On Thursday, January 6, 2022 at 5:29:12 PM UTC, bitrex wrote:> On 1/6/22 11:50 AM, Don Y wrote: > > Advice on good ("respectable") candidates for the above? > > Not keen on a guitar-wannabe "instrument"... > > > > (Also, not keen on anything that will not be able to > > completely restore axe to original condition when no > > longer needed/wanted) > If you want to add a thing to add to a regular electric guitar to get > polyphonic MIDI output without modifying the guitar significantly I > think it limits the options to something like the hex pickups you can > non-destructively mount: > > <https://www.sweetwater.com/store/detail/GK3--roland-gk-3-divided-pickup> > > Then you'll need something to convert the hex signal to MIDI output and > there are solutions that come with sounds and some without e.g. this > package: > > <https://www.sweetwater.com/store/detail/GR55BK--roland-gr-55-guitar-synth-with-gk-3-pickup> > > Different companies have made a number of boxes like that over the years > in pedal and rackmount form. I have a Yamaha G50 rack unit that you can > plug the Roland hex pickup into it's just a MIDI converter, no onboard > sounds. It's about 25 years old but still works very well with pretty > low latency when you get it adjusted, IIRC the selling point at the time > is that it uses a neural network to detect notes before a full cycle has > elapsed; even on the low E of a guitar (much less a bass) if you wait a > whole cycle it introduces objectionable latency. I expect nowadays the > good systems all have their own proprietary algorithms for that. > > There are probably other systems that's just the setup I'm most familiar > with. > > GraphTech makes an all-in-one where the hex pickup is integrated into a > hard tail or Floyd Rose bridge and you can add a piezo amplifier for > pseudo-acoustic sounds and/or a hex driver for MIDI conversion > internally to the guitar. that can be an expensive way to go, the > bespoke Floyd Rose is almost $300 and the electronics is another $300, > kinda absurd for what you get which is a bare PCB with pin-headers and > about $5 worth of components: > > <https://graphtech.com/products/ghost-hexpander-preamp-kit>Yeah, Pat Metheny is one well known user of the Roland guitar synth setup... https://equipboard.com/pros/pat-metheny?gear=effects-pedals = RS
Guitar MIDI controller/converter - bitrex?
Started by ●January 6, 2022
Reply by ●January 15, 20222022-01-15
Reply by ●January 15, 20222022-01-15
On Saturday, January 8, 2022 at 10:13:46 PM UTC, Don Y wrote:> On 1/8/2022 12:57 PM, John Walliker wrote: > > On Friday, 7 January 2022 at 00:15:40 UTC, Don Y wrote: > >> On 1/6/2022 12:50 PM, bitrex wrote: > >>>> I've also seen a product or two that look at the *audio* and try to reconstruct > >>>> the fingering from that. Theoretically possible but a fair bit of work to > >>>> accomplish in real-time. Not sure how that latency stacks up with the > >>>> rest of the processing chain... > >>> > >>> I haven't tried one of those in some time but the last I did I wasn't > >>> impressed. The G50 is reported to have an average latency of about 10 ms on the > >>> low strings and feels pretty acceptable to me (haven't measured it myself) for > >>> guitar playing that's not speed metal riffs at least. > >>> > >>> For playing USB keyboards though, like with virtual instruments, it has to be > >>> lower than that, I find around 2ms is about what I can tolerate, I can tell the > >>> difference between a 2ms response and a 5 ms, and 10 ms starts to feel like an > >>> eternity. > >> Well, at least there is an explicit (industry-wide) acknowledgement that there > >> is a significant issue, there. And, attempts to put a number on it. Too many > >> other interfaces just leave that unspoken -- /caveat emptor/. > >> > >> I question whether the physical-virtual instruments ("funky UI's that try to > >> LOOSELY resemble a real instrument") end up altering *how* you play. The > >> "physical presence" of an instrument seems to be a part of the performance > >> experience; would a "sousaphone emulator" (that you can hold in one hand) > >> *play* like a real sousaphone? Even if the emulator was a 100% faithful > >> reproduction... > > > > How about a pneumatically controlled pipe organ in a large church where > > the latency could be around 200ms? > Most wind instruments have high latency -- esp if excited by > human breath. > > My above comment was wrt the *feel* of the instrument-wannabe vs. the > real instrument. What's it like to play a "little box with a mouthpiece" > that emulates a tuba vs. having to carry the tuba on your shoulders > while trying to summon up enough air to excite it? Does the little > box represent the same "impedance" to your windpipe as the genuine > article? Do you have to move the same amount of air to produce a > particular sound? > > [I.e., playing trills on a trumpet is a piece of cake compared to > a trombone, baritone horn, etc. Does a trombone-emulator present > the same PHYSICAL playing challenges as a real trombone?]Latency is hard to avoid unless youre putting a pressure sensor right at the stimulus point. For a guitar, that would be under each fret position, for each string. i.e. ~15 x 6 = 90 sensors. Not use how to do *that*, reversably & non destructively. and while not interefering with normal acoustics, ergonomics, etc. = RS
Reply by ●January 15, 20222022-01-15
On 1/15/2022 6:01 PM, Rich S wrote:> On Saturday, January 8, 2022 at 10:13:46 PM UTC, Don Y wrote: >> On 1/8/2022 12:57 PM, John Walliker wrote: >>> On Friday, 7 January 2022 at 00:15:40 UTC, Don Y wrote: >>>> On 1/6/2022 12:50 PM, bitrex wrote: >>>>>> I've also seen a product or two that look at the *audio* and try to reconstruct >>>>>> the fingering from that. Theoretically possible but a fair bit of work to >>>>>> accomplish in real-time. Not sure how that latency stacks up with the >>>>>> rest of the processing chain... >>>>> >>>>> I haven't tried one of those in some time but the last I did I wasn't >>>>> impressed. The G50 is reported to have an average latency of about 10 ms on the >>>>> low strings and feels pretty acceptable to me (haven't measured it myself) for >>>>> guitar playing that's not speed metal riffs at least. >>>>> >>>>> For playing USB keyboards though, like with virtual instruments, it has to be >>>>> lower than that, I find around 2ms is about what I can tolerate, I can tell the >>>>> difference between a 2ms response and a 5 ms, and 10 ms starts to feel like an >>>>> eternity. >>>> Well, at least there is an explicit (industry-wide) acknowledgement that there >>>> is a significant issue, there. And, attempts to put a number on it. Too many >>>> other interfaces just leave that unspoken -- /caveat emptor/. >>>> >>>> I question whether the physical-virtual instruments ("funky UI's that try to >>>> LOOSELY resemble a real instrument") end up altering *how* you play. The >>>> "physical presence" of an instrument seems to be a part of the performance >>>> experience; would a "sousaphone emulator" (that you can hold in one hand) >>>> *play* like a real sousaphone? Even if the emulator was a 100% faithful >>>> reproduction... >>> >>> How about a pneumatically controlled pipe organ in a large church where >>> the latency could be around 200ms? >> Most wind instruments have high latency -- esp if excited by >> human breath. >> >> My above comment was wrt the *feel* of the instrument-wannabe vs. the >> real instrument. What's it like to play a "little box with a mouthpiece" >> that emulates a tuba vs. having to carry the tuba on your shoulders >> while trying to summon up enough air to excite it? Does the little >> box represent the same "impedance" to your windpipe as the genuine >> article? Do you have to move the same amount of air to produce a >> particular sound? >> >> [I.e., playing trills on a trumpet is a piece of cake compared to >> a trombone, baritone horn, etc. Does a trombone-emulator present >> the same PHYSICAL playing challenges as a real trombone?] > > Latency is hard to avoid unless youre putting a pressure sensor right > at the stimulus point. For a guitar, that would be under each fret > position, for each string. i.e. ~15 x 6 = 90 sensors. > Not use how to do *that*, reversably & non destructively. > and while not interefering with normal acoustics, ergonomics, > etc. = RSBut *players* of particular instruments already have a "feel" for the inherent characteristics of an instrument and adjust their play, accordingly. Change that "feel" (because you're using an approximation of the instrument) and your "play" changes, as well. E.g., The basics of playing any horn are similar. Yet, the actual results vary considerably as the physics of the instrument vary (e.g., with size, required air volume, etc). I'd imagine playing a ukulele to be significantly different than a double bass.
Reply by ●January 15, 20222022-01-15
On Sunday, January 16, 2022 at 1:09:31 AM UTC, Don Y wrote:> On 1/15/2022 6:01 PM, Rich S wrote: > > On Saturday, January 8, 2022 at 10:13:46 PM UTC, Don Y wrote: > >> On 1/8/2022 12:57 PM, John Walliker wrote: > >>> On Friday, 7 January 2022 at 00:15:40 UTC, Don Y wrote: > >>>> On 1/6/2022 12:50 PM, bitrex wrote: > >>>>>> I've also seen a product or two that look at the *audio* and try to reconstruct > >>>>>> the fingering from that. Theoretically possible but a fair bit of work to > >>>>>> accomplish in real-time. Not sure how that latency stacks up with the > >>>>>> rest of the processing chain... > >>>>> > >>>>> I haven't tried one of those in some time but the last I did I wasn't > >>>>> impressed. The G50 is reported to have an average latency of about 10 ms on the > >>>>> low strings and feels pretty acceptable to me (haven't measured it myself) for > >>>>> guitar playing that's not speed metal riffs at least. > >>>>> > >>>>> For playing USB keyboards though, like with virtual instruments, it has to be > >>>>> lower than that, I find around 2ms is about what I can tolerate, I can tell the > >>>>> difference between a 2ms response and a 5 ms, and 10 ms starts to feel like an > >>>>> eternity. > >>>> Well, at least there is an explicit (industry-wide) acknowledgement that there > >>>> is a significant issue, there. And, attempts to put a number on it. Too many > >>>> other interfaces just leave that unspoken -- /caveat emptor/. > >>>> > >>>> I question whether the physical-virtual instruments ("funky UI's that try to > >>>> LOOSELY resemble a real instrument") end up altering *how* you play. The > >>>> "physical presence" of an instrument seems to be a part of the performance > >>>> experience; would a "sousaphone emulator" (that you can hold in one hand) > >>>> *play* like a real sousaphone? Even if the emulator was a 100% faithful > >>>> reproduction... > >>> > >>> How about a pneumatically controlled pipe organ in a large church where > >>> the latency could be around 200ms? > >> Most wind instruments have high latency -- esp if excited by > >> human breath. > >> > >> My above comment was wrt the *feel* of the instrument-wannabe vs. the > >> real instrument. What's it like to play a "little box with a mouthpiece" > >> that emulates a tuba vs. having to carry the tuba on your shoulders > >> while trying to summon up enough air to excite it? Does the little > >> box represent the same "impedance" to your windpipe as the genuine > >> article? Do you have to move the same amount of air to produce a > >> particular sound? > >> > >> [I.e., playing trills on a trumpet is a piece of cake compared to > >> a trombone, baritone horn, etc. Does a trombone-emulator present > >> the same PHYSICAL playing challenges as a real trombone?] > > > > Latency is hard to avoid unless youre putting a pressure sensor right > > at the stimulus point. For a guitar, that would be under each fret > > position, for each string. i.e. ~15 x 6 = 90 sensors. > > Not use how to do *that*, reversably & non destructively. > > and while not interefering with normal acoustics, ergonomics, > > etc. = RS > But *players* of particular instruments already have a "feel" for > the inherent characteristics of an instrument and adjust their > play, accordingly. Change that "feel" (because you're using an > approximation of the instrument) and your "play" changes, as well. > > E.g., The basics of playing any horn are similar. Yet, the actual > results vary considerably as the physics of the instrument vary (e.g., > with size, required air volume, etc). > > I'd imagine playing a ukulele to be significantly different than > a double bass.Sure, most acoustical instruments have an inherent acousto- mechanical "latency" and ergonomic factors that set a upper- limit on fluent note change. You could retrofit one of these types with electronics that make it outperform the original instrument. :-) For a horn (lip-vibrated aero- phone), we could trigger on finger- or other actions, and not rely on the lip-buzz. cheers, RS
Reply by ●January 15, 20222022-01-15
On 1/15/2022 6:49 PM, Rich S wrote:> On Sunday, January 16, 2022 at 1:09:31 AM UTC, Don Y wrote: >> On 1/15/2022 6:01 PM, Rich S wrote: >>> On Saturday, January 8, 2022 at 10:13:46 PM UTC, Don Y wrote: >>>> On 1/8/2022 12:57 PM, John Walliker wrote: >>>>> On Friday, 7 January 2022 at 00:15:40 UTC, Don Y wrote: >>>>>> On 1/6/2022 12:50 PM, bitrex wrote: >>>>>>>> I've also seen a product or two that look at the *audio* and try to reconstruct >>>>>>>> the fingering from that. Theoretically possible but a fair bit of work to >>>>>>>> accomplish in real-time. Not sure how that latency stacks up with the >>>>>>>> rest of the processing chain... >>>>>>> >>>>>>> I haven't tried one of those in some time but the last I did I wasn't >>>>>>> impressed. The G50 is reported to have an average latency of about 10 ms on the >>>>>>> low strings and feels pretty acceptable to me (haven't measured it myself) for >>>>>>> guitar playing that's not speed metal riffs at least. >>>>>>> >>>>>>> For playing USB keyboards though, like with virtual instruments, it has to be >>>>>>> lower than that, I find around 2ms is about what I can tolerate, I can tell the >>>>>>> difference between a 2ms response and a 5 ms, and 10 ms starts to feel like an >>>>>>> eternity. >>>>>> Well, at least there is an explicit (industry-wide) acknowledgement that there >>>>>> is a significant issue, there. And, attempts to put a number on it. Too many >>>>>> other interfaces just leave that unspoken -- /caveat emptor/. >>>>>> >>>>>> I question whether the physical-virtual instruments ("funky UI's that try to >>>>>> LOOSELY resemble a real instrument") end up altering *how* you play. The >>>>>> "physical presence" of an instrument seems to be a part of the performance >>>>>> experience; would a "sousaphone emulator" (that you can hold in one hand) >>>>>> *play* like a real sousaphone? Even if the emulator was a 100% faithful >>>>>> reproduction... >>>>> >>>>> How about a pneumatically controlled pipe organ in a large church where >>>>> the latency could be around 200ms? >>>> Most wind instruments have high latency -- esp if excited by >>>> human breath. >>>> >>>> My above comment was wrt the *feel* of the instrument-wannabe vs. the >>>> real instrument. What's it like to play a "little box with a mouthpiece" >>>> that emulates a tuba vs. having to carry the tuba on your shoulders >>>> while trying to summon up enough air to excite it? Does the little >>>> box represent the same "impedance" to your windpipe as the genuine >>>> article? Do you have to move the same amount of air to produce a >>>> particular sound? >>>> >>>> [I.e., playing trills on a trumpet is a piece of cake compared to >>>> a trombone, baritone horn, etc. Does a trombone-emulator present >>>> the same PHYSICAL playing challenges as a real trombone?] >>> >>> Latency is hard to avoid unless youre putting a pressure sensor right >>> at the stimulus point. For a guitar, that would be under each fret >>> position, for each string. i.e. ~15 x 6 = 90 sensors. >>> Not use how to do *that*, reversably & non destructively. >>> and while not interefering with normal acoustics, ergonomics, >>> etc. = RS >> But *players* of particular instruments already have a "feel" for >> the inherent characteristics of an instrument and adjust their >> play, accordingly. Change that "feel" (because you're using an >> approximation of the instrument) and your "play" changes, as well. >> >> E.g., The basics of playing any horn are similar. Yet, the actual >> results vary considerably as the physics of the instrument vary (e.g., >> with size, required air volume, etc). >> >> I'd imagine playing a ukulele to be significantly different than >> a double bass. > > Sure, most acoustical instruments have an inherent acousto- > mechanical "latency" and ergonomic factors that set a upper- > limit on fluent note change. You could retrofit one > of these types with electronics that make it outperform > the original instrument. :-) For a horn (lip-vibrated aero- > phone), we could trigger on finger- or other actions, and > not rely on the lip-buzz. > cheers, RSYes, but the point of "instrumenting" an instrument is to record your *current* playing proficiency. Change the way the instrument plays/"feels" and do you end up turning a prodigy into a neophyte? The instrument is an integral part of the performance, alongside the performer.
Reply by ●January 16, 20222022-01-16
On 1/15/22 8:01 PM, Rich S wrote:> On Saturday, January 8, 2022 at 10:13:46 PM UTC, Don Y wrote: >> On 1/8/2022 12:57 PM, John Walliker wrote: >>> On Friday, 7 January 2022 at 00:15:40 UTC, Don Y wrote: >>>> On 1/6/2022 12:50 PM, bitrex wrote: >>>>>> I've also seen a product or two that look at the *audio* and try to reconstruct >>>>>> the fingering from that. Theoretically possible but a fair bit of work to >>>>>> accomplish in real-time. Not sure how that latency stacks up with the >>>>>> rest of the processing chain... >>>>> >>>>> I haven't tried one of those in some time but the last I did I wasn't >>>>> impressed. The G50 is reported to have an average latency of about 10 ms on the >>>>> low strings and feels pretty acceptable to me (haven't measured it myself) for >>>>> guitar playing that's not speed metal riffs at least. >>>>> >>>>> For playing USB keyboards though, like with virtual instruments, it has to be >>>>> lower than that, I find around 2ms is about what I can tolerate, I can tell the >>>>> difference between a 2ms response and a 5 ms, and 10 ms starts to feel like an >>>>> eternity. >>>> Well, at least there is an explicit (industry-wide) acknowledgement that there >>>> is a significant issue, there. And, attempts to put a number on it. Too many >>>> other interfaces just leave that unspoken -- /caveat emptor/. >>>> >>>> I question whether the physical-virtual instruments ("funky UI's that try to >>>> LOOSELY resemble a real instrument") end up altering *how* you play. The >>>> "physical presence" of an instrument seems to be a part of the performance >>>> experience; would a "sousaphone emulator" (that you can hold in one hand) >>>> *play* like a real sousaphone? Even if the emulator was a 100% faithful >>>> reproduction... >>> >>> How about a pneumatically controlled pipe organ in a large church where >>> the latency could be around 200ms? >> Most wind instruments have high latency -- esp if excited by >> human breath. >> >> My above comment was wrt the *feel* of the instrument-wannabe vs. the >> real instrument. What's it like to play a "little box with a mouthpiece" >> that emulates a tuba vs. having to carry the tuba on your shoulders >> while trying to summon up enough air to excite it? Does the little >> box represent the same "impedance" to your windpipe as the genuine >> article? Do you have to move the same amount of air to produce a >> particular sound? >> >> [I.e., playing trills on a trumpet is a piece of cake compared to >> a trombone, baritone horn, etc. Does a trombone-emulator present >> the same PHYSICAL playing challenges as a real trombone?] > > Latency is hard to avoid unless youre putting a pressure sensor right > at the stimulus point. For a guitar, that would be under each fret > position, for each string. i.e. ~15 x 6 It sensors. > Not use how to do *that*, reversably & non destructively. > and while not interefering with normal acoustics, ergonomics, > etc. = RS >It's been done before, you don't really need strings at all once you do that. But that ends up becoming a new type of instrument that you have to learn, more like a strange keyboard instrument or high-tek accordion than a guitar. The hex pickups + good MIDI conversion boxes/algorithms let you do the usual stuff you can do with a guitar with real strings like string bends, slides, hammer ons and pull-offs and often can interpret those string-specific techniques reasonably successfully, it still feels like you're playing a guitar for the most part
Reply by ●January 16, 20222022-01-16
On 1/15/22 9:11 PM, Don Y wrote:> On 1/15/2022 6:49 PM, Rich S wrote: >> On Sunday, January 16, 2022 at 1:09:31 AM UTC, Don Y wrote: >>> On 1/15/2022 6:01 PM, Rich S wrote: >>>> On Saturday, January 8, 2022 at 10:13:46 PM UTC, Don Y wrote: >>>>> On 1/8/2022 12:57 PM, John Walliker wrote: >>>>>> On Friday, 7 January 2022 at 00:15:40 UTC, Don Y wrote: >>>>>>> On 1/6/2022 12:50 PM, bitrex wrote: >>>>>>>>> I've also seen a product or two that look at the *audio* and >>>>>>>>> try to reconstruct >>>>>>>>> the fingering from that. Theoretically possible but a fair bit >>>>>>>>> of work to >>>>>>>>> accomplish in real-time. Not sure how that latency stacks up >>>>>>>>> with the >>>>>>>>> rest of the processing chain... >>>>>>>> >>>>>>>> I haven't tried one of those in some time but the last I did I >>>>>>>> wasn't >>>>>>>> impressed. The G50 is reported to have an average latency of >>>>>>>> about 10 ms on the >>>>>>>> low strings and feels pretty acceptable to me (haven't measured >>>>>>>> it myself) for >>>>>>>> guitar playing that's not speed metal riffs at least. >>>>>>>> >>>>>>>> For playing USB keyboards though, like with virtual instruments, >>>>>>>> it has to be >>>>>>>> lower than that, I find around 2ms is about what I can tolerate, >>>>>>>> I can tell the >>>>>>>> difference between a 2ms response and a 5 ms, and 10 ms starts >>>>>>>> to feel like an >>>>>>>> eternity. >>>>>>> Well, at least there is an explicit (industry-wide) >>>>>>> acknowledgement that there >>>>>>> is a significant issue, there. And, attempts to put a number on >>>>>>> it. Too many >>>>>>> other interfaces just leave that unspoken -- /caveat emptor/. >>>>>>> >>>>>>> I question whether the physical-virtual instruments ("funky UI's >>>>>>> that try to >>>>>>> LOOSELY resemble a real instrument") end up altering *how* you >>>>>>> play. The >>>>>>> "physical presence" of an instrument seems to be a part of the >>>>>>> performance >>>>>>> experience; would a "sousaphone emulator" (that you can hold in >>>>>>> one hand) >>>>>>> *play* like a real sousaphone? Even if the emulator was a 100% >>>>>>> faithful >>>>>>> reproduction... >>>>>> >>>>>> How about a pneumatically controlled pipe organ in a large church >>>>>> where >>>>>> the latency could be around 200ms? >>>>> Most wind instruments have high latency -- esp if excited by >>>>> human breath. >>>>> >>>>> My above comment was wrt the *feel* of the instrument-wannabe vs. the >>>>> real instrument. What's it like to play a "little box with a >>>>> mouthpiece" >>>>> that emulates a tuba vs. having to carry the tuba on your shoulders >>>>> while trying to summon up enough air to excite it? Does the little >>>>> box represent the same "impedance" to your windpipe as the genuine >>>>> article? Do you have to move the same amount of air to produce a >>>>> particular sound? >>>>> >>>>> [I.e., playing trills on a trumpet is a piece of cake compared to >>>>> a trombone, baritone horn, etc. Does a trombone-emulator present >>>>> the same PHYSICAL playing challenges as a real trombone?] >>>> >>>> Latency is hard to avoid unless youre putting a pressure sensor right >>>> at the stimulus point. For a guitar, that would be under each fret >>>> position, for each string. i.e. ~15 x 6 = 90 sensors. >>>> Not use how to do *that*, reversably & non destructively. >>>> and while not interefering with normal acoustics, ergonomics, >>>> etc. = RS >>> But *players* of particular instruments already have a "feel" for >>> the inherent characteristics of an instrument and adjust their >>> play, accordingly. Change that "feel" (because you're using an >>> approximation of the instrument) and your "play" changes, as well. >>> >>> E.g., The basics of playing any horn are similar. Yet, the actual >>> results vary considerably as the physics of the instrument vary (e.g., >>> with size, required air volume, etc). >>> >>> I'd imagine playing a ukulele to be significantly different than >>> a double bass. >> >> Sure, most acoustical instruments have an inherent acousto- >> mechanical "latency" and ergonomic factors that set a upper- >> limit on fluent note change. You could retrofit one >> of these types with electronics that make it outperform >> the original instrument. :-) For a horn (lip-vibrated aero- >> phone), we could trigger on finger- or other actions, and >> not rely on the lip-buzz. >> cheers, RS > > Yes, but the point of "instrumenting" an instrument is to > record your *current* playing proficiency. Change the way the > instrument plays/"feels" and do you end up turning a prodigy > into a neophyte? The instrument is an integral part of the > performance, alongside the performer.It's pretty remarkable what can be done with physical modeling synthesis of wind instruments and a breath controller nowadays: <https://youtu.be/YJr9rmrMVpU>
Reply by ●January 18, 20222022-01-18
"Rich S" wrote in message news:fa1c2bfe-a14c-419e-8b0e-d2a9e0d4c74an@googlegroups.com... On Saturday, January 8, 2022 at 10:13:46 PM UTC, Don Y wrote:> On 1/8/2022 12:57 PM, John Walliker wrote: > > On Friday, 7 January 2022 at 00:15:40 UTC, Don Y wrote: > >> On 1/6/2022 12:50 PM, bitrex wrote: > >>>> I've also seen a product or two that look at the *audio* and try to > >>>> reconstruct > >>>> the fingering from that. Theoretically possible but a fair bit of > >>>> work to > >>>> accomplish in real-time. Not sure how that latency stacks up with the > >>>> rest of the processing chain... > >>> > >>> I haven't tried one of those in some time but the last I did I wasn't > >>> impressed. The G50 is reported to have an average latency of about 10 > >>> ms on the > >>> low strings and feels pretty acceptable to me (haven't measured it > >>> myself) for > >>> guitar playing that's not speed metal riffs at least. > >>> > >>> For playing USB keyboards though, like with virtual instruments, it > >>> has to be > >>> lower than that, I find around 2ms is about what I can tolerate, I can > >>> tell the > >>> difference between a 2ms response and a 5 ms, and 10 ms starts to feel > >>> like an > >>> eternity. > >> Well, at least there is an explicit (industry-wide) acknowledgement > >> that there > >> is a significant issue, there. And, attempts to put a number on it. Too > >> many > >> other interfaces just leave that unspoken -- /caveat emptor/. > >> > >> I question whether the physical-virtual instruments ("funky UI's that > >> try to > >> LOOSELY resemble a real instrument") end up altering *how* you play. > >> The > >> "physical presence" of an instrument seems to be a part of the > >> performance > >> experience; would a "sousaphone emulator" (that you can hold in one > >> hand) > >> *play* like a real sousaphone? Even if the emulator was a 100% faithful > >> reproduction... > > > > How about a pneumatically controlled pipe organ in a large church where > > the latency could be around 200ms? > Most wind instruments have high latency -- esp if excited by > human breath. > > My above comment was wrt the *feel* of the instrument-wannabe vs. the > real instrument. What's it like to play a "little box with a mouthpiece" > that emulates a tuba vs. having to carry the tuba on your shoulders > while trying to summon up enough air to excite it? Does the little > box represent the same "impedance" to your windpipe as the genuine > article? Do you have to move the same amount of air to produce a > particular sound? > >> [I.e., playing trills on a trumpet is a piece of cake compared to >> a trombone, baritone horn, etc. Does a trombone-emulator present >> the same PHYSICAL playing challenges as a real trombone?]>Latency is hard to avoid unless youre putting a pressure sensor right >at the stimulus point. For a guitar, that would be under each fret >position, for each string. i.e. ~15 x 6 = 90 sensors.Oh dear....er.... that will be 24 x 6 mate..... :-) Kevin Aylward https://www.kevinaylward.co.uk/gr/index.html http://www.anasoft.co.uk/ SuperSpice http://www.kevinaylward.co.uk/ee/index.html
Reply by ●January 18, 20222022-01-18
On 1/18/2022 1:49 PM, Kevin Aylward wrote:> > > > > "Rich S" wrote in message > news:fa1c2bfe-a14c-419e-8b0e-d2a9e0d4c74an@googlegroups.com... > > On Saturday, January 8, 2022 at 10:13:46 PM UTC, Don Y wrote: >> On 1/8/2022 12:57 PM, John Walliker wrote: >> > On Friday, 7 January 2022 at 00:15:40 UTC, Don Y wrote: >> >> On 1/6/2022 12:50 PM, bitrex wrote: >> >>>> I've also seen a product or two that look at the *audio* and try to >>>> >> reconstruct >> >>>> the fingering from that. Theoretically possible but a fair bit of >>>> >> work to >> >>>> accomplish in real-time. Not sure how that latency stacks up with the >> >>>> rest of the processing chain... >> >>> >> >>> I haven't tried one of those in some time but the last I did I wasn't >> >>> impressed. The G50 is reported to have an average latency of about 10 >>> >> ms on the >> >>> low strings and feels pretty acceptable to me (haven't measured it >>> >> myself) for >> >>> guitar playing that's not speed metal riffs at least. >> >>> >> >>> For playing USB keyboards though, like with virtual instruments, it >>> >> has to be >> >>> lower than that, I find around 2ms is about what I can tolerate, I can >> >>> tell the >> >>> difference between a 2ms response and a 5 ms, and 10 ms starts to feel >> >>> like an >> >>> eternity. >> >> Well, at least there is an explicit (industry-wide) acknowledgement >> >> that there >> >> is a significant issue, there. And, attempts to put a number on it. Too >> >> many >> >> other interfaces just leave that unspoken -- /caveat emptor/. >> >> >> >> I question whether the physical-virtual instruments ("funky UI's that >> >> try to >> >> LOOSELY resemble a real instrument") end up altering *how* you play. >> The >> >> "physical presence" of an instrument seems to be a part of the >> performance >> >> experience; would a "sousaphone emulator" (that you can hold in one >> hand) >> >> *play* like a real sousaphone? Even if the emulator was a 100% faithful >> >> reproduction... >> > >> > How about a pneumatically controlled pipe organ in a large church where >> > the latency could be around 200ms? >> Most wind instruments have high latency -- esp if excited by >> human breath. >> >> My above comment was wrt the *feel* of the instrument-wannabe vs. the >> real instrument. What's it like to play a "little box with a mouthpiece" >> that emulates a tuba vs. having to carry the tuba on your shoulders >> while trying to summon up enough air to excite it? Does the little >> box represent the same "impedance" to your windpipe as the genuine >> article? Do you have to move the same amount of air to produce a >> particular sound? >> >>> [I.e., playing trills on a trumpet is a piece of cake compared to >>> a trombone, baritone horn, etc. Does a trombone-emulator present >>> the same PHYSICAL playing challenges as a real trombone?] > >> Latency is hard to avoid unless youre putting a pressure sensor right >> at the stimulus point. For a guitar, that would be under each fret >> position, for each string. i.e. ~15 x 6 = 90 sensors. > > Oh dear....er.... that will be 24 x 6 mate..... :-)Actually, "depends on the guitar"...
Reply by ●January 18, 20222022-01-18
Don Y <blockedofcourse@foo.invalid> wrote in news:sr76lj$6qo$1@dont- email.me:> Advice on good ("respectable") candidates for the above? > Not keen on a guitar-wannabe "instrument"... > > (Also, not keen on anything that will not be able to > completely restore axe to original condition when no > longer needed/wanted) >Do not know what you want but I have found the best tuner app there is. It has voice analysis too, and we're talking down to fine grain tonal resolution. It is called: TE Tuner It has a metronome, it has different instrument envelopes. I set it to sawtooth and get FOOL spectrum. ;-) Actually that widens it out past all the instruments it has in it. Very exacting, useful item. I also have the Moog apps. WOW. A full bore synth right in my hands! So many sounds in it that one recognizes from movies and music of the past. I also paid the $25 whatever for "Garage Band" which is also awesome. The Moog and the iPad app Garage Band have midi stuff in them. I am sure the Moog will let you feed in your stuff, modify it and feed it back out. Anyway. I just wanted to mention the tuner. That is an excellent application for apple hardware. I do not know if it is on other platforms.
