Question re MTurboReverb

[dmbaer]

My question has to do with the Input and Output parameters in the Late panels.

Let’s say you use three Late generators enabled. Input to 1 and 2 is 100% dry, 0% prev and outputs are 0% out, 100% prev. If input to 3 is 0% dry, 100% prev, then the signal it gets is an equal mix of the outputs of 1 and 2, correct?

Now, let’s say (just for purposes of understanding the internal workings) we had 1 and 2 with inputs as before (100% dry, 0% prev). Output of 1 is 0% out, 100% prev. Output of 2 is 100% out, 0% prev. Now, input to 3 will be just the output of 1. Is that right? In other words, by sending all of 2’s output to out, we don’t break the chain of prev content leaving silence as input to 3, do we?

Lastly, this scheme would have the input to 1 from prev be silence in all cases. Again, is this the way it works?

Thanks in advance for enlightenment.

[MeldaProduction]

My question has to do with the Input and Output parameters in the Late panels.

Let’s say you use three Late generators enabled. Input to 1 and 2 is 100% dry, 0% prev and outputs are 0% out, 100% prev. If input to 3 is 0% dry, 100% prev, then the signal it gets is an equal mix of the outputs of 1 and 2, correct?

Not at all. You set the output of the first 2 as "0% out" which means 0% of it goes out . So the first 2 do nothing. Since there's some confusion, I won't analyze the rest of it.

Anyways: the input and output parameters are completely independent. The structure of the plugin is:

in ->ER1 -> ER2 ->.. -> LR1 -> LR2 ->... out

And every stage can mix the dry input into it as well and can control how much of its own output gets out and how much of it's input gets out intact.

So:
Input=100% prev 0% dry & Output = 100% out 0% prev => this module is executed serially, in other words it is processing the output of the previous module and replaces it completely for the next stages

Input=0% prev 100% dry & Output = 100% out 100% prev => this module is executed in parallel, in other words, it is processing only the dry signal, doesn't care about output from the previous module, and it mixes its output with the output from the previous module

[DarkStar]

<< too slow, but I've typed / painted it now >>

I understand it differently (from the OP). I have put together a signal path schematic (which may be completely wrong )



In your first example:
-- LR1 gets the dry input and its output is not used
-- LR2 gets the dry input and its output is not used
-- LR3 gets 100% Previous, which is the original (dry) signal (as it is not affected by LR1 or LR2)

To get an equal mix of LR1 and LR2 only into LR3 I think I would use:
-- LR1 in: 0% Prev, 100% In (presumably Previous for LR1 is silence, or is it the output from the last ER?))
-- LR1 out: 100% Out, 0% Prev
-- LR2 in: 0% Prev, 100% In
-- LR2 out: 100% Out, 100% Prev (to mix LR1 and LR2)
-- LR3 in: 100% Prev

I have illustrated them here:



In your second case, the input to LR3 is just that from LR2. I do not think that the output from LR1 only can be passed into LR3.

[spencerlee]

I have to say the signal flow makes no sense to me. I cannot make everything run in parallel with the correct levels.
Why can't we have an internal mix bus and an input bus.

Then you either select the input from the input bus (parallel) or from the previous module (serial)
If you are running serial the output to the mix bus would be turned off.
If you are running parallel you the send the output to the mix bus.

So If you want to run everything in parallel, which you normally would, you just select the input from the input bus and send the output to the mix bus. Then you could get the correct levels. Then the early reflections don't have to go through the late reflections.

Very simple and easy to understand. The only downside is that if you are using all modules you will have to watch the output level to the mix bus. But we are engineers and musicians, so whats the problem.

I have spent a week trying to make sense of this. I give up I am beaten. I even redesigned the easy window to stop the 7 clicks to get anything done. Picture below. But we cannot store the text for the algorithm so my presets wont work.
Spencer

Reverb-Additions.jpg

[MeldaProduction]

DarkStar: Nice images!! Exactly!!

Spencer: Sorry, but you tend to sort of overcomplicate things . In most cases one needs only one ER and one LR, so there's nothing to do. If you need more, it all works just fine and is done the way it is for optimization purposes (and simplicity). I won't comment on the algorithm anymore, you know how it is...

[DarkStar]

In that case, let me try a couple more :

(Edited, see the discussion below):

[MeldaProduction]

The only thing I don't understand is the 50 in bottom right

[DarkStar]

I wondered about that too.

My logic was:
-- the Out Mix for LR2 is set to 100% Prev and 100%Out, mixing LR1 and L2 equally (50% each),
-- so I thought that in the Out Mix for LR3 should be 100% Prev (i.e. 50% LR1 and 50% LR2) and 50% Out too
-- if not then should it be 100% ?
-- or, perhaps, 3 in parallel is nonsense

[dmbaer]

And every stage can mix the dry input into it as well and can control how much of its own output gets out and how much of it's input gets out intact.

So:
Input=100% prev 0% dry & Output = 100% out 0% prev => this module is executed serially, in other words it is processing the output of the previous module and replaces it completely for the next stages

Input=0% prev 100% dry & Output = 100% out 100% prev => this module is executed in parallel, in other words, it is processing only the dry signal, doesn't care about output from the previous module, and it mixes its output with the output from the previous module

So, if I may ask in another way ...

The Prev output value says how much of the input to the Late module is to be passed out to the Prev bus and none of the modules processed sound goes out on Prev? Then if Prev output is 0, there's nothing on the Prev bus for the next consecutive Late module at all?

If I (now ) understand things, then the following hypothetical configuration is not possible: Late 1 output gets passed to two parallel late modules that then send their output to the final output and none of Late 1 output goes directly to final output. In other words, parallel operation is only possible when all the modules are "up front".

By the way, thank you DarkStar for all the work you put into diagramming your take on this. I expect I'm not the only one here who will benefit from it.

[jmg8]

I wondered about that too.

My logic was:
-- the Out Mix for LR2 is set to 100% Prev and 100%Out, mixing LR1 and L2 equally (50% each),
-- so I thought that in the Out Mix for LR3 should be 100% Prev (i.e. 50% LR1 and 50% LR2) and 50% Out too
-- if not then should it be 100% ?
-- or, perhaps, 3 in parallel is nonsense

Actually to get parallel, LR3 should be 66% and LR4 on 50%

LR3 would be made up of 3 signals LR1, LR2 and LR3. To split them evenly they must each take a third. So 66% previous is a combo of LR1 and LR2 (33% and 33%) then last 3rd is from LR3.

Same logic for LR4 but with quarters instead.

[DarkStar]

Almost ...

If you move the Out control from its minimum value to its maximum, you'll see that it starts at
0% Out, 100% Previous and goes through 50% Out, 100% Previous, to 100% Out, 100% Previous,
to 100% Out, 50% Previous and ends at to 100% Out, 0% Previous
so there is always a signal present. If the Prev part of the Output control is 0%, then the Out part will be 100%.

I don't think that your case above is possible either - as far as I can see, parallel LRs can only work with the Dry signal.

You're welcome - I put the diagrams together so that I could understand and I hope they help others too.

[DarkStar]

I wondered about that too.

My logic was:
-- the Out Mix for LR2 is set to 100% Prev and 100%Out, mixing LR1 and L2 equally (50% each),
-- so I thought that in the Out Mix for LR3 should be 100% Prev (i.e. 50% LR1 and 50% LR2) and 50% Out too
-- if not then should it be 100% ?
-- or, perhaps, 3 in parallel is nonsense

Actually to get parallel, LR3 should be 66% and LR4 on 50%

You may well be right - could you explain your thinking?

[jmg8]

I edited my post.

[MeldaProduction]

Ha, what??? If you want all stuff in parallel (as many as you want), just set input to 100% dry and output to 100% out, that's all, no black magic there

[jmg8]

But In LR3 if 100% prev is made up of LR1 and LR2 surely they only use 50% each? So LR3 with 100% out would be twice as loud?