Monitored input louder than recorded sound

Cubase
#1

I’m confused here… I’ve got a sync’d drum machine playing a pattern and being monitored on a stereo track.
If I flip between live live input and the recorded signal (clicking the orange monitor button) , the live input is about 3db louder. (based on the Supervision metering)
Muting the track completely mutes the signal, so I don’t think there’s any direct monitor stuff going on. (and direct monitoring is not checked anyway)
Interface is RME UFX… so using the RME Total Mix… but I can’t find anything odd there. (and besides, like I said… muting the track in Cubase completely mutes the signal so I think the issue lies in Cubase somehow)

Any advice? Thanks.

#2

Hi,

Is the Constrain Delay Compensation option enabled, by any chance?

#3

No, thanks. Turning Constrain Delay on and off has no change.

I’m trying to think of any other useful info here.

Basically, all I can think of is maybe I’m feeding my Outputs 1-2 from another source by accident. I can’t find this though. (and like I said… turning monitoring on and off immediately changes the level by about 3db.
Like I said, I don’t use direct monitoring in Cubase and I’ve also disabled it in Totalmix. So I can’t imagine that being a factor.

I don’t have any sends enabled.

I’ve gone thru every output channel in Totalmix and made sure there is no other channel feeding Outputs 1-2.

I’ve tried disabling the headphones in the control room and this doesn’t have any effect either.

I don’t know if I added anything useful here, but thanks again.

#4

Hi,

Do you use Monitor bus in the Control Room? Is the Audio Connections > Outputs routed to any Audio Device’s output?

#5

What is the ‘monitor bud’? Here are some screenshots. (Thanks again)

image

image

image

#6

Hi,

Sorry a typo… This should be Monitor bus.

It seems LYD48 is your Monitor bus. The settings seems to be correct on Cubase.

#7

I’ve noticed this over the years as well. Probably some simple bug, like the pan-law isn’t taken into account when monitoring. Should be easy to test …

#8

Hi,

Just for test, could you try in the Cubase Safe Start Mode [Disable preferences], please?

#9

I’m not at my desk… but if there is a way to safestart without wiping my preferences, I’ll do that ASAP.

Fishcorp… that’s super interesting. 3db is the default there isn’t it?

#10

Fishcorp… I don’t have the best grip on panlaw… but panning the channel to one side would be a way to test this right?

#11

… it’s a massive bug! I’m writing up a repro as we speak …

EDIT:
actually - let me rephrase that:
it’s two things: an oversight ~ and a poor design-decision …

#12

Is it a panlaw thing after all?

#13

So … two issues:

When monitoring a mono source through a stereo-bus, the project pan-law is apparently disregarded! Repro for this:

  • Create an empty project
  • Set project pan-law to -6db (just to make the issue more obvious!)
  • Create a mono input bus
  • Put a test-generator vst on this input bus, sine @ 440hz, -12db.
  • Create a stereo audio track and set the input to the mono test bus.
  • Enable record + input monitoring on the audio track!
  • Set up a 4 bar loop and record this …
  • Play back the 4 bar loop while toggeling input monitoring on and off - and observe the -6db difference between the recorded vs. the monitored signal!

This is an easy fix … but there is a problem burried underneath:

It seems that pan-law is applied at the start of the signal-chain, instead of at the panner position in the chain - and this is something i have to examine a bit more thoroughly, before ranting on the internet :slight_smile: …

Best, Rune (aka FishCorp)

1 Like
#14

Amazing. Thanks for the effort there. Now that I know I’m not doing anything wrong, I can live with the issue. :slight_smile:

#15

This is quite annoying.

Is there a feature request to have this fixed?