you did take the time to benchmark these processors you would find the M4 processor performs extremely well against other processors, including ones that suck up more juice for sure, but this isn't too surprising overall. Apple is already on the TSMC N3E process, whereas AMD is currently using TSMC N4P and Intel is currently using TSMC N3B on their most cutting edge chips. So on top of any advantages they might have for other reasons (like jamming the RAM onto the CPU die, or simply better processor design) they also have a process node advantage.

GeekyBear 2y ago
SPEC has been the industry standard benchmark for comparing the performance of systems using different instruction sets for decades now. Traditionally, Anandtech would have been the first media outlet to publish the single core and multicore integer and floating point SPEC test results for a new architecture, but hopefully some trusted outlet will take up the burden. For instance, Anandtech's Zen 5 laptop SKU results vs the M3 from the end of July: > Even Apple's M3 SoC gets edged out here in terms of floating point performance, which, given that Apple is on a newer process node (TSMC N3B), is no small feat. Still, there is a sizable deficit in integer performance versus the M3, so while AMD has narrowed the gap with Apple overall, they haven't closed it with the Ryzan AI 300 series. https://www.anandtech.com/show/21485/the-amd-ryzen-ai-hx-370-review/4 https://www.anandtech.com/show/21485/the-amd-ryzen-ai-hx-370... Zen 5 beat Core Ultra, but given that Zen 5 only edged out the M3 in floating point workloads, I wouldn't be so quick to claim the M4 doesn't outperform Zen 5 single core scores before the test results come out.
- spoaceman7777 2y ago
  just considering the number of simple calculations a CPU can compute isn't a very good comparison. apple's chips are using an ARM architecture, which is a Reduced Instruction Set Computer (RISC) setup, vs x86-64, which is a Complex-ISC (CISC). The only good comparison is to judge a variety of real world programs compiled for each architecture, and run them.
  - erik_seaberg 2y ago
    https://en.wikipedia.org/wiki/SPECint https://en.wikipedia.org/wiki/SPECint talks about a dozen programs they selected for single-core logic and discrete math, including gcc and bzip2 (there are more than a dozen others using floats). Over time, RISC and CISC borrowed from each other: https://cs.stanford.edu/people/eroberts/courses/soco/projects/risc/developments/index.html https://cs.stanford.edu/people/eroberts/courses/soco/project...
  - GeekyBear 2y ago
    > The only good comparison is to judge a variety of real world programs compiled for each architecture, and run them. I'm guessing that you don't realize that you are describing SPEC? It's been around since the days when every workstation vendor had their own bespoke CPU design and it literally takes hours to run the full set of workloads. From the same Anandtech article linked above: > SPEC CPU 2017 is a series of standardized tests used to probe the overall performance between different systems, different architectures, different microarchitectures, and setups. The code has to be compiled, and then the results can be submitted to an online database for comparison. It covers a range of integer and floating point workloads, and can be very optimized for each CPU, so it is important to check how the benchmarks are being compiled and run. More info: > SPEC is the Standard Performance Evaluation Corporation, a non-profit organization founded in 1988 to establish standardized performance benchmarks that are objective, meaningful, clearly defined, and readily available. SPEC members include hardware and software vendors, universities, and researchers. SPEC was founded on the realization that "An ounce of honest data is worth a pound of marketing hype". https://www.spec.org/cpu2017/Docs/overview.html https://www.spec.org/cpu2017/Docs/overview.html
throw_that_away 2y ago
Remember too back in the day when you're looking at a Mac with 1/4 the power of a PC, and it's 4x the price. I think we're starting to see those ratios reversed completely. And at the same time, the power, heat, etc.. is just sitting at the floor.
- sho_hn 2y ago
  Yeah, this rings true. I'm not an Apple customer, but I certainly remember the days when Mac users had to justify the much worse bang-for-the-buck of Apple hardware (then) with it being the only ticket to their preferred software experience. These days it appears more that the hardware is fantastic, especially in the laptop form factor and thermal envelope, and perhaps the downside is a languishing macOS.
  - bogeholm 2y ago
    I use macOS daily for dev and office work, and to me it doesn’t feel languishing at all. Add Homebrew and Docker or Podman, and we’re off. The only places I can see there could be features missing are: - IT management type stuff where it looks like Apple are happy just delegating to Microsoft (eg. my workstation is managed with InTune and runs Microsoft Defender pushed by IT), - CUDA support if you’re into AI on NVIDIA - Gaming I hear, but I don’t have time for that anyway :) Of course this is biased, because I also generally just _like_ the look and feel of macOS
    - pipodeclown 2y ago
      Just basic customisation functionality is missing in MacOS. Recently bought a MacBook and am constantly amazed about the stuff you can't do. Oh you want do use normal scrolling on your external Bluetooth mouse (like everybody that uses a mouse does) and natural scrolling on the trackpad? Nah bro, choose one for both. You wanna just plug in some generic usb hub connected to multiple monitors like you did with your 500 euro laptop? No can't do. Want to specify the SDR content brightness on an external display separately from HDR content brightness when enabling HDR on the display? Nah bro we're just gonna blast full brightness at you all the time, and I can go on. Hardware is second to none but the software is really.holding the machine back
mabedan 2y ago
> you really need to be able to compare CPUs is a healthy mix of synthetic benchmarks and real-world workloads I like how Apple got roasted on every forum for using real world workloads to compare M series processors to other processor. The moment there’s a statistic pointing to “theoretical” numbers, we’re back to using real world workload comparison
- jchw 2y ago
  You need both, this is just standard practice. Check for example how CPUs are compared on outlets like Gamer's Nexus or Phoronix. Apple didn't get roasted for presenting real world performance, they got roasted for doing the kinds of things that marketing people do: making vague blanket claims about performance that couldn't actually be reasonably validated. (Intel and AMD routinely get roasted for similar things.)
adrian_b 2y ago
The higher single-thread GB6 scores are from overclocked Intel or AMD CPUs. The M4 core @ 4.5 GHz has a significantly higher ST GB6 performance than Lion Cove @ 5.7 GHz or Zen 5 @ 5.7 GHz (which are almost equal at the same clock frequency, with at most a 2% advantage for Lion Cove). Having higher GB6 scores should be representative for general-purpose computing, but there are application domains where the performance of the Apple cores has been poor in the past and M4 is unlikely to have changed anything, i.e. the computations with big integer numbers or the array operations done on the CPU cores, not on the AMX/SME accelerators. Nevertheless, I believe that your point about the existence of higher ST GB6 scores is not weakened by the fact that those CPUs are overclocked. For a majority of the computer users in the world, the existence of higher ST performance that can be provided by either overclocked Intel/AMD CPUs or by CPUs made by Apple is equally irrelevant, because those users would never choose any of these 2 kinds of CPUs for their work.
- jchw 2y ago
  Honestly, I do agree, it is kind of ridiculous that it is fighting neck and neck with relatively overclocked AMD and Intel CPUs. I always do a little bit of overclocking but some of these overclocks are pretty high. Still, I do think some of Apple's inherent advantages make it less surprising that they're able to win on benchmarks. Again, the process node, the RAM directly on the chip. And hell, they are probably also able to get here because of targeting AArch64 with no 32-bit ARM compatibility. Either way, Apple appears to be a couple years ahead of the competition right now when it comes to efficient processors, just like they were with the M1.

by jchw 2y ago

GeekyBear 2y ago

SPEC has been the industry standard benchmark for comparing the performance of systems using different instruction sets for decades now. Traditionally, Anandtech would have been the first media outlet to publish the single core and multicore integer and floating point SPEC test results for a new architecture, but hopefully some trusted outlet will take up the burden. For instance, Anandtech's Zen 5 laptop SKU results vs the M3 from the end of July: > Even Apple's M3 SoC gets edged out here in terms of floating point performance, which, given that Apple is on a newer process node (TSMC N3B), is no small feat. Still, there is a sizable deficit in integer performance versus the M3, so while AMD has narrowed the gap with Apple overall, they haven't closed it with the Ryzan AI 300 series. https://www.anandtech.com/show/21485/the-amd-ryzen-ai-hx-370-review/4 https://www.anandtech.com/show/21485/the-amd-ryzen-ai-hx-370... Zen 5 beat Core Ultra, but given that Zen 5 only edged out the M3 in floating point workloads, I wouldn't be so quick to claim the M4 doesn't outperform Zen 5 single core scores before the test results come out.

spoaceman7777 2y ago
just considering the number of simple calculations a CPU can compute isn't a very good comparison. apple's chips are using an ARM architecture, which is a Reduced Instruction Set Computer (RISC) setup, vs x86-64, which is a Complex-ISC (CISC). The only good comparison is to judge a variety of real world programs compiled for each architecture, and run them.
- erik_seaberg 2y ago
  https://en.wikipedia.org/wiki/SPECint https://en.wikipedia.org/wiki/SPECint talks about a dozen programs they selected for single-core logic and discrete math, including gcc and bzip2 (there are more than a dozen others using floats). Over time, RISC and CISC borrowed from each other: https://cs.stanford.edu/people/eroberts/courses/soco/projects/risc/developments/index.html https://cs.stanford.edu/people/eroberts/courses/soco/project...
- GeekyBear 2y ago
  > The only good comparison is to judge a variety of real world programs compiled for each architecture, and run them. I'm guessing that you don't realize that you are describing SPEC? It's been around since the days when every workstation vendor had their own bespoke CPU design and it literally takes hours to run the full set of workloads. From the same Anandtech article linked above: > SPEC CPU 2017 is a series of standardized tests used to probe the overall performance between different systems, different architectures, different microarchitectures, and setups. The code has to be compiled, and then the results can be submitted to an online database for comparison. It covers a range of integer and floating point workloads, and can be very optimized for each CPU, so it is important to check how the benchmarks are being compiled and run. More info: > SPEC is the Standard Performance Evaluation Corporation, a non-profit organization founded in 1988 to establish standardized performance benchmarks that are objective, meaningful, clearly defined, and readily available. SPEC members include hardware and software vendors, universities, and researchers. SPEC was founded on the realization that "An ounce of honest data is worth a pound of marketing hype". https://www.spec.org/cpu2017/Docs/overview.html https://www.spec.org/cpu2017/Docs/overview.html

throw_that_away 2y ago

Remember too back in the day when you're looking at a Mac with 1/4 the power of a PC, and it's 4x the price. I think we're starting to see those ratios reversed completely. And at the same time, the power, heat, etc.. is just sitting at the floor.

sho_hn 2y ago
Yeah, this rings true. I'm not an Apple customer, but I certainly remember the days when Mac users had to justify the much worse bang-for-the-buck of Apple hardware (then) with it being the only ticket to their preferred software experience. These days it appears more that the hardware is fantastic, especially in the laptop form factor and thermal envelope, and perhaps the downside is a languishing macOS.
- bogeholm 2y ago
  I use macOS daily for dev and office work, and to me it doesn’t feel languishing at all. Add Homebrew and Docker or Podman, and we’re off. The only places I can see there could be features missing are: - IT management type stuff where it looks like Apple are happy just delegating to Microsoft (eg. my workstation is managed with InTune and runs Microsoft Defender pushed by IT), - CUDA support if you’re into AI on NVIDIA - Gaming I hear, but I don’t have time for that anyway :) Of course this is biased, because I also generally just _like_ the look and feel of macOS
  - pipodeclown 2y ago
    Just basic customisation functionality is missing in MacOS. Recently bought a MacBook and am constantly amazed about the stuff you can't do. Oh you want do use normal scrolling on your external Bluetooth mouse (like everybody that uses a mouse does) and natural scrolling on the trackpad? Nah bro, choose one for both. You wanna just plug in some generic usb hub connected to multiple monitors like you did with your 500 euro laptop? No can't do. Want to specify the SDR content brightness on an external display separately from HDR content brightness when enabling HDR on the display? Nah bro we're just gonna blast full brightness at you all the time, and I can go on. Hardware is second to none but the software is really.holding the machine back

mabedan 2y ago

> you really need to be able to compare CPUs is a healthy mix of synthetic benchmarks and real-world workloads I like how Apple got roasted on every forum for using real world workloads to compare M series processors to other processor. The moment there’s a statistic pointing to “theoretical” numbers, we’re back to using real world workload comparison

jchw 2y ago
You need both, this is just standard practice. Check for example how CPUs are compared on outlets like Gamer's Nexus or Phoronix. Apple didn't get roasted for presenting real world performance, they got roasted for doing the kinds of things that marketing people do: making vague blanket claims about performance that couldn't actually be reasonably validated. (Intel and AMD routinely get roasted for similar things.)

adrian_b 2y ago

The higher single-thread GB6 scores are from overclocked Intel or AMD CPUs. The M4 core @ 4.5 GHz has a significantly higher ST GB6 performance than Lion Cove @ 5.7 GHz or Zen 5 @ 5.7 GHz (which are almost equal at the same clock frequency, with at most a 2% advantage for Lion Cove). Having higher GB6 scores should be representative for general-purpose computing, but there are application domains where the performance of the Apple cores has been poor in the past and M4 is unlikely to have changed anything, i.e. the computations with big integer numbers or the array operations done on the CPU cores, not on the AMX/SME accelerators. Nevertheless, I believe that your point about the existence of higher ST GB6 scores is not weakened by the fact that those CPUs are overclocked. For a majority of the computer users in the world, the existence of higher ST performance that can be provided by either overclocked Intel/AMD CPUs or by CPUs made by Apple is equally irrelevant, because those users would never choose any of these 2 kinds of CPUs for their work.

jchw 2y ago
Honestly, I do agree, it is kind of ridiculous that it is fighting neck and neck with relatively overclocked AMD and Intel CPUs. I always do a little bit of overclocking but some of these overclocks are pretty high. Still, I do think some of Apple's inherent advantages make it less surprising that they're able to win on benchmarks. Again, the process node, the RAM directly on the chip. And hell, they are probably also able to get here because of targeting AArch64 with no 32-bit ARM compatibility. Either way, Apple appears to be a couple years ahead of the competition right now when it comes to efficient processors, just like they were with the M1.