Hello, I’m having issues gain staging from Groove Agent SE 5 in Cubase Pro 12 and Pro 13. (Latest versions for all)
I design drum samples in Xfer Serum, export wave files, and load them into Groove Agent to create kits. While my samples individually have sufficient headroom, I just ran into an issue where a kick drum sample quickly repeats as 32nd notes in the production and the kick drum track peaks at +0.5. I haven’t ran into this issue since I have started making my own samples.
Within GA, I assign outputs to all my samples. I do not use “Kit Mix” or the “Agent” buses. How do I gain stage when sending the samples to Cubase? Is this possible?
My gut feeling is mixer > rack > pre > gain. Then simply drop the gain so I’m not clipping. But would this work or would my kick drum sample still technically be clipping?
I know this would make it quieter, but would it also truly stop it from ever clipping in the first place?
Where is the first place that I can turn down the signal to safely create headroom without my sample still clipping later down the signal chain? Is there something else I need to do within GA that I’m missing?
Worst case, I supposed I can assign the hot sample to a bus within GA mixer and turn down a fader there. But, ideally, I feel like there has to be a way to properly gain stage the sample the way my outputs are currently setup.
How would you correctly gain stage a sample in this situation?
I really appreciate any help and knowledge! Thanks!
There are many ways to check if your drum samples clip. One quick and easy way is to use Statistics Window.
The reason your GA output is exceeding 0dB while your samples are not, is likely that you’re triggering them so fast that the samples overlap and the sum of them yields a higher volume output. In such a case I would simply turn down the kick sample in GA.
PS. It might be a good idea to export your samples a little quieter. In a mix, you seldom you seldom want your snare e.g. to peak at near 0dB.
I really appreciate your reply! Thanks for letting me know about the Statistics Window. I was unaware of it!
Unfortunately, I think there might be a miscommunication. Sorry, if I’m not being clear. I’m not sure how to turn down the kick in GA. Since I’m not using the “Kit Mix” or any “Agent” buses, none of the faders in the GA mix section have any impact on the level. Also, none of the meters within GA visually display any audio signal. Do you know if there’s another way to adjust the level within GA?
Since my samples are assigned to different outputs in Cubase, I’m beginning to assume this is why no fader within GA affects the level. That’s why I was thinking that pre-gain might be the best way to back off the level as I considered the output that it’s assigned to would be the equivalent of an agent bus. But I’m not sure if it’s meant to work this way.
If it does work this way, it would conflict with how the outputs work when I use other drum samplers. For instance, when using Kontakt in Cubase, if samples are coming out hot and I use multi-out routing, if I turn down the faders in Cubase mixer then they’re still clipping. I have to turn down the level within Kontakt so they don’t come out hot. I assumed it would be the same when using GA. But I don’t see a way to lower the level in GA.
Is it possible that it doesn’t matter how the sample is lowered? Similarly to how some Steinberg stock plugins, like frequency, you can go in and out hot as long as it’s eventually lowered before the signal hits something that has headroom restrictions?
Thank you! I have no idea how I missed something so obvious. I guess I was so busy looking for something in the mixer that I completely overlooked this. Thanks again
Even though the actual questions is answered, I’d like to confirm this particular thought of yours.
Indeed as long as your signal is “in the box” and the actual values are in a floating point format (every DAW nowadays uses floating point numbers for their audio engine) you cannot realistic clip an audio signal.
The indication of a clipped signal actually only is a warning that, if you leave the realms of floating point values, you’d get a clipped audio signal. Practically that means clipping occurs only when the signal leaves the box (when Cubase passes the audio data over to the audio interface driver or if you render an audio file with 16 or 24 bit resolution), not before.
In theory you could mix everything to +100 dBFS and there would be no clipping of any kind as long as you put the final output fader to -100 dB.
NB: Before anybody starts an argument why you cannot mix to +100 dB - I know why but these are all “artificial” limitations programmed by plugin vendors and DAW makers, not physical limitations like in the analog world. Not that there’s anything wrong with that.
Thanks! Now I don’t feel as bad that I already exported the project stems before I got the solution (deadline) by turning down the pre gain on the kick output channel lol. Turns out it would be the same either way
I always save/render any audio file, that is not the final product, in a floating point format, I find 32bit float to be sufficient. This way I can always lower a “too hot” signal at any time.
