Mono or stereo samples?

[lmv]

I am working on a pet project that involves reusing some very old samples I have, almost everything drum-related. I have them in mono, but depending on what software I use to play them, they sound like they're stereo. They usually sound "better," but they often sound just different, and I'm not sure whether this "different" is better or worse. Sometimes, they sound "better" as mono, by "better" I mean they seem to contain more audio information than their stereo playback counterpart. They sound a little weird because they only play into one ear, but do seem to have a little more detail.

Now, I can choose between using them as mono or resampling them as stereo, but I'm not aware of whatever implications there might be ahead. Are there any? Which one should I use?

[Unaspected]

Converting a mono file to stereo will only take up more hard drive space.

If you are talking about processing them with a stereo effect and then rendering the result as a stereo file, then be sure to check for mono compatibility.

However, I wouldn't. I would leave them as mono and then treat them as required when mixing.

[lmv]

What does "mono compatibility" mean?

[Unaspected]

What does "mono compatibility" mean?

That your stereo signal collapses to a mono signal without any loss of character.

Some processes which make mono signals appear to be in stereo do so in ways that might cause wave cancellation when the two signals are summed as one.

[BertKoor]

[...] be sure to check for mono compatibility. [...]

Anticipated question: how does one test for mono compatibility?
Answer: for instance with a Mid/Side util plugin, select only the Mid channel.

[lmv]

Well, I'm using the desktop Caustic application to play and recapture the samples, and it plays them in stereo. I have the samples in mono, but I used to compose rough sketches in Caustic with certain changes applied to them.



I want to reuse those old sketches and do something with them in a proper DAW. It's very important that they sound exactly the same. Using just the raw samples will be very difficult because of the tweaks. Recreating pan is very easy, recreating tune is quite possible but imprecise, and recreating "punch" and decay may be well nigh impossible. I don't even know what "punch" is. I know it often sounds better, but I don't know what actual effect it is. And how much "decay" is really applied there? What is the application's idea of decay anyway?

But the answer I was looking for is whether "converting" the samples from mono to stereo involves any kind of liability.

[BertKoor]

Punch & Decay sounds like it's controlling a 2-knob compressor. Feed it a properly designed test signal, and it will become apparent what it does.

[...] whether "converting" the samples from mono to stereo involves any kind of liability.

Liability as in the copyright on the samples? Processing in any form does not change that.
But I think you're talking about something else.

[BertKoor]

From https://singlecellsoftware.com/docs/cau ... atbox.html

12. Punch knob: Controls how quickly the sample fades in. Can be used to remove the aggressive attack spike on most percussions.
13. Decay knob: Controls how quickly the sample fades out.

No mention of any mono-stereo processing. Again, do a proper test to (dis)prove it. I don't really hear anything like that (on Android)

[lmv]

I'm not sure what you call "mono-stereo processing." I hear it in stereo in the android app and in the Windows desktop version.

[whyterabbyt]

I'm not sure what you call "mono-stereo processing." I hear it in stereo in the android app and in the Windows desktop version.

When audio software that outputs a stereo mix for a stereo (speaker or headphone) setup plays a mono sample, that mono signal will be panned somewhere in the notional stereo field. That will typically be in the centre, which means that equal amounts of that signal are being sent to both left and right. outputs.

When you say you are hearing it in stereo, if that's what you're hearing, equal volume in both headphones, then that is the normal, expected behaviour because you're listening in stereo.

Just because you're hearing something come out of both speakers, that doesnt mean what you're hearing is a stereo signal. Its likely to be a mono signal panned in stereo.
And the only way you could hear it in actual mono is to only use one speaker.

However, if the mono signal has been processed in some way to actually make it some sort of actual stereo signal (ie effectively turning it two different simultaneous signals, one for each of the left and right outputs) then that's a different matter.
That process would be called 'mono-stereo processing'. As in 'mono to stereo processing'. There are various ways this can be done in software, but it'd be exceptionally strange to find it being done automatically, with no control by the user.

[lmv]

What I hear is sound coming out of two speakers. In my case, earbuds. I don't have speakers.

I run Caustic for desktop and play the samples individually. I capture them with an application that will capture whatever sound the computer is playing. Those captured samples are *some kind of* stereo, i.e. two speakers/earbuds.

I cannot play the samples as mono from Caustic. Caustic for desktop plays them as "some kind of stereo." I have the original mono samples, but when I capture them as "whatever the computer is playing," they become stereo. That is confirmed by Audacity. The original samples are identified as mono, the captured samples are identified as stereo.

The captured samples also sound different from the original samples because they are played by Caustic with tweaks. It's because of the tweaks that I'm doing this. I want the samples to sound just the way they do in Caustic without going through the trouble of tweaking them (and, most likely, never achieving the exact same result).

[Unaspected]

If there is no specific stereo processing going on with the output - with the only stereo effect being the pan control - you could feasibly sample them on a single channel and save as mono files.

Probably doesn't matter much these days as everyone has surplus space on their hard drives and fast internet connections for file sharing - but I still think that understanding redundancy is important in these cases - it is best practice: Essential if creating products but less important if mixing your own work without needing to share project files.

However, if you are planning on mixing these sounds then I would recommend tracking with all channels panned to centre and saving them as mono channel signals. If you are sampling them out to use in another sampler then I would definitely recommend doing this.

You can always perform a null test by inverting one channel and summing the two - with all sounds panned to centre. There should be complete silence if there is no additional stereo processing.

[whyterabbyt]

What I hear is sound coming out of two speakers. In my case, earbuds. I don't have speakers.

Yes, and as Ive said, that will always be the case for any mono sound which is played back in stereo. (Unless you specifically pan it fully left or fully right.)

I have the original mono samples, but when I capture them as "whatever the computer is playing," they become stereo.

Yeah, that's also covered by my explanation. The Caustic software is playing back in stereo; your mono sample is being mixed between the stereo outputs. This is normal to DAWs and other audio software, as I said.

If you've got Audacity, and you want to make a stereo file mono, you can export one or other channels. Split the stereo file to mono, mute unwanted channels, export.

[lmv]

The channels are not identical. How am I supposed to choose which one to lose?

[BertKoor]

That is confirmed by Audacity. The original samples are identified as mono, the captured samples are identified as stereo.

The only thing Audacity is looking at is the number of channels. One channel = mono, two channels = stereo, more than 2 = surround.

The captured samples also sound different from the original samples because they are played by Caustic with tweaks. It's because of the tweaks that I'm doing this. I want the samples to sound just the way they do in Caustic without going through the trouble of tweaking them (and, most likely, never achieving the exact same result).

And what exactly is the difference? What have you done already to investigate? Do you have any theory on what's happening there?

The channels are not identical.

I very much doubt so, you have not provided any proof yet. Placebo effect is usually subtle enough. I currently don't have a working Windows machine, otherwise I would do some proper tests to (dis)prove this. Either a nulling test or view the side channel, both are easy, technically the same.

How am I supposed to choose which one to lose?

You have more options:

• full stereo (why not)
• left channel only
• right channel only
• mid channel only (summed left & right)

Not my project but yours. So you pick what you think is best