Greetings. Was messing around with the NLS as I read it causes aliasing. Any chance of supersampling getting added? It would be a killer addition.
I certainly would like to see this too, but it only really poses a problem if the artefacts become audible. This extreme example results with aliasing down below -60dB which certainly has the potential to be audible, especially after compression. However, it is an extreme example.
A real world scenarios should see far less Although, many instances may amount to something more noticeable. If it’s a concern you can always roll off the super highs where it matters. before it hits the NLS. Personally, I’ve never really noticed it to be a problem running at 96k, but then, I like to use gates and LP roll-offs alot as well. Not because of aliasing, but it seems like it’s a good strategy nonetheless less.
Hi Simon, thanks for pitching in.
This test is actually just a clean 800Hz sinewave, with a modest drive setting and without clipping the NLS channel. Didn’t take an impossible amount to get the aliasing into the sub frequencies and below, even on 96kHz SR. So at 8kHz we’re already producing considerable aliasing down into the mids, on a single instance, even on this modest setting. Add NLS across an entire mix, crank up the drive, and one can start to imagine why some people find NLS to sound flatter than competitors solutions, it’s probably the aliasing.
But your tips are sound and you are right: We’re talking nuances here with light settings and responsible use. This is however what’s keeping me from setting up a virtual console with NLS, as I’d like to be able to use a lot of concurrent tracks, and feel free to actually drive those beautifully recorded consoles! My finicky workaround is to use a wrapper with supersampling, but this kills workflow if I want to change settings.
Edit: I’m sure 96kHz samplerate projects work nicely, it’s just not feasible in my case.
It’s a catch 22 as oversampling can help minimise this, but at an increased overhead. Making it an option and putting it in the decision of the user’s hand is probably the most intelligent way to handle it.
Though, no recording system has ever been perfect, analogue or digital. You just and up swapping one set of artefacts for another. People might say the analogue artefacts are more pleasant, but that’s true only at extremely low levels. It can be just as damaging to to music too,
To be honest I’ve never noticed NLS to be “flatter” myself, but that might be the 96k thing. Sure it isn’t significantly better, but better is still better.
I would also love to see oversampling on NLS! It still is a great tool, that I use on a daily basis on hundreds of mixes.