Advanced reverbing.

Phryq

AlbertMcKay.com
I've been experimenting a lot with revert lately (as in the past 3 years).

I'm going firstly for good sounding, and secondly for realism (realism is important, but goodness moreso). Here's something I'm working on. Does it sound real?

Different ERs (All Teldex, but 5m, 7m, 10m)
https://www.dropbox.com/s/ro7q8wnyojwcic3/Strings Test 84 Part1 v3.mp3?dl=0

All 7m ERs (but positioned differently).
https://www.dropbox.com/s/n14ona9k0z6f16k/Strings Test 84 Part1 v4.mp3?dl=0

All 7m ERs with a Tail
https://www.dropbox.com/s/1l17fdnvocz6v8w/Strings Test 84 Part1 v5 Tail.mp3?dl=0

I know the instruments aren't panned traditionally; I though they sound better with this panning, but maybe not... In this recording, I use a separate ER on each track, and Altiverb's spacial positioner for ERs. Then I send everything to an LR, and put adaptiverb before altiverb as the ER. I use adaptiverb to 'filter' unharmonic sounds, but don't actually use it's reverb, as I like Altiverb better. I'm wondering if the tail is too short.

Also, what your thoughts are on

1 - having separate ER and LR sends.
2 - Sending the ER send to the LR.
3 - EQing ER, and LR (especially high pass and low pass)
4 - Should ER be 100% wet?
5 - Spacial positioning within a verb. Altiverb has it's own positioner, and it's recommended to only use it 100% wet? What is the consequence of being less than 100% wet? I was thinking to send a mono signal to the positioner, but have my own separately mixed stereo dry-signal. Separate sends for ER, LR, and EQs.

I just got Altiverb, and I should say... it's much better than the algos I've been using... maybe I just don't know how to use them.

Chris Hein's violin has some cool "body reverbs", short IRs that change the instrument's tone. I've been trying to simulate such a thing using B2, but it's not so effective.
 
Last edited:
OP
Phryq

Phryq

AlbertMcKay.com
Right, by late reflections I just mean Tail.

Usually I high-pass and low-pass the send with the Tail, I do so to the ER, not not as much.

I'm wondering if others EQ their ER send, or just insert the ER onto each track individually, and if they're EQing ERs, how so?

One problem I find is that close-mics pick up a lot of noise, especially string bow-noise, which sounds bad in a reverb. Some you can get rid of with EQ (I love that Embertone has a knob to remove it). Adaptiverb has some ability to get rid of it as well, so I put that before the Tail, but what about ER?

I'm afraid of losing realism by tampering too much with the ER.

I'm also thinking about modulation; I read an article once "modulated reverb tails don't actually occur naturally", but in my piano recordings, I can hear lots of movement; maybe my mics just weren't steady... but I've never really enjoyed the sounds of algorithmic reverb modulation... again maybe I can't tweak it properly.
 

muk

Senior Member
In my opinion you are making things much more complex than they need to be, and in a field where there is little to be gained from that. I find reverb to be one of the more firgiving parts of a mockup, and it's significance pales completely in comparison to the programming of the parts. If your midi programming is excellent, even a much too dry mockup will sound good (way too much reverb can still kill it, but that's easily avoided). But if the programming is sloppy or bad, even with perfect reverb it will sound ugly.

In my opinion you would be best served with a simple approach to reverb. Pan your instruments manually (I read that you have to be careful with Altiverbs stage placement. But I don't have Altiverb so I can't comment). Then use something like this free tool to create depth:

http://www.tokyodawn.net/proximity/

If you want you can do that manually too, but it's a bit more complicated than panning, and Proximity works well. Finally, create three group tracks 'front', middle', 'rear'. Use any reverb you like, on 'front' set the least amount of tail, and on 'rear' the most. Route your instruments to the according group, and you're all set. If you are using libraries with differing levels of reverb baked in, use any ER you like to bring the drier ones closer to the wet ones. Doesn't need to be matched perfectly, just approximate it by ear.
If you want you can set another instance of reverb on the mixbus, with just a bit of tail (be careful not to make the mix too wet).

You should be able to create such a setup in a reasonable amount of time, and it will work just fine. After that, focus heavily on you programming skills, as that's by very far the single most important aspect of any mockup.
 

Joram

Active Member
Right, by late reflections I just mean Tail.
Usually I high-pass and low-pass the send with the Tail, I do so to the ER, not not as much.
Why not? The sound of the early reflection quite some influence on the "tone" of the room.
 

JohnG

Senior Member
unless you are using very dry samples (I don't know Chris Hein's library), I would skip early reflections completely and use convo on the stereo mix for a homogenous string piece like this. And @Phryq -- your panning is a bit extreme.
 

KEnK

Senior Member
wrote this response yesterday but forgot to send it.
Having read what other people are posting,
I'd like to say that I've also spent a great deal of time exploring how my various devices work-
reverb schema, compressor arrays, the subtle difference between this and that type of saturation, etc.
Sure, one can say "you're making something simple far to complex",
but the end result has been I know my tools and can get exactly what I want when I need it.
In my view, worth the effort.
well, her ya go.

Pretty close to "real" but a string quartet (or trio) is really hard to get over.
The middle voice seems a tad lower than a real performer might play it

listened to all 3 versions
v4 (All 7m ERs (but positioned differently)
has the greatest sense of depth front to back) (just my opinion)
Also, what your thoughts are on

1 - having separate ER and LR sends.
2 - Sending the ER send to the LR.
3 - EQing ER, and LR (especially high pass and low pass)
4 - Should ER be 100% wet?
5 - Spacial positioning within a verb. Altiverb has it's own positioner, and it's recommended to only use it 100% wet? What is the consequence of being less than 100% wet? I was thinking to send a mono signal to the positioner, but have my own separately mixed stereo dry-signal. Separate sends for ER, LR, and EQs.
No personal experience w/ Altiverb-
but I do 1. through 4. about 95% of the time.
Seems like those have become all "standard practice" these days, but exceptions are often viable.
Mono reverbs can be very cool- I don't use them in a orchestral stage setting though.
I like them in rock, funk or weird pop.

k
 

Paul T McGraw

Senior Member
@Phryq I also am struggling with trying to learn basic mixing and engineering. I often have a hard time trusting my own ears and second guessing myself. I can not open your third example. But the second sounds much better than the first.

To answer your question, yes, it sounds very close to "real". The sound of the instruments is good, but to my ears the mix is not perfect. But I am ignorant of exactly how to fix it. The instruments lack body, especially the viola and cello, yet later in the clip, the bass starts to sound boomy, so I don't think you can just boost the lower mid range. I am guessing that better Early Reflections are needed, but that is just a guess. Nice quartet writing.
 
OP
Phryq

Phryq

AlbertMcKay.com
The place where it gets boomy, that's actually where the 'bass' comes in; before that it's just cello.

I think the last version with the tail sounds best now, listening with fresh ears, but maybe the bass should be totally dry, or without any ER.

I do do some 'extreme panning'. I like it, but... other people have commented. Does that panning sound *bad* or just unusual? Because if it sounds bad, I will try to realign my taste with the rest of the world.
 

Paul T McGraw

Senior Member
The place where it gets boomy, that's actually where the 'bass' comes in; before that it's just cello.

I think the last version with the tail sounds best now, listening with fresh ears, but maybe the bass should be totally dry, or without any ER.

I do do some 'extreme panning'. I like it, but... other people have commented. Does that panning sound *bad* or just unusual? Because if it sounds bad, I will try to realign my taste with the rest of the world.
I was able to open your third mix this morning. Possibly the difficulty was mine? Anyway, it does indeed sound the best of the three. The instruments sound very, very believable. The bow noise is extremely convincing. I would certainly never have guessed this was VI's if I just heard this on the radio. Perhaps that is a failing on my part.

One thing we have to remember is the extreme variety of listening environments of our internet audience. Some are using headphones, from cheap to pro grade. Some are using cheap desktop speakers, some premium speakers, and a few have a pro studio setup with perfectly flat monitors.

Your third mix sounds really good to me. I listened on both HD600 headphones and on audioengine 5+ speakers. In the final analysis all we can do is listen for ourselves on several different setups, then after getting some feedback, trust our own judgement.
 

Paul T McGraw

Senior Member
This new mix is my favorite of the ones you have posted. It sounds as if the listener is sitting in a chair in the midst of the players, within say 4 or 5 feet of them. If that was your goal, you did great!
 
OP
Phryq

Phryq

AlbertMcKay.com
Yes, that was exactly my goal :)

I'm so happy. But can't stop obsessively tweaking... so I never finish anything.
 

JohnG

Senior Member
I do do some 'extreme panning'. I like it, but... other people have commented. Does that panning sound *bad* or just unusual? Because if it sounds bad, I will try to realign my taste with the rest of the world.
Do what you like!!

I find it distracting, but that's my take. Don't listen to anyone if you like it.