Hi. I was trying to equalize the guitar volume I was getting from direct monitoring vs what I was getting from Jamulus when connected to my local server. With my guitar in the left channel, Audio quality set to High, channel mix volume set to maximum, and Mono audio channels selected (selecting mono-in/stereo-out didn't make a difference), the only way I could achieve this was by moving the Local audio pan slider to the upper-most R-50 position. Now my guitar volume is very close if not identical to the same level I get when direct monitoring. But then I realized, with both guitar and vocal mic plugged in, there is no way to reach maximum volume for each input - I can only get 50% volume for each with Local pan in the middle position. I can get 100% guitar but no vocal, or 100% vocal but no guitar, etc. Is this expected behavior in Jamulus or am doing something wrong?
After messing around some more, the only way I could get maximum signal volume from both guitar and vocal (to match direct monitoring levels) is to run 2 Jamulus sessions simultaneously - one with guitar hard-panned to R-50, and one with vocal hard-panned to L-50. Is this not possible to do within a single Jamulus session?
Any help is appreciated.
Last edit: Al Udell 2020-06-12
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
You are right, in Mono mode I mix both channels together and then attenuate the final signal by 3 dB to avoid getting overloads. This is a very conservative mixing attenuation. I think I should change this. I just created a Github Issue for that: https://github.com/corrados/jamulus/issues/353
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Just thinking about this more Volker, it occured to me that center attenuation affects people using only one mono input. So when using one instrument in left chanel and nothing in right (or vise-versa), and using center pan, the output is by default attenuated 3dB - so users are not getting the full volume of their signal, and are forced to increase their gain - possibly into distortion, to compensate.
I suspect most users like me just use one input for instrument and one for vocal. When not singing, users tend to just use a mic to verbally communicate before, between, after songs - not during. In that case, center attenuation is reducing their instrument signal while jamming. Even when singing and playing an instrument, I suspect center attenuation may not be necessary? But I'm no expert in dB gain or overload.
Perhaps giving users control over pan rules would be appropriate - e.g. 0, 3, 4.5, 6 dB center attenuation, etc? Or doing no center attenuation and providing an output overload indicator?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I think I was wrong, it is 6 db loss in power (3 dB in amplitude). Anyway, what I do is kind of cross-fade between the channels. The advantages of that are:
1. If you have the same signal on both channels and move the Pan, the level stays constant.
2. You can be sure that the signal is never overloaded.
So I think I should keep the Pan implementation. What would help you would maybe some kind of an input boost. If there would be a 6 dB power boost and you put the Pan in the middle, you would have the two of your input signals with the original level. What do you think?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
No worries Volker. I wouldn't want you to change the pan rule just to accomodate me. At least I know now that center pan equals 3dB attenuation. My only concern, if you decide to stay with the current rule, is that many mono users may not realize they're not getting full signal volume from their 2 inputs when using center pan.
I've also noticed that when using the -R switch to record sessions, the center pan attenuation also affects recorded mono signal strength. I compared mono guitar signal strength while recording in Reaper. First using direct input from my USB audio interface, then second by using Jamulus with center pan and another at R-50 pan. The R-50 pan signal matched the direct input signal strength exactly. So, just to expand on my point, many users may not be aware that they're not getting full signal volume in their session recordings.
I think I'll continue to run 2 Jamulus sessions simultaneously (hard-panned R-50 for guitar, hard-panned L-50 for vocal). That way I'll get full signal with no attenuation. Besides, I really like having separate guitar and vocal faders in the mixer window.
Last edit: Al Udell 2020-06-14
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Some minor requests regarding pan: 1. Can we have a centre detent (so that anything under +/-3% or so goes to zero pan? 2. Can the knob be made larger? Thanks for your consideration, Volker.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi, I'm referring to the round knobs above each musician's fader. Would be nice if they are larger, and go to centre if the setting is close enough to centre.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I also mentioned that previously: https://github.com/corrados/jamulus/pull/145#discussion_r415902476 but tarmoj the developer of that feature said: "In practice I don't think it is very crucial, if the pan is set to the middle by hand more or less, it sounds in center enough.". I think I agree to him in this regard.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi. I was trying to equalize the guitar volume I was getting from direct monitoring vs what I was getting from Jamulus when connected to my local server. With my guitar in the left channel, Audio quality set to High, channel mix volume set to maximum, and Mono audio channels selected (selecting mono-in/stereo-out didn't make a difference), the only way I could achieve this was by moving the Local audio pan slider to the upper-most R-50 position. Now my guitar volume is very close if not identical to the same level I get when direct monitoring. But then I realized, with both guitar and vocal mic plugged in, there is no way to reach maximum volume for each input - I can only get 50% volume for each with Local pan in the middle position. I can get 100% guitar but no vocal, or 100% vocal but no guitar, etc. Is this expected behavior in Jamulus or am doing something wrong?
After messing around some more, the only way I could get maximum signal volume from both guitar and vocal (to match direct monitoring levels) is to run 2 Jamulus sessions simultaneously - one with guitar hard-panned to R-50, and one with vocal hard-panned to L-50. Is this not possible to do within a single Jamulus session?
Any help is appreciated.
Last edit: Al Udell 2020-06-12
You are right, in Mono mode I mix both channels together and then attenuate the final signal by 3 dB to avoid getting overloads. This is a very conservative mixing attenuation. I think I should change this. I just created a Github Issue for that: https://github.com/corrados/jamulus/issues/353
Thanks for doing that Volker.
Just thinking about this more Volker, it occured to me that center attenuation affects people using only one mono input. So when using one instrument in left chanel and nothing in right (or vise-versa), and using center pan, the output is by default attenuated 3dB - so users are not getting the full volume of their signal, and are forced to increase their gain - possibly into distortion, to compensate.
I suspect most users like me just use one input for instrument and one for vocal. When not singing, users tend to just use a mic to verbally communicate before, between, after songs - not during. In that case, center attenuation is reducing their instrument signal while jamming. Even when singing and playing an instrument, I suspect center attenuation may not be necessary? But I'm no expert in dB gain or overload.
Perhaps giving users control over pan rules would be appropriate - e.g. 0, 3, 4.5, 6 dB center attenuation, etc? Or doing no center attenuation and providing an output overload indicator?
I think I was wrong, it is 6 db loss in power (3 dB in amplitude). Anyway, what I do is kind of cross-fade between the channels. The advantages of that are:
1. If you have the same signal on both channels and move the Pan, the level stays constant.
2. You can be sure that the signal is never overloaded.
So I think I should keep the Pan implementation. What would help you would maybe some kind of an input boost. If there would be a 6 dB power boost and you put the Pan in the middle, you would have the two of your input signals with the original level. What do you think?
No worries Volker. I wouldn't want you to change the pan rule just to accomodate me. At least I know now that center pan equals 3dB attenuation. My only concern, if you decide to stay with the current rule, is that many mono users may not realize they're not getting full signal volume from their 2 inputs when using center pan.
I've also noticed that when using the -R switch to record sessions, the center pan attenuation also affects recorded mono signal strength. I compared mono guitar signal strength while recording in Reaper. First using direct input from my USB audio interface, then second by using Jamulus with center pan and another at R-50 pan. The R-50 pan signal matched the direct input signal strength exactly. So, just to expand on my point, many users may not be aware that they're not getting full signal volume in their session recordings.
I think I'll continue to run 2 Jamulus sessions simultaneously (hard-panned R-50 for guitar, hard-panned L-50 for vocal). That way I'll get full signal with no attenuation. Besides, I really like having separate guitar and vocal faders in the mixer window.
Last edit: Al Udell 2020-06-14
Some minor requests regarding pan: 1. Can we have a centre detent (so that anything under +/-3% or so goes to zero pan? 2. Can the knob be made larger? Thanks for your consideration, Volker.
Are you talking about the round Pan know above the fader or the local fader slider?
Hi, I'm referring to the round knobs above each musician's fader. Would be nice if they are larger, and go to centre if the setting is close enough to centre.
I also mentioned that previously: https://github.com/corrados/jamulus/pull/145#discussion_r415902476 but tarmoj the developer of that feature said: "In practice I don't think it is very crucial, if the pan is set to the middle by hand more or less, it sounds in center enough.". I think I agree to him in this regard.
Volker, how about adding a new checkbox in the Main or Settings screen to toggle mono center pan attenuation on/off?
Are you using the Mono-In/Stereo-Out mode?
I have just finished my implementation, see https://github.com/corrados/jamulus/issues/353#issuecomment-646184246
I've been using Mono mode mostly. Using Mono-in/Stereo-out mode makes alot of sense for no attenuation. I thank you sir!