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

VST, AU, AAX, CLAP, etc. Plugin Virtual Instruments Discussion
Post Reply New Topic
RELATED
PRODUCTS

Post

I bought this synth soon as it was available for sale. :hihi: Couldn't help myself.

The only beef I have with it is that the PitchBend action could be a little smoother. It sounds 'stepped' now, so there's a 'ladder' effect as the pitch rises when you're bending it to a full octave. Would be sweet if it went from one octave to the next without the steps. Maybe AQ can smooth it out for 1.0 though. :shrug:

Either way, fine synth indeed. Right up there with my MiniMonsta :love:

Peace 8)

Post

The spec looks very good on this synth.
But the GUI just makes it impossible for me to use it.
The perspective gives me a headache and the colour scheme is all wrong, ie: the white 'centre' marks on some knobs mix with white text making it an effort just to see where it is.

I do like the hardware look otherwise. Tassman, aims to recreate that hardware look and yet is very clear and self explanatory.

Also, the filter, when turned using a mouse is heavilly stepped. I didn't look further to see if an envelope filter sweep would eliminate the steps, since, as I mentioned, the GUI is difficult to read.

I'm sorry for the negative comments, I hope they are taken as constructive criticism, as I do like the specification. :(

Post

I finally got round to downloading and installing poly-anna as it's the last chance before the release ... but it doesn't show up in the browser in Live6.05 on my set up. AMD 1.8 Athlon and XP ... most dissapointing :(

Post

TotcProductions wrote:Hey Admiral,

The volume setting isn't saved within the PAP files...is it supposed to be that way? If so, any chance of changing that?

Thanx 8)
Yep, meant to be that way. You should adjust your patch's volume with the Gain knobs in the Amps.

Any control that's not surrounded by a rounded box isn't saved with the patch, so Volume, Tune, the Pitch Wheel, and the new Hold button are non-recalled parameters. (though it IS saved when you save .fxb .fxp files from your host. And it IS recalled with your projects.)

Post

thecontrolcentre wrote:I finally got round to downloading and installing poly-anna as it's the last chance before the release ... but it doesn't show up in the browser in Live6.05 on my set up. AMD 1.8 Athlon and XP ... most dissapointing :(
I'll double check that TCC, thanks for the report. Live users, is there anything like a "scan for new plugins" feature that might help here?

Post

himalaya wrote:The spec looks very good on this synth.
But the GUI just makes it impossible for me to use it.
The perspective gives me a headache and the colour scheme is all wrong, ie: the white 'centre' marks on some knobs mix with white text making it an effort just to see where it is.

I do like the hardware look otherwise. Tassman, aims to recreate that hardware look and yet is very clear and self explanatory.

Also, the filter, when turned using a mouse is heavilly stepped. I didn't look further to see if an envelope filter sweep would eliminate the steps, since, as I mentioned, the GUI is difficult to read.

I'm sorry for the negative comments, I hope they are taken as constructive criticism, as I do like the specification. :(
No worries himalaya, your comments are appreciated.

Note that you can download different background color schemes on the Poly-Ana page: http://www.admiralquality.com/products/Poly-Ana/

And yes, the GUI is on a lower priority thread than the sound engine, so if your CPU is taxed making sounds, this will reduce the frequency that it updates the GUI, and create the jerky knob motion. The knobs should move smoothly when the CPU load is light. Try reducing Poly-Ana's quality setting to 1X oversampling while you're performing, then raise it back up when rendering your tracks. That might give you enough CPU power back to smooth out the interface.

(Question: What kind of graphics adaptor do you have in your machine?)

Oh and totc, you mentioned "zipper noise" in the Pitch Bend. Yes, there's no smoothing applied but this was intentional as a good controller should send a dense enough stream of pitch bend messages. The other issue is latency, larger host buffers reduce the timing accuracy of MIDI events (at least in realtime playing). They sound quite smooth on my system with 3ms output buffers. So it could be your controller, your latency, or a combination of both.

Tell me, do they sound any smoother when a sequence is played back vs. real time play? Or when rendered to an audio track?

I'll think about adding an option to enable PB smoothing. I didn't want it because it reduces the performance accuracy, but to users with larger buffers its a bigger issue.

Post

AdmiralQuality wrote:
thecontrolcentre wrote:I finally got round to downloading and installing poly-anna as it's the last chance before the release ... but it doesn't show up in the browser in Live6.05 on my set up. AMD 1.8 Athlon and XP ... most dissapointing :(
I'll double check that TCC, thanks for the report. Live users, is there anything like a "scan for new plugins" feature that might help here?
Poly-Ana won't open for me with Savi Host either ... I'm wondering if this is an SSE issue? I know my CPU doesn't support SSE ...

Post

thecontrolcentre wrote:
AdmiralQuality wrote:
thecontrolcentre wrote:I finally got round to downloading and installing poly-anna as it's the last chance before the release ... but it doesn't show up in the browser in Live6.05 on my set up. AMD 1.8 Athlon and XP ... most dissapointing :(
I'll double check that TCC, thanks for the report. Live users, is there anything like a "scan for new plugins" feature that might help here?
Poly-Ana won't open for me with Savi Host either ... I'm wondering if this is an SSE issue? I know my CPU doesn't support SSE ...
Yep, that's it.

Hmmmm. I guess a non-SSE version would be the best way to take care of this. Let me get back to you in a day or two on this.

(And don't worry, if it hasn't been able to run then your evaluation period hasn't started yet, so you'll still get a full week evaluation when we get you your working copy.)

Post

AdmiralQuality wrote:
himalaya wrote:The spec looks very good on this synth.
But the GUI just makes it impossible for me to use it.
The perspective gives me a headache and the colour scheme is all wrong, ie: the white 'centre' marks on some knobs mix with white text making it an effort just to see where it is.

I do like the hardware look otherwise. Tassman, aims to recreate that hardware look and yet is very clear and self explanatory.

Also, the filter, when turned using a mouse is heavilly stepped. I didn't look further to see if an envelope filter sweep would eliminate the steps, since, as I mentioned, the GUI is difficult to read.

I'm sorry for the negative comments, I hope they are taken as constructive criticism, as I do like the specification. :(
No worries himalaya, your comments are appreciated.

Note that you can download different background color schemes on the Poly-Ana page: http://www.admiralquality.com/products/Poly-Ana/

And yes, the GUI is on a lower priority thread than the sound engine, so if your CPU is taxed making sounds, this will reduce the frequency that it updates the GUI, and create the jerky knob motion. The knobs should move smoothly when the CPU load is light. Try reducing Poly-Ana's quality setting to 1X oversampling while you're performing, then raise it back up when rendering your tracks. That might give you enough CPU power back to smooth out the interface.

(Question: What kind of graphics adaptor do you have in your machine?)

Oh and totc, you mentioned "zipper noise" in the Pitch Bend. Yes, there's no smoothing applied but this was intentional as a good controller should send a dense enough stream of pitch bend messages. The other issue is latency, larger host buffers reduce the timing accuracy of MIDI events (at least in realtime playing). They sound quite smooth on my system with 3ms output buffers. So it could be your controller, your latency, or a combination of both.

Tell me, do they sound any smoother when a sequence is played back vs. real time play? Or when rendered to an audio track?

I'll think about adding an option to enable PB smoothing. I didn't want it because it reduces the performance accuracy, but to users with larger buffers its a bigger issue.

Couple of Points:

1. I think Himalaya is talking about the SOUND of the filter being stepped not the knob movement. Could be wrong but that's the impression I got.

2. How can a Pitch Bend Wheel send a "Denser" stream of Data than the Midi Spec allows?

Post

AdmiralQuality wrote:I guess a non-SSE version would be the best way to take care of this. Let me get back to you in a day or two on this.
that's very kind ... thank you :D

Post

Teksonik wrote:Couple of Points:

1. I think Himalaya is talking about the SOUND of the filter being stepped not the knob movement. Could be wrong but that's tha impression I got.
Yes, but if the GUI is slow to refresh, then the knob will move in larger and less frequent increments, and you'll directly hear these as there's (intentionally) no smoothing applied. Longer buffers can also be responsible for this, as with the pitchbend case.

2. How can a Pitch Bend Wheel send a "Denser" stream of Data than the Midi Spec allows?
The MIDI spec doesn't specify anything about the density actually. Other than the baud rate, which is 31250 bps http://en.wikipedia.org/wiki/Midi#Elect ... onnections

10 bits per MIDI byte (due to start and stop bits)

A PB message takes 3 bytes. One status and two data bytes.

So 31250 bps / 30 bits = 1041.666... Pitchbend messages per second.

Trust me, if you send 1000 PB messages per second, you'll get very smooth sounding pitchbends from Poly-Ana. In actuality, most controllers reduce this data stream, and exactly how much is a non-constant. And this limit is only on a MIDI cable. Your host can actually time MIDI messages down to sample accuracy on playback, or the buffer size for real time performance. (Yes, performances sound more accurate on playback than during performance! Yet another reason to always work with the smallest buffer you can.)

That's all I meant. I don't in fact know yet if that's the cause of totc's observation.

Cheers!
Last edited by AdmiralQuality on Sun Feb 18, 2007 2:14 am, edited 1 time in total.

Post

I'll probably buy this synth before the price rise because....well I'm a synth slut. Is there any way to re-set the one week time limit? I'd like to try the latest demo to see if it will fit now on my 20" monitor at 1280x1024.

Post

Teksonik wrote:I'll probably buy this synth before the price rise because....well I'm a synth slut. Is there any way to re-set the one week time limit? I'd like to try the latest demo to see if it will fit now on my 20" monitor at 1280x1024.
The interface still runs after expiry. Just no sound. So if all you want is to twiddle the knobs, you can do that for free, forever. :hihi:

Oh, Poly-Ana also is a free GUI keyboard that you can use to play other instruments from (as it provides a MIDI output too!) Shhhhh! Don't tell anyone! ;)

Post

I wonder if there is a way to determine how many Pitch Bend messages per second a controller sends if this indeed the case. So theoretically setting the PW to bend +/- 2 semitones would result in less or no zippering and setting the PW to +/- an octave or more would result in more zipper effect. Totc can you try that and see if it holds true?

Post

Depends on your host.

In fact, if Totc can send me a .mid file, we can count em and I can hear it on my system. In fact, I can render it to .wav and we can compare.

Keep in mind, just because it's going to 1.0 soon, tweaks and improvements like these will continue throughout the product lifetime. "Software is never done".

EDIT: Oh and I'd recommend we set the Poly-Ana patch to +/- 4 Octaves. (Yes, Poly-Ana has 8 octaves of pitchben range). That'll always be the worst case, regardless of the other parameters.
Last edited by AdmiralQuality on Sun Feb 18, 2007 2:31 am, edited 1 time in total.

Post Reply

Return to “Instruments”