Poly-Ana 1.0 New VA softsynth. 1 week full functioning demo

[AdmiralQuality]

Hi Paulie

Can you send me the patch that's causing clicking so I can hear it? (Or if its a preset, just let me know which one.) Also, could you send me a short .wav of the click sound? Or does it not happen when you render your tracks, only in real time?

No plans to increase the maximum polyphony, but you might be able to work around it with 2 or more instances of Poly.

Thanks for the kind words everyone, great to hear you're liking her!

[Paulie Phonick]

The patch I find it easiest to reproduce this behavior with is the 'Bassmaster' preset, with attack time on ENV1 raised to 0.0309s to avoid the initial click. Play really fast (= ~32nd at 120 BPM) legato (=slightly overlapping) bass notes eg. in the G1-G2 range and the clicks will become immediately apparent (especially in headphones). At a close look at the recording, the clicks look as skips in the output (weird). Their placement is directly corellated with positions of note-on events. I've emailed a .wav recording containing these clicks to support at admiralquality.com (can't put it online at the moment).

The clicks are present when playing live as well as on renderings, regardless of the oversampling setting. I'm using PolyAna in energyXT 1.4, with EMU 1616M at 44.1 kHz and 256 samples latency. The CPU load does not exceed 15%.

[AdmiralQuality]

Hi Paulie,

(Here's the .wav Paulie sent me for reference: http://www.admiralquality.com/mp3/polya ... clicks.wav )

Thanks for the good reporting! I definitely hear the clicks you mention. At first I had some trouble reproducing this myself but now I hear it, only on notes where the pitch of the note changes (but not every time).

Let me look at a few things so I can understand exactly what's causing this and get back to you on it. At first I thought it might just be the sudden jump in pitch but if that was the case, portamento should smooth it out and I don't hear that happening. Might be another tweak to the envelope code is needed. (In fact I think this is it, notice that setting the Emv-1 VEL LEVEL to zero gets rid of the click. Already fixed an issue with velocity level and the release segment being interrupted by a new attack... I think this might be the same thing but from when the sustain segment is interrupted by a new attack. If so, it'll be an easy fix.)

Good catch! And don't worry, I'm sure it's fixable. Will get back to you soon on this.

[meeks]

Is the five days up? I'm getting message that my time is up. Is there going to be another going up to try it out?

[funkychickendance]

And did you get to the bottom of the "click" problem?

Interestingly, I finally encountered this glitch, too...on the clavinet patches, if you play real fast. I thought edging up the buffer size would fix it, but it didn't...seems to be an attack/release issue. But you're the expert, so you tell me.

/funxi

[AdmiralQuality]

Yep funxi, there will be a version 0.91 coming soon (Monday?) and will address that click issue.

meeks, your evaluation period was 7 days, not 5. (Actually it's 8 days!) Sorry but sounds like your time's up. There may be an evaluation reset at some point in the future but not before the 1.0 release, so if you liked Poly-Ana and want the $60.00 off deal, now is the time.

And now by popular demand... a firm release date for 1.0: April 2nd. (Weather permitting. ) There, you have a bit more time now to take advantage of the Beta offer, as a few of you had requested higher up in the thread.

Cheers!

[AdmiralQuality]

And did you get to the bottom of the "click" problem?

Interestingly, I finally encountered this glitch, too...on the clavinet patches, if you play real fast. I thought edging up the buffer size would fix it, but it didn't...seems to be an attack/release issue. But you're the expert, so you tell me.

/funxi

BTW funxi, the "Clean Clavinet" patch won't be affected by the 0.91 fix. The reason that patch is clickky is because I like clicks. They're funky! But you can soften it a bit by turning the attack time on Env-1 up a bit (like to about .004 second instead of .0001) and the same thing with Env-2's release time.

[meeks]

meeks, your evaluation period was 7 days, not 5. (Actually it's 8 days!) Sorry but sounds like your time's up. There may be an evaluation reset at some point in the future but not before the 1.0 release, so if you liked Poly-Ana and want the $60.00 off deal, now is the time.

Right....what I'm saying is, I installed it, tried to open it, and instantly got a pop up message that my time had expired. There was no evaluation, unless I did something wrong was I suppose to enter something? Yeah, my time is up...best of luck to ya.

[AdmiralQuality]

Sorry meeks, I didn't understand from your post that you didn't get any evaluation at all. You seemed to be talking about "5 days" so I assumed you did have some evaluation time, as the expired message doesn't mention anything about the number of days, it just says it's expired. If you got a message at some point saying "5 days" that was because you were on day 3 of your evaluation.

So if your time isn't up, something must be wrong.

You're absolutely sure you haven't run Poly-Ana before? Also, what version of Windows are you running? And just a guess here: Is is possible that you've installed it from an account that lacks administrative privileges on your system?

Let me know, and I may PM you with some more things to look at. Believe me, if there's a bug in the evaluation period code, I want to find it and get it working on your system.

Cheers.

[funkychickendance]

And did you get to the bottom of the "click" problem?

Interestingly, I finally encountered this glitch, too...on the clavinet patches, if you play real fast. I thought edging up the buffer size would fix it, but it didn't...seems to be an attack/release issue. But you're the expert, so you tell me.

/funxi

BTW funxi, the "Clean Clavinet" patch won't be affected by the 0.91 fix. The reason that patch is clickky is because I like clicks. They're funky! But you can soften it a bit by turning the attack time on Env-1 up a bit (like to about .004 second instead of .0001) and the same thing with Env-2's release time.

On examination, this can be 'played around' by not pushing my high-speed trills to wanky Eddie van Halen velocities on a QWERTY keyboard. The Env-1/Env-2 change helped soften it up a bit, too. And it's hard to detect at all from a regular MIDI k'board.

So, it's a buy. Edit: done, and very simple it was too! Time to rotate the GUI to something else now...the oatmeal-colored one makes my playing too restful

/funxi

[meeks]

Sorry meeks, I didn't understand from your post that you didn't get any evaluation at all. You seemed to be talking about "5 days" so I assumed you did have some evaluation time, as the expired message doesn't mention anything about the number of days, it just says it's expired. If you got a message at some point saying "5 days" that was because you were on day 3 of your evaluation.

So if your time isn't up, something must be wrong.

You're absolutely sure you haven't run Poly-Ana before? Also, what version of Windows are you running? And just a guess here: Is is possible that you've installed it from an account that lacks administrative privileges on your system?

Let me know, and I may PM you with some more things to look at. Believe me, if there's a bug in the evaluation period code, I want to find it and get it working on your system.

Cheers.

Never ran it before...Brand new Vista laptop....FL7. I have full admin rights. It does this....

Auth Error 3 [ok]
Auth Error 3 [ok]
Thank you for eveluating....etc [ok]

[AdmiralQuality]

Never ran it before...Brand new Vista laptop....FL7. I have full admin rights. It does this....

Auth Error 3 [ok]
Auth Error 3 [ok]
Thank you for eveluating....etc [ok]

Thank you very much for the details meeks. This very probably is Vista, and a known (or at least suspected) issue. New security policies in Vista have broken a lot of software installs and I have yet to build a Vista system to test on so I'm not surprised if there's issues. You may very well be the first to try it on Vista. So thanks again for the report.

Is anyone using Poly-Ana on Vista successfully?

I'll get back to you ASAP on this meeks, obviously it has to be fixed as Vista isn't going away.

[thecontrolcentre]

Well, I've enjoyed my week with the nonSSE version of Poly-Ana ... I've been quite busy, so I only got to spend a couple of evenings tweaking, but both sessions have produced some great samples This is a very special sounding VSTi ...
@Admiral Quality ... I was wondering if I buy Poly-Ana now will you be continuing to develop the nonSSE version, or is the current version "it"?

[AdmiralQuality]

Well, I've enjoyed my week with the nonSSE version of Poly-Ana ... I've been quite busy, so I only got to spend a couple of evenings tweaking, but both sessions have produced some great samples This is a very special sounding VSTi ...
@Admiral Quality ... I was wondering if I buy Poly-Ana now will you be continuing to develop the nonSSE version, or is the current version "it"?

Glad you like her TCC. Right now, as far as I know, you're the only one usiung the SSE-disabled version so this all actually hinges on you. If you buy it I'll continue to release SSE-disabled versions for you as long as you need (if I recall from your previous post, you are planning an upgrade soon, correct?) If you don't then I'll just let popular demand drive this one. If a customer needs it, I'll produce it. The non-SSE version isn't exactly hard to do, but it does add about an hour to a process that's already taking most of a day (getting a release out.)

Speaking of which, the 0.91 release that I was hoping to have out today is close, but not quite there yet, and I have some other things to take care of so it might be tonight or tomorrow before it's out. Just FYI.

Also note that the clicking issue has been discovered and it's not quite what I thought it was, though it's close. What's happening is the velocity level affects the amplitude of the entire envelope. So when playing a Unison sound legato with legato triggering set to ON, and with a different velocity to the new note than the previous note, you get a sudden jump as you move from the decay/sustain segment of the 1st note's envelope to the 2nd's note's new level. This problem never came up on classic mono-synths as they didn't have velocity sensitivity.

I can't think of many ways this could be dealt with without applying some kind of smoothing to the envelope levels (which I think is ungraceful, would interfere with intentionally clickky sounds, and would just add more CPU load). But maybe the idea that makes most sense is if legato triggered notes don't take the new velocity... instead they maintain the velocity of the note that originally opened the envelopes. This would only happen when Legato Triggering is selected on a Unison sound (Legato Triggering switch has no effect on Polyphonic sounds anyway.)

Make sense?

[AdmiralQuality]

Beta 0.91 (RC2) is out!

What's new:

- Legato triggering of unison sounds now maintains original note velocity. Eliminates clicks in patches like Bassmaster when played legato with varying velocity. Polyphonic sounds and/or sounds without Legato Trigger on will not be affected by this change.

- Added pitchbend smoothing and option to disable. There you go totc, no complaints now, right?

And meeks, could you download and try this new v0.91 on Vista please and get back to me by email or PM? I fixed the problem you had (I think) but there may be others further down the line so we'll take it one error at a time.

Send me any remaining bug reports! The 1.0 release is scheduled for April 2, you have until then to take advantage of the $69.95 offer. ($129.95 USD after April 2, 2007.)

Thanks! Enjoy everybody.