KVR Audio

V'ger · Post by **V'ger** » Fri Apr 06, 2012 7:35 pm

Thanks, really digging this compressor. Can really be pushed hard without bad things happening.

gremildo · Post by **gremildo** » Wed Apr 11, 2012 4:39 am

Hi Fabien, this comp is now my favourite (+teh DDMF NY comp *and DC8C) for clean/ ,mastering purposes.- altough your is cleaner than the NY.

Difference RMS / peak is a lot more subtle than I'm use to hear compared with other comps - I'm using the RMS btw, works fine.

I prefer to not use the Low and Hi harmonics.
Only downside of 4.0 version , ratios above 3 or 4 doens't sounds good as the previous version.It was perfect on 3.0 (Imo...)

Thank you - excellent job, don't believe it's free !

FabienTDR · Post by **FabienTDR** » Wed Apr 11, 2012 9:34 am

Hey gremildo,

yes, the RMS/Peak this is very subtle. I'm still playing around with more detection options and will try to make both modes more obvious and replace the switch with a "RMS/Peak" knob instead.

The ratio is very strong since the last update (0.4.0). The reason is that I added some kind of a "turbo charger" for ratios above ~3:1. Feedback compressors basically can't compress much more than 3:1, I used a trick to enable ratios up to 10:1. Here's how internal ratios map to effective ratios in the TDR Feedback Compressor (due to the feed-back structure):

Feed-forward ratio => Effective ratio for feed-back structure

1:1 => 1:1
2:1 => 1,5:1
4:1 => 2:1
oo:1 => 3:1
-8:1 => 10:1 <- turbo charger here

...You see the "-" sign before "8:1"?

I suppose you can now easily understand why ratios above 3:1 sound that aggressive.

I'm experimenting with a "ratio range" switch, something like "mastering/tracking" or just "turbo". So that one can easily switch between the "old" ratio behavior (0.3.0) and the new ratio "feel" (0.4.0). But I'm not sure about it.

I'll maybe combine the RMS mode with the simple ratio and the PEAK mode with the turbo charged ratio. This could elegantly fix both issues you mentioned (too strong ratio and to subtle RMS/PEAK).

BASSDRIVE · Post by **BASSDRIVE** » Thu Apr 12, 2012 4:47 am

Thank you, sounds very clean!

FabienTDR · Post by **FabienTDR** » Mon Apr 23, 2012 3:19 pm

Update v 0.4.04 beta available!

V'ger · Post by **V'ger** » Tue Apr 24, 2012 1:23 am

Thanks, what's been updated?

And I see on your to-do list a VST3 version to get external sidechain, but hope you do for VST2 version as well since it's of course possible there too.

oh and maybe an idea to put up web links in your sig and profile so it's easier to find

FabienTDR · Post by **FabienTDR** » Tue Apr 24, 2012 1:51 am

V'ger wrote:Thanks, what's been updated?

And I see on your to-do list a VST3 version to get external sidechain, but hope you do for VST2 version as well since it's of course possible there too.

oh and maybe an idea to put up web links in your sig and profile so it's easier to find

-Harmonics section completely re-factored and renamed "COLOR". The effect is now compression dependent

-"BLEND" knob mapping and display changed to dB scale. The range is limited to max -20dB. You'll now have to use the BYPASS switch to achieve lower values

I'm playing around with VST2 side-chaining, but this feature has low priority. The main problem is that this feature creates compatibility problems in many hosts and the algorithm CPU load will blow up (because the side-chain input needs to be oversampled too). It's also not really useful in the context of mastering, so I'm focusing other improvements before.

camsr · Post by **camsr** » Tue Apr 24, 2012 3:38 am

The "Blend" value, that is a trim for the wet signal? It is very confusing.

I am seeing far better CPU usage now! Something like 66% of the first version.

camsr · Post by **camsr** » Tue Apr 24, 2012 6:51 am

Cheers for the Color function! That has just the right amount of lift to it.

FabienTDR · Post by **FabienTDR** » Tue Apr 24, 2012 10:53 am

@camsr: Glad you like the new color function! Yes, I also improved the CPU performance substantially (without sonic compromises).

About the blend feature. Yes, it is a dry/wet function, but only over a limited range of 20dB. I used a linear "0% - 100%" scale before but switched it to dB scale for more accurate interaction. Also, the "0%" option made no sense IMO.

[Make-Up] -> [Blend] -> [Output Gain]

The make-up is meant to compensate the level losses introduced by compression, that is, it should be set to the maximum amount of gain reduction indicated by the meter. With such a proper gain staging, the blend function will then be able to provide well controllable upward/parallel compression. The final output gain's only purpose is to control gain staging with the next processors in the chain. It should be easy once you get used to it, at least, I hope so!

Note that the whole processor itself is designed in such a way that it doesn't need proper "analogish" gain staging. The compressor can equally handle very low (~ -30dB VU) and crazy high input levels (> +100dB VU), the sonic results are exactly the same. In other words, the input level doesn't "drive" the unit.

V'ger · Post by **V'ger** » Tue Apr 24, 2012 2:49 pm

Interesting. Btw you said the compressor was originally made for a particular need you had, has it done the job iyo? Was it for group or master bus?

FabienTDR · Post by **FabienTDR** » Wed Apr 25, 2012 3:58 pm

V'ger wrote:Interesting. Btw you said the compressor was originally made for a particular need you had, has it done the job iyo? Was it for group or master bus?

Of course I'm happy with it

And I hope to make useful for others thanks to your help (and all the other kind beta testers).

I personally like it on the master bus, 3/4 of the material I use to test the audio parts are stereo mixes. I run a label so I have access to a lot of "raw" material from different producers and many experienced users as well.

I do nearly zero tracking, so don't expect too much from this side. However, I have over a dozen prototypes waiting to be packaged for the public. Some of them are more mix oriented. The next one will probably be my high quality transient/envelope shaper. Also on the list: A heavy-weight EQ, a mastering oriented de-esser, a limiter/maximizer, a stereo processor, an exciter/color-box and more. I have most already in use, but have bad cpu performance, no gui and low stability outside my little world.

scalawag · Post by **scalawag** » Wed Apr 25, 2012 5:36 pm

So, Tokyo Dawn Records is yours?? What a damm GREAT label!
Congratulations and thank you for this GREAT compressor.

FabienTDR · Post by **FabienTDR** » Wed Apr 25, 2012 9:12 pm

scalawag wrote:So, Tokyo Dawn Records is yours?? What a damm GREAT label!
Congratulations and thank you for this GREAT compressor.

Thanks man! TDR is not "mine", just the half of it

eytanmich123 · Post by **eytanmich123** » Wed Apr 25, 2012 11:04 pm

FabienTDR wrote:
scalawag wrote:So, Tokyo Dawn Records is yours?? What a damm GREAT label!
Congratulations and thank you for this GREAT compressor.
Thanks man! TDR is not "mine", just the half of it

So, do you own the "Tokyo Da" half, or the "wn Records" ?

Can you elaborate on the new timings a little ? (the FF --> FB thing)

Anyone tried the new TDR Feedback Compressor?