Do Volume handles affect S/N ratio?

I am in a discussion with another engineer whom I sent files to. He wants me to re-bounce the files at a higher volume. He is a ProFools, sorry Pro Tolls user. I suggested he just pull up the volume handle on the file in question, His argument is it will increase the noise floor. My argument is that i get lower volume files sent to me quite often, I import the in and pull up the volume handle. I have never noticed and noise generated in this manner.

Can someone enlighten me in this matter? my feeling is that it is a digital file and by increasing the volume in this manner will not affect the integrity of the files as opposed to boosting the file on the bounce by possibly having to use a compressor/expander or some other method to get the file file amplitude where he wants it.

I know someone out there has the answer and has surely dealt with this before.

Thanks
Michael

just watching

Me too :wink:

PS. The S/N ratio for the file is set in stone once it is recorded (no bouncing up or down in volume can change that).

Now, I’m just watching… and waiting :wink: :wink:

This is a trick question. ANY gain will bring up the noise floor on a digital file, if there is a noise floor in that file to begin with. The question is whether you recorded an analog source, i.e. something with a noise floor - if so, that noise floor is going to come up no matter which method you use. If it is purely virtual instruments, or for example a lot of effects using convolution (which in theory can have noise in the reverb tail etc) - it can get a little more complicated - but still, only very subtly so.

Non-realtime gaining without summing is the simplest DSP operation there is - you are basically adding numbers, and it doesn’t incur distortion (within limits). Heeding your engineer friend’s request to re-bounce it is exactly the same as moving a volume handle on a file. Put differently: in the digital world, adding is adding (assuming no outboard gear or processing that would alter it otherwise!)

Judging by your comment “he wants me to re-bounce the files at a higher volume”, and trying really hard not to go to the “ProTools users are filled with pseudo-scientific insanity” place (lol), his/her request smacks of an analog way of looking at things - you increase the gain of the source material at the first amp stage to get out of the (analog) noise floor when you record/export.

But to reiterate: if the audio is already recorded, i.e. you are working with audio clips in your session, and have no outboard processing, or dither etc, and it is a purely digital signal path, it is a totally moot point: it doesn’t matter whether you, or he, or anyone else raises the gain, and it doesn’t matter at which point.

Hope that helps :slight_smile:

Thanks, those were my thoughts exactly, Yes if I had recorded through an analogue desk to tape and the signal was too low, yes the HISS is going to be increased, but that is not the case. I have never experienced any added noise when increasing the volume handles of files I have been sent. So I basically just opened the track, raised the volume handles so the .wav file ‘looked’ bigger, approx +6dB and exported them. This is a client that has taken her project to another studio that is more convenient and her words “more professional because he doesn’t treat her like a friend but a client”, oh and charges double my rate, so he must be better, right? I’m sure he’s trying to impress her with some BS techno babble…My discount rate just increased if she comes back to my studio

LOLZ!

In my “old age” (lol) I’ve learned a trick: tell them you’re doing the thing they want you to do, while secretly doing the thing you know is the real answer. You then get a happy client AND the professional result :slight_smile: I knew a guy that would pretend to raise the gain on the mixing board during guitar solos, and the guitarist felt like he was tracking louder…in reality the fader never moved lol. So evil :smiling_imp:

Customers - its all about signal to noise ratio

That’s some fancy social engineering going on behind the console!

The ratio will always be the same when just adjusting gain.

We have all had some broken cables or what not, when working with a struggling artist. To have some artificial breaks, distracting the artist away from nerves and tension.
Yes, psychology is a big part of being a producer and or engineer when working with artists.

And yes, the S/N ratio will always be the same no matter how you gain it up or down. Clean gain that is :wink:

thanks to everyone for confirming my beliefs

was that answer a Yes or a No? :slight_smile: I think its a yes. My brain hurts.

Z

it was a tape hiss and movement problem ,it was better to always record sounds much louder than tape noise,i think the S/N ratio is a term that has just stayed with us from the tape and analogue days .no such problems with digital to the point where people now add tape noise. just make sure there are no background noise or hums when recording.

It is still analogue days :wink:

My studio is full of analogue gear, just not tape anymore. S/N ratio is still a valid technical term, and most pro gear are calibrated so that the best S/N ratio correspond with 0dBVU (+4dBU = 1.23Vrms). To be picky, it also consider the “signal to distortion” ratio.
Much misused analogue gear, pre-amps in particular, these days. Due to “bit hogging”. No problem in the digial domain, if not clipping.

But in the analogue signal prior to A/D convertion there are many a pre-amp being pushed to hard out there, adding distortion. It may be ok for a few tracks (like snare drum), but not across all tracks.
So for a better and cleaner mix/recording read up on your analogue equipments specs. At least learn to “respect” the VU Meter and the 0dBVU standard. Often corresponding with the converter calibration set to 0dBVU = -18dBFS (RMS).

So to sum up, the S/N Ratio is still a valid technical term (as the audio has to get in and out of the digital domain to be of any use).

PS. All sound is analogue, no matter how it is stored and/or manipulated. Except for the voices in your head :smiley:

https://www.youtube.com/watch?v=s9seoFdfHYA

If I play this digital recording through my system (specs below) it works fine at low to medium volumes, but breaks up at higher volumes. Other videos play fine at any reasoable volumes. I thought that digital recordings did not suffer in this way? Is it to do with the original recording level

how do you mean breaks up , distorts your speakers ? it`s more likely to do with the frequency that is more prominent, in this case ,it sounds 300 hz heavy .

Perhaps the engineer in question was just not in the mood to do anthing that day, and was just being procrastinative.

Not here. I have the yt vol. slider up all the way, and my room vol. at a decent level - no clipping/distortion/breaking up here. Play the vid, and look at the meters on your sound card. Are they in the red?

No, it does not go into red. On closer listening its only the voice over, and not the original recording, that break up
Must be the level of the original mic recording I think
Z

Signal to Noise ratio in an analog circuit is a measure of the constant noise generated by an analog circuit over the actual audio signal.

In the digital realm no such issue exists. “Noise” can be generated on purpose by plugins. Various types of distortion can result from sample rate conversion and other processes and be interpreted as noise. But this is not generally a function of level. And “raising the volume handles” does nothing to the S/N ratio in a digital file. Just makes everything louder.

HOWEVER: The lower the level of a digital audio file (below 0DBFS) the less number of bits are being used to represent that audio signal. This is overcome somewhat using floating point math (like Cubases’ internal processes). But .wav files are not floating point and the lower the level the lower the resolution of that file. A 24 bit .wav file at a low enough level may actually be USING only 8 bits. Raising up the volume handles of such a file DOES NOT increase the resolution.

So . . though I wouldn’t call the issue “signal to noise”, there IS a reason to output digital files at as high a level as possible while still maintaining the desired dynamic range and conforming to any standards.

Hugh