Adjusting monitoring

[jonetsu]

Due to some events I used headphones for some time. Now I'm back with the HS5s and the HS8 subwoofer and now have proper monitor stands instead of having the HS propped up with books on the desk. I listened to Bitwig sketches I have, to Mixbus mixes and to Soundcloud uploads. The SC uploads seems to have lost a part of the bass frequency range. When the same mix is played in Mixbus32C there's more bass. Both sounds good in the headphones though.

That leaves me with a great puzzlement about what people actually hear of my music out there. So the question is, how do you calibrate the monitoring speakers so that they yield a 'best' sound that represents a fair standard of what people, using fair audio systems, can hear ? Each speaker has a volume control and the subwoofer has cutoffs for itself and for the sends to the two other speakers. I do not know at which level the subwoofer should be set for instance so that it reflects a bass that 'makes sense'.

Also, what I currently do is to have the wav file from the final Mixbus stereo export converted to ogg then uploaded to SC. In that conversion and in the processing done by SC (if any, looks like they're not doing much), could there be a loss in frequencies in the bass range ?

Cheers.

[Matt]

Not sure why you're losing bass, but setting up monitors is more about preference and learning them. I have a JBL monitor controller that has a set up utility for smoothing out the frequency response (MSC-1 maybe?). It works but you still need to listen and compare to your favorite best sounding commercial music as much as possible. Until you can just "feel" when it's right. And even then don't trust yourself and compare often.

[jonetsu]

Right. I forgot about using commercial releases as I was looking for a technical way to assure a 'standard' while it's a question of how it feels. Following this it'd be probably a good thing to 'calibrate' according to a similar commercial piece (or several) when beginning on a new mix. Eventually, to create a 'signature', the calibration would be done using previous original pieces, if those were done right in the first place.

[jonetsu]

When playing commercial music to calibrate the monitoring speakers, is playing a compressed format (ogg, mp3) not ideal for frequency response ?

[Matt]

Anything is better than nothing. But I would try and use lossless files if I could. It's not as much about frequency response, as it is about detail and depth. Dynamics is also at play.

My other best tip, that's taken me a long time to get my head around is, mix the tracks to each other. Don't worry as much about how a track sounds soloed, but more about how it compares to the other tracks. For example your listening to the vocal and the guitars. One sound more up front then the other. And you're going for a more laid back sound. So cut around 1-2khz on the track that's more forward. Or wherever in the frequency spectrum one is in front of the other. Keep in mind the overall tonality of the mix, but compare and contrast to the tracks in the mix. Not just how one particular track sounds. The one possible exception is "the main track" of the song. Normally the lead vocal. Make it sound like you want it to sound, and then balance the other tracks around it.

Once you get the frequencies balanced between the tracks, then you can use overall EQ to help with any deficiencies caused by less than perfect monitoring. Because no matter what you do, there isn't a perfect "standard". Everything is a compromise. So you've got to learn and adapt to the anomalies of your monitoring situation.

[English Guy]

What is ironic is that you are mixing on good gear but the recipient at the other end may not. It does pay to listen on other devices, headphones etc.