Rendering volume

Started by herodotas, January 11, 2020, 08:45:43

Previous topic - Next topic

herodotas

Hello,
I have some questions about:
Why rendered sample volume is not the same like original in pattern? It is not possible to do 1 by 1 rendering?
And also noticed about metering plugins ( I use voxengo SPAN and Melda loudness analyzer) do not work with VST instruments, only with samples.
life is darker than it seems

herodotas

OK I found answer about rendering. But whats about plugins?
life is darker than it seems

Saga Musix

QuoteWhy rendered sample volume is not the same like original in pattern? It is not possible to do 1 by 1 rendering?
That is exactly what OpenMPT does. What are you comparing the output against?
The only reason why the volume could be different is that your output is clipping. If you export to unclipped floating-point audio, playback in another application will most likely be affected by the limiter in Windows' mixer.

QuoteAnd also noticed about metering plugins ( I use voxengo SPAN and Melda loudness analyzer) do not work with VST instruments, only with samples.
SPAN works with any source. I assume that you try to apply it to the pattern channel on which the instrument is playing. This does not work because plugin routing cannot be "split up" between channels. Imagine your instrument playing on two different pattern channels, one pattern channel sending its output to SPAN and the other one doesn't. How is that supposed to work?! OpenMPT only gets a single output stream from the plugin, it cannot figure out which part of that stream was coming from the one pattern and channel and which came from the other.
The solution is to set SPAN in the "Output to..." dropdown of the instrument plugin.
» No support, bug reports, feature requests via private messages - they will not be answered. Use the forums and the issue tracker so that everyone can benefit from your post.

herodotas

Short example module about rendering
life is darker than it seems

Saga Musix

The sample is just as loud as it should be. But your global sample volume is set to 48 (-8dB), so if you import the sample back into OpenMPT, it will be played at -8dB, not at 0dB.
» No support, bug reports, feature requests via private messages - they will not be answered. Use the forums and the issue tracker so that everyone can benefit from your post.

herodotas

Thanks, I found this to. Maybe is logical volume sliders set 0 dB by default in .mptm. I dont like to use templates.
life is darker than it seems

Saga Musix

QuoteMaybe is logical volume sliders set 0 dB by default in .mptm.
It isn't really, because that means that as soon as you play more than one sample, your mix is almost guaranteed to clip.
Yes, you could "fix" that by setting any of the other sample volumes to half their value by default, but due to the very limited range of those settings (0...64), this would result in a loss of precision of volume commands. It's something that we could do in the (far) future if we extend MPTM's volume settings to be more fine-grained.

Quote from: herodotas on January 12, 2020, 10:49:54
I dont like to use templates.
Why? That is not a very rational thing to do when using a template solves exactly this problem for you.
» No support, bug reports, feature requests via private messages - they will not be answered. Use the forums and the issue tracker so that everyone can benefit from your post.

herodotas

#7
I dont know why I dont like templates, maybe it is the head problems, but anyway I found another solution, make all sliders default volume 64 (-6.02 dB) and now you have some headroom for avoid clipping. But the best option  would be to add proper mixer in .mptm  ::)

No, thats not work :(
life is darker than it seems