Different Mix Laws for Delay and Reverb?

S

Seven2Eleven

Inspired
Hey everyone,

I'm coming from the Helix world and I'm just trying to understand the mix laws for delays and reverbs in Fractal world.

So after reading the wiki and searching the forum, my understanding is that...

  • When running delays in series, the dry level stays constant with the mix between 0-50% and then decreases linearly to zero after. Consequently, there is no need to compensate for dry level with the level knob and can be left at 0 db when running in series.
  • Reverbs on the other hand, don't behave this way. The mix knob actually decreases the dry level from 0-100% and therefore it's necessary to compensate by setting the level to +3 db when running in series. You can avoid this by running in parallel instead and running mix at 100%, use the input gain knob as an "fx send" or the level as an "fx return".

My question is why the different rules for delay and reverbs? They're both time based effects. Why can't they follow the same mix laws? And why 3 db specifically? Doesn't the dry level get lower and lower as raise the mix with reverbs? Wouldn't the higher mix mean you'd have to compensate more and more with the level?

Or maybe I'm misunderstanding something.

Any help would be appreciated. Thanks!
 
They don't follow the same mix law cuz the way the perceived volume changes when summing wet and dry signal is different.
A delay is usually used to produce few repeats that are clearly distinct from the dry signal so the perceived volume of the dry remains the same.
A reverb on the other hand, often has very short pre-delay times and a lot of repeats which sum with the dry signal, so the perceived volume increases (in most cases).

3dB is the boost you need to apply to have unity gain on the dry signal when the mix control is set to 50%
 
DLC86 said:
They don't follow the same mix law cuz the way the perceived volume changes when summing wet and dry signal is different.
A delay is usually used to produce few repeats that are clearly distinct from the dry signal so the perceived volume of the dry remains the same.
A reverb on the other hand, often has very short pre-delay times and a lot of repeats which sum with the dry signal, so the perceived volume increases (in most cases).

3dB is the boost you need to apply to have unity gain on the dry signal when the mix control is set to 50%
Click to expand...
Ok that makes sense I guess as for the difference between delay and reverb. But I'm still confused as to why 3db? Why not just have 50% be unity gain?
 
DLC86 said:
For the reason above 😅
Click to expand...
I guess something still isn't clicking then. You said the "perceived volume" increases with reverbs, but not the actual db levels? And if the perceived volume is increasing, wouldn't you want to decrease the level to compensate, not increase?
 
To further clarify the mix laws see this pic.
IMG_20230620_175614.jpg

Red line is the wet signal level for the delay block, blue line is for the reverb block and black line for other blocks that have a 6dB attenuation at 50% (e.g. MultiDly, plex delay, compressor, etc..).
Dry level is the dotted black line for the 6dB law, for the other two I couldn't draw them cuz I'm on my phone, but imagine them as specular to their "wet" lines counterparts.

All three mix laws try to keep the perceived volume of the summed dry+wet signal constant across the range of the mix knob but, depending on the type of effect, a different mix law is required to achieve that.
All depends on how the dry and wet signals are correlated to each other.
 
PS: if you've ever used a DAW you probably already know what correlation means.

Let's say you have recorded a guitar track, if you duplicate it and play both track simultaneously your output level will increase by 6dB, that's cuz the two signals are identical so every sample gets doubled in level.

Now erase the duplicated track and instead record again by playing the same part once again. This time the two tracks are similar but not identical, so the level increase when playing both will be around 3dB.

Then erase the second track again but this time record a totally different part. When you play back them together you'll probably get 1 or 2 dB increase in the output meters but the perceived volume of the first track will remain the same.

These 3 different mix laws basically try to counteract this phenomenon for each of these scenarios, but mind you that it also depends on the settings you use, this perceived volume compensation doesn't work for every possible setting of the block.
 
DLC86 said:
To further clarify the mix laws see this pic.
View attachment 122294

Red line is the wet signal level for the delay block, blue line is for the reverb block and black line for other blocks that have a 6dB attenuation at 50% (e.g. MultiDly, plex delay, compressor, etc..).
Dry level is the dotted black line for the 6dB law, for the other two I couldn't draw them cuz I'm on my phone, but imagine them as specular to their "wet" lines counterparts.

All three mix laws try to keep the perceived volume of the summed dry+wet signal constant across the range of the mix knob but, depending on the type of effect, a different mix law is required to achieve that.
All depends on how the dry and wet signals are correlated to each other.
Click to expand...
Ok I get that different effects merit different mix laws. I hope not being dense but I'm still trying to understand the need for the plus 3db for reverbs. It sounds like it's to compensate for something intrinsic to the specific mix law that's hard coded into the reverb block. Then why not also hard code +3db as well for the level?
 
Seven2Eleven said:
Ok I get that different effects merit different mix laws. I hope not being dense but I'm still trying to understand the need for the plus 3db for reverbs. It sounds like it's to compensate for something intrinsic to the specific mix law that's hard coded into the reverb block. Then why not also hard code +3db as well for the level?
Click to expand...
It's not a need, it's to be used only if you want to have the dry signal always at the same level, regardless if the reverb block is active or not.

If you leave the output level at 0dB, basically the dry+wet signal will have about the same overall volume as when the block is bypassed, but the dry signal is attenuated by 3dB.

If you set the output level at +3dB the dry signal (with the mix at 50%) will have the same level as when the block is bypassed, but the overall volume (dry+wet) will be higher by 3 dB

PS: I suggest experimenting with this a bit cuz the best setting depends on how you use the block.
First of all, if the bypass mode is set to "mute fx in" all this becomes irrelevant cuz the dry signal is always affected by the mix knob, even when the block is bypassed.
If you use the input gain to set your wet level, the "perceived volume compensation" will be thrown off. And in this case it's best to set the output level at +3dB imho.

And so on..
 
DLC86 said:
It's not a need, it's to be used only if you want to have the dry signal always at the same level, regardless if the reverb block is active or not.

If you leave the output level at 0dB, basically the dry+wet signal will have about the same overall volume as when the block is bypassed, but the dry signal is attenuated by 3dB.

If you set the output level at +3dB the dry signal (with the mix at 50%) will have the same level as when the block is bypassed, but the overall volume (dry+wet) will be higher by 3 dB

PS: I suggest experimenting with this a bit cuz the best setting depends on how you use the block.
First of all, if the bypass mode is set to "mute fx in" all this becomes irrelevant cuz the dry signal is always affected by the mix knob, even when the block is bypassed.
If you use the input gain to set your wet level, the "perceived volume compensation" will be thrown off. And in this case it's best to set the output level at +3dB imho.

And so on..
Click to expand...

Ok that makes things little more clear but I guess what I'm still thinking is that wouldn't most people want the default setting to be that the dry volume is the same as when the effect is bypassed? I get that everyone has different use cases but that's a reasonable assumption I think.

And with all this considered, if I wanted to run both my delays and reverbs in series (I don't mind delays being affected by reverb), and not worry about any level issues when bypassing them, I'd have to set the delay to 0 db and the reverb to +3db, and control the amount of delay and reverb with the mix knob?
 
Seven2Eleven said:
Ok that makes things little more clear but I guess what I'm still thinking is that wouldn't most people want the default setting to be that the dry volume is the same as when the effect is bypassed? I get that everyone has different use cases but that's a reasonable assumption I think.

And with all this considered, if I wanted to run both my delays and reverbs in series (I don't mind delays being affected by reverb), and not worry about any level issues when bypassing them, I'd have to set the delay to 0 db and the reverb to +3db, and control the amount of delay and reverb with the mix knob?
Click to expand...
Running the Reverb in series with +3dB and 50% Mix is the same net result as running it in parallel.

I would guess that most people running Reverb in series do not do this. Level at 0dB and Mix adjusted to taste would be typical, I think.

Don't overthink it... And use your ears to find what sounds good to you.
 
I'd seen this chart before and knew about the +3dB setting. I didn't care enough to investigate, because I rarely go above 20-25% reverb mix anyway. But now that the question has been asked, I'm along for the ride.

My theory was that it was borrowed from the panning laws used built into recording consoles or DAW software. If I remember correctly, the -3dB attenuation in the center made it easy to provide equal output power to the speaker no matter where a signal was panned. It's been a while since I studied pan laws - but I think that's the gist of it.

So I just assumed that Fractal, being the forward thinking geniuses that they are, would have wanted to maintain that equal power principal so they could avoid any possible left/right power differentials as the signal leaves the Fractal and goes into a stereo speaker/amp.

I could easily be wrong. Maybe they just ran out of coffee on the morning they ported the reverbs into the firmware :)
 
You must log in or register to reply here.
Back
Top Bottom