Hi, I couldn't find any related topic in here ... it seems all audioz' users who run Cubase are doing great
I mostly use reaper but I like to keep an eye on good old Cubase as well. It runs decently on my machine (i7-13650) as long as I set ASIO guard to high (but I still get a few peaks once in a while, up to 50/80% at times). However, without ASIO guard on, I randomly peak at max within a minute or so, even doing "nothing".

I have been trying a few things on C12 and now C13 with limited success.
Disabling specific services (automatic updates, prefetch etc.) does not seems to do very much in my case. I added exception for Cubase folder/VST3/libraries, but that also seems to have no effect.
Disabling C-state in BIOS is what had the most noticeable impact, but then that's not a change I want to have permanent.

has anyone found anything else that works well?

thanks for your input

 

Last edited by a moderator: Jan 3, 2024

I’ve tried everything from disabling devices, power schemes, cpu core priority ect ect. Cubase & Nuendo have always sucked ass with random spikes… I recently tried out Reaper and was amazed how lean the entire program was on my resources. But I can’t have pre faded post pan sends in Reaper so I went back to the old dog Nuendo.

Ps/ Disabling ASIO guard helped a bit in my case

 

not really a feature I use but have you seen this thread? https://forums.cockos.com/showthread.php?t=159413

the lower I can get is "normal". Below that I get random peaks quite easily, unless I disable C-states, in which case peaks remain somewhat under control. Thanks for your input at any rate.

 

@lysergyk yessir I’ve seen that post in the reaper forums. I just don’t like the idea of making duplicates of tracks for a signal flow routing that’s been available for 2 decades in Slow Tools… But I understand every saw has their quirks.

 

Got some time yesterday so I did a bit more testing just out of curiosity and found that Cubase 12 offers way better performances than 13 for some reason.
From what I read online it seems C13 is generally better (which would make sense) but in my case that's when I started noticing problems, so I went back to C12.
Basically on my computer, C12 peaks at less than 25%, when C13 with the same settings is already above 50%...weird! I tested with the latest versions (12.0.70/13.0.20)
Playing with a lot of services and processes, I also found the Nvidia Container service had a pretty negative impact too, it even affected Reaper (though hardly noticeable). After disabling it, C13 behaves better but nothing like C12 anyway.
Is it the case for anyone else here?

 

Disabling Network Adaptor or Wi-Fri is best as this is the main culprit most of the time if you are connected to the internet.
Also you can run LatencyMon (Latency Monitor) to see which Processes may be causing issues.
https://resplendence.com/latencymon
Steinberg wrote something about Microsoft, AMD and Nvidia here:
Graphical User Interface errors in Cubase 13 / Nuendo 13 – Steinberg Support

 

I do not doubt your post; me not being on Windows at all. But that is some BS. That should never happen, even if it is what is happening.

 

lol wtf Steinberg.

 

wifi/network has 0 incidence in my case. This may have been the case a few years ago but I believe things have moved on these days
(I use wifi6).
LatencyMon gives all green light, but it is too basic to actually be helpful in this type of situation. However, Microsoft has excellent tools. The excellent (and free) sysinternals suite is what I use mostly, but the problem in my case seems to stem from C13, since C12 is doing pretty well. Certainly not as good as Reaper, but Reaper is nearly unbeatable when it comes to performance, so that's not the point here.

 

That's exactly what I said...how come C12 is doing better than its revamped brother? Shouldn't it be the other way around... I mean did they change something in the way C13 works?
my computer isn't old so there should not be any issue

 

Last edited: Dec 13, 2023

I would switch to Reaper in a heartbeat if it had pre fader post pan sends. I’m not about to make a template with double the tracks simply for a feature that Pro Tools has had since day 1… On a similar note. Steinberg and their 8 send limit is pathetic.

 

New Cubase 13 update today not sure if they fixed any of those issues with difference between C12 and C13

 

I gave it a try already....with the same services/processes running in the background, C12 performs still significantly better than C13 in my case.

 

Have a read through here also
Cubase 13: random CPU spikes still here, is maddening - Cubase - Steinberg Forums

 

Yeah it really shouldn't, but one thing that can happen is another USB device like a network adapter, wifi/bluetooth dongle sharing the same bus as the audio interface and eating up the bandwidth, because its transferring data, keeping the connection up or looking for new networks Never had that problem on Windows or macOS, tho.

 

Maybe Kenny found a solution for pre or post pan send in Reaper.

 

Sorry I forgot this...

I did help someone on here with a Nvidia related problem actually. The Crashdump file he supplied showed it was actually caused by the Nvidia driver. https://audiosex.pro/threads/problems-with-v-r-and-r2r-versions-of-cubase-12.64317/#post-713883

What GPU is in the machine? This is the last thing I would suspect when using a DAW software.

 

Last edited: Dec 13, 2023

Oh, absolutely no need to apologize That a GPU can act up is also news to me. I also forgot something. Doesn't Steinberg products still require a Syncrosoft e-license USB dongle? If it's legit, lysergygk could try to unplug all USB devices, that are not needed except the dongle and audio interface or switch ports around of those two.
Another speculation: maybe the cracked Cubase is looking for the dongle and polling a specific USB port like crazy, but runs anyway without it. Like DAWs and plugins are trying desperately to reach a server in the background, even when there is no network. Switching just the port of the audio interface or even running Cubase without a interface as a test could rule that out. Otherwise I am out of ideas.

 

No, that protection method was scuttled. That's how the sister site version releases can happen so fast since v12.

 

On first test I get the same performance on 12,13 I get better performace with ASIO guard on medium , did not try yet disabling it entirely. But i have an AMD . Apparently there were some changes in 13 concerning the intel e/p cores architecture . Maybe this needs further updates