[sdiy] Question for those with musical ears

Mike Bryant mbryant at futurehorizons.com
Mon Apr 19 18:12:01 CEST 2021


Interesting thanks.

I started with a 100uS slide on the frequency bin windows, which are several cycles of the frequency being detected, but have now gone to 1mS with no real noticeable effects on some quite extreme pitchbends (Keith Emerson - 3/4s into Knife Edge).  
I'll give 5mS a try.

-----Original Message-----
From: Synth-diy [mailto:synth-diy-bounces at synth-diy.org] On Behalf Of Roman Sowa
Sent: 19 April 2021 11:38
To: synth-diy at synth-diy.org
Subject: Re: [sdiy] Question for those with musical ears

Here's a little recording I made couple of years ago for testing how fast CV must change not to be heard as steps. It also used full 14-bit resolution PitchBend for this like in Mike's experiment.
In this case PitchBend range was 10 octaves, so less than a cent per step. It's definitely too fast sweep to even consider looking for any stepping, but it was great fun making it.
The result of my test was that as long as frequency is updated every 5ms or faster, then I'm good, whatever the step was.

https://soundcloud.com/roman-sowa/midi-pitchbend

Roman
PS. watch out for volume, it's earsplitting at upper end.

W dniu 2021-04-16 o 18:32, Mike Beauchamp pisze:
> On 4/14/21 7:47 PM, Mike Bryant wrote:
>> Consider a continuous glissando played on an analogue synth - the 
>> frequency rises smoothly between the start and end pitches.
>>
>> But on a digital synth there will always be discrete steps between 
>> successive frequencies as the frequency is gradually stepped between 
>> the start and end pitches.
>>
>> Question is, what is the minimum step (in cents) needed such that the 
>> best musical ears cannot tell the frequency isn't rising smoothly as 
>> on the analogue synth, but in many discrete steps.
>>
>> It may be that the rate of change can effect it so if so please 
>> assume a glissando starting at middle C rising at 1 second per semitone.
>>
>> Thanks
>>
>> Mike
>>
> 
> Interesting question Mike..
> 
> I have built a continuous pitch instrument that covers 4 octaves, 
> while it is all analog it also has a MIDI output. So those 4 octaves 
> get turned into a 14bit pitch bend message which should be .2 cent 
> resolution. Previously I was only getting 12 useable bits out of my 
> ADC which is a little more than 1-cent resolution and that's as low of 
> resolution as I'm personally OK with going.
> 
> It would be very easy to test the perception of this by just lowering 
> the ADC resolution until I hear some stair stepping in the pitch. But 
> like all the other perceptions, I'm very reluctant to assume that my 
> threshold is on par with other people that might use this gear. If 
> you've ever heard me sing, you know that pitch isn't something I'm 
> most sensitive to :)
> 
> But as others have mentioned as well, the context probably matters - 
> so if another instrument is holding a steady note while this one is 
> being changed or if there are long reverb tails or delays to leave the 
> original note for the new note to beat against then the perceptions of 
> the pitch changes would be heard in these beatings and that could make 
> things very noticeable.
> 
> I'd imagine this gets more sensitive at higher pitches too:
> 
> - A 1 cent pitch difference at C2 gives a beat frequency of .5Hz. Great.
> - A 1 cent pitch difference at C8 gives a beat frequency of 41Hz. Yikes!
> 
> 
> When it comes to turning something into discrete steps, I think we 
> should be aiming for close-to-ideal instead of barely-perceptible when 
> hardware allows.
> 
> The pitch resolution is usually an after-thought since most synths are 
> played with keyboards, so the pitch resolution is only important for 
> the tuning granularity and pitch-bend-wheel flourishes. But more 
> alternative controllers are around that are pushing for constant 
> continuous pitch (like the giant Roli) and I think digital synths need 
> to be designed with this in mind, allowing for large pitch-bends (at 
> least 4 octaves) and with a very high resolution.
> 
> Mike
> 
> 
> 
> 
> 
> 
> 
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy
> Selling or trading? Use marketplace at synth-diy.org
_______________________________________________
Synth-diy mailing list
Synth-diy at synth-diy.org
http://synth-diy.org/mailman/listinfo/synth-diy
Selling or trading? Use marketplace at synth-diy.org



More information about the Synth-diy mailing list