New filters

[Toltekradiation]

@pikseliahky : yes! here it is !

[Toltekradiation]

and here is the version with volume compensation fix for the SVF filters .

[matrix12x]

@Toltekradiation, Really great work.

[Toltekradiation]

well, it seems i can't help trying things , and i find those filters really useful :
lowshelf & highshelf from https://cytomic.com/files/dsp/SvfLinearTrapOptimised2.pdf

[Toltekradiation]

adcIn version attached.

[baxgas]

@Toltekradiation Thanks, I'm really eagerly waiting to hear and test them! :-)

One thing I noted is that when I save the patches with the Filters, in the new version they are messed up, probably because of the numbered positions of the algorithms, and that they change eg. there are more filters before the Distortions?

Also, with the new OS, the parameter changes seem much more sluggish, eg. I have 4 Channels sequenced and I try to change a paramter and it increments only a few steps then stops, also the Inst change is much slower, with a blink of the LCD. Maybe these are already too much for the processor to handle? :-(

Can the filename of the bins be renamed so the PreenFm2 truncation of the characters is a bit more informative?

[Toltekradiation]

@baxgas, sorry for the messy project handling . Yes, that's because i added the new filters before distortions..
the processor is now a bit more busy than it was, i fear. i will now try to optimize my code more.

[baxgas]

@Toltekradiation I understand there is a limitation in processing power... btw. you used the overclocked version (210o), but maybe the "regular" version is a bit less demanding? (just an idea)

[Toltekradiation]

hi @baxgas, thanks for the suggestion, i tried the regular version, but no luck..
so i changed my testing procedure, to be closer from real use case, and, yes, sadly, the cpu started to overflow very quickly. so, some radical optimisations needed !
i removed oversampling on lot of filters ; in a mix you won't hear the change i guess
+ removed lot of waveshaping, well, the good news is that it sound cleaner now (or cheaper )
now it seems to work well with a 4 track sequencing. it still may overflow if you really insist, so i will try to free more cpu cycle if i can..
i may remove lp2 & hp2 as these are still oversampled, and now lp & lp3 seems to be enough.

[Toltekradiation]

good news, i found the "float cast" optimization and it make a big difference ,
i could bring back oversample on the lp3, hp3, bp3, peak filters !
for a test, i could have 4 parts running fast with per step program change, and it kept smooth !
the other optimizations was a good thing anyway..

[Toltekradiation]

here is the adcIn version

[baxgas]

@Toltekradiation Thank you for the improvements, now it works really smoothly at my PreenFM2, fantastic 
I will test it thoroughly and let you know if I find some more tips for improvement, but for now it seems really-really great!
One thing I noticed with the file-naming is that the OS Flasher recognises the .bin only if it's filename is starting with p2_ Otherwise it is great you renamed them, so the filename truncation is much more informative!

Just a question, will the position of the effects move in some future version (planning adding some more), so I can track which parameters were saved with the patches? Now the old patches load in with a bit different FX, I guess because it saves the ID, and it changes if some new are added before.

Thank you so much for the great job you done with these improvements!

[kyoto]

good work - thank you very much

[Toltekradiation]

hi guys!
i'm very glad you're enjoying my work !
i may leave the version like that, but don't hesitate to suggest ideas or requests,
it seem usable enough, so, i don't plan to add any other fx or change the fx list.
maybe polish it a bit, but would have better to really dive into sound design now .

[maki]

Hello, could you then make a tracking generator for each slot in Modulation Matrix?