Cubase 12 does ignore the spare cpu cores when it actually is hunger for more. I don’t know why, maybe they want to keep my house from a fire disaster.
Well, that’s depressing,
I’m off to get my one-core computer then…
The following is true for playback and recording (I never looked at the CPU usage during rendering, but I will check that asap):
If you route all your tracks to a Master/Mix buss, which has insert plugins loaded, then most of the load is on one core. That is just how it works, not only Cubase is doing that, but every other DAW too. The more processing is on busses, the more the load wont be spread over different cores.
I did some testing some time ago with different DAWs and they all more or less use the same amount of CPU, Cubase being on the more efficient side. So all these claims, that Cubase is “old code”, “unefficient” or “much slower” than other DAWs is just BS. If you think like this, check out other DAWs and do some comparable testing.
I have the same CPU as the OP, I am gonna check that as soon as I can. I usually have 200-300 tracks which are routed to busses with FX on them and all of these busses are routed to a Master buss with additional FX on them. I call these “heavy projects”, buss processing, mixbuss processing and mastering (ProL2 limiter with 16x oversampling=heavy load) is done all in one project.
So if you have a mix structure where there is a master buss with processing on it, you better look for better single core performance when looking for CPUs.
Now I think Cubase should use all cores when rendering, as that is not a time critical process, so there must be something wrong on the OPs system. But I’ll report back after checking it on mine.
I have a 18 cores (36 threads) and Cubase used core 1 and 14 only on export…very perplexed…
Are you guys saying that Reaper would use all 18 cores if need be??
If so, why is not Steinberg implementing this function??
Computer science tells us that there is a fundamental limit to parallelization and thus getting all of your CPU cores fully loaded. It’s called Amdahl’s Law, and I’m pasting a link to the Wikipedia entry on it below. From that article:
For example, if a program needs 20 hours to complete using a single thread, but a one-hour portion of the program cannot be parallelized, therefore only the remaining 19 hours’ (p = 0.95) execution time can be parallelized, then regardless of how many threads are devoted to a parallelized execution of this program, the minimum execution time cannot be less than one hour. Hence, the theoretical speedup is limited to at most 20 times the single thread performance.
Im quite interested in this new CLAP protocol, and I’m hoping the developers of this protocol succeed with their goal – which is specifically to improve audio performance on multicore CPUs, and make plugins easier to deal with as a whole. Time will tell.
So… it is on Cubase not being able to fully utilize multi-core…? um, meaning, I shall not spend extra money on 20-core Mac Studio Ultra…? I am almost a page away from ordering one and these posts hold me back lol… man, cause I don’t wanna waste if it is not fully functioning right…
Does this mean, also, single-core CPU power is very important especially if someone is heavily inserting on ‘one’ track…?
What would be the best balanced bet for someone with intel i7 6-core 3.2 chip with 64gb ram, and to buy for the next desktop…?!
Sorry if I am being too annoying, leaving comments on everywhere! I just want to be sure of it
Better Performance From Modern CPUs
Developed with modern CPUs in mind, CLAP takes multi-thread management to a new level, with a clear and efficient allocation of roles between plug-in and host. Specifically, CLAP allows collaborative multicore support between plug-in and host through a so-called “thread-pool”, also allowing hosts to manage CPU-threading for plug-ins that provide their own multicore support. Preliminary tests show significant performance gains compared with current solutions.
Hilarious marketing. Don’t get me wrong. We need a proper standard. But let it be based on proper engineering and scientific principles.