did i write 5-6 seconds of delay? i guess when you're playing at a normal volume, sounding like 'stepping 5 or 6 meters away from their guitar amp' doesn't really matter
On Fri, Feb 5, 2021 at 10:35 PM Zhiguang Zhang <[email protected]> wrote: > cool to see re: your work, although i still don't see how as much as 5-6 > seconds of delay in a pitch shifter pedal is at all acceptable for any > guitarist in a live setting. what kind of pitch shifter would that be? > some weird boutique pedal? i don't know .. > > On Fri, Feb 5, 2021 at 10:02 PM robert bristow-johnson < > [email protected]> wrote: > >> > On 02/05/2021 4:05 PM Zhiguang Zhang <[email protected]> wrote: >> > >> > >> > So the guitarist using the pitch shifter might have a delay that is >> like stepping 5 or 6 meters away from their guitar amp. >> > >> > that sounds like an acceptable amount of delay in any live situation, >> just from using a guitar pedal? does such a guitar pedal even exist? >> > >> >> there's a few. >> >> >> https://www.sweetwater.com/store/detail/Whammy5--digitech-whammy-5-pitch-shift-pedal?mrkgadid=3316424859&mrkgcl=28&mrkgen=gpla&mrkgbflag=0&mrkgcat=guitars&&acctid=21700000001645388&dskeywordid=92700046937561040&lid=92700046937561040&ds_s_kwgid=58700005285197088&ds_s_inventory_feed_id=97700000007215323&dsproductgroupid=545466810937&product_id=Whammy5&prodctry=US&prodlang=en&channel=online&storeid=&device=c&network=g&matchtype=&locationid=9003010&creative=303058207581&targetid=aud-994727059085:pla-545466810937&campaignid=1465808371&gclid=CjwKCAiA9vOABhBfEiwATCi7GLbL_eakqEtKvyJ0lbn0T79ldvAbH0LjTJqKyHsSEgv0WJJZvoRi-RoCdJEQAvD_BwE&gclsrc=aw.ds >> >> my work ended up in rack gear, but it doesn't matter. if the low end of >> a guitar is a waveform with 12.1 ms period, there just ain't enough >> information to ever detect a note that low without waiting for at least a >> period to see some periodicity and estimate the period length and from that >> the fundamental frequency and pitch. >> >> i think 13 ms is as quick as the AXON guitar synth could acquire the >> pitch of a plucked E6 string. i am not saying i ever did it that fast. >> >> i *usually* like to compare one entire period to another entire period, >> so that would be 2 periods of delay (24 ms). but you can compare a segment >> of waveform that is less than a period to another segment of the same >> length and offset by one period. so that makes it longer than 12.1 ms and >> shorter than 24.2 ms before you get the minimum amount of information to >> discern pitch. i have had decent results at 15 ms and better results at 17 >> ms delay. >> >> what the Whammy has for delay, i dunno. >> >> -- >> >> r b-j [email protected] >> >> "Imagination is more important than knowledge." >> >> > >> > On Fri, Feb 5, 2021 at 1:04 PM robert bristow-johnson < >> [email protected]> wrote: >> > > > On 02/05/2021 4:00 AM Andy Farnell <[email protected]> >> wrote: >> > > > >> > > > >> > > > Another way of thinking is that maybe your DAC jitter and/or >> > > > channel skew is acceptable from a practical engineering POV >> > > > if you trade off a little time against spatial accuracy and >> > > > use statistical methods to correct. I'm guessing for a VR/music >> > > > application that something in the order of 20ms is your >> > > > critical time frame. You can do a lot of remedial calculation >> > > > in that time. >> > > >> > > For inter-channel skew, 20 milliseconds sounds horrible. You can't >> have that. >> > > >> > > But for the delay of some live algorithms, 20 ms **might** be >> acceptable. I've done a few real-time pitch shifters, and if you put in a >> decent pitch detector so that the splices in the pitch shifter are good and >> smooth splices, that pitch detector will have a 13 or 15 ms delay. Add that >> to the mean delay of the splicing procedure and you might have 20 ms of >> delay. So the guitarist using the pitch shifter might have a delay that is >> like stepping 5 or 6 meters away from their guitar amp. >> > > >> > > But delay is accumulative. We have what we call a "delay budget" for >> live audio processing that we have to worry about. And even moreso for >> monitor feedback. This is why a lotta bands that tour insist on a, or bring >> their own, analog mixing board for the monitors. It's also why yo might see >> an earplug in an artist's ear. >> > > >> > > cell phones, the delay is absolutely horrible. some engineers should >> be lined up against the wall and shot. >> > > >> > > -- >> > > >> > > r b-j [email protected] >> > > >> > > "Imagination is more important than knowledge." >> > > >> >
