So what level should I be recording at???

[spong]

I tend to record as hot as I can get without clipping. However, I have read some threads that suggest that I should leave more headroom so the mix buss doesn't get overloaded.

So what level should I be aiming at for individual tracks?

I am using HD3/192 with the dithered mixer.

[DJS]

spong,

Not knowing what sort of music you record or the methods you use, I won't say 'do this or that,' but I'll tell you how I do it.

My experience is in recording 'acoustic' ensembles, mostly jazz and classical. I never use compression when I track, nor do I rely on close-micing very much. Most of my 'sound' comes from a stereo pair (spaced omni's, X/Y of some flavor, M/S, and more recently ribbons (fig. 8)), and the rest from spot mics on individual players or sections.

When I set levels, I do so when the ensemble is rehearsing the loudest part of the piece, so that the peaks are around -2 db VU. My converters are calibrated to 0 db VU = -18 dbFS. Recording 24-bit audio at this calibration level, I've never had a digital clip, though it's come close a few times. Since I rarely use more than 12 tracks on a project, I have no issues with PT mixer configurations that share DSP chips, which seems to be one of the general complaints about the PT mixer architecture.

I'm really not qualified to comment on how much level should hit plug-ins because I use them so infrequently, and even then only for "touch-up" (HP flitering, etc.) on spot mics within larger jazz ensembles.

The sort of music I record demands an accurate representation of dynamic range, so I don't know if my experience is all that relevant to this topic, but I'm happy to add my $.02.

Doug

[masmit]

Spong, my approach is like this. I have my ADs set to 0VU=-14dbfs. I use external VU meters on the way in and aim to get most things pekaing at around 0 VU. This will often mean a slightly lower level than the approach you've been taking, but in my experience, the results are easier to mix, and everything stays comfortably under control. For what it's worth, I generally mix to 1/4", and I aim to get my mixes peaking at 0 VU, or maybe a db or two over if I like how it sounds...

This is working for me.

Cheers, Mark

[PTUser NYC]

Observing 0dbVU is not just about increased mixbus headroom. There are arguments on either side of that one.

Whether or not the mixbus needs more headroom, if you record hotter than 0dbVU (which may be -14, -18, -20db etc on your converters) then you are gonna be asking your analog electronics to work above their normal operating level, and you will get some distortion or color. Sometime this is a good thing, but across all channels it can often lead to a degredation in sound.

My advice is to track lower, around 0dbVU, and to set your converters up so that 0dbVU is around -14dbFS, so you can get more resolution than you would at 0dbVU = -20dbFS (FS means full scale).

This way, you get a good sound from your analgo gear, and you can still go back out of your converters into analog compressors or EQs during mixdown without blowing them up.

Then, after your mix is finished, if you want to normalize, or increase the output gain of your mixbus compressor in order that the final mix is as loud as possible on CD, go ahead.

You will be surprised at how much more open everything will sound. I think tracking too hot is the number one culprit behind the 'bad Pro Tools sound'. Inferior converters, like the 888|24 are a close second.