I noticed this problem when I first tried a Voyager at a music store about four years ago. I was really excited to use an actual analog synthesizer, and I couldn't wait to max the filter's resonance and do a full, slow turn of the knob. I was constantly frustrated by my Korg MX's digital stepping, and I was really excited to finally hear a smooth analog sweep, so you can imagine my shock and chagrin when I heard digital stepping on a Moog synthesizer!
I was very disappointed, and even a little disillusioned. I ended up getting a Macbeth M5N, and I fell in love with its pure analog glory. However, when the Voyager OS came out (I always refer to it as the Model E), I was very happy to become its owner. To be honest, and I know this is sacrilege in some circles, I traded my M5N for it (and the price difference), and the reason I did so is that I like the Model E's sound much better. It's warmer, smoother, and creamier (I know those terms are becoming a Moog cliche, but they're just so apt), and it complements my mostly acoustic, instrumental music much better.
To get back on topic, I was very dissatisfied with this shameful digital artifact. Doesn't the Voyager have a Real Panel Control setting, though? Does using it fix the problem? I think if I had the Voyager I would always use the Real Panel Control, and then I would save my patches digitally. That seems like a good way to use it. But I guess you'd have to turn off Real Panel Control when you recall your patches...
The Voyager's a fine instrument, but I'm glad I went with the Model E.
Noise artifacts when changing cutoff
Because the cutoff controller needs such a wide range, would it be possible to replace the rotary encoder with a normal pot, and then hang the rotary encoder off the back of the pot and have it actuated by the pot's shaft? Doing this would allow you to directly control the cutoff with a CV when twisting the knob during a performance, and yet have a midi cc message available for storage of the parameter when the voice is saved. Of course, if one uses the mod wheel to control the cutoff, then the cutoff would need to be able to be controlled with a midi cc message.RL wrote:Hi,psicolor wrote:nice!
Do you know, why controlling the cutoff with a controller does NOT produce artifact noise although the cutoff pot does?
Where's the difference?
If it was the scanning, you could turn scanning off afair.
the voltage of the external pot controls directly the filter without processor scanning.
Mittelfranken gruesst Unterfranken
That all being said, unless you are listening to cutoff changes under a small number of specific circumstances, you're not likely to hear any stepping.
There are no encoders used for the Moog Voyager !
These are normal pots, that are digitalized by measuring the given voltage drop they cause. And this is done about 500 times per second!
This is the way most hybrid synths are build: Prophet 5, PPG waves, Memory Moog, ...
These are normal pots, that are digitalized by measuring the given voltage drop they cause. And this is done about 500 times per second!
This is the way most hybrid synths are build: Prophet 5, PPG waves, Memory Moog, ...
keep on turning these Moog knobs
Prodigy * minimoog '79 * Voyager * MF102 * MF103 * MF104z * MP201 * Taurus 3 * Minitaur * Sub Phatty * MF105 * Minimoog 2017+ MUSE * One 16 (sold)
Prodigy * minimoog '79 * Voyager * MF102 * MF103 * MF104z * MP201 * Taurus 3 * Minitaur * Sub Phatty * MF105 * Minimoog 2017+ MUSE * One 16 (sold)
Thanks for schooling me on that, Till.till wrote:There are no encoders used for the Moog Voyager !
These are normal pots, that are digitalized by measuring the given voltage drop they cause. And this is done about 500 times per second!
This is the way most hybrid synths are build: Prophet 5, PPG waves, Memory Moog, ...
Would it be correct to say that the sampled voltages are used to create midi cc values? Would it also be correct to say that the midi cc values are then used to set the value of the cutoff in real time?
If the above is true, then could the PC board be altered to feed CVs from the pot to the traces that feed analog CVs directly to the cutoff freq without eliminating the scanning of the CV from the cutoff pot? Perhaps the software could then be updated to alter the existing subroutine for the cutoff knob so that it just stored the scanned cutoff cc value without actually making the change to the cutoff filter itself. This would prevent the software from changing the cutoff value when the knob is turned, allow the analog CV to change the cutoff value instead, and keep the cutoff cc value available for the mod wheel to use if it is routed to the cutoff, as well for saving with the preset.
Of course, this is all pure conjecture on my part, as I have absolutely no in-depth knowledge of the circuits or the software involved. However, perhaps this might be the germ of an idea that leads to a solution?
Humbly submitted.
Edited to clarify a couple of things...
Last edited by mayidunk on Wed Dec 09, 2009 7:18 pm, edited 2 times in total.
-
RL
- Posts: 770
- Joined: Wed May 19, 2004 4:17 am
- Location: http://www.moogmusic.de/
- Contact:
Hi Bob,mayidunk wrote:If the above is true, could the PC board be altered to also feed CVs from the pot to the traces that feed analog CVs directly to the cutoff freq? And perhaps the software could then be updated to alter the existing subroutine for the cutoff knob so that it just stored the cutoff cc value without actually making the change to the cutoff filter itself.
that's right yes. But you need an electronic switch and this is called RAC (I think).
Rudi
What does the electronic switch do?RL wrote:Hi Bob,mayidunk wrote:If the above is true, could the PC board be altered to also feed CVs from the pot to the traces that feed analog CVs directly to the cutoff freq? And perhaps the software could then be updated to alter the existing subroutine for the cutoff knob so that it just stored the cutoff cc value without actually making the change to the cutoff filter itself.
that's right yes. But you need an electronic switch and this is called RAC (I think).
Rudi
My thoughts on the stepping of the filter cutoff:
Wouldn't it be possible to add a slew between the analog voltage derived from the digital value scanned by the cutoff pot and the actual CV going to the filter? I doubt you could hear it with large knob turns, but it would smooth out the stepping with small knob turns. Just a few milliseconds should be enough. The Virus TI does a wonderful job of smoothing when changing cutoff and pitch, which gives it an organic feel in some way. And that's in the digital domain. In the analog world, this is easily done with a resistor and capacitor.
This slew of course only affects the cutoff input from the cutoff knob, the other filter cutoff inputs should not be altered.
Can this really be that hard?
B
Wouldn't it be possible to add a slew between the analog voltage derived from the digital value scanned by the cutoff pot and the actual CV going to the filter? I doubt you could hear it with large knob turns, but it would smooth out the stepping with small knob turns. Just a few milliseconds should be enough. The Virus TI does a wonderful job of smoothing when changing cutoff and pitch, which gives it an organic feel in some way. And that's in the digital domain. In the analog world, this is easily done with a resistor and capacitor.
This slew of course only affects the cutoff input from the cutoff knob, the other filter cutoff inputs should not be altered.
Can this really be that hard?
B
Not midi CCs, but digital data.mayidunk wrote:...
Would it be correct to say that the sampled voltages are used to create midi cc values? Would it also be correct to say that the midi cc values are then used to set the value of the cutoff in real time?
This would involve massive changes on the knob board. And not only software changes.If the above is true, then could the PC board be altered to feed CVs from the pot to the traces that feed analog CVs directly to the cutoff freq without eliminating the scanning of the CV from the cutoff pot? Perhaps the software could then be updated to alter the existing subroutine for the cutoff knob so that it just stored the scanned cutoff cc value without actually making the change to the cutoff filter itself....
First you would have to change all the knobs, so their output would govern the full range of needed CV span. Just think of the filter cutoff covering many octaves. Would for example 5 Volt span be enough? And 5 V would be way to much on the detune. Then each pot would have to be connected with the right point on the analog board.
But then, the voltage would be present all the time. So you have to cut this voltages when selecting a new preset. But this switch has to check, if you edit the sound afterward to do the connection again.
So you either end up doing something without a memory (like the Voyager Old School), or you have to redesign the synth. And then you have to pay for all the edition hardware needed. This is why Rudi (RL) wrote about the cost, that the Little Phatty method is way to expensive on such a big synth with so many knobs.
Believe me, if it would be a easy thing to do, Moog would have done it. But neither Moog nor other synth manufactures did it like you propose it.
keep on turning these Moog knobs
Prodigy * minimoog '79 * Voyager * MF102 * MF103 * MF104z * MP201 * Taurus 3 * Minitaur * Sub Phatty * MF105 * Minimoog 2017+ MUSE * One 16 (sold)
Prodigy * minimoog '79 * Voyager * MF102 * MF103 * MF104z * MP201 * Taurus 3 * Minitaur * Sub Phatty * MF105 * Minimoog 2017+ MUSE * One 16 (sold)
Till,till wrote:Not midi CCs, but digital data.mayidunk wrote:...
Would it be correct to say that the sampled voltages are used to create midi cc values? Would it also be correct to say that the midi cc values are then used to set the value of the cutoff in real time?
This would involve massive changes on the knob board. And not only software changes.If the above is true, then could the PC board be altered to feed CVs from the pot to the traces that feed analog CVs directly to the cutoff freq without eliminating the scanning of the CV from the cutoff pot? Perhaps the software could then be updated to alter the existing subroutine for the cutoff knob so that it just stored the scanned cutoff cc value without actually making the change to the cutoff filter itself....
First you would have to change all the knobs, so their output would govern the full range of needed CV span. Just think of the filter cutoff covering many octaves. Would for example 5 Volt span be enough? And 5 V would be way to much on the detune. Then each pot would have to be connected with the right point on the analog board.
But then, the voltage would be present all the time. So you have to cut this voltages when selecting a new preset. But this switch has to check, if you edit the sound afterward to do the connection again.
So you either end up doing something without a memory (like the Voyager Old School), or you have to redesign the synth. And then you have to pay for all the edition hardware needed. This is why Rudi (RL) wrote about the cost, that the Little Phatty method is way to expensive on such a big synth with so many knobs.
Believe me, if it would be a easy thing to do, Moog would have done it. But neither Moog nor other synth manufactures did it like you propose it.
I figuered that, since an analog pedal can be plugged into the filter jack and manipulate the cutoff value without there being any stepping noise, that the incoming CV from the pedal was not being scanned, and so was directly manipulating the cutoff. Supposing this to be true, I then thought that it might be possible to have the cutoff knob potentiometer's CV sent directly to the cutoff, the same as the pedal's CV.
As for the magnitude of the range of CVs that the cutoff pot needs to produce, i.e., the CV span that you mention above, aren't those CVs already being produced by the cutoff knob's pot? These are the voltages that are being sampled and digitized by the scanner, no? I guess I don't follow why you're saying that all of the pots would need to be altered to achieve the range of CVs the cutoff filter needs, if the cutoff knob's pot is already producing these wide ranging control voltages. Unless that pot is not doing what I think it's doing...
The question here is that, since the pedal is driving the cutoff with an analog CV, and since the the cutoff knob's potentiometer is presumably already dropping the full range of control voltages needed for it to cover all of the octaves it needs to cover, then why couldn't the analog CV voltage being produced by the cutoff knob's pot be used to drive the cutoff point the same as the pedal's analog CV would? Of course, I'm not taking into account any voltage summing that may be going on in the digital circuitry that keeps the different CV controllers (i.e., the mod wheel, the pedal, any other pot that may be mapped to the cutoff...) from arbitrarily altering the resultant CV at the cutoff whenever they are brought into the mix. This is probably where my idea falls apart.
Thanks for straightening me out regarding the scanner only capturing voltage samples as digital data points, and not doing midi cc conversions with them. This is a good discussion, I appreciate yours' and Rudi's input on this. I may just be learning something, here!
Thanks.
Bob
Last edited by mayidunk on Thu Dec 10, 2009 12:51 pm, edited 1 time in total.
Hi Bob,
please keep in mind, that the Voyager is able to store parameters. So the stored data is transformed into a steady control voltage that sets the filter to the certain cutoff. To control these stored data, the data is digitally changed. And the data used for the change is set by the cutoff knob. So there is not yet a direct voltage connection to the cutoff CV in of the filter. The changed voltages at the pots are A/D converted and then the digital board changes the set value for the cutoff digitally. And this digital information is D/A converted to voltages that actually change the full analog sound generation.
The parameter storage and the parameter changes are digital. But result and produced by analog voltages.
The CV filter input jack is not programmable at all. So this voltage is always add to the control voltage that sets the stored cutoff point. And so these stored/edited voltage is generated very different from the live and real voltage input.
The range of voltage change is measured. To make things easier, all the knobs use the very same voltages. On the Waldorf WAVe for example all knobs use a 10k linear pot. So one scanning electronic is used (most likely multiplexed) to measure the current knob position and creates a digital representation of the knob position. This now digital value is checked, if it was changed against the stored patch values after the call of the current patch (=edited). If the knob was moved (and the sound edited), the digital value is converted into voltage to drive the real analog CV parameters. If it was not touched (= no edit), the stored digital parameter value is transformed into the needed analog voltage to drive a parameter on the analog board. But the voltage range used to change the analog CV may have a wider or more narrow range of voltage change. And it does not depend on the pot on the front panel. This panel knobs are just linking interface parts for us to communicate with the digital board in the old fashioned way of knobs.
The knob is changing from zero to some voltage. And this voltage is NOT representing the actual control voltage. So the range of the pots (their resistant values) is the same on all knobs. No matter, if the filter is changes several octaves of 1 V/oct or just the little needed voltage change (a few hundred mV of range!) for detuning.
Think of the pots as controllers. Not used for midi (but also creating it), but giving data to the digital board. And the digital board plus the analog input jacks and the interface input are the only analog voltages to "talk" to the analog board where _ALL_ sounds are then created based on these external feed in voltages. This is the main concept of analog synths with digital memory for about 30 years.
Please believe me, that your idea is simply not working. And it would take serious redesign of all the panel knobs and boards, the digital board and would need many, many addition parts. So this is not a thing you could do for an upgrade. Or a custom job. It would be a very expensive new Voyager with plenty of additional parts.
And the concept of the Little Phatty is different. And because of the high cost for real knob mode, it does not feature specific pots for all the functions. You may assign a few pots to a bigger number of parameters. Just to save high costs. But this was pointed out before by Rudi, the one and only software developer for the Voyager.
And please name me a single digital stored sound synth, that uses real knobs to direct control the CVs voltages? I am into synths for 28 years. And I did not heard about this yet (Little Phatty excluded).
And by the way: I am not a technician. I am a hobby synth player and a master carpenter. But I always like to know what is going on into things I use. I dislike the idea of some kind of "black box".
please keep in mind, that the Voyager is able to store parameters. So the stored data is transformed into a steady control voltage that sets the filter to the certain cutoff. To control these stored data, the data is digitally changed. And the data used for the change is set by the cutoff knob. So there is not yet a direct voltage connection to the cutoff CV in of the filter. The changed voltages at the pots are A/D converted and then the digital board changes the set value for the cutoff digitally. And this digital information is D/A converted to voltages that actually change the full analog sound generation.
The parameter storage and the parameter changes are digital. But result and produced by analog voltages.
The CV filter input jack is not programmable at all. So this voltage is always add to the control voltage that sets the stored cutoff point. And so these stored/edited voltage is generated very different from the live and real voltage input.
The range of voltage change is measured. To make things easier, all the knobs use the very same voltages. On the Waldorf WAVe for example all knobs use a 10k linear pot. So one scanning electronic is used (most likely multiplexed) to measure the current knob position and creates a digital representation of the knob position. This now digital value is checked, if it was changed against the stored patch values after the call of the current patch (=edited). If the knob was moved (and the sound edited), the digital value is converted into voltage to drive the real analog CV parameters. If it was not touched (= no edit), the stored digital parameter value is transformed into the needed analog voltage to drive a parameter on the analog board. But the voltage range used to change the analog CV may have a wider or more narrow range of voltage change. And it does not depend on the pot on the front panel. This panel knobs are just linking interface parts for us to communicate with the digital board in the old fashioned way of knobs.
The knob is changing from zero to some voltage. And this voltage is NOT representing the actual control voltage. So the range of the pots (their resistant values) is the same on all knobs. No matter, if the filter is changes several octaves of 1 V/oct or just the little needed voltage change (a few hundred mV of range!) for detuning.
Think of the pots as controllers. Not used for midi (but also creating it), but giving data to the digital board. And the digital board plus the analog input jacks and the interface input are the only analog voltages to "talk" to the analog board where _ALL_ sounds are then created based on these external feed in voltages. This is the main concept of analog synths with digital memory for about 30 years.
Please believe me, that your idea is simply not working. And it would take serious redesign of all the panel knobs and boards, the digital board and would need many, many addition parts. So this is not a thing you could do for an upgrade. Or a custom job. It would be a very expensive new Voyager with plenty of additional parts.
And the concept of the Little Phatty is different. And because of the high cost for real knob mode, it does not feature specific pots for all the functions. You may assign a few pots to a bigger number of parameters. Just to save high costs. But this was pointed out before by Rudi, the one and only software developer for the Voyager.
And please name me a single digital stored sound synth, that uses real knobs to direct control the CVs voltages? I am into synths for 28 years. And I did not heard about this yet (Little Phatty excluded).
And by the way: I am not a technician. I am a hobby synth player and a master carpenter. But I always like to know what is going on into things I use. I dislike the idea of some kind of "black box".
keep on turning these Moog knobs
Prodigy * minimoog '79 * Voyager * MF102 * MF103 * MF104z * MP201 * Taurus 3 * Minitaur * Sub Phatty * MF105 * Minimoog 2017+ MUSE * One 16 (sold)
Prodigy * minimoog '79 * Voyager * MF102 * MF103 * MF104z * MP201 * Taurus 3 * Minitaur * Sub Phatty * MF105 * Minimoog 2017+ MUSE * One 16 (sold)
Hi Till,
Great explanation! So, the voltages produced by all of the pots, cutoff included, are only for the purpose of depicting the position of the knob, they don't produce the CVs themselves. All of the CVs are created in the digital board based upon the relative positions of the knobs. And, like you said, the CVs produced by the pedal plugged into the filter jack are just added to the CVs already being sent to the cutoff. So, in the end, I guess the cutoff pot is not doing what I thought it was. (Silly wabbit!) Thanks for taking the time to explain that to me.
Do you know if the OS's interface uses scanners on the panel pots like the Voyager's, or do the pots produce the CVs directly? I've been reading that there are no stepping artifacts in the filter knob like there are on the Voyager's. Is the OS truly old school in every way?
Thanks,
Bob
Great explanation! So, the voltages produced by all of the pots, cutoff included, are only for the purpose of depicting the position of the knob, they don't produce the CVs themselves. All of the CVs are created in the digital board based upon the relative positions of the knobs. And, like you said, the CVs produced by the pedal plugged into the filter jack are just added to the CVs already being sent to the cutoff. So, in the end, I guess the cutoff pot is not doing what I thought it was. (Silly wabbit!) Thanks for taking the time to explain that to me.
Do you know if the OS's interface uses scanners on the panel pots like the Voyager's, or do the pots produce the CVs directly? I've been reading that there are no stepping artifacts in the filter knob like there are on the Voyager's. Is the OS truly old school in every way?
Thanks,
Bob
Hi Bob,
the Voyager Old School is using a direct pot to analog sound board approach. Just like the old Minimoog used to do.
Adding a digital board between the knobs and the analog sound engine would app to the cost and would be without a use at all.
The only digital thing in the Voyager appears to be the keyboard logic, because the used Fatar keyboard does not produce voltages by using an array of resistors. It is the very same keyboard as used in the normal Moog Voyager and some other companies synths. Fatar is on of the biggest keyboard supplier. But I am not sure if this is really a digital thing or something like analog logic board. Maybe some could give better details about this here.
So there are no steps when using a Voyager Old School knob.
the Voyager Old School is using a direct pot to analog sound board approach. Just like the old Minimoog used to do.
Adding a digital board between the knobs and the analog sound engine would app to the cost and would be without a use at all.
The only digital thing in the Voyager appears to be the keyboard logic, because the used Fatar keyboard does not produce voltages by using an array of resistors. It is the very same keyboard as used in the normal Moog Voyager and some other companies synths. Fatar is on of the biggest keyboard supplier. But I am not sure if this is really a digital thing or something like analog logic board. Maybe some could give better details about this here.
So there are no steps when using a Voyager Old School knob.
keep on turning these Moog knobs
Prodigy * minimoog '79 * Voyager * MF102 * MF103 * MF104z * MP201 * Taurus 3 * Minitaur * Sub Phatty * MF105 * Minimoog 2017+ MUSE * One 16 (sold)
Prodigy * minimoog '79 * Voyager * MF102 * MF103 * MF104z * MP201 * Taurus 3 * Minitaur * Sub Phatty * MF105 * Minimoog 2017+ MUSE * One 16 (sold)
The only digital anything in the OS is the scanning board for the keys...the rest of it is analog, which is why I want one so very badly...but I cannot afford it because of my Taurus III purchase.mayidunk wrote:Hi Till,
Great explanation! So, the voltages produced by all of the pots, cutoff included, are only for the purpose of depicting the position of the knob, they don't produce the CVs themselves. All of the CVs are created in the digital board based upon the relative positions of the knobs. And, like you said, the CVs produced by the pedal plugged into the filter jack are just added to the CVs already being sent to the cutoff. So, in the end, I guess the cutoff pot is not doing what I thought it was. (Silly wabbit!) Thanks for taking the time to explain that to me.
Do you know if the OS's interface uses scanners on the panel pots like the Voyager's, or do the pots produce the CVs directly? I've been reading that there are no stepping artifacts in the filter knob like there are on the Voyager's. Is the OS truly old school in every way?
Thanks,
Bob
Minitaur, CP-251, EHX #1 Echo, EHX Space Drums/Crash Pads, QSC GX-3, Pyramid stereo power amp, Miracle Pianos, Walking Stick ribbon controller, Synthutron.com, 1983 Hammond organ, dot com modular.
Thanks, Till! And thanks to all for letting me hijack the thread for a few posts. This was a good discussion! And, Voltor, there's an OS out there with your name on it. Besides, 3 hours after those pedals arrive you'll forget all about getting an OS, you'll be too busy figuring out how to put the house back on its foundation! 
Another way around the noise problem (w/o having to build a black box) is to use your VX-351. W/ no pedal plugged into Mod 1, use the Mod 1 out on the vx to an attenuator input. Take the attenuator output to the filter input on the back of your Voyager. Voila! Use the attenuator knob as your new cutoff knob (and keep the c/o knob on the front panel all the way counter clock wise). Just tried this, and it works great!
-andrew bunny
-andrew bunny