Hi !
I'm on a patch based on FFT, what I want to do is to record a loop and play with it.
so far, I manage to use the "store array" to memorise sequenced FFT : It work fine with 128 memo slots, but I need much more !
I tried 1024 memory slots but HH3 crashes !
my guess is that I would have more luck with a SDK module than with a script. (all I need is 4 x 1024 slots of 1024 elements = 4,194,304 floats to store, that is less than 2 minutes of audio)
what do you think ?
cheers
store array script and FFT
http://oli-lab.org
Win11 Ryzen9/32GB RAM - RME MADIFACE - SSL alpha link 4-16 - OSC capable interfaces
follow OLI_LAB adventures on Mastodon
@olivar_premier@mastodon.social
Win11 Ryzen9/32GB RAM - RME MADIFACE - SSL alpha link 4-16 - OSC capable interfaces
follow OLI_LAB adventures on Mastodon
@olivar_premier@mastodon.social
i *feel yes sdk would behave much better than script for this, still pretty wild amount of data hehe. 
it should handle play/rec in 'ram' if arrays were allocated i suppose, what might be more tricky maybe is if need hard offline/store/recall, might be possible with preset but i do feel preset manager won't like thart much recalling 16 millions bytes.. have to try
just an idea that prob won"t fit your needs here, but an alternative way could be to use a 1024 asio block, and feed that 'fft frame' as audio block to rec/play in sampler.
or for ex 512 would go to 2 samplers, 256 to 4 ect, then on play would recombines the arrays. or maybe there is a way one shot in a multichannel sampler, i don't remember if we can rec multi channel.
it should handle play/rec in 'ram' if arrays were allocated i suppose, what might be more tricky maybe is if need hard offline/store/recall, might be possible with preset but i do feel preset manager won't like thart much recalling 16 millions bytes.. have to try
just an idea that prob won"t fit your needs here, but an alternative way could be to use a 1024 asio block, and feed that 'fft frame' as audio block to rec/play in sampler.
or for ex 512 would go to 2 samplers, 256 to 4 ect, then on play would recombines the arrays. or maybe there is a way one shot in a multichannel sampler, i don't remember if we can rec multi channel.
yeah ! thanx for your answer.
I was thinking of some blocksize way to store and play the data, just like with audio, should be more optimized this way.
I will also do more testing with the phase vocoder, that give more pleasing sounds than the FFT, even at smaller sizes.
cheers
Olivar
I was thinking of some blocksize way to store and play the data, just like with audio, should be more optimized this way.
I will also do more testing with the phase vocoder, that give more pleasing sounds than the FFT, even at smaller sizes.
cheers
Olivar
http://oli-lab.org
Win11 Ryzen9/32GB RAM - RME MADIFACE - SSL alpha link 4-16 - OSC capable interfaces
follow OLI_LAB adventures on Mastodon
@olivar_premier@mastodon.social
Win11 Ryzen9/32GB RAM - RME MADIFACE - SSL alpha link 4-16 - OSC capable interfaces
follow OLI_LAB adventures on Mastodon
@olivar_premier@mastodon.social
Who is online
Users browsing this forum: No registered users and 248 guests
