Voxengo Premium Membership - All Voxengo Plugins For a Fixed One-Time Fee
Forums     Plugins     Marquis Compressor Program-Dependent Release Contour

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.

OK - things are looking up!  If I could sum up what we have so far (?)

The 3 release params comprise a system.  It's also established that this "Program-Dependent Release Contour" system can be tuned using your ears.  The values associated with each Param are more or less arbitrary (0.10 thru 10) but the graph gives a relative idea of how much program dependence we have dialed in overall via the slope.  We also know that even though PDRC param values 10,10,0.1 and 0.1,10,10 may look the same on the graph they may not sound the same.

The PDRC params also seem to affect the gain reduction meter (obviously!) is that an additional way to tune the PDRC system?  If so then maybe the meters could be slowed down a touch to better see which rms level the release is affecting (if that's the case).  I'm starting to get used to sweeping the release params but it does take a minimal set of monitoring to do it (I'm in an apartment so during off hours I would depend on metering more - if that's possible for this).


This is right, PDRC do affect gain reduction meter - its params affect the release stage behavior and thus they affect gain reduction meter as well.

If you are talking about slowing down meter ballistics, I do not think this will help much.  In fact, I think it's hardly possible to tune PDRC via gain reduction meter.


Ok, so far, I get the idea that the different parameters are adjustments for different filters.  Is it possible to say, what each filter is?  For example, why would I turn up Param 1 instead of Param 3?

Aleksey Vaneev: This is right, PDRC do affect gain reduction meter - its params affect the release stage behavior and thus they affect gain reduction meter as well.

Aleksey Vaneev: If you are talking about slowing down meter ballistics, I do not think this will help much.  In fact, I think it's hardly possible to tune PDRC via gain reduction meter.


Ok, since you want to know.  I do not have a good set of monitors.  I have learned to shoot straight with a crooked gun basically.  I am not mixing with computer speakers or anything like that, but I do have to use multiple references in order to achieve a good mix.  I just don't have the 1000 bucks or so I need for a decent set.  But trust me, it is top on my list, as mixing the way I am drives me nuts.  I am doing all of my work out of my apartment and the last bit of expenses that put me in debt was better placed on a machine that would handle all the tracks and effects/softsynths I could throw at it.  So it is not likely I will be hearing subtle differences.  Especially through a set of ear buds while I am here at work.  But I will try again in a second.

I asked a pretty straight question.  Yes?  I know how programming works, I do software for a living.  So when you assign a parameter a value, even if the value is arbitrary, it has a purpose.  I am not asking the world here, just want to know what characteristic of the release contour is changing when I turn Parameter 1, Parameter 2, or Parameter 3.  That's all.  They all obviously change the slope.  Subtle differences or not, they have specific functions.  And I simply want to know what they are.  Ok...if it is a filter...what kind of a filter?

bmanic: How many times do people have to tell you the answer?  I've already said it many times so I will do one last try (sorry if the following text seems rude but I want to make it absolutely sure you understand what I say and that you know I'm not saying stuff out of ignorance but rather trough hearing it, ABX validated):

bmanic: 0.1, 10, 10 is not sounding the same as 10, 10, 0.1.  You can clearly hear this in the clips that I posted.  Listen and compare the release behaviour of the two clips.  The cymbals 'pump & breathe' differently in the clips.  Want scientific 'proof'?  Phase reverse one of the clips and summ them in a wave editor, you will have the difference.

bmanic: Either A) you lack experience, B) have a bad monitoring system, C) you don't even want to understand what is being told to you

bmanic: Which one is it?  Truly, this is not about any pissing contest but rather answering your question but how do you show a blind man that the wall is painted yellow?

bmanic: Bottom line: the release parameters set to X, Y, Z does NOT equal them set to Z,Y,X or Y,Z,X and this is audible , more or less so depending on the source material

bmanic: I can't state it more clear than this.  Now if Aleksey isn't too bored he will provide the exact equation or something.

bmanic: - bManic


Ok....I must wholly apologize for lashing out at you bmaniac.  I downloaded the wavs so I could switch between them at a faster pace.  And yes, I could hear the difference in the releases on the cymbals.  Even through these crappy ear buds here at work.  I notice it particularly in the second crash hit.  One is released to a slightly "shraper" sound for lack of a better term.  SO again, I apologize for jumping to conclusions in my frustration.

However, I am still a person who likes to know what they are tweaking.  And a description of the filter behavior, no matter how mind boggling, would be preferred.  If I don;t get it....oh well.  I will just tweak until satisfied.

fgimian: bManic, I respect your opinion and your ears mate.  I'm simply going by exactly how Aleksey described the function of these 3 knobs.  You see, with Aleksey's description, the fact should actually be that there is no difference between the 2 samples.

bmanic: That's cool, but he is wrong then.  I trust my ears explicitly and in this case it's not even very difficult to hear the difference (the sonalksis debate was much harder to convince people and very few believed me until the developers themselves confirmed it).

bmanic: Just summ the clips, one phase reversed and it will even be quite clear in what way the knobs work.  The clue lies within the cymbals of those two clips.

bmanic: - bManic


I think I've described it all.  Also urosh gave you an idea - these filters is the same as having three peaking filters tuned to the same Q and gain, but to different frequencies.  You can always substitute peakfilter A for peakfilter C, etc.  What really matters is the overall *frequency response* (read *release contour*).

Elvenking, it's ok.  To hear dynamics well one really needs proper monitoring (any kind of headphones/earbuds are a big no-no, as it's very hard to hear compression on headphones in general).

I'll try to describe my non-scientific view on how the release controls "sound", that is, how I perceive them to work.  I have not the faintest idea of any technicalities but I'll describe what I hear.. tomorrow.  Right now I'm too tired and need to get some sleep. :)

Cheers!

bManic


Ok, as far as getting a technical angle on this.  I am throwing in my cards.  I totally understand the comparison between the filters having identical Q's and gains.  Obviously, if they represent different frequencies, they have a profoundly different effect.  The main variable here being...frequency.  If you adjust knob number one, you are going to get a very different boost than if you adjust knob number two.

Here is what I understand about compressor release and this feature.  Release time is how long the compressor will take to return to normal input level when it finds that the input signal is no longer above the threshold.  Ok.  And the Release contour shapes, how the compressor will release the signal over time.  Very cool feature.

First of all...correct me if I am wrong on the above.

Observation 1: The graph does not have any labels, so I can only guess what it is representing.

Observation 2: From what has been said here, the graph doesn't show any differences when say 10,1,10 or 1,10,10 is set.  Ok...no big deal, there actually is a sonic difference though.

Final question: Considering that the two main variables of compressor release are 1)Time and 2)Level(db), AND that this feature will allow to to create a curve so that the release time is not simply linear over the set release time.  My question is, are we actually varying the curvature of the release contour at different points of the release time with the different Parameters.  For example, my release time is 350ms, I adjust param1 because I want a steep initial release time for the first 120ms, and then I adjust Param 2 becuase now I want the levels to stop rising so fast back to the level of the input signal.  This I could understand.

Note: If I am the only one who is not crystal clear on this after all the discussion, one of the forum members can freely whack me over the head and set me straight at any time by explaining this in detail.  I fail to believe that there are things about this compressor that cannot be put into words.

Also, what settings will yield a simple linear release time?

Aleksey Vaneev: I think I've described it all.  Also urosh gave you an idea - these filters is the same as having three peaking filters tuned to the same Q and gain, but to different frequencies.  You can always substitute peakfilter A for peakfilter C, etc.  What really matters is the overall *frequency response* (read *release contour*).


deleted

Dabuek: * Well if you keep the end points at max and min then a single knob could easily be made to bend the line

This is not correct.  Please do not write something if you are not sure.

I'm not forcing you to give me your money, especially if the product is not satisfying.  I also do make refunds withing the first month of purchase.  So, if something bothers you, send me a note and I'll refund your order.

It's a talk about nothing.  I've made my comments and explanations.  Discover and explore.

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.