mix quality --> recording quality

ambi's post touched on some key reasons why perfection in the tracking stage, even though its an ideal way to go, cant always be followed to the letter in a DIY environment. You can spend tons of hours trying to get the right sound and have it sit well in the mix, but unfortunately you often lose something in the concentration. For example, I spent a hell of a long time just trying to get the right sound for the lead guitar in a song.. drums, bass, and rhythm guitars were tracked.. I found the right sound and tracked it, but I noticed that even after all that work, I realized that I will still have to make some EQ cuts because somehow there are frequency conflicts between the rhythm and lead that I didnt notice before. The track was far too well performed to re-track, and it maintained the character crucial to that part of the song.. It's been a couple weeks now.. and I'm really digging the sound of the lead.. it really does go well.. just not perfectly.. so this is definately a case of "Fix in the Mix".

Another example is mids on guitars.. I have a tendency to want the sound of scooped mids, but I don't dare track it like that: your ears can turn on you later, telling you that scooped mids does not fit your song.. I find it best to track with more mids than my ear initially wants me to, and then I tell myself that I will cut the mids in the mix later.. which I don't, because after a few weeks, I start liking the overall sound with the mids in it...

I also do not use compression or reverb during tracking. I don't have the experience to be able to determine at the outset how much is too much, so I'd rather wait until all the pieces are together until I start applying effects.

Sometimes there are other occasions where you can't decide on the smallest but still important details of the song you are recording. Some examples: whether a cymbal crash belongs here, whether I should fill in that space there with hushed voices, whether I should use guitar solo A with this sound or B with that sound.. Sometimes all the options sound good.. So what do you do? Create different mixes with different combinations of effects and alliterations and choose from the pool which one's going on the final product. This is another benefit to recording dry. You may in fact realize that certain groups of instruments should be compressed together or share the same fx.. if you applied fx beforehand, you may not have the option to group instruments..

my point is.. as much as I work towards the perfect recordings before mixing, "Fixing in the Mix" isn't always avoidable, but can be a great tool.

Cy
 
I guess I will show my ignorance here and ask how you do this:
-----------------------------------------------------------------------------------
Fader math" errors are also destructive to sound quality... Recording as close to mix level as possible is your best bet if using the computer. Then leave the faders at unity gain.
------------------------------------------------------------------------------------

If you want to record so that you don't have to move the sliders in your software recording program, how can you keep the record levels up on all tracks? Wouldn't you have one track at -1 db and another track at -10 db or maybe even lower? Could someone please give more details on how you would record say 3 vocals, lead and 2 backup, and have record levels high enough on each track so they are already (almost) mixed?

Thanks,
larrye
 
Well, I definately agree with the "track it right from the beginning" school of thought, but...

I think our job as recording and mixing engineers (and frequently the talent, as well, in the home studio) is to learn to recognize the trade off point.

At what point is your quality of performance maxed out and at what point is your ability to capture it exhuasted?

I've spent hours and days trying to "track it right" by attempting to perform a part better than I could or capture the part better than my skill or equiptment was capable of doing.

I've spent just as much. and probably more, time trying to make a track sound better with plugs, eq, and mixing tricks when it just wasn't possible.

In both cases I learned a TON about my skills and my equipment, but it took a lot of time. When I get serious about a project, or I'm doing work for someone else, I stay well away from the bleeding edge of my skills and equiptment and try to compromise between "tracking it right" to the best of my ability and leaving some things to the mix.

The more I work with this stuff, the better I get at knowing where the trade of is and where to concentrate mt efforts. It really pays off when dealing with people who can't play their instruments or are unfamiliar with the recording process- you HAVE to know how to compensate for them with BOTH tracking and mixing techniques.

For me its a matter of making the best of what you have- skills and equipment. That's what makes this such a fun challenge, right? :)

Take care,
Chris
 
larrye said:
I guess I will show my ignorance here and ask how you do this:
-----------------------------------------------------------------------------------
Fader math" errors are also destructive to sound quality... Recording as close to mix level as possible is your best bet if using the computer. Then leave the faders at unity gain.
------------------------------------------------------------------------------------

If you want to record so that you don't have to move the sliders in your software recording program, how can you keep the record levels up on all tracks? Wouldn't you have one track at -1 db and another track at -10 db or maybe even lower? Could someone please give more details on how you would record say 3 vocals, lead and 2 backup, and have record levels high enough on each track so they are already (almost) mixed?

Thanks,
larrye

I don't know about the fader-math issue. Maybe it falls into the 'every move you make kills your audio' area.(?)
In application though, it would seem to be pretty tough to get near a dead on recorded-mix. (Maybe he'll come back and help...)

The Nika's theory (Pipeline's going to jump all over me :p ) is that you can do this (record lower) in digital, and it doesn't hurt -and, it might make plugins act better.
Beyond that... I'm in over my head.
:D
 
I agree with most of what is said here, but the one thing I disagree on completly is tracking it at the levels you would have it at in the mix. No way. That goes against my every grain. If your fader math is bad, get new software and start tracking it right. Sonar sounds great. Protools needs work. Dunno about the others. The thing is, recording at the levels you think you will be mixing at is, in my opinion, a recipe for disaster. Its just not right to me. Your tracks should all be just shy of reds door when they are all at 0. You are wasting headroom and raising the noise floor recording any other way.
 
That's a pretty brilliant idea. :D Should I also record it already panned the way I want it, too? Hell, I can save myself a ton of mixing time. :D Thanks to whomever's idea that was!
 
Could someone go into a bit of detail on gain staging? I think in my particular case my signal chain is so simple that it isn't really an issue, but it's not something I'm familiar with in the recording setting (I'm more familiar with gain balancing in home audio/theater equipment).

I'd love to learn more about this... never know when you'll need that bit of knowledge. :)
 
Do a search under my name, using "gain staging" as keywords in the newbies forum... its a tip-of-the-week thread...
 
tubedude said:
... The thing is, recording at the levels you think you will be mixing at is, in my opinion, a recipe for disaster. Its just not right to me. Your tracks should all be just shy of reds door when they are all at 0. You are wasting headroom and raising the noise floor recording any other way.

It goes against everything we have grown up to learn, and our intuition. But it could be completly true. If your 'analog' dynamic range + about 15-20 db fits inside of the 110db window (or is it 144db? I always screw the numbers up), it just doesn't mater. The error extra noise and % distortion is swamped by our analog world's noise.
Or so it goes, +/- 3db.:D
I'm just trying to gather and apply as I go.
:)
Wayne
 
Ever since moving to 24 bits, I don't worry about having every track "just kissing the red".

I don't think my tracks have suffered.

If a signal is too loud or too soft, adjust the preamp gain. If the dynamic range is way out of control, patch in a compressor.

Or maybe by now I've forgotten what the question was...
 
Oh yeah, I remember now...

(This isn't an argument with anyone in particular - just a different perspective...)

It's fine to take all the time in the world to experiment with mic positions and equipment selection to find out how the recorded sound changes - and it's a critical part of the learning experience.

But do it on your own time. In real life recording situations, taking hours to find the right mic'ing scheme will kill the energy of a session and guarantee a less than optimal musical result.

It is likely that, at least in the beginning, you will not be dealing with top pros or top budgets. If you develop a reputation for being efficient and not "wasting" time and energy you will be rewarded with a lot of word of mouth business. I can't tell you how many times I've seen or heard about "engineers" who interrupt a slamming take because they "need" to make a technical adjustment. Talk about coitus interruptus!!!

I guess my point is, if the argument is "track it right, don't fix it in the mix..." , my point is that the law that supercedes that is:

Capture the performance at any cost. Don't be the factor that destroys the creative groove...

I'd rather try to fix a slightly technically flawed brilliant performance in the mix, than to have a perfectly tracked bland performance.
 
If, and say IF, Nika's theories on bit rate are correct, youll have nothing to loose if you set levels so they mix themselves leaving the return faders at zero, as long as they are somewhere " UP there" in the first place. Youll have 48 dB to play with up or down before you even go below 16 bits. And furthermore, IF his theories are correct, being below 16 bits is also no big deal, as it is ALWAYS 6 dB per bit, most stuff would sound fine in 8 bit or even less

not that I believe his theories :) but its worth looking into...I DO believe that bits have "precision" and " resolution" contrary to Nika, but his stuff is perfect on paper, not to be dismissed
 
littledog said:
Oh yeah, I remember now...

(This isn't an argument with anyone in particular - just a different perspective...)

It's fine to take all the time in the world to experiment with mic positions and equipment selection to find out how the recorded sound changes - and it's a critical part of the learning experience.

But do it on your own time. In real life recording situations, taking hours to find the right mic'ing scheme will kill the energy of a session and guarantee a less than optimal musical result.

It is likely that, at least in the beginning, you will not be dealing with top pros or top budgets. If you develop a reputation for being efficient and not "wasting" time and energy you will be rewarded with a lot of word of mouth business. I can't tell you how many times I've seen or heard about "engineers" who interrupt a slamming take because they "need" to make a technical adjustment. Talk about coitus interruptus!!!

I guess my point is, if the argument is "track it right, don't fix it in the mix..." , my point is that the law that supercedes that is:

Capture the performance at any cost. Don't be the factor that destroys the creative groove...

I'd rather try to fix a slightly technically flawed brilliant performance in the mix, than to have a perfectly tracked bland performance.

What he said.
Good save Littledog.
 
Back
Top