Cherry Audio Releases Polymode Synthesizer Plugin!

[cherryDan]

Hi there,

This thread is about Polymode Synthesizer, so these questions about CA2600 and Voltage Modular are best asked in a more appropriate location. To address them quickly:

Could you please have a look at the aliasing issues when using osc sync on both ca 2600 and voltage modular.

When our next CA2600 update is released, possibly as early as next week, please re-examine the issue and let us know if you still see a problem. Oscillator sync by it's very nature creates complex harmonics, unless two oscillators are perfectly in tune with one another. I can't speak to how other manufacturers build their products, but our oscillators are never perfectly in tune with one another, and drift in subtle ways, so there will always be complex frequencies produced when you sync one oscillator with another.

Also another question , why was ca 2600 original sample rate of48 khz the same as voltage modular ?

Our original design was the keep the internal processing at a fixed sample rate, so that CPU usage would be consistent regardless of the host's sample rate. For example, running a vintage synthesizer plugin at 96 kHz or 192 kHz can use a great deal more CPU than is necessary, with no major payback in sound quality. Ultimately, we weren't happy with the latency introduced in order to do high-quality sample-rate conversion, and there was a CPU cost as well, so in the end it was a wash. We reworked the plugin's DSP to operate at the native sample rate after all. Not every experiment pays off, unfortunately.

- Dan @ Cherry Audio

[gentleclockdivider]

I have reported it in detail on the cherry audio forum , with soundexamples , numerous times.
https://forums.cherryaudio.com/viewtopi ... 1&start=10
It’s stone cold digital aliasing , and souns nasty, nothing nice about it .
This is what I mean with customer support , you act like you hear this aliasing issue for the first time .
To me it seems is has been ignored from day one .

[wwjd]

this is a POLYMODE thread

[Reefius]

Some people just don't know when to quit.

[gentleclockdivider]

Cherry Dan is now aware of the issue , let’s see what happens .
I’ll shut up now .

[plexuss]

I did a little study. I put a track together with 10 instances of CA2600 including the drums. I used compression and EQ on the tracks and some delay on specific tracks. I asked people for feedback on the sound quality. Here is a summary of what people said. I don't think anyone claimed anything bad related to aliasing, do you?

The question I wanted to explore was does aliasing matter in a practical sense in a mix? From what people are saying, no one specifically heard any issues related to aliasing. The blur comments I think were likely due to the sound character over-all of CA2600. Thoughts?

Here is the study thread: viewtopic.php?f=14&t=560343

I distilled the feedback down, from the thread, into categories:

MIX RELATED
emergent parts come through nicely
presence, clarity of individual parts
kick is subby
frequencies between the instruments stepping on each other
way too much going on imho
one big blur **
bit "dirty" **
bit "fizzy" in the highs
unbalanced and "dirty"
gives an uneasy feeling
disoriented
blurs together
**I had way to to much reverb on the mix commented on here. reverb removed and reposted

SOUND CHARACTER RELATED
HIGH
High end is missing
lack treble

MID
mids are fairly consistent
The mid range is nice

LOW
lacks bass definition
low end is lacking
lack bass
low end is overpowering
gap between the low end and mids

Disclaimer: This was not meant to be a scienfic study, but rather an exploration of the topic to see what people would say and if there was any correlation or useful data to be extracted. Results will be a matter of opinion and not fact.

[Tj Shredder]

I just demoed Polymode and like it. The part which is different to the usual subtractive synths is clearly the resonance filters.
In case someone (I don’t call names...) still wants to talk about aliasing, it should be related to Polymode and nothing else...! The demo should be enough to test.
I never had a Polymoog, though a colleague had, but I can’t compare the sound with my memory. I still do hear high frequencies, but its well possible that my age helps to be not too sensible regarding aliasing. Whoever is, can report here. I don’t care too much about parts in the sound only measurement can find though...
My aliasing tests just bend high notes, if I hear the highs going down while the pitch moves up, it would be aliasing. Setting the oscillators to 4’ I don’t hear those typical artifacts...(I do hear them on my beloved CA2600 at 48kHz but that can produce also much higher frequencies and I would never use it in that range, and switching to 96kHz eliminates it...)

[focusrite]

Disclaimer: This was not meant to be a scienfic study, but rather an exploration of the topic to see what people would say and if there was any correlation or useful data to be extracted. Results will be a matter of opinion and not fact.

I see the good intention but this is what I'm hearing:

[Dark Fiber]

No one specifically heard any issues related to aliasing.

That's a rather strange conclusion. As if anyone can tell that the unwanted extra frequencies are caused by aliasing. You'll have to provide an A-B test if you want to make such a conclusion.

If people are telling that the mix is "dirty", that the high frequencies are "fizzy", than you could also conclude that exactly that is caused by aliasing.

Just a reminder that foldback aliasing creates extra non-musical high frequencies and as a result of these extra frequencies the intermodulation distortion will increase. IMD is often heard in the lower frequencies. To do a better test you'll have to make another version: all the CA2600 instances in metaplugin with 8x oversampling. Then you'll have a dirty version and a cleaner version.

[giant_panda]

nevermind

[plexuss]

No one specifically heard any issues related to aliasing.

That's a rather strange conclusion. As if anyone can tell that the unwanted extra frequencies are caused by aliasing.

Its an accurate conclusion : no one specifically called out aliasing. That is at least one conclusion that can be drawn from the study.

As to whether what the listeners heard is caused by aliasing, you are correct a more controlled study would have to done. The debate continues...

[vurt]

I did a little study. I put a track together with 10 instances of CA2600 including the drums. I used compression and EQ on the tracks and some delay on specific tracks. I asked people for feedback on the sound quality. Here is a summary of what people said. I don't think anyone claimed anything bad related to aliasing, do you?

The question I wanted to explore was does aliasing matter in a practical sense in a mix? From what people are saying, no one specifically heard any issues related to aliasing. The blur comments I think were likely due to the sound character over-all of CA2600. Thoughts?

Here is the study thread: viewtopic.php?f=14&t=560343

I distilled the feedback down, from the thread, into categories:

MIX RELATED
emergent parts come through nicely
presence, clarity of individual parts
kick is subby
frequencies between the instruments stepping on each other
way too much going on imho
one big blur **
bit "dirty" **
bit "fizzy" in the highs
unbalanced and "dirty"
gives an uneasy feeling
disoriented
blurs together
**I had way to to much reverb on the mix commented on here. reverb removed and reposted

SOUND CHARACTER RELATED
HIGH
High end is missing
lack treble

MID
mids are fairly consistent
The mid range is nice

LOW
lacks bass definition
low end is lacking
lack bass
low end is overpowering
gap between the low end and mids

Disclaimer: This was not meant to be a scienfic study, but rather an exploration of the topic to see what people would say and if there was any correlation or useful data to be extracted. Results will be a matter of opinion and not fact.

ive never felt so used in my life!

[DMG68]

come on vurt, I’m sure you been used more than that in your life, you just enjoyed it more before so didn’t feel so used...
At least we were part of the great experiment!

[vurt]

[WasteLand]

an aliasing probe