cubase performance on macos vs windows 10/11

[rardier]

Hello,
i would like to know if someone has noticed (benchmark) a difference between cubase 11/12 on macos vs on windows 10/11 on the same computer (latest intel based mac or hackintosh or ryzentosh), i saw some cubase benchmark on the latest M1 pro/max and i found the result really curious (80 tracks of retrologue on M1 pro/max vs 95 tracks of retrologue on ryzen 5950x) and now i wonder if one of the reason is that cubase performs better on MacOs.

regards

[Serhii Kot]

It has better performance on my M1 Pro (10 cores, 16gb ram) than my 5800x(32gb ram) ryzen PC. Cubase 12.

[Trensharo]

It has better performance on my M1 Pro (10 cores, 16gb ram) than my 5800x(32gb ram) ryzen PC. Cubase 12.

The 5950X has twice the core count of the 5800X, Higher Clock Speeds and a higher TDP.

There is no comparison between these two CPUs. They are on different levels in terms of performance ceiling, especially in anything moderately SMT-optimized.

Keep in mind, that these are ALL P-Cores. Ryzen doesn't have eCores, so every core is the equivalent of an Intel P-Core or a Performance Core on Apple Silicon (in terms of function). This can be a reason why some people with higher end CPUs (than yours) may not notice a huge differences when doing CPU load tests. It's possible that the higher core count x86 CPUs have more processing headroom than lower core count M-Series CPUs.

This is why Apple has M1 Max and M1 Ultra (basically Dual M1 Max).

Anything mildly SMT-optimized is going to give drastically better performance on a 5950X in comparison to a 5800X (which was always mediocre value - better to go to the 12-Core 5900X for not much more investment).

Additionally, due to Infinity Fabric design, Ryzen needs 3600MHz DDR4 RAM to reach optimal performance. Most people are running 3000-3200MHz RAM. You should also have a PCIe 4 NVMe system/application drive for optimal system responsiveness. I actually can feel the difference between the two with Boot Up/Shut Down/Sleep, Loading Heavier Applications (or those that load lots of components - particularly when paired with a better CPU) and things like populating directories with lots of files in Windows Explorer, etc.

Keep in mind that most people are not going to be benchmarking their machines while producing music. They're just going to be using them, and computers spend a lot of time doing nothing but waiting for us to press a button. It really isn't that big a deal.

I have a Ryzen 9 4000-series laptop that tackles everything I can throw at it. The same exact projects that I do on my Desktop or M1 Pro MBP, running the same plug-ins, synths, samplers and sample libraries, etc. That's an 1st Gen M1-era Ryzen Laptop. I never even think about performance (due to noticing what may be a lack of performance) when using that machine.

It does all the same stuff my desktop does... fast. And I prefer not to get sucked into the "perpetually benchmarking performance while using my machines" mindset

[Trensharo]

Also, Cubase has a reputation for running better on Windows. It may have something to do with everything running through Steinberg's preferred ASIO system on that platform. I dunno.

I don't really notice a difference because I'm not really into looking for them.

[rardier]

Also, Cubase has a reputation for running better on Windows. It may have something to do with everything running through Steinberg's preferred ASIO system on that platform. I dunno.

I don't really notice a difference because I'm not really into looking for them.

yes i m quite sure cubase performed better on windows arounds ten years ago (i saw benchmark on that),
but there were also windows changes introduced (mmcss restrictions for example) that had impact on cubase performances.
i have a ryzen 7 4800hs laptop and ryzen 9 5950x desktop i see the difference ,while i can finish my project on my laptop , it takes more time (freeze track , bounce,direct offline processing,render in place increasing buffer size etc etc) the workflow is less fluid, so yes it s a matter of comfort
The thing is , if it s became a fact that cubase performs better on mac os than on windows it could be a reason to choose a pc with hackintosh ability (depending on how much better cubase is on macos) or go the macbook pro m1 way which is a bit too expensive for my taste but also bring other stuff on the table like good onboard dac / cpu power stays quite the same on battery / stays relatively quiet , but it s also brings drawback if you are not only using your computer for cubase (not everything is arm/m1 compatible).

[Trensharo]

Nothing you said makes any sense. this is no different than getting a new Intel i7 1300-H series Laptop and claiming Cubase runs better on Intel than AMD.

You're comparing a mid-range Ryzen 7 APU to a high end Ryzen 9 CPU. How do you expect that to pan out?

You can make an M1 look bad by comparing it to an M1 Max, as well. This is practically what you are doing.

I have a Ryzen 9 4900HS laptop here. I have no issues with Buffers, Direct Offline Processing, Track Freezing, etc. on it. This is just not a thing for me.

Ryzen did have issues on Windows 11 after launch, but a Chipset Update fixed that within a month. That was a long time ago. You shouldn't be running an unpatched chipset unless you are REALLY bad with PC maintenance habits (applies to both Macs and PCs).

Some of these issues can be due to drivers in other areas - outside of Cubase. No one here knows enough about your machine to be able to see where the issues are coming from.

If you're using a Focusrite Interface on that laptop, you can start by replacing that.

[rardier]

i know my english could be better but...i m claiming nothing, i didn t say i had issues, my two system works fine, but i just said that on my laptop i have to free some cpu more often by freezing track or direct off line processing... i m using a babyface pro fs and i am on win 10. Good for you if you are perfectly fine with your laptop, i am comparing the ryzen 7 4800hs and the ryzen 9 5950x because it s what i use, who said the 4800hs was bad? i just said i felt the difference..if that makes no sense for you ..i won t argue... i just asked if someone had made a cubase benchmark on the same system but with the 2 different os (windows 10/11 and macos) just to know if cubase perfoms better on mac os , i don t even need to explain why i want to know just want to share but if you feel i m stupid don t need to talk with me.

[Trensharo]

You're going to feel the difference because the 4800HS is very low end-compared to a 5950X. Even compared to a Ryzen 7 3700X, you will feel the difference. The desktop Ryzen is just... better, in every way.

You are writing as if you think you should be able to load a [relatively heavy] project on any machine and see comparable performance. When you load up a project on a weaker CPU, you have a less compute capacity.

The differences you notice are expected and predictable. They are not remarkable - or even worth mentioning. Everyone just assumes that will be the case when they think about two machines with that level of disparity in CPU Compute Capacity.

It's like someone saying they noticed they have to Freeze More Tracks on their 9th Gen i7 Laptop vs. their 12th Gen i7 Desktop when running the same relatively heavy projects. Well... Duh?

And frankly, I don't really run into issues with CPU capacity on mine. One would have to see what types of projects you're running, what Virtual Instruments/Samplers/Libraries you're using, what plugins you're using, and also what your ASIO settings are. Some things like VST3 "Don't process when no audio signal" are turned off by default in Cubase, so that can also factor into things, as well.

Frankly, there are more things external to Cubase that can affect its performance - positively or negatively - than Cubase itself, at this point.

[rardier]

You are writing as if you think you should be able to load a [relatively heavy] project on any machine and see comparable performance. When you load up a project on a weaker CPU, you have a less compute capacity.

The differences you notice are expected and predictable. They are not remarkable - or even worth mentioning. Everyone just assumes that will be the case when they think about two machines with that level of disparity in CPU Compute Capacity.

It's like someone saying they noticed they have to Freeze More Tracks on their 9th Gen i7 Laptop vs. their 12th Gen i7 Desktop when running the same relatively heavy projects. Well... Duh

You extrapolate, i never said or wrote that i should be able or expect to run the same project the same way between my laptop and my desktop , i didnt wrote that the differences were unexpected
... you said it s not worth mentionning but you said this:
"
I have a Ryzen 9 4000-series laptop that tackles everything I can throw at it. The same exact projects that I do on my Desktop or M1 Pro MBP, running the same plug-ins, synths, samplers and sample libraries, etc. That's an 1st Gen M1-era Ryzen Laptop. I never even think about performance (due to noticing what may be a lack of performance) when using that machine.

It does all the same stuff my desktop does... fast"
and this is ambiguous, it could be understood as ryzen laptop/desktop/m1 pro all perform the same when it just means that the least powerful of them is already enough for your need.

But your need isn't my need.

I just wrote that for my workflow /my project the the power of ryzen 5950x is quite perfect and the ryzen 4800hs (which is not far from your r9 4900) is ok but slows me down a bit

[Trensharo]

I am not sure how I am extrapolating anything?

I am simply stating that the Ryzen is a 2020-era down-volted chip designed for smaller gaming laptops where the dGPU in those machines do a lot of heavy lifting. You kept mentioning it as if it were a huge factor, when it was a basic expectation.

If you want a laptop that can do it better, then you always had to buy one with a better CPU in it.

Comparing to Apple, or Intel, or newer AMD is always going to have that laptop looking mediocre because 12th Gen Intel, Ryzen 6000-series and M1 Pro/M2 series CPUs are just better.

Like I said, even Zen 3 desktop CPUs (Ryzen 7/9) destroys those APUs, so they were never going to keep up with any desktop you had with a newer Ryzen 6/7/9 CPU in it. It's like expecting a car to go 90km/h when its top speed is 70km/h. Your Ryzen 9 goes at 100km/h, your Laptop goes at 60km/h. Anything that requires more speed than the laptop can offer will feel throttled there, but will often fly on the desktop.

The chances of a Ryzen 7 4800HS keeping up with a decent desktop, even one from 2020, was [ALWAYS] non-existent.

Beyond that, the only application that I've used on Windows that performs better on macOS is DaVinci Resolve Studio. REsolve is simply better on macOS, even though it's still amazing on Windows when given the hardware to do its thing. But, in a laptop form factor, this necessitates going with at least a 3070 Ti. Anything with that card in it will not look like a 13 or 14" MacBook Pro, and if it did it probably will get 3 hours of battery max life off the charger even when running Windows Notepad.

For some reason, Windows machines do not know how to limit GPU processing to the iGPU when the device is unplugged. It will allow any and everything to keep accessing the dGPU, so the only way to get something that runs off the battery like a MBP, is to get an Intel Machine with an iGPU only (e.g. some Dell XPS are nice).

[rardier]

i mean you extrapolate when you think that i expected that on same os the ryzen 4800hs should perform as good as the ryzen 5950x.
you extrapolate everytime you write " you write/say this AS IF"
i have no "AS IF" in mind.
But what i want to know for Cubase and Cubase only , because it depends also on Daw used, is : given the SAME system (cpu/motherboard/souncard) does cubase performs better on mac os? i m asking that after viewing this video and the results post by user on the comment :

[Trensharo]

No. I'm wondering why you think it performing worse on a 4800HS is remarkable enough to write in a forum post when asking about performance... as if it's a relevant data point.

I've been pretty clear about that, using every analogy that comes to mind... but you don't seem to be able to read and understand/get this.

Macs run specific spec packages for their machines, with highly tuned drivers. You will never be able to get a Mac with the same level of drivers (Chipset, Audio, Graphics, etc.) when throwing together a "Hackintosh."

There is no reliable way to test the way you want people to assume performance will be compared. Most people comparing Windows and Mac these days will be comparing an x86 PC to an M-Series Mac. Even comparing to Intel Macs makes little sense, since those CPUs are 2-3 generations old, at this point.

This would have made more sense in 2019. It is completely inviable for getting any useful data in 2023.

The newest Ryzen 9 (6000-series) APUs will lose to an M1 Max, while probably beating out the i9 in the last 16" Intel MacBook Pro. It's a senseless comparison... and throwing macOS on a machine with x86 hardware and expecting it to be a sensible performance comparison would be quite a choice...

Again: Comparing the two with old hardware and unsupported driver stacks atop worse performing components isn't reliable, due to how all of these factors interact with each other in a computer.

OS comparisons like this were only totally reliable when Macs were x86 and you could Boot Camp Windows natively onto the machines and run it with a supported (1st party supplied) full driver stack.

[rardier]

"No. I'm wondering why you think it performing worse on a 4800HS is remarkable enough to write in a forum post when asking about performance... as if it's a relevant data point.

I've been pretty clear about that, using every analogy that comes to mind... but you don't seem to be able to read and understand/get this."

I am able to read and understand, but you might think that you are smarter than every one here or you want to have last word and not seems to be honest and not really respectful enough to understand what other want to say=> once again because you are stubborn => i m not like you tryinv to be "remarkable" when i share an info or what i do or what i want, i just said i use an ryzen 4800hs and ryzen 5950x , and i m more comfortable with the 5950x and yes it s EXPECTED but you also give perfect exemple (or counter exemple) that is also depend on the usage when for your usecase a m1 pro/ryzen desktop/ryzen laptop performs the same

"
OS comparisons like this were only totally reliable when Macs were x86 and you could Boot Camp Windows natively onto the machines and run it with a supported (1st party supplied) full driver stack"

i kind of agree with that it s easier with that setup, but maybe also can be done with a dual boot hackintosh / windows.

[Proteinshake]

i would like to know if someone has noticed (benchmark) a difference between cubase 11/12 on macos vs on windows 10/11 on the same computer

In fact I have today. Using the 3XS benchmark, Cubase 12, I9 10850k .. I could have 11(!) more tracks on the hackintosh side, that's 70 with Win11 vs 81 with Big Sur.

[rardier]

i would like to know if someone has noticed (benchmark) a difference between cubase 11/12 on macos vs on windows 10/11 on the same computer

In fact I have today. Using the 3XS benchmark, Cubase 12, I9 10850k .. I could have 11(!) more tracks on the hackintosh side, that's 70 with Win11 vs 81 with Big Sur.

thank you very much Proteinshake!