The thing about MPowersynth's Reverb...

[Breeze]

I read a lot of complaints about MPS's algorithmic reverb and I myself have had some difficulty getting good sounds out of it. I took a good look (and listen!) at it today and I think I've figured out a few things that should help. In general I find this reverb better at rooms and reflectivity, but not so much for large artificial space. The Convo reverb fares better but some of the following applies to it as well.

- Right off the bat, most of the time the reverb shouldn't be used inline with source. The Wet/Dry mixer of the Reverb does increase reverb input as it's moved toward "wet" but using it inline doesn't give you the same level of fine control over what goes into the reverb as using it on a separate output channel. Once there you can use another insert to manage what's fed into the reverb. And once there you can now mute the main signal to really hear what the reverb chain is doing without having to alter the input wet/dry balance.

- A particularity of this reverb is that it's particularly "bumpy": you can hear echoing rather than ultra smooth tails. Using two reverbs one after the other can help smooth that out significantly. Three?

- EQ is your friend. You can really shape both the input and the output of the reverb and the Dynamic EQ is handy to dynamically control resonances both before and after the reverb chain.

- Compression at the output of the reverb chain can also help keep things in control and bring out the reverb. Also good for those crazy big EDM reverbs. Compression at the input can also tame crazy transients.

- And finally, the Gain control in the insert at the end of the Reverb chain is your remix fader.

Some observations on the reverb itself:

- The COLOR section seems badly organized; there are six parameters and it would make sense to see them like this:
LOW DAMPENING - HIGH DAMPENING - MODULATION
LOW FREQUENCY - HIGH FREQUENCY - MODULATION FREQ

- While there are subtle differences between the Reverb Modes, they really don't stand out as very different. It might be interesting to include other more dramatically different modes here in the future.

- I'd like to see slower modulation than 0.1 Hz. 1/10th of a second is still pretty fast for a reverb.

As much as I like many of the Melda processors, I have to admit that they're not my first pick for reverbs. And while a lot of VSTi's have reverbs with far less options in in their settings, somehow many have managed to implement something pleasing with very few controls. In MPS, it seems more work is often required to get to the same place, and at the cost of CPU cycles. Sometimes sound trumps flexibility.

[dmbaer]

>slower modulation than 0.1 Hz. 1/10th of a second is still pretty fast for a reverb

For what it's worth, 0.1Hz is a 10 second period, not 1/10th of a second. But, I agree, I might use something even a little longer than that on rare occasions.

[Breeze]

>slower modulation than 0.1 Hz. 1/10th of a second is still pretty fast for a reverb

For what it's worth, 0.1Hz is a 10 second period, not 1/10th of a second. But, I agree, I might use something even a little longer than that on rare occasions.

You're right, of course. It's actually hard to hear what the modulation is doing as the reverb bounces around so much even without it. Without a smoother tail, slower speeds aren't going to be very effective, but it would be useful to have deeper modulation in the time and pitch domains.

[MeldaProduction]

Thanks for the exhaustive info breeze!
Now, first I don't really follow the first thing about dry/wet - could you explain a little more?
About the "bumpy", smoothness... could you post some audio examples? From my personal experiences the reverbs are the most personal of all audio processors , some people just like that and others hate it etc. So I'd like to understand what you do like and what these things mean. If you have time to create a few examples of course .

[Breeze]

Now, first I don't really follow the first thing about dry/wet - could you explain a little more?

Sorry, you're right: looking at it more closely, the wet/dry control does increase the reverb input as it moves away from the dry side. I revised my post as a result.

About the "bumpy", smoothness... could you post some audio examples? From my personal experiences the reverbs are the most personal of all audio processors , some people just like that and others hate it etc. So I'd like to understand what you do like and what these things mean. If you have time to create a few examples of course .

I actually don't have a lot of time these days, but I invite you to listen to the Lexicon Hall, or Valahalla Vintage Verb. These reverbs have smooth tails where the decay isn't sensed as echoes as much as it is in MPS. And before you mention the obvious matter of pricing, to a lesser degree I find this behavior similar in the reverbs of Hive, Sylenth, Spire, and other VSTi's with built-in reverbs.

I haven't been able to create that pleasing smoothness in MPS with a single Reverb in the chain so I suggested a number of ways of getting more out of the current reverb algos. BTW, all of the above also seem to have Fq-dependent behaviors that are flattering to the source. The basic reason for my OP was that I find I have to massage MPS's reverb to get satisfying results whereas this seems much simpler with other VSTi's.

You're right that reverb taste is personal, but judging by the number of people who seem to have issues with MPS's reverb there is perhaps something there that can be improved. Or maybe it's just the misfits complaining...

[ZentralmassivSound]

I am chiming in here with a few examples.

https://dl.dropboxusercontent.com/u/334 ... everbs.mp3

First three are MReverb, fourth one is Valhalla Room. All of the MReverb examples have a rather unpleasant "repetition" or "loop" effect in the last part of the tail (I boosted it very much here with a comp/limiter so one can better hear it). It sounds almost like a bad sampler loop. One stumbles upon this in many presets, and in the treble-rich presets its especially noticeable (example 1). It has a metallic, dead, robotic character.

In the Valhalla example you can hear a smoother tail, it seems to have more "diffusion", it is softer.

I noticed that the "loop" effect in MReverb depends on almost all of the settings but especially size, diffusion, width, widening, absorption. By dampening the highs it generally sounds smoother, but not quite perfect. As others pointed out, it costs some trial and error until you get something more or less smooth, while oftentimes it sounds metallic. Maybe the smoothness could be improved by design?

[Breeze]

Thanks, Z... That's pretty much what I'm talking about. You can also hear it much more distinctly and exaggerated by using these unnatural settings:

- Dry/Wet at 100% Wet
- Early/Late to 100% Late
- PreDelay 0%
- Diffusion and Reverb Diffusion to 0%
- Reverb time over 7000 ms
- High and Low Damping at 0%

[MeldaProduction]

Thanks folks, I'll check it out