1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Overkill with 7 reverb sends?

Discussion in 'Mixing, Post-Production, and Effects' started by kimarnesen, Jun 12, 2018.

  1. kimarnesen

    kimarnesen Composer

    859
    449
    Feb 12, 2014
    as I have both a lot of Spitfire stuff and dry samples I first send all the dry ones to a short convolution reverb to give them
    some pre-delay and some space to match the Spitfire samples. This is 3 different sends with different amount of pre-delay (close, mid, far).

    Then these + Spitfire samples are all sent to one convolution and one algorithmic reverb to place all of them in the same room. Except some drums that are sent to a studio reverb, and as an effect I sometimes send a track to a cinematic very long tail reverb.

    The main question is if the first step is really necessary, if the early reflection/pre-delay on the dry samples makes a difference and especially if 3 different ones are necessary.
     
  2. Gerhard Westphalen

    Gerhard Westphalen Scoring Mixer

    I'll normally put something extra on dry tracks but I always do that as an insert so that it gets sent with it. A normal reverb will behave very differently if it receives a dry signal rather than a signal that's already "placed" so it's more difficult to get good results.
     
    nas, aaronventure, JohnG and 3 others like this.
  3. ryanstrong

    ryanstrong Senior Member

    1,630
    274
    Oct 15, 2011
    +1
     
  4. OP
    OP
    kimarnesen

    kimarnesen Composer

    859
    449
    Feb 12, 2014
    That makes sense, I'm gonna try that.
     
  5. Puzzlefactory

    Puzzlefactory Senior Member

    1,122
    730
    Nov 22, 2016
    I use two. One algorithmic and one convolution and then blend the two (although sometimes I don’t bother with the convolution at all).

    If I were in a position where I had to send stems then I would have 14. One algorithmic and one convolution for each of my groups, Strings, brass, woodwinds, percussion, sound design, vocals and keys.

    I personally don’t bother with separate reverbs for long and short articulations.
     
    kimarnesen likes this.
  6. OP
    OP
    kimarnesen

    kimarnesen Composer

    859
    449
    Feb 12, 2014
    Do you bother with an extra reverb for early reflections on dry samples?
     
  7. Puzzlefactory

    Puzzlefactory Senior Member

    1,122
    730
    Nov 22, 2016
    No.
     
  8. storyteller

    storyteller Senior Member

    630
    554
    Dec 26, 2015
    I suppose this is going to get into how sensitive your ears are. I've gone down numerous paths, but the paths have collided into the example below. While it is much more complex than just slapping a surround reverb on it, I prefer these results the best (however, I do take shortcuts from time to time).

    On each DRY section (e.g. Violins 1, Violins 2, Violas, Cellos, Basses, etc), I place an algorithmic insert (Phoenix Verb fyi) that is purely pre-delay (no tails) to mimic the section placement in the virtual hall I am creating. Blend this to taste and try not to make it too obvious. This is more like a sweetener for placement. Step 2 is to send each section (with the in-line pre-delay included) to separate IRs for Stage, Tree, Far/Balcony. The particular IRs I like to use include omni and condenser microphone IRs for each recording position. Each of these IRs are loaded on separate tracks. So that actually makes 6 Stereo Reverb sends. And, with the main fader for each section representing the Dry signal (with pre-delay), that makes 7 verbs per section... When blending, I add one more glue verb to everything to blend all of the sample libraries together.

    Clearly, that takes up a ton of post-processing resources for a full orchestra, but it is my favorite way to recreate a hall at the moment. It sounds complex, but the 6 verb sends plus inline pre-delay just mimic the recording process of a well-recorded, multi-mic library like Spitfire. I also try other methods from time to time, but I constantly return to this setup as having the most accurate reproduction for a virtual hall.
     
    kimarnesen likes this.
  9. JohnG

    JohnG Senior Member

    6,941
    2,674
    Nov 13, 2007
    California
    Whether or not one needs early reflections depends on how dry the "dry" samples are.

    Are you talking about VSL or just close mics from other libraries? Even within Spitfire some of the libraries are much drier than others.
     
  10. OP
    OP
    kimarnesen

    kimarnesen Composer

    859
    449
    Feb 12, 2014
    You are right, very few of my samples are completely dry so I don’t think 3 ER-sends
    is necessary after all.
     
  11. JohnG

    JohnG Senior Member

    6,941
    2,674
    Nov 13, 2007
    California
    If you are trying to recreate a symphonic / large scale / amphitheatre / "big" orchestral and/ or hybrid sound, why not start with samples that already sound like that?

    For so many years I've heard people talk about having "control" over the sound. That's great for a pop record or something highly processed. It's also super cool for stuff like the Dark Knight score, in which the sonic perspective is consciously and deliberately shifted all over the place (still one of the coolest scores in every way for decades, along with its brothers, "Batman Begins" and the third one).

    But man it's a lot of work to do that. And if you start with samples recorded from 3 meters away, not always successful.

    Hall samples already have "real" reverb in them, so you don't need to be the Antonio Stradivari of engineers to make it sound decent. By fiddling with mic positions, you can dispense with some or all of this early reflections stuff that I see so many people getting stuck on. HZ Strings offers about a hundred mic combinations, maybe more, so you can fiddle about to your heart's content.

    Maybe mine is a minority position, but all the really lovely libraries seem to have at least three to five mic positions. Mess around with those maybe, instead of spending time on all this.
     
  12. nas

    nas Senior Member

    311
    99
    Dec 28, 2015
    Jordan
    If you're using dry samples, I think 2-3 reverbs would suffice. 1 ER and 2 different Hall tails to give a little complexity. Other than that for wet samples maybe just 1 or 2 reverb tails on wet samples to add some glue if you're blending sections and different libraries... and don't forget panning (either through your mixer or from the VI, if available)
     
    kimarnesen likes this.
  13. OP
    OP
    kimarnesen

    kimarnesen Composer

    859
    449
    Feb 12, 2014
    Yes, that’s the approach I’m using now.
    But it turned out very few of my samples are completely dry. Even some of the “dry” libraries have some room and early reflections, so they would just need more of the tail reverb.

    I usually pan every instrument a little bit even with libraries recorded “in situ” as I’ve found a little extra panning helps with the overall stereo image.
     
  14. robgb

    robgb I Have Strong Opinions

    1,853
    1,535
    Apr 6, 2016
    I use Hornet Spaces as an insert on the dry instrument track and a send to either a convolution reverb or to Soundtoy's Little Plate, which is (was?) a phenomenal free reverb. Sometimes I'll add a reverb to the instrument in Kontakt using the factory "delayish" preset instead of using Hornet Spaces.
     
  15. nas

    nas Senior Member

    311
    99
    Dec 28, 2015
    Jordan
    Yes, I do the same as you.

    You could also consider what overall sound you're going for. A more vintage sound might call for slightly darker tones with the highs rolled off, a more collapsed stereo image, and maybe some tape saturation and a little wow and fluter to give a more vintage vibe. So it really depends on the context and what kind of soundscape you're trying to create.
     
  16. Divico

    Divico Senior Member

    Imo those 3 sends are maybe a bit too much. Id probably settle with one verb to match the dry ones and go on from there. Maybe 3 or 4 verbs for the main groups respectively.

    I was playing around with my template today and came up with this:
    Proximity on group channels (this vst is awesome. Gives you a lot of psychoacoustic effects to create depth. One of them being 2 models of early reflections)
    Each group has its own instance of So. Cal in Spaces
    Glue Verb (trying out Little Plate for this) to add tail and audible reverberation (to taste)
     
    Last edited: Jun 16, 2018

Share This Page