MDynamics - upward compression - no GR graph?
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Hehe, or don't go carefuly and use normal compression with range 
-
- KVRist
- 75 posts since 5 Oct, 2005
Jep, indeed, normal compression is easier to handle. But as said, upward can be used as a nice sound design tool. I've experimented today with several sources.
On the one hand you're totally right, Vojtech: With recorded material (like vocals), one has to use a gate or remove the noise floor in pre-processing as it's really loud.
But using it on softsynths or ready made loops is funny (as there's mostly no noticeable noise floor)!
I've also achieved to "reverse" the dynamic of a signal. So the initial loud attack becomes quieter than the initial calm part of a loop for example. Sounds very "artifical"...
And in some cases I perceived the upwarded result was louder than the normal compression result (measured at same levels on the output).
I'll do a video about it the next time and post it here...
Thinking "unusual" is great...isn't that the motto of Vojtech?
On the one hand you're totally right, Vojtech: With recorded material (like vocals), one has to use a gate or remove the noise floor in pre-processing as it's really loud.
But using it on softsynths or ready made loops is funny (as there's mostly no noticeable noise floor)!
I've also achieved to "reverse" the dynamic of a signal. So the initial loud attack becomes quieter than the initial calm part of a loop for example. Sounds very "artifical"...
And in some cases I perceived the upwarded result was louder than the normal compression result (measured at same levels on the output).
I'll do a video about it the next time and post it here...
Thinking "unusual" is great...isn't that the motto of Vojtech?
http://www.wildcafe.com -> artist website | http://www.danaandwild.com -> artist website
http://www.phonicfusion.com -> label website | http://www.bedengler.com -> parent enterprise
http://www.web-kasse.at -> web project
http://www.phonicfusion.com -> label website | http://www.bedengler.com -> parent enterprise
http://www.web-kasse.at -> web project
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Yes, but if you use normal compression with range, you'll have the same effect, but without the need to perform any gating
.
-
- KVRian
- 703 posts since 28 Oct, 2014
theres a way around this, and this is how i use multiband dynamics.wildcafe wrote:With recorded material (like vocals), one has to use a gate or remove the noise floor in pre-processing as it's really loud.
instead of raising the bottom left node, you leave that bottom node at the bottom, and create another node further up at the point you want the upward comp to stop compressing.
then you simply raise that curve shape handle to about 40 - 60%.
this will give the compression, and still keep the noise floor at zero.
the beauty of this technique (which you cant do the range way), is the simplicity of hearing the compressed vs uncompressed areas, when setting up.
so to try this, leave the bottom node at the bottom, then create a 2nd node half way up, or whatever (this node marks where no upward comping will happen).
now create another node just to the left of the half way node (node 3), and drag N2 right to the bottom. you have now created a gate. so move N2 and N3 around, until you can only hear the areas you dont want compressed (the transients).
when you are happy that N3 is set at the area you want compression to stop, delete N2, so the gate is now removed and no compression is taking place.
now raise the curve handle between N1 and N3 by 40 - 60%, and you will have a nice upward comp, that doesnt raise the noise floor.
you can also change the curve to freestyle, so you can get a more suitable shape of curve.
i also asked vojtech for th OPs request some years ago, by email, but it seems like he's not happy with people using his comps in that way, which seems strange.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Hmmm, could you post a screenshot of this? I'm actually quite lost in it
.
Anyways folks, I cannot repeat strongly enough - upwards compression is generally wrong and there's a trivial way to make it happen without the flaws - just use normal compression and range. That's all! And if you worry about the changed loudness, enable AGC or maximization.
Anyways folks, I cannot repeat strongly enough - upwards compression is generally wrong and there's a trivial way to make it happen without the flaws - just use normal compression and range. That's all! And if you worry about the changed loudness, enable AGC or maximization.
-
- KVRian
- 703 posts since 28 Oct, 2014
here we go.MeldaProduction wrote:Hmmm, could you post a screenshot of this? I'm actually quite lost in it.
its a simple concept, but its the easiest and most reliable way i've found for using multiband.
here's picture A. i add one node somewhere in the middle: ________
pic B, create a node just to the left of mid node, and drag it down to create a gate. now move those two mid nodes left and right until you can only hear the audio you dont want compressed (the transients): ________
pic C, once point is set, delete that lower node to remove gate: ________
pic D, raise curve handle to compress upward without removing punch, or raising noise floor
You do not have the required permissions to view the files attached to this post.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Aaaaaha! Calling it "multiband" is a little problematic - it would mean using a MMultiBandDynamics, hence splitting the signal into multiple frequency bands. This is more like multiple segments.
-
- KVRian
- 703 posts since 28 Oct, 2014
well yes i was using multiband, and i've only just noticed that the OP isnt, but technique works on all melda comps!
but yes, i shouldnt have been using the word multiband in that post
but yes, i shouldnt have been using the word multiband in that post
