I've always wanted to mess with pitch to midi but never got around to it. I have a Roland GR-50 guitar synth that uses a GK2 hexiphonic pickup and 13 pin connector. The GKx hexiphonic pickups aren't too expensive... I've considered picking up a broken GR-50 or similar for parts...
Just a thought!
On the pickup subject... I would point out that Chip's 'Fun with Goertzel' looks like it runs at a 1 khz sample rate, and that appears to be randomly chosen. I have done no math on this, and it is still half-baked.. but the sample rate can probably run MUCH faster... The strings are metal, why not try to do a pickup that can be etched on a PC board? Maybe even end up with something superior... You would only need 1/2 of the demo per string.
Hehehe, this idea made me giggle like a little girl. The sampling rate can certainly run much faster. Hopefully the signal to noise ratio will be good enough. There's a lot that can be done with signal processing. I don't know how linear the physics of a capacitive pickup would be, or if it matters. But a superior pickup that can be easily installed on any stringed instrument? Hehehe.
I've always wanted to mess with pitch to midi but never got around to it. I have a Roland GR-50 guitar synth that uses a GK2 hexiphonic pickup and 13 pin connector. The GKx hexiphonic pickups aren't too expensive... I've considered picking up a broken GR-50 or similar for parts...
Just a thought!
Ah, of course, why didn't I think of the GK2! It appears that there is now the GK3.
The fretboard could also be a controller for a backing-track MP3 player. Nominate one or two random notes that would be silent, to perform a control function. Right now we have to tap-dance on start/stop switches.
On the pickup subject... I would point out that Chip's 'Fun with Goertzel' looks like it runs at a 1 khz sample rate, and that appears to be randomly chosen. I have done no math on this, and it is still half-baked.. but the sample rate can probably run MUCH faster... The strings are metal, why not try to do a pickup that can be etched on a PC board? Maybe even end up with something superior... You would only need 1/2 of the demo per string.
Hehehe, this idea made me giggle like a little girl. The sampling rate can certainly run much faster. Hopefully the signal to noise ratio will be good enough. There's a lot that can be done with signal processing. I don't know how linear the physics of a capacitive pickup would be, or if it matters. But a superior pickup that can be easily installed on any stringed instrument? Hehehe.
Welcome to the forum!
Thanks for the welcome. Everyone!
Hmmm, I just can't put this half-baked idea down. I'm thinking fairly wide electrodes between the strings, alternating 0 and 90 degrees carrier phase, A longish aspect ratio for good coupling. Then the 'Pickup' electrodes offset along the length, so they can also be wide.
This should be tolerant to noise because we are only interested in the resultant angle.. Or.. Now I wonder what effects may be possible also looking at the magnitude, maybe mixing. It would all be there, and take few or no additional components..
On the pickup subject... I would point out that Chip's 'Fun with Goertzel' looks like it runs at a 1 khz sample rate, and that appears to be randomly chosen. I have done no math on this, and it is still half-baked.. but the sample rate can probably run MUCH faster... The strings are metal, why not try to do a pickup that can be etched on a PC board? Maybe even end up with something superior... You would only need 1/2 of the demo per string.
Hehehe, this idea made me giggle like a little girl. The sampling rate can certainly run much faster. Hopefully the signal to noise ratio will be good enough. There's a lot that can be done with signal processing. I don't know how linear the physics of a capacitive pickup would be, or if it matters. But a superior pickup that can be easily installed on any stringed instrument? Hehehe.
Welcome to the forum!
Thanks for the welcome. Everyone!
Hmmm, I just can't put this half-baked idea down. I'm thinking fairly wide electrodes between the strings, alternating 0 and 90 degrees carrier phase, A longish aspect ratio for good coupling. Then the 'Pickup' electrodes offset along the length, so they can also be wide.
This should be tolerant to noise because we are only interested in the resultant angle.. Or.. Now I wonder what effects may be possible also looking at the magnitude, maybe mixing. It would all be there, and take few or no additional components..
Welome!
Sounds like we have an educated person involved with this endeavor.
On the pickup subject... I would point out that Chip's 'Fun with Goertzel' looks like it runs at a 1 khz sample rate, and that appears to be randomly chosen. I have done no math on this, and it is still half-baked.. but the sample rate can probably run MUCH faster... The strings are metal, why not try to do a pickup that can be etched on a PC board? Maybe even end up with something superior... You would only need 1/2 of the demo per string.
Hehehe, this idea made me giggle like a little girl. The sampling rate can certainly run much faster. Hopefully the signal to noise ratio will be good enough. There's a lot that can be done with signal processing. I don't know how linear the physics of a capacitive pickup would be, or if it matters. But a superior pickup that can be easily installed on any stringed instrument? Hehehe.
Welcome to the forum!
Thanks for the welcome. Everyone!
Hmmm, I just can't put this half-baked idea down. I'm thinking fairly wide electrodes between the strings, alternating 0 and 90 degrees carrier phase, A longish aspect ratio for good coupling. Then the 'Pickup' electrodes offset along the length, so they can also be wide.
This should be tolerant to noise because we are only interested in the resultant angle.. Or.. Now I wonder what effects may be possible also looking at the magnitude, maybe mixing. It would all be there, and take few or no additional components..
Welome!
Sounds like we have an educated person involved with this endeavor.
Which is what I find amazing about this community. 👍
There's always a tension between lowering the barrier to entry - bringing in more people - and keeping the education level and quality of contribution high.
There's always a tension between lowering the barrier to entry - bringing in more people - and keeping the education level and quality of contribution high.
I barely made it out of high school with a 2.34 GPA. At the behest of my parents, I went to college and got a 0.95 GPA and was put on academic probation. So, I fell out the bottom and just focused on Parallax stuff. At one point, when we had about a dozen employees, my girlfriend (at the time) encouraged me to go back to school so I could get a REAL job. It just recently occurred to me, "Wait a minute... Those twelve people all thought THEY had real jobs."
My experience of the Arduino crowd is that there's a prevalence of 'I don't know much about electronics, but now I don't need to', whereas the Parallax crowd seems to be better educated, even if that is through self-education.
My experience of the Arduino crowd is that there's a prevalence of 'I don't know much about electronics, but now I don't need to', whereas the Parallax crowd seems to be better educated, even if that is through self-education.
Right on! We are living in a time when almost anything can be learned for free. Everybody can know as much as they want to know. Some people don't want to know much.
I never did college either, and I am also learning new things all the time. I've been teaching myself things since I was a child, and I'll continue until I can't (as in I'm dead).
Anyone can learn anything these days, it's fantastic.
Right on! We are living in a time when almost anything can be learned for free. Everybody can know as much as they want to know. Some people don't want to know much.
Too true. I am mostly self-educated. It is probably true that I had a fool for a teacher!
When I finally jumped through the hoops and got my degree, there was this fella in the classes who always had his earbuds in. He took the courses over and over until he 'passed'. In lab, he couldn't light an LED, yet eventually graduated with the same degree I earned...
My experience of the Arduino crowd is that there's a prevalence of 'I don't know much about electronics, but now I don't need to', whereas the Parallax crowd seems to be better educated, even if that is through self-education.
Right on! We are living in a time when almost anything can be learned for free. Everybody can know as much as they want to know. Some people don't want to know much.
I find for many people it's not a lack of desire to know a lot, but a case of not wanting to put in the time and effort to acquire the knowledge. For those who want to learn and are willing to put in the time and effort we are living in a golden age where we can learn every day.
I have a theory that GPA somewhat correlates with intelligence (obviously), but that there is an actual inversion that occurs at a certain point. These people with 3.95 GPA's etc, there is an actual mental illness at work when you consider what it takes to achieve it.
So Yamaha ended up buying Line 6 and stuck the Variax electronics in one of their own guitars but no further development. Desktop setup software is out of date and they never got around to implementing auto-tune.
I just came across a NIB Peavey AT-200 couldn't believe my luck. Now I have four AT-200s, a Variax JTV-59 and a Variax 700.
The market needs a P2-based system that can be adapted to any guitar. @pik33 ????
Those robots aren't gonna be needed...you need a new project
@Mickster said:
So Yamaha ended up buying Line 6 and stuck the Variax electronics in one of their own guitars but no further development. Desktop setup software is out of date and they never got around to implementing auto-tune.
I just came across a NIB Peavey AT-200 couldn't believe my luck. Now I have four AT-200s, a Variax JTV-59 and a Variax 700.
The market needs a P2-based system that can be adapted to any guitar. @pik33 ????
Those robots aren't gonna be needed...you need a new project
Craig
I have no guitar and I don't know how to play one. Keyboards only. No hardware to test.
I think a guitar effect is possible. We have ADCs. They are somewhat limited, but I think they are good enough. We have DACs that are good. So, of course, an audio effect processor is possible.
Well, as I have electric Guitars and do tinker with Guitar effects and P2 also.
The most difficult part in my experience is the extremely wide dynamic range of the Guitar pickup signal. You want to cover a clean signal. On the other hand to get the standing tone of a lead sound, there is lots of gain necessary. This gain also is active for any kind of noise. Grounding, Radio, Mobile Phones, Power Supply, Digital Spi I2C,..... Everything is creeping into the signal.
The builtin Adc is certainly not good enough for this application.
The device which is discussed here would need the full dynamic range at its output too. So P2's builtin DAC would not be good enough for this.
About state of the art: I have just bought a Ampero Stomp 2. They claim 24bit Adc with 127dB Signal to noise. Thats 21 valid bits. I cannot verify this number, but it is really good.
In my own experiments the best way was to use an analog compressor and emphasis before the Adc. And a noise gate which can be done with the processor.
So P2's builtin DAC would not be good enough for this.
When a P2 is powered by a clean power supply and is mounted on a proper board (=Eval), the SNR is over 100 dB. While it accepts 16-bit input only, a noise shaper can add several bits (=24 bit input and conversion is possible) as the sample rate is way over 1 MHz at 16 bit resolution.
Edge+breakout however is much more noisy.
So P2's builtin DAC would not be good enough for this.
When a P2 is powered by a clean power supply and is mounted on a proper board (=Eval), the SNR is over 100 dB. While it accepts 16-bit input only, a noise shaper can add several bits (=24 bit input and conversion is possible) as the sample rate is way over 1 MHz at 16 bit resolution.
Edge+breakout however is much more noisy.
Well, I don't know. How did you measure these 100dB?
I think, SNR will be more easy, if you have some physical distance with shielding and separate power supplies between P2 and ADC/DAC.
My homebrew setups are very far away from 100dB. The Kiss board is not optimised for this sort of application.
I have been wondering, if I should experiment with differential inputs/outputs to eliminate some noise.
I dunno. There are six transducers, one for each string. I get the impression that all the DSP gets is the frequency of the string oscillation and then adds or subtracts to that frequency, depending on what is required. The tone shaping happens down the line. Or am I oversimplifying? On the AT-200, each string is sampled at 8KHz.
Feel free to shoot me down on this but the so-called "tone" contributed-to by the furniture is a total myth.
In synth terms, I see the strings as Manually Controlled Oscillators
Well, I don't know. How did you measure these 100dB?
I "measured" it using a 3-bit "sine wave" and my ears. I connected an Eval to a 2x200W amplifier, set the amplifier to the maximum volume and generated this 3-bit wave using 16-bit noise dac and PWM dac modes. The goal was to compare them.
What I hear, was: sound - clear with PWM mode, not so clear with noise DAC (strange distortions added too in this mode) and a noise floor.
While ear is not a good measurement tool, the SNR in this case was at least 30..40 dB. This means the overall SNR for a 16-bit signal is much over 100 dB. I was impressed a lot.
This was however a short program working in one cog that generated this wave and nothing else.
While I tried to hear a DAC noise floor using an Edge with a working audio player, the result was not so good. I can hear a lot of "digital noise" generated by the rest of the connected hardware. I have to repeat the experiment with an Eval AND a big working program that uses a lot of cogs as I don''t know where is this noise generated. Maybe in a P2 itself, maybe it si a board problem (capacity, etc). However, even a Raspberry Pi with a DAC hat (Hifiberry) does the same: you can hear the digital noise at a similar level.
To make an audio related hardware, the PCB and power suply for I/O has to be carefully designed and I think there should be no USB and video connected to the audio processing P2. This kind of stuff needs to be placed on another controller placed on another PCB.
The communication protocol between "digital" and "analog" board should be designed in a way that prevents the accoustic band signals to be present there.
Comments
Just a thought!
Welcome to the forum!
Ah, of course, why didn't I think of the GK2! It appears that there is now the GK3.
Bingo!
Auto-tune
Alternate tuning
Emulation of other guitars including acoustics, 12-string, Sitar, Banjo, Dobro.
Pitch-to-MIDI
Open source
This would be a huge crowd-funding success.
Thanks for the welcome. Everyone!
Hmmm, I just can't put this half-baked idea down. I'm thinking fairly wide electrodes between the strings, alternating 0 and 90 degrees carrier phase, A longish aspect ratio for good coupling. Then the 'Pickup' electrodes offset along the length, so they can also be wide.
This should be tolerant to noise because we are only interested in the resultant angle.. Or.. Now I wonder what effects may be possible also looking at the magnitude, maybe mixing. It would all be there, and take few or no additional components..
Welome!
Sounds like we have an educated person involved with this endeavor.
Which is what I find amazing about this community. 👍
I barely made it out of high school with a 2.34 GPA. At the behest of my parents, I went to college and got a 0.95 GPA and was put on academic probation. So, I fell out the bottom and just focused on Parallax stuff. At one point, when we had about a dozen employees, my girlfriend (at the time) encouraged me to go back to school so I could get a REAL job. It just recently occurred to me, "Wait a minute... Those twelve people all thought THEY had real jobs."
My experience of the Arduino crowd is that there's a prevalence of 'I don't know much about electronics, but now I don't need to', whereas the Parallax crowd seems to be better educated, even if that is through self-education.
Right on! We are living in a time when almost anything can be learned for free. Everybody can know as much as they want to know. Some people don't want to know much.
Anyone can learn anything these days, it's fantastic.
Too true. I am mostly self-educated. It is probably true that I had a fool for a teacher!
When I finally jumped through the hoops and got my degree, there was this fella in the classes who always had his earbuds in. He took the courses over and over until he 'passed'. In lab, he couldn't light an LED, yet eventually graduated with the same degree I earned...
Knowledge is always what you make of it!
I liked learning, but I didn't like homework.
+1, +1
I find for many people it's not a lack of desire to know a lot, but a case of not wanting to put in the time and effort to acquire the knowledge. For those who want to learn and are willing to put in the time and effort we are living in a golden age where we can learn every day.
That and chasing you-know-what....😁😁😁👍
https://www.roland.com/global/products/gk-3/
A few more ideas
ElektroCaster
An open, modular guitar-design with some nifty features.
https://hackaday.io/project/161675-elektrocaster
So Yamaha ended up buying Line 6 and stuck the Variax electronics in one of their own guitars but no further development. Desktop setup software is out of date and they never got around to implementing auto-tune.
I just came across a NIB Peavey AT-200
couldn't believe my luck. Now I have four AT-200s, a Variax JTV-59 and a Variax 700.
The market needs a P2-based system that can be adapted to any guitar. @pik33 ????
Those robots aren't gonna be needed...you need a new project
Craig
I have no guitar and I don't know how to play one. Keyboards only. No hardware to test.
I think a guitar effect is possible. We have ADCs. They are somewhat limited, but I think they are good enough. We have DACs that are good. So, of course, an audio effect processor is possible.
Well, as I have electric Guitars and do tinker with Guitar effects and P2 also.
The most difficult part in my experience is the extremely wide dynamic range of the Guitar pickup signal. You want to cover a clean signal. On the other hand to get the standing tone of a lead sound, there is lots of gain necessary. This gain also is active for any kind of noise. Grounding, Radio, Mobile Phones, Power Supply, Digital Spi I2C,..... Everything is creeping into the signal.
The builtin Adc is certainly not good enough for this application.
The device which is discussed here would need the full dynamic range at its output too. So P2's builtin DAC would not be good enough for this.
About state of the art: I have just bought a Ampero Stomp 2. They claim 24bit Adc with 127dB Signal to noise. Thats 21 valid bits. I cannot verify this number, but it is really good.
In my own experiments the best way was to use an analog compressor and emphasis before the Adc. And a noise gate which can be done with the processor.
When a P2 is powered by a clean power supply and is mounted on a proper board (=Eval), the SNR is over 100 dB. While it accepts 16-bit input only, a noise shaper can add several bits (=24 bit input and conversion is possible) as the sample rate is way over 1 MHz at 16 bit resolution.
Edge+breakout however is much more noisy.
Well, I don't know. How did you measure these 100dB?
I think, SNR will be more easy, if you have some physical distance with shielding and separate power supplies between P2 and ADC/DAC.
My homebrew setups are very far away from 100dB. The Kiss board is not optimised for this sort of application.
I have been wondering, if I should experiment with differential inputs/outputs to eliminate some noise.
@"Christof Eb."
I dunno. There are six transducers, one for each string. I get the impression that all the DSP gets is the frequency of the string oscillation and then adds or subtracts to that frequency, depending on what is required. The tone shaping happens down the line. Or am I oversimplifying? On the AT-200, each string is sampled at 8KHz.
Feel free to shoot me down on this but the so-called "tone" contributed-to by the furniture is a total myth.
In synth terms, I see the strings as Manually Controlled Oscillators
I "measured" it using a 3-bit "sine wave" and my ears. I connected an Eval to a 2x200W amplifier, set the amplifier to the maximum volume and generated this 3-bit wave using 16-bit noise dac and PWM dac modes. The goal was to compare them.
What I hear, was: sound - clear with PWM mode, not so clear with noise DAC (strange distortions added too in this mode) and a noise floor.
While ear is not a good measurement tool, the SNR in this case was at least 30..40 dB. This means the overall SNR for a 16-bit signal is much over 100 dB. I was impressed a lot.
This was however a short program working in one cog that generated this wave and nothing else.
While I tried to hear a DAC noise floor using an Edge with a working audio player, the result was not so good. I can hear a lot of "digital noise" generated by the rest of the connected hardware. I have to repeat the experiment with an Eval AND a big working program that uses a lot of cogs as I don''t know where is this noise generated. Maybe in a P2 itself, maybe it si a board problem (capacity, etc). However, even a Raspberry Pi with a DAC hat (Hifiberry) does the same: you can hear the digital noise at a similar level.
To make an audio related hardware, the PCB and power suply for I/O has to be carefully designed and I think there should be no USB and video connected to the audio processing P2. This kind of stuff needs to be placed on another controller placed on another PCB.
The communication protocol between "digital" and "analog" board should be designed in a way that prevents the accoustic band signals to be present there.