This topic was created before release of the latest product version, and it may contain details irrelevant to this version. Replying is disabled for this topic.
I know you've been busy with CurveEQ and LampThruster releases all week and haven't had a chance to deal with this enhancement.
I'm just giving it a friendly bump in case it fell too far down the list !
Thanks for your great support and quality products.
OK, I've put a newer version (it's not on Downloads page yet): http://www.voxengo.com/files/VoxengoSonicFinalizerVST_12a_setup.exe
Instead of RMS or average I've implemented envelope follower detector.
I don't know whether it suits your goal or not, but Env mode alone sounds simply great.
Aleksey - I don't know if the Envelope function is what kylen was looking for, but it *made* a bass track I've been working on. It rocks! You rock! Thank you so much.
I'm working with the new Sonic Finalizer w/Envelope Follower detector now and dialing in some sounds.
Is it possible to say more about which envelope you are following without giving away trade secrets?
Just curious, it's not an rms or average envelope. I'm kind of new to this so I'm just trying to understand how the envelope is obtained and what it is an envelope of? I see in my reasearch the term volume envelope used quite a bit as a control signal but I don't know what they mean by "Volume".
Has something to do with loudness I guess...
Maybe I'll email you with a more specific question later on or bump into the answer on the internet.
Thank you for the enhancement!
Envelope follower kind of follows the envelope of the signal. Of course, it captures the changes in the volume, so it must be close to RMS and average in this aspect. Also, it has some inertia (like RMS and average have).
I implemented envelope follower simply due to CPU usage which is very low for it. RMS and average will require much more CPU.
Thanks for the explaination and new feature Aleksey.
I'm afraid this one's going to keep me busy for a while calibrating it and having fun with it.
I've run some pink noise thru it and using CurveEQ as a spectrum analyzer can see how it works somewhat.
I've set the High gain to 3.5db and the Low gain to 2.6db to give me a flat curve on CurveEQ (using +3db/octave slope). Then with the fastest attack time and medium release time I drop both high and low threshold at the same time (using mouse alt/click) until I get a light amount of orange across Sonic Finalizer gain reduction spectrum (ratio is highest, 8.5:1).
Now I change the detector input from Peak to Envelope and the orange lights turn off from about 1KHz on up thru 22.5K. In other words the compressor is paying attention to the part of the signal with the most power now which is what I want.
Now I can put another Sonic Finalizer in the chain to control the peaks that the Envelope Follower SF doesn't pay attention to.
I'm trying this on music also and so far I can get much more control than just using the Peak setting. So far it kind of seems like the kind of control you get when you put an equalizer on a compressor in the side-chain and lower all the bands from 1KHz on up. All I'm saying here is that I have to use it a lot more before I'll have a feel for the sensitivity of the Envelope follower and how to set the thresholds better.
Anyone else care to share some compressor calibration tips or links also? I'm kind of a rookie on this stuff - you don't just start twisting knobs do you ? What is your method is what I'm wondering.
I have to get used to the controls more because the high and low threshold, attack, and ratio (I think) interact which each other somewhat. Not just this SF version but all that I've used (just an observation).
Thanks for the great update Aleksey! This seems to be what I was asking for. All for now. I agree with John, up above, you rock!
p.s. Aleksey when you put this out if you like there's a new Cool Edit Pro forum that's started up on http://www.separatedband.com/cep/index.php
where you can announce the release under the Third Party Plugins folder.
Even though Cool Edit doesn't officially support VST adapters VST works 99% of the time now using Chainer, VST Adapter (Cakewalk) should work shortly, and Directixer.
Thanks for the forum link - I'll join in when it's crowded enough.
As for Env mode, I have not investigated it a lot myself, that was simply an experimental try... Since I've still have not put the latest beta on the public downloads, I can make various adjustments if you feel it works in some odd way. Just post your comments!
I seem to have noticed that the attack and possibly the threshold and release controls seem to have an effect outside of the range of their perspective Hi or Low band.
I haven't had a chance to fully explore this but it seems like I can use the Low band threshold and attack to affect the gain reduction lights in the High zone - or where I think the high zone should be. That may be due to the design of the machine itself and the "slope" of the controls effectiveness. You don't actually have split signals and crossovers in there do you ? I thought Sonic Finalizer might be more like a stack of band-limited compressors in a rack - 32 of them.
Anyway, if I get more info on the control interaction I'll send it along. I don't think it's new to this version though.
Right now I'm trying to see how I can fine tune the Envelope Follower and get a song under control when the average 'loudness or volume' swings around too much. I've got no problems squishing or controlling peaks from the top down. I'm working on controlling loudness from the inside at the loudness envelope - that seems to be where the envelope follower is closer to. Unless you wanted to take an rms reading from time to time and re-adjust the dynamics in the next pass. Kind of like CurveEQ spectrumatch. THis wouldn't be a dynumatch but a dynuadjust. If a song I like has an dynamic swing of min-rms to max-rms during a verse then I could set the loudness of my song to the same averages of min/max.
Oh, boy this is getting way off. I need more coffee.
Thanks Aleksey. I'll keep digging and enjoying!
kylen, of course high and low' bands controls will affect each other in some way due to the interpolation. e.g. increasing high end's attack will increase all intermediate bands' attacks too. It's all pretty simply: if you have 10ms attack for the low end, and 1ms attack for the high end, you get about 5ms attack for the mid range.
Yes, SF uses a stack of 32 bands.
In any case, before you can dial-in something, you have to imagine what do you actually need. Glued mix, airy bass end, compressed highs, etc... If you know what your mix lacks of, tuning SF becomes a pretty easy task.