As a very basic an example I'll start with a short audio loop, duplicate it a few times, drag them off to the left and the right to chop it up*, run a script to copy the arrangement and double it, go back to my individual slice layers, duplicate them, change their position in the second "bar", run the script again, etc.A_SN wrote: Just being curious, in what way does doubling the image help you create complex sequences?
Sometimes I'll open a blank document crop it to .3 seconds, draw a bit of percussion, make a new layer with a synth type noise, and do the same process. The phrase "less is more" could almost be the official slogan of photosounder. I mean, generally when making synth type noises to get usable sounds I have to get very detailed on a very small image. I guess for me more "intelligent" subtractive tools would help.
Between the move tool, flipping and the various blend modes you can add quite a bit of complexity. Now, this is a new way of working for me and it often sounds like something out of the BBC radiophonic workshop. I mean I'm generally used to playing on a keyboard live to do most of my sequencing so it's a matter of figuring out what types of visual arrangements are sonically pleasant.
Another important point is that I have all of the layer command and tool commands mapped to shortcuts.
Hope that makes sense. I should make a video of how I use photosounder, hopefully someday.
*I'm not sure if this behavior is a bug, but it's incredibly useful, please keep it! It's also a great way to make selection shapes. I make a new layer make a dot with my white spray tool the distance from the 0,0 point on the x,y axis I want my rectangle to be then I fill maxima left and bottom. I also made shortcuts for these.
Yeah, both are useful. I still think someday expanding on the magnet functionality optionally to snap to timing or things like semitones, whole tones, etc. would help.A_SN wrote: Well, for rhythm you've got the 4/4 rhythm overlay, I know it doesn't change the image but it can help you transform it accordingly. As for harmonic rules you have the operation to group everything to the nearest semitone, which is pretty useful when used on a layer that only represent a base frequency.
Yeah, convolution needs a bit of a warning or automatic lowering of the spray width. Sometimes I accidentally click while it's set at 60 or something and it takes 2-3 minutes.A_SN wrote: No need for a script to do that, you can use a convolution layer with a bunch of tiny dots on it.
A_SN wrote:Not sure what what you're mentioning does, feel free to mail me examples.
If you use a script along the lines of:
Duplicate layer
Move layer +3 % x +4 %
Duplicate layer
Move layer +5 % x +4 %
It will depend on the size of the layer how fast it happens. Or you can do other scripts that scale such as
Duplicate layer
Move layer +9 % x +4 %
Duplicate layer
Move layer +10 % x +4 %
or
Duplicate layer
Move layer +3 % x +8 %
Duplicate layer
Move layer +5 % x +8 %
My ears are the final arbiter, but it can be useful to use math like this as well.
RE: rendering audio on another computer on a network
Especially for lossless I think I'd find it very useful. When I'm feeling "musical" having to sit around and wait on my computer can be excruciating.A_SN wrote:Sure I guess that could be done. But would it really be convenient?
Oh, that's great! For some reason I thought it'd have to copy the license file over.A_SN wrote: Well you can have it on a USB key and run it from there. It won't even leave any tracks on the host machine.
I think I do almost everything the "wrong" way in photosounder.A_SN wrote: It's funny that I don't really have any crashes myself. Except a few days ago when I loaded a 10 min sound and then turned on live synthesis, but it displayed a clear error message saying it couldn't allocate the ~200 MB buffer it had to. Since I'm on a 64-bit system now I should try to make a 64-bit build, at least for Windows.