peaks or not when mixing

K

keerus

New member
ok, i'm mixing some general rock music. some indie, some punk, whatever. i'm new to mixing so i need some tips. when you're at the mixing stage, which is better: let the waveform breathe, having no peaks and not all crushed next to 0dB (and let the mastering engineer make it "loud"), or carefully tune the mix so it comes close to 0dB but no peaks (without the use of a hard limiter or anything else that would force it), or make it as loud as possible with peaks and all during the mixing stage? i am guessing the latter, but when i see the red light go on, it bugs me = ). just wondering what all your input was. thanks.

justin
 
Blue,


You may have to excuse my ignorance here but:

I read the article. Absolutely correct all the way down and something people need to understand. But it seems that if one were to actually understand level structure, and one were using the 24 bit format:

1) One would understand that -15dB FS = 0dB VU (Which you already stated).

1b) 0dB FS then equals a whopping +22dB VU (As you know)

2) At 24 bit, there is no need for bit cramming. Bit cramming will inevitably get you ultra hot +22VU signals that have to be brought down with either individual faders or the master fader....WAY down.

3) If one has observed good level practices and is now mixing, why would one need to push the board THAT hard? I realize you want to use all available bits but you stated that at -12 or 15 on the masterlink meters is not anywhere near loud enough basicly.......when was the last time you heard of anyone printing an analog mix at +22dB VU?!

3b) I dont care what board you use, you will get an invariably different sound when you push that hard on the op amps...maybe good, maybe not.....it may even be out of the "sweet range" of some of the much better boards. What do you do? Will printing that mix at (what is essentially) 0 to +6 VU be a bad thing?

What do you generally like to print a mix at? What is the general peak level? What is the general RMS level? Safe guess here.........I know it depends.


Am I way off base here? Am I missing something? Does this make sense at all? Its a bit late right now......I'm not trying to come off testy...I'm actually trying to put this against my current idea of level practices and something needs to be straightened out.....I'm in no way implying that it is obviously you.

Let me know your thoughts.....



heylow
 
Actually - the article is kind of backwards - I was describing what was going on in a level-mismatch scenario that I encountered.

You're absolutely correct when you say you don't want to be "bit-cramming" (good phrase, BTW!)... and I intentionally didn't mention that specifically because the whole point of the article was that everyone's signal chain is different, and to make informed decisions about *your* levels, you have to know *your* level-chain and adjust signal accordingly.

Maybe I misunderstood something but one thing that isn't clear from your comments though is your equating of 24-bits to a higher value of 0dBFS....

More bits doesn't mean your output level at 0dBFS is any hotter, just that your bit "depth" is wider at lower levels....
 
Blue Bear Sound said:
You're absolutely correct when you say you don't want to be "bit-cramming" (good phrase, BTW!)... and I intentionally didn't mention that specifically because the whole point of the article was that everyone's signal chain is different, and to make informed decisions about *your* levels, you have to know *your* level-chain and adjust signal accordingly.

Maybe I misunderstood something but one thing that isn't clear from your comments though is your equating of 24-bits to a higher value of 0dBFS....

More bits doesn't mean your output level at 0dBFS is any hotter, just that your bit "depth" is wider at lower levels....
Click to expand...


OH! NO!......I did not mean to infer that at all. 0dBFS is 0dBFS whether youre at 24 bits or 8 bits. Yes....we agree on that last sentence. 16 bits has the same 0dBFS but we often "bit crammed" so that our signal did not fall into lower resolution whereas with 24 bit there is (for lack of a better term) headroom because there is higher resolution at even lower levels.

I believe that 16 bits gave us poor understanding of levels because we needed to bit cram to keep signal integrity, yet we were tracking at a whopping +20-22dBVU......it was a rock and a hard place, in theory.

I believe that lack of understanding has carried over into 24 bit and people are still needlessly cramming bits. I suppose if you stay all digital it doesnt matter that much but even there, there is some theory that says if you track up at 0FS all the time, you subtly compromise quality because there will inevitably be "spikes" that are too quick for the meters and clip the converters. Just a thought.....

All in all, a great article though and necessary too. I believe that proper levels and the real meaning of 16 vs. 24 bits are probably the most misunderstood ideas in home recordingland.....sometimes even in Pro recordingland!

I just like to test my knowledge constantly on these subjects because, like most, I learned about matching analog and digital gear the hard way. I'm currently working on my own project, which is being tracked on an Alesis HD24 and taken to be mixed at a very nice little analog studio complete with Neve board and and Mike Spitz modified Ampex 2 track so, needless to say, I have to have my level theory correct the first time!

Just keeping on my toes!


heylow
 
Interesting article Bruce, but I think it has raised more questions than it has answered for me.

Background info (which may or may not be relevant)

I have a S/craft Spirit Studio 24/8 (similar to the Mackies in it's layout and functions) and it is set up with 16 channels hardwired for mic inputs, direct outs and tape returns. I record on a Fostex D160 and if necessary have the ability to record all 16 trks similtaneously. The console doesn't have a meter bridge, so I have to rely on the meters on the D160's control panel...........it's a pain in the arse but better than nothing.:)

OK, a question..............when tracking, I have the choice of using either the channel pre-amp gain or channel fader to increase/decrease the signal going to the recorder. In your opinion, is there a right or wrong way to approach the use of these, and/or, what are the pros and cons of using one of these controls rather than the other. At mixdown I have the choice of the fader or a tape trim, so I guess that should be considered too.

ChrisO :cool:
 
The idea is that you maintain signal levels throughout your gear chain that maximize the signal level, relative to the specific gear's intended operating level.

Send a steady tone throughout your chain and see where your meters read on each meter'd piece of gear. Unless you've mixed gain structures, this will give you a "line of sight" to use as a meter-level reference.

For adjusting gain on pres, it's always better to set channel fader to zero (or nominal), the use trims to adjust for input level (same with the tape returns if you have trims for them - I don't, which can be a pain).... the reason to keep the channel fader at the 0 mark is because it is in this position where the fader response is most linear and usually reflects the mid-point of the gear's intended signal operating range.
 
Once you do all that Bear suggested, you still may find yourself occasionally needing to "ride" the volume while tracking especially dynamic singers. At that point, I would recommend using the faders (not the trim pots) for any fine tuning adjustments made while tracking. It's easier to move them in small increments, it's easier to visaully note their exact positions, and often moving them is less noisy than adjusting the trim pots.
 
Cool, you have confirmed what I was thinking re the 0 setting for faders. You just managed to explain it better than I was "visualising" it.

:cool:
 
You must log in or register to reply here.
Back
Top