[Can skip this post and go to the one below for what I think the bottom line is … ]
Hello - I am not understanding what/why I’m hearing with REVerence, but I’m pretty sure about what I’m hearing after scores of A/B tests - it seems to be making a vocal track softer. To be honest, I can’t quite tell if softer is the right word … maybe something like “more muffled”, or “not as bright” is more accurate, but it definitely is doing something to the track to make it sound “less” in some manner.
REVerence: As a send effect from a vocal track (100% wet) - LA Studio, predelay 100 msec, Autogain, all others are the default setting.
A/B Test Type #1 - REVerence bypass toggle on/off
As expected, If I put the send at negative infinity, there is no difference in the sound when bypass is toggled.
BUT, if I put the send even at a small amount, -20 for example, REVerence seems to lower the volume of the track (or, as above, something that I’m perceiving similar to that) compared to when it is bypassed.
A/B Test Type #2 - REVerence toggled with Reverb A
I thought maybe what I was hearing was normal reverb behavior (I’m a newb in many ways, this is one of my first “detailed critical listening” activities). So I brought up Reverb A, and found that at subjectively similar reverb levels, there was NO decrease in volume (or whatever it exactly was I was hearing). Here are the Reverb A settings (as an insert): Wet/Dry = 20/100, same 100 msec pre-delay, Hi Cut filter was -3dB, the Low Cut filter was -2dB. the room size was 76, and reverb time was 0.75 sec.
Finally, the vocal track itself is a combination of a dry track with it’s parallel compressed track. After the Reverb it is going to low levels of “Da Tube”, and then Cubase “Maximizer”. Oh, and there’s just a bit of Cubase de-esser on the return from the parallel compression, and on the dry track itself before it sends to the parallel compression.
Can anyone help me understand why REVerence seems to have that “less” or “lower volume” or “more muffled/not as bright” effect, where Reverb A doesn’t?
I’m digging up this old thread because I’m not able to figure out what is going on … can someone help please?
Every reverb I’ve tried in Cubase (Reverence, Room Works, and Reverb A) phase cancels noticeably my with dry track, both subjectively and also measurably (dropping the volume by 2dB or more). Is this just on my system, or is that expected?
What I did:
Dry track is a vocal, and I also tried an 800 Hz sine wave.
Dry Track routed to “Non-Verb” Group routed to FInal Vox Group. This is so I can monitor the volume of the signal routed through the reverb when it is toggled to bypass and back.
Dry Track Sent to a Group with inserts of Reverence/RoomWorks/Reverb A, in turn routed to “Verb” Group, then routed to the same Final Vox Group.
For each inserted reverb, I varied the pre-delay value, pretty much from 0msec to 40 msec. For Reverence it was only the early reflections of “LA Studio” (no tail); also I tried another random reverb in Reverence, and a random one in Room Works.
Here’s what happens:
If I bypass any of the reverbs, I get a signal in Final Vox that peaks up to 2dB higher, and sounds even louder than that, compared to when the reverb is not bypassed. Also, the tonal quality of the Final Vox signal is different than when the reverb is not bypassed. And I have to say, in this very sparse and exposed song, the change in vocal tone is not one I like.
Altering the time of the pre-delay, millisecond by millisecond, does change how much of an effect there is, but it is still very noticeably no matter what value I slide it to.
Is this typical, that volumes drop and the tone changes when adding an early reflection to a dry signal? How can I get around that?
Where do you bypass the reverbs?
By bypassing the insert(s) on the FX channel itself or by bypassing the sends from the source channel (dry track in your case)?
If the first is the case, the behavior is normal, because in fact you twice have an unprocessed signal, which obviously will result in increasing the level.
This kind of stuff confuses me sometimes too. I’ll bypass my send effect, and the signal will be louder.
I guess it is doubling the signal. But as soon as I initiate the plugin the signal is normal. I think even when plugin is not processing, ie. high threshold on comp.
I’m actually bypassing within the plug-in itself. Doubting everything I do at this stage, I also recall the increase isn’t a 6dB one, it’s about 2dB, so I think that makes my recollection correct, but I will check when I get home.
Two other things come to mind, one is that the source channel is actually routed to a channel with a De-esser inserted in it … that De-esser channel has no output routing, but is Sent on to the channel with the reverb inserted in it. Does anyone think that would cause a problem? I will troubleshoot that when I get home too.
The other thing is, I am EQng, I can’t remember if it’s the input to the verb/output from the verb or both. Surely the phase delay associated with EQ isn’t causing problems? I’ll turn off all the EQ just to be sure.
Finally, and here is where I’m getting all turned around: I’m wondering why comb-filtering with associated volume/tone changes with every verb aren’t the expected thing - after all, a source is being combined with a version of itself that has only minor changes (those caused by the reverb engine). I’m just a newbie, but not having read about that anywhere I figure it must be wrong, but can’t figure out why.
That’s what causing an increased level.
When you don’t disable the send to the FX channel on the dry channel, the FX channel still receives a signal from that send. The only difference is, it isn’t processed because you disabled the insert on the FX channel. So it’s identical and will result in an increased level. The same will happen when you set the mix level of the insert to 0%.
Experimenting by treating two or more signals differently by sending it to different busses in fact is a very powerful and popular production technique (parallel processing).
Why you run in unwanted comb filtering issues is hard to say, it depends on a lot of factors (type of VST, type of used filters, amount of VST in the chain etc.)
Thanks, Niles. in this case, though, the processing (i.e., the verb) isn’t affecting the volume of the signal leaving the reverb group channel, I can say that because I am monitoring that volume independently as I toggle the reverb bypass on/off. All the volume decrement and tone changes occur only when the “parallel” signal (with the verb toggled on) is added to the dry voice.
I think I am hearing comb filtering, but mainly I’m wondering how I can use these verbs w/out these additional effects/artifacts.
So when you playback the vocal track with the FX channel muted (not bypassed) it’s louder then when you playback the vocal track with the FX un-muted?
You don’t happen to have the phase flipped somewhere in the FX chain?
Not the FX channel muted, but the plug in bypassed that is an insert in that channel. As near as I can tell, like what mpayne described as his experience, above:
It’s exactly as mpayne describes. The signal is doubled when it’s unprocessed (send on, inserts bypassed) because it’s identical. If you want to check the difference of the FX enabled or disabled, the right way is to disable the send instead of the insert on the FX channel, or mute the FX channel.
When you enable the effect on the FX channel the signal is processed, so most likely it isn’t identical to the source anymore. Measuring the peak levels tells you nothing about the nature of the signal
Comb filtering appears when the same signal arrives at the same point but with a delay - which is what early reflections are - several delays of the original signal. If you hear that in a negative way, you need to change teh level or algorithm or pre-delay…
We’re good, Niles, thanks! Part of the problem is I had the preference on to “Mute Sends” when muting the channel, but after making changes back and forth, I got to hear what I had to - thank you!!
OK, TC, that is the conclusion I (after geologic periods of time of painful experimentation ) came to. I also just found this reference Q. How can I achieve a wider vocal sound? where Mike Senior says the same thing, so you are in good company (sorry, HE is in good company ):
" … there is the potential for this short reverb/ambience/early reflection patch to phase-cancel with the dry sound somewhat, so careful adjustment of the distribution of the earliest reverb reflections can pay dividends."
I’ve worked through this with both your help, and have something I can live with easily, and that is a good start - so thank you!
) came to. I also just found this reference 
):