Stereo Tool

That reminds me of a similiar concern which I expressed in a post in the generic News section of the forum. When loud vocals (doesn't matter male/female) are present, the combined action of clipping and bringing the level down to set AGC threshold, renders weird artifacts - sometimes deforming the stereo-image. This is more pronounced when the clipping level is set very close to 100% mark (even more if channel separation is high).
The same effect can be seen (I mean heard) in single band compressor/limit when used in AGC style.

It's true that the MB doesn't exhibit such artifacts (atleast not when the preamp is not set too high ... ~+6dB ).
But 'radio' folks are not going to like this approach given the fact that this comes at an expense of uniformity in volume levels as MB is not RMS based.

_________________
visit website

I'm gradually backing down the preamp too with what I'm testing. I've dropped to +12 from +15.66 already...

That should be covered by compressor before/after MB and MB. Firts one we don't have and second one cannot do that, atleast not with this kind compress mode.

Well, from what I my ears can tell, I've had multiband set up pretty good (not perfect, but pretty good), but the AGC has been fighting against what I want, while I thought it was helping.

To give an example, here is HypersonicV6_Web with preamp reduced to +6dB, AGC disabled, preemphasis reduced, and up to 83% volume (5/8ths) from 80% (4/5ths), and nothing else changed.

The AGC seems to impact voices, seemingly imparting extra bass, which is probably why the Bass AGC got added.

Bss AGC ON ?

No, I never used the Bass AGC, although from what I remember experimenting with it last night, voices still got suppressed / bassified... I'll test more later today.

Edit: The NoAGC preset is showing a significant drop in RMS and increase in DR (2-2.5 dB, based on some initial testing) on many tracks. Voices are no longer squashed as much, so oddly enough the track being discussed here (The Pretender) is showing an increase in RMS and a drop in DR, probably due to the uncoiling of the midrange, and so I'm going to have to redo the preset based on not using the AGC.

With respect to having it all wrong for desiring lower RMS values in the mix, it's interesting that Alan Parsons, an often-lauded industry mixing and mastering specialist, who did the mastering for Pink Floyd's "The Dark Side of the Moon", is not fond of making mixes with higher RMS levels.

It is also of interest that Ian Shepherd, another mastering engineer, was able to convince Bob Katz, another mastering engineer with a published book and a few Grammy awards, that the TT offline meter does indicate DR satisfactorily.

I also think at issue here is the peak vs. RMS nature of multiband, and the lack of the ability to control the degree of compression or limiting on a per-band basis.

Here is a new version of Lucidity FX 5.6
Are you love sound characteristics of my preset?
There is new updated version for you! It has just got better dynamic. Better volume consistency (no big jumps).
Lucidity FX 5.6 (with ST6.22 beta) is close to what I want to hear!

The main changes from version 5.0:
- Prepared for SA Stereo Tool 6.22 (beta)
- More dynamics! 99.5% of tested (300) songs achieved at least DR8!
- "Volume bursts" reduced to minimum. (new AGC filtering in ST 6.22 and MB tuning)
- and... eh...because my English is bad, so in short: It just sounds better!

Updates can be downloaded in the first post: viewtopic.php?p=14419#p14419