Presswerk 3rd Beta (now Rev 3225, Dec 9th)

[3ee]

Yes, I also have the GR behavior but I didn't consider it a "bug" ...I tough it was a side-effect of the auto release feature or something...

[paterpeter]

Yes, I also have the GR behavior but I didn't consider it a "bug" ...I tough it was a side-effect of the auto release feature or something...

This also happens with minimal release time and adapt=0 (no auto release).

[xx JPRacer xx]

I think this could be related with hosts "disabling" plugins that don't receive audio. For example in Renoise there's an option called "auto suspend plugin when silent". I'm sure there's something similar in Reaper.

[paterpeter]

The needle does go down after a second or so when no audio is received, so I doubt the DAW is disabling the plugin. Also I doubt that any host would disable a plugin the exact sample the audio input stops. Most delays would stop to work.

[Turello]

Hi URS and all folks, I've tried, very amazing, almost perfect but...

- some latency not allow to put it as parallel comp or in single channel (but I think that this latency is due to some programming code common to all u-he products) but only as comp in master channel

- some CPU loud allow to use it only as master comp (same for latency)

when expires the beta testing s.number?

Hope to not offence no one, aniway the plug is on beta phase so he need some pratical suggestion to grown\improve...

[wuh]

I've been using the first beta in a few of my tracks, love it. I was wondering, will this new beta alter the sound in any way when it replaces the previous instance?

[sascha]

- some latency not allow to put it as parallel comp or in single channel (but I think that this latency is due to some programming code common to all u-he products) but only as comp in master channel

Since any latency is reported, the host should be able to compensate for it, no?

[sascha]

I've been using the first beta in a few of my tracks, love it. I was wondering, will this new beta alter the sound in any way when it replaces the previous instance?

The latest changes only affect the UI and the workflow. No changes to the DSP part were made.

[3ee]

- some latency not allow to put it as parallel comp or in single channel (but I think that this latency is due to some programming code common to all u-he products) but only as comp in master channel

Since any latency is reported, the host should be able to compensate for it, no?

That's the thing, some DAWs are terrible of doing that, several of them won't do that properly so we, at least I tend to avoid latency inducing plugs...

[paterpeter]

Sascha:

1. I suppose it's by design that DPR gets applied to the dry signal, too?
2. When HQ is enabled, I can't null (a polarity inverted) Presswerk with the unprocessed version of the signal. (Scenario: track 1 - dry signal, sending to track 2 with Presswerk, track 2 has polarity inverted), even with the Dry:Wet turned to fully dry. I get difference across the whole frequency range.
3. With HQ disabled and the same scenario, I can't null either, but the difference is mostly low-frequency content.

Are 2 and 3 supposed to happen? Or should we expect perfect cancellation?

All tested in latest Reaper.

[xx JPRacer xx]

The needle does go down after a second or so when no audio is received, so I doubt the DAW is disabling the plugin. Also I doubt that any host would disable a plugin the exact sample the audio input stops. Most delays would stop to work.

Ok, sorry, misread your post. You're right, this as nothing to do with what I've said.

[xx JPRacer xx]

Little bug, sorry if it was already reported... When using a zoom level greater than 100% the description of preset are not displayed correctly in the preset browser.

[sascha]

Sascha:

1. I suppose it's by design that DPR gets applied to the dry signal, too?

Yes, rotation needs to be on the entire signal, otherwise it would cause phase issues.

2. When HQ is enabled, I can't null (a polarity inverted) Presswerk with the unprocessed version of the signal. (Scenario: track 1 - dry signal, sending to track 2 with Presswerk, track 2 has polarity inverted), even with the Dry:Wet turned to fully dry. I get difference across the whole frequency range.

In HQ mode, we're applying oversampling (4x at 44.1/48k, 2x at 88.2/96k). What you're hearing is the slight delay of the brickwall/half-band antialiasing filters. Alas, there's no free lunch.

3. With HQ disabled and the same scenario, I can't null either, but the difference is mostly low-frequency content.

At what level (RMS), specifically?
Please ensure that saturation is dialed down completely for this test. Also, when checking the 'dry' part, 'dry hp' needs to be at 1Hz (which still filters, a possible cause for the LF delta signal).

[paterpeter]

Sascha:

1. I suppose it's by design that DPR gets applied to the dry signal, too?

Yes, rotation needs to be on the entire signal, otherwise it would cause phase issues.

I expected as much, thanks.

2. When HQ is enabled, I can't null (a polarity inverted) Presswerk with the unprocessed version of the signal. (Scenario: track 1 - dry signal, sending to track 2 with Presswerk, track 2 has polarity inverted), even with the Dry:Wet turned to fully dry. I get difference across the whole frequency range.

In HQ mode, we're applying oversampling (4x at 44.1/48k, 2x at 88.2/96k). What you're hearing is the slight delay of the brickwall/half-band antialiasing filters. Alas, there's no free lunch.

I expected the oversampling to be the cause.

3. With HQ disabled and the same scenario, I can't null either, but the difference is mostly low-frequency content.

At what level (RMS), specifically?
Please ensure that saturation is dialed down completely for this test. Also, when checking the 'dry' part, 'dry hp' needs to be at 1Hz (which still filters, a possible cause for the LF delta signal).

Input: white noise at -14dBFS RMS
After cancellation: around -57dBFS RMS (it's moving between -60 and -48 I'd say)
Saturation dialed down, all Dry, no DPR, no HQ, no compression taking place. It's not that it's causing audible issues when run in parallel as far as I've heard. Also, I probably wouldn't run it as a send effect anyway but use the dry/wet dial instead. But as you were asking for this, here you go

[sascha]

Saturation dialed down, all Dry

There we have it. I guess it's the dry-part's highpass, set at 1Hz. Mix it in iverted and you have a lowpass@1Hz (6dB slope). One could probably turn it off completely, but I thought 1Hz would never be an issue in studio practice. Is it, really?