I run Cubase Pro 12. I don’t feel like it’s able to utilize my CPU in full. Read somewhere in an old thread it’s not good at using the benefits of multicore processors (mine has 12 cores). My question is: has cubase adressed performance optimization in the later updates? Will my problem be less or worse if I upgrade?
I have a Windows 11 PC with 64 Gb DDR4 RAM and a AMD Ryzen 9 3900 12-core CPU (3,1GHz-4,3GHz/70MB CACHE/AM4). It should be performing better than it is..
It depends on what CPU you have, but I would suggest following these threads, which discuss what’s going on in much more detail:
Have you considered that the work you are asking the CPU to do only requires a fraction of the CPU’s capacity.
I am asking this question do to Cubase maxing out and audio starts lagging. It can’t run my projects without freezing every track and having audio buffer size on 2048. And it’s still on the edge of it’s capabilities. My CPU should be able to handle the processes, but somethings seems to be wrong..
try the c15 demo
Don’t be coy, tell us which one.
Which is ? I ask this, with also more precisions about the OS, disk drives and RAM available if possible, because…
… something seems to be wrong indeed, but with your setup rather than Cubase, unless you have projects with 167 tracks, using an amount of 374 plug-ins.
I have an AMD Ryzen 9 3900 12-core CPU (3,1GHz-4,3GHz/70MB CACHE/AM4). it runs on a Windos 11 PC with 64 Gb DDR4 RAM
And the project right now is 75 tracks and 123 plugins…
Almost every single track is frozen. So its just the plugins on the buses that are running real time
Ok there is your issue, how you use the busses, how many busses and what plugins are on them. Be exact!
It would be more useful to see the CPU threads from the task manager menu as I”m sure despite your CPU saying only 12% being used, I bet you’ll have a thread or 2 maxing out.
M
This. ^^^^^
Yes, I’ve done extensive research during the day and optimized my system according to several sources I’ve found and realized I need to use something like Audiogridder. Which I’ve installed and fooled around with for a bit to understand. It seems like the effects on the CPU usage is great, but it also seems tedious to work with. Not a great interface and functions like showing the frequency curve from several tracks inside one Pro-q does not work.
What I’m wondering now is how other people do it to not have to use such software. Do people really have that effing powerful individual cores in their processors? Or am I missing some keypart in the system optimization? Guess I gotta go deeper into the rabbit hole to find out.
One of my best albums was recorded on a computer less powerful than my current phone!!!
M
Yes I have smaller projects as well. But sometimes you just need much layering and processing…
This looks like a good example to share over in the other thread I mentioned above. Link here for convenience: Core handling in coming updates for Win 11?
We know that Matthias at Steinberg has been reading that thread, so I suggest you link over to this thread from that thread so that he can see yet another example of the issue.
IMO it’s learning about how to structure a project so that you don’t gang up on a single core. This largely has to do with how buses work, as well as how plugins are loaded across tracks, and of course also learning which plugins are sucking up your CPU like crazy and planning accordingly. With good planning, I find Cubendo to work pretty well on complex projects and I can get some good, balanced core/thread distribution.
ALSO, lately I’ve been trying to change the way I look at projects… I’ve gotten WAY too lazy in recent years! So lately I TRY (in theory) to avoid doing EVERYTHING in one project file. Back in the “old days” you had different phases of a song’s production, from songwriting to tracking to producing to mixing to mastering, etc.. I TRY to break projects into similar phases like that instead of falling into the temptation that ALL phases of a song, from beginning to end, can be executed in one project file. I find when I break projects into phases like this, it has also helped my creativity and productivity.
Having said that, sometimes it’s not possible to split projects up into phases like that, and you can’t avoid having a very very complex project file, and then eventually you may run into the lopsided core usage situation that will cripple even the fastest CPUs. Steinberg DOES indeed need to address this and make things better. BUT, we live in the NOW, and not some fantasy land, so I tend to look for practical solutions that work TODAY. So in that case, YES, I do use other approaches like Audiogridder, VE Pro (locally, and also across more than one machine), and sometimes I have manually synced up two machines old school, and distributed projects across them via the old master/slave approach. (Am I allowed to use those terms these days? 
)
I will also mention this is one reason why I keep Reaper installed as well, for extreme cases, due to its better core/thread management. In fact, Reaper also has a feature called “subprojects” which is just pure genius on top of everything else, which allows you to compartmentalize extremely complex projects into a various nested subprojects, or “chunks” of any level of complexity, and YES, this has come in very handy to me over the years. (It’s also great for sound design and film cue management.)
The superior thread management, deceptively brilliant subprojects feature, and of course excellent click-and-drag ripple editing, are three key reasons why I keep Reaper around, actually, despite Cubendo’s other superior (IMO) features and workflows. So Reaper remains a “backup” secondary DAW for me. And BTW I have requested all three things from Steinberg over the years, over and over again, and agreed with threads that mention them.
But setting Reaper aside, in general I can build very complex projects without a sweat most of the time, even on older hardware, with Cubendo. BUT again, I just go back to the things I mentioned at the beginning of this post, structuring the project in a more optimal way, and then I get FAR better core distribution.
And BTW, this is NOT a big deal to have to do in most cases IMO. Planning and structuring your project based on the tools/plugins/tracks/buses that you need has ALWAYS been part of the production process, long before DAWs existed. We have always been limited by resources – tracks, channels, mixing hardware, outboard gear, number of buses, and so forth. And even in the first wave of DAWs and digital mixers, we were likewise limited. So we are now positively SPOILED for power and flexibility these days, and frankly, it’s contributed to a lot of people having very bad habits… including myself. But I’ve found that sometimes thinking of how to configure your project in advance can be helpful to the creative process too, instead of just stacking more and more and more on top of the system.
Anyway, there’s my long answer to your question. 
YMMV, good luck!
And yes, there’s always that. Throwing money at the situation to get a more powerful CPUs can help, BUT not always… there are a lot of reasons why this can be counterproductive sometimes too! For example, new CPU architectures may show remarkable improvements in benchmark scores, but thread scheduling and other kinds of issues can pop up that can actually make things worse in some cases for something like a finely tuned DAW. For example, when Intel released the hybrid P/E core architecture of the Alder Lake series (12th gen), Steinberg actually suggested for people to hold back buying them at first due to the thread issues that showed up, until they worked things out (which they did, and now they behave well of course). But if you had been an early adopter, you would have been having issues in the early days. So it’s a mixed bag when you go for the latest gen sometimes…
