Does Signal Follower = A Sidechain Compressor?

Im confused.

Say I got 2 tracks, both tracks have a different breakbeat. If I have the native compressor on track 2. and feed the signal from the (track 1s) signal-follower into track 2s compressor, what parameter do I need to assign to get a sidechain compressor?.. using my ear I guesed the Ratio… am I right? it sounded prety pumpin’n’bumpin after I did that

You add a gainer to track 2 and you assign the Signal Follower to that (or if you want you can assign it to the tracks ‘vol’). And it’s the sidechain compression. Can you imagine you don’t even have to use a compressor for that :P

If the signal follower had a checkbox to invert it’s output you could just connect it to the volume control of track two. It hasn’t, but I believe it was suggested in another thread.

Simply invert min and max values and voilà, you got your sidechain like operation.

Thanks kRakEn/gORe, I didn’t work that out when experimenting last night :) Great work by the way!

thanks guys

so awesome!

ya… from my understanding, you can’t actually sidechain a compressor in Renoise. Using signal follower, you can take the volume output parameter of one track to automate the parameter of another dsp or other attributes but it’s automation, not actual signal routing. To have compressing pumping, you need to take the volume output of one track and feed it as the input signal to one of the vst sitting on another track (a compressor). Right now, you can tweak every knobs on a vst with the input from another track but not actually have it go into the input signal of the vst so they went 99% of the way but not quite there.

There’s nothing in the compressor parameters except the actual input you can put your signal from the other track in to make it a true sidechain compressor. All parameters are supposed to be static as you pump dynamic signal into a compressor input.

In practice it’s not a good idea to use a Signal Follower as trigger for a sidechain sim. Why? Because there’s a delay of the passed trigger signal with a time amount dependend on your audio buffer size.

When you’re running a 50ms buffer with your ASIO driver, the trigger signal passed by the Signal Follwer will be 50ms too late. Accordingly with any other buffer times.

That’s at least what happens on Windoze. Can’t tell, if it’s the same case on other OSs.

I doubt that: That latency is usually for the overall output (What Renoise feeds to the ASIO card in total through the master track) and not just for one track. But perhaps your ASIO card allows you to set different latency values per output channel and you have different output channels assigned to separate tracks? The signal follower listens to the signal before the audio is output. But you might be right if the signal follower is influencing a DSP parameter on a track that is on the very far right end of the pattern whereas the signal follower is put on one of the most left tracks. Renoise processes tracks from left to right and the more latency is caused by processing time of the DSP chains on the tracks in between, the more it might affect the moment of parameter change of the DSP. I know PDC tackles the audio stream delay but i’m personally not sure if PDC tackles parameter triggering delay as well. (If all logic is done right, it should tackle it)

Just try it yourself. Here’s a setup, that layers noise on a group, triggered by a signal follower. Just use a low buffer setting for one turn and a huge buffer setting for another turn. I could provide several other setups, that prove the same behavior for all of them. There’s a difference in the time gaps between using DX and ASIO drivers, but still both have those gaps. And there’s really nothing complex or special processed here.

Edit: Okay, checked this with a single CPU instead of 4 (or any other “multiple”-CPU) and got it back in sync then. So this also seems to be related to multi-CPU usage. But as I told, it’s also related to the buffer size then, when using multiple CPUs.

I’ll be damned (tested it with Dsound, the latency buffer adjustment of my ASIO driver only affect recording latency), not quite logical behavior though, specially not when rendering.
Learned something new.

Agreed. ;)

It is, but if you are depending on solutions like Asio4All, i doubt you can minimize it to 1-2msecs. 5 to 10 is a more realistic number.