Jump to content
Why become a member? ×

Recommended Posts

On 09/01/2018 at 00:21, ped said:

Yes but you said there is latency with a GK + Vbass, which there isn’t :)

But  there is : you maybe don't hear it anymore or are so accustomed to anticipate it that you don't notice it anymore... :drinks:

Link to comment
Share on other sites

8 minutes ago, Hellzero said:

But  there is : you maybe don't hear it anymore or are so accustomed to anticipate it that you don't notice it anymore... :drinks:

There’s none. Only as much as using a normal pickup in a normal bass. Ask Roland if you must!

Link to comment
Share on other sites

  • 3 weeks later...
  • 5 months later...
  • 3 years later...
On 08/01/2018 at 23:21, ped said:

Yes but you said there is latency with a GK + Vbass, which there isn’t :)


That’s because the Vbass isn’t doing pitch-to-MIDI conversion on the signal, but instead processing it heavily to do what it does.
 

If you want MIDI output, on the other hand, pitch detection must insert some latency, as already mentioned. If you wait for one full cycle of a low E, which is around 42 Hz, that will take 1/42 seconds, plus the time to calculate pitch, then transmit MIDI, then produce a new note. You will definitely hear that much latency. 
 

So Roland has their way around the problem with the Vbass: don’t rely on pitch detection and MIDI. The Fretsense folks have another solution, which also doesn’t require pitch detection. 

Link to comment
Share on other sites

34 minutes ago, bnt said:


That’s because the Vbass isn’t doing pitch-to-MIDI conversion on the signal, but instead processing it heavily to do what it does.
 

If you want MIDI output, on the other hand, pitch detection must insert some latency, as already mentioned. If you wait for one full cycle of a low E, which is around 42 Hz, that will take 1/42 seconds, plus the time to calculate pitch, then transmit MIDI, then produce a new note. You will definitely hear that much latency. 
 

So Roland has their way around the problem with the Vbass: don’t rely on pitch detection and MIDI. The Fretsense folks have another solution, which also doesn’t require pitch detection. 

Yep I believe we said that earlier

Link to comment
Share on other sites

13 hours ago, bnt said:

If you want MIDI output, on the other hand, pitch detection must insert some latency, as already mentioned. If you wait for one full cycle of a low E, which is around 42 Hz, that will take 1/42 seconds, plus the time to calculate pitch, then transmit MIDI, then produce a new note. You will definitely hear that much latency. 

 

It's actually worse than that. The very best pitch detection systems require a minimum of one and a half waveform cycles to begin to identify the note. Most take much longer. 

 

IMO pitch detection is a dead-end for live performance (it might have a limited use for programming where the latency can be compensated for once the note data has been captured). Signal processing is the way forward because it allows all aspects of stringed instrument playing technique to be utilised.

Link to comment
Share on other sites

I nearly bought a Roland (G-33 or G-88?) bass last week, but I decided to sleep on it and then it was gone.

I've got a Bartolini multi-pole pickup and a Roland control panel to put in a Ric when I get around to doing the project.

9z4m1Nj.jpg

6A7U1FQ.jpg

It'll feed into a Roland GM-70 MIDI converter.

Of course, being a pitch to note converter, plus old tech, there will be some delay there.

Plus, I've no idea what I'll do with it when I have it working!

I expect I'll use my bass to trigger a Minitaur or Model D, but then I'll probably realise that I need a bass line too and my interest will wander elsewhere.

In the meanwhile, I've got a Source Audio C4, which triggers pretty well.

  • Like 1
Link to comment
Share on other sites

On 03/05/2022 at 11:42, BigRedX said:

 

It's actually worse than that. The very best pitch detection systems require a minimum of one and a half waveform cycles to begin to identify the note. Most take much longer. 

 

IMO pitch detection is a dead-end for live performance (it might have a limited use for programming where the latency can be compensated for once the note data has been captured). Signal processing is the way forward because it allows all aspects of stringed instrument playing technique to be utilised.

 

Shouldn't it only require two zero crossings, which could be anything between half a cycle and very very nearly a full cycle?

Link to comment
Share on other sites

11 hours ago, tauzero said:

 

Shouldn't it only require two zero crossings, which could be anything between half a cycle and very very nearly a full cycle?

 

It's a lot more complicated then that because the waveform of a plucked string is unlikely to be symmetrical about the zero point. There are plenty of standard waveforms that display asymmetry - consider the pulse wave on a synth which is a square wave where the mid zero crossing is not equidistant between the two either side. For that you would definitely need 3 zero crossings to get a stable pitch and more if the pulse width changed over time. Real-life waveforms are even more complex than electronically generated ones, and that's before you take into account other variables like the fact that simply plucking the string will cause the first few cycles to be sharp.

 

Of course you can get around some of this by using pitch quantisation, but then you also lose a lot of the expressiveness that comes from a playing a guitar or bass.

Link to comment
Share on other sites

49 minutes ago, BigRedX said:

 

It's a lot more complicated then that because the waveform of a plucked string is unlikely to be symmetrical about the zero point. There are plenty of standard waveforms that display asymmetry - consider the pulse wave on a synth which is a square wave where the mid zero crossing is not equidistant between the two either side. For that you would definitely need 3 zero crossings to get a stable pitch and more if the pulse width changed over time. Real-life waveforms are even more complex than electronically generated ones, and that's before you take into account other variables like the fact that simply plucking the string will cause the first few cycles to be sharp.

 

Of course you can get around some of this by using pitch quantisation, but then you also lose a lot of the expressiveness that comes from a playing a guitar or bass.

Well; the concept of zero crossings is a bit simplistic anyway; you really need something like a spectrogram & FFT to analyse the frequencies in an audio signal (https://en.wikipedia.org/wiki/Spectrogram).

 

321px-Spectrogram_of_violin.png

Link to comment
Share on other sites

2 hours ago, prowla said:

Well; the concept of zero crossings is a bit simplistic anyway; you really need something like a spectrogram & FFT to analyse the frequencies in an audio signal (https://en.wikipedia.org/wiki/Spectrogram).

 

321px-Spectrogram_of_violin.png

 

But for pitch to CV or MIDI you don't want to extract all the harmonic frequencies in the signal. You need to identify the fundamental as quickly as possible and in this case zero crossings are actually useful.

Edited by BigRedX
Link to comment
Share on other sites

There is a kind of circuit called a Phase Locked Loop (PLL) that locks on to an input signal and produces a new square wave signal, at the same frequency and in phase. It also outputs a voltage proportional to the frequency that other circuits can use. I played with the 4046 chip version years ago, not for pitch detection but more in search of nasty sounds & effects e.g. using dividers on the voltage to get subharmonics. It was finicky, the circuit needs tuning, and it was slow to lock on to a bass signal, so I don’t think e.g. Roland used a general purpose PLL like that. I might get in to it again, but I see someone else has done what I was looking for and more, in pedal form: 

 

 

Link to comment
Share on other sites

On 04/05/2022 at 07:16, prowla said:

I nearly bought a Roland (G-33 or G-88?) bass last week, but I decided to sleep on it and then it was gone.

I've got a Bartolini multi-pole pickup and a Roland control panel to put in a Ric when I get around to doing the project.

9z4m1Nj.jpg

 

 

It'll feed into a Roland GM-70 MIDI converter.

 

 

Never seen one of those, wasn't aware that one existed!

 

Link to comment
Share on other sites

6 minutes ago, bnt said:

There is a kind of circuit called a Phase Locked Loop (PLL) that locks on to an input signal and produces a new square wave signal, at the same frequency and in phase. It also outputs a voltage proportional to the frequency that other circuits can use. I played with the 4046 chip version years ago,

 

Bringing back my nightmares there - I used to spend a large part of my working day fixing a 4046 genlock circuit on an embedded computer. When it went wrong it was a nightmare to work with.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...