I've fooled around with VCV Rack some, but it was with a specific purpose: coming up with an accordion-like sound. I found it really difficult. It seems like there are some instruments that are...
I've fooled around with VCV Rack some, but it was with a specific purpose: coming up with an accordion-like sound. I found it really difficult. It seems like there are some instruments that are pretty easy to do (like getting a decent bass sound) and others, I hardly know where to begin.
Have you tried aiming for specific sounds or is it more an exploratory process?
At the time I bought an app called Syntorial. Although I didn't finish the lessons, I still recommend it. (It's not specifically about modular synths, but I think the ear training would still be...
At the time I bought an app called Syntorial. Although I didn't finish the lessons, I still recommend it. (It's not specifically about modular synths, but I think the ear training would still be very useful.)
However, it was mostly helpful for reproducing sounds that were made on a synthesizer originally. If you hear a synthesizer used in a song and wonder how to make that sound, it seems like the right training to have. Reproducing natural instruments, not so much.
Yeah, an exact replica isn’t something I care about and physical modeling is what I’d ideally want to do. The problem is that it requires research and it’s only been done for more popular...
Yeah, an exact replica isn’t something I care about and physical modeling is what I’d ideally want to do. The problem is that it requires research and it’s only been done for more popular instruments. Sure, it’s been done for strings, tubes, and percussion. I also saw a series in Sound on Sound about emulating a violin.
I went looking for research on the physics of accordion sound and it’s pretty thin. I saw a citation for one paper that might be useful but it’s not online.
The practical approach is probably to give up and use a soundfont.
Uh, thanks! Don't feel obligated but maybe it will be fun? Some background: I've recorded samples of notes from my accordion and looked at the FFT analysis in Audacity and other tools like Sonic...
Uh, thanks! Don't feel obligated but maybe it will be fun? Some background:
I've recorded samples of notes from my accordion and looked at the FFT analysis in Audacity and other tools like Sonic Visualizer. It's neat in a way but didn't really help me figure out what to do in VCV Rack.
A problem with FFT is you need a fairly long sample to get much detail in the frequency domain, and it doesn't tell you about the transients or how the sound evolves, which seems important because accordion notes are often played staccato. I think if the sound doesn't evolve then it's just going to sound like a buzz saw?
Another tool that seems pretty neat for looking at how the sound changes is SPEAR, though I'm not sure what to do with it and there is no Linux version.
The sound I'd really like to make though, is from diatonic accordions, which isn't what I play. The opening of this song for example.
If I got into again I'd probably look at other free-reed instruments and see what's been done. Here's a web page with lots of links about the scientific study of harmonicas. The Journal of the Acoustical Society of America (JASA) seems to have a lot of the best papers, but the papers aren't online as far as I can tell.
I've fooled around with VCV Rack some, but it was with a specific purpose: coming up with an accordion-like sound. I found it really difficult. It seems like there are some instruments that are pretty easy to do (like getting a decent bass sound) and others, I hardly know where to begin.
Have you tried aiming for specific sounds or is it more an exploratory process?
At the time I bought an app called Syntorial. Although I didn't finish the lessons, I still recommend it. (It's not specifically about modular synths, but I think the ear training would still be very useful.)
However, it was mostly helpful for reproducing sounds that were made on a synthesizer originally. If you hear a synthesizer used in a song and wonder how to make that sound, it seems like the right training to have. Reproducing natural instruments, not so much.
Yeah, an exact replica isn’t something I care about and physical modeling is what I’d ideally want to do. The problem is that it requires research and it’s only been done for more popular instruments. Sure, it’s been done for strings, tubes, and percussion. I also saw a series in Sound on Sound about emulating a violin.
I went looking for research on the physics of accordion sound and it’s pretty thin. I saw a citation for one paper that might be useful but it’s not online.
The practical approach is probably to give up and use a soundfont.
Uh, thanks! Don't feel obligated but maybe it will be fun? Some background:
I've recorded samples of notes from my accordion and looked at the FFT analysis in Audacity and other tools like Sonic Visualizer. It's neat in a way but didn't really help me figure out what to do in VCV Rack.
A problem with FFT is you need a fairly long sample to get much detail in the frequency domain, and it doesn't tell you about the transients or how the sound evolves, which seems important because accordion notes are often played staccato. I think if the sound doesn't evolve then it's just going to sound like a buzz saw?
Another tool that seems pretty neat for looking at how the sound changes is SPEAR, though I'm not sure what to do with it and there is no Linux version.
The sound I'd really like to make though, is from diatonic accordions, which isn't what I play. The opening of this song for example.
If I got into again I'd probably look at other free-reed instruments and see what's been done. Here's a web page with lots of links about the scientific study of harmonicas. The Journal of the Acoustical Society of America (JASA) seems to have a lot of the best papers, but the papers aren't online as far as I can tell.