[Csnd] Spectral filtering
| Date | 2007-12-15 12:24 |
| From | jpff |
| Subject | [Csnd] Spectral filtering |
I have written and tested (basically) two opcodes to filter in the
spectral domain.
fsigout pvsbandp fsigin, klowcut, klowfull, khighfull, khighcut
fsigout pvsbandr fsigin, klowcut, klowfull, khighfull, khighcut
for bandpass and bandreject filters, with a bandpass like
klowfull __________________________ khighfull
/ \
/ \
/ \
/ \
/ \
________/ \______________
klowcut khighcut
and the reject is the complement of that
In the CVS now. Now to write the manual pages .....
==John ffitch
"This research is of insufficient standard"
|
| Date | 2007-12-15 23:26 |
| From | Tim Mortimer |
| Subject | [Csnd] Re: Spectral filtering |
This is really cool, Do-able with the existing opcode set (although i believe the original request came from someone concerned about real time latency...) I was planning on putting this "feature" in my own PVOC instruments, but had not because they were starting to get a "bit complex" & unwieldly.... now i can just add one line of code! thanks mr ff. jpff wrote: > > I have written and tested (basically) two opcodes to filter in the > spectral domain. > > fsigout pvsbandp fsigin, klowcut, klowfull, khighfull, khighcut > fsigout pvsbandr fsigin, klowcut, klowfull, khighfull, khighcut > > for bandpass and bandreject filters, with a bandpass like > > > klowfull __________________________ khighfull > / \ > / \ > / \ > / \ > / \ > ________/ \______________ > klowcut khighcut > > and the reject is the complement of that > > > In the CVS now. Now to write the manual pages ..... > > ==John ffitch > > "This research is of insufficient standard" > > > Send bugs reports to this list. > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe > csound" > > -- View this message in context: http://www.nabble.com/Spectral-filtering-tp14350379p14356328.html Sent from the Csound - General mailing list archive at Nabble.com. |
| Date | 2007-12-15 23:32 |
| From | Tim Mortimer |
| Subject | [Csnd] Re: Spectral filtering |
can the low cut value extend to < 0, & the highcut above the highest bin? No audio effect, but retaining stability of operation in that case would enable them to be used on the fly as lowpass & higpass as well, without the need to "recalculate" the notch width.... you could add lines to prevent this of course, but i thought may as well run with the "one line of code reqd only" approach & accomodate any "erroneous values" as simply "at the limit of the spectral domain" Tim Mortimer wrote: > > This is really cool, > > Do-able with the existing opcode set (although i believe the original > request came from someone concerned about real time latency...) I was > planning on putting this "feature" in my own PVOC instruments, but had not > because they were starting to get a "bit complex" & unwieldly.... > > now i can just add one line of code! > > thanks mr ff. > > > > jpff wrote: >> >> I have written and tested (basically) two opcodes to filter in the >> spectral domain. >> >> fsigout pvsbandp fsigin, klowcut, klowfull, khighfull, khighcut >> fsigout pvsbandr fsigin, klowcut, klowfull, khighfull, khighcut >> >> for bandpass and bandreject filters, with a bandpass like >> >> >> klowfull __________________________ khighfull >> / \ >> / \ >> / \ >> / \ >> / \ >> ________/ \______________ >> klowcut khighcut >> >> and the reject is the complement of that >> >> >> In the CVS now. Now to write the manual pages ..... >> >> ==John ffitch >> >> "This research is of insufficient standard" >> >> >> Send bugs reports to this list. >> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >> csound" >> >> > > -- View this message in context: http://www.nabble.com/Spectral-filtering-tp14350379p14356339.html Sent from the Csound - General mailing list archive at Nabble.com. |
| Date | 2007-12-16 03:26 |
| From | "Hector Centeno" |
| Subject | [Csnd] Re: Re: Spectral filtering |
| Attachments | None |