Tracking compression....

I am probably a bit heavy handed when it comes to tracking compression ;D ;D  but finding sometimes this is not the best option. What are some peoples ideas on compression when tracking

Thanks
AC

Squish it n' squash it,


    Bish it n' bosh it,



      just don't over do it!

It is essential that your monitor mix be as close as possible to the finished mix, so you can judge how your track sits in the mix.
I use moderate amounts of GR, so generally I compress again in the mix.
I use fairly fast Attack and release times, typically less than 5 miliseconds attack and about 100 release, ratio less than 2:1, so it never pumps (that's what is impossible to get rid of later).
This way, I have never been in the situation where I wished I had not compressed so much when tracking.

Nice!

    I've given some thought about this too, that every op amp/transistor/whats it, in a circuit is not linear so keeping things compressed to a more managable degree can help things sound better because the audio signal is staying on a smaller section of the non-linear "whats it" and therefore staying more uniform to itself. It that makes sense.

I know tape isn't linear, I know we like to think high quality digital is totally linear, but its not.

Does that makes sense? . Having a signal whose voltage max and min difference is only +/-1v (compressed) compared to a signal +/-3 volts. It makes sense to me that in the real world you would be recording a more linear unto itself because you not going to such extremes on the circuitry.

Have I eaten one to many paint chips? or is my understanding somewhat correct

Thanks
AC

abechap024 said:

Does that makes sense? .

Click to expand...

I think it makes more sense when talking about electro mechanical things like speakers and phono cartridges. That's the weak link. They want to move as little as possible. High acceleration with high velocity is a killer.

I go for it, in the beginning. Old school method.
I'm not one of these..64 tracks, play it safe, save it for later, thinger's...

After hearing Ken Scott speak at a lecture, and listening to a few of the other guys...my mind changed about it. You have to be careful, but with some practice, you get better.

ie: decisions up front. Get mix now. Not later...

Me likey.

I've stopped compressing drums on the way in. I had a "back to basics" moment a few years back when I realized that I really didn't remember what just plain old drums sounded like. I had gotten so used to squashing the kick and snare indescrimenantly that I was losing the envelope of the actual instrument.

Vocals and guitars are a different thing. Me squishy.

Yep, whatever sounds best. Compressed, uncompressed...etc.

It is essential that your monitor mix be as close as possible to the finished mix, so you can judge how your track sits in the mix.
I use moderate amounts of GR, so generally I compress again in the mix.
I use fairly fast Attack and release times, typically less than 5 miliseconds attack and about 100 release, ratio less than 2:1, so it never pumps (that's what is impossible to get rid of later).
This way, I have never been in the situation where I wished I had not compressed so much when tracking.

Click to expand...

+1

First it obviously depends on the music.  My biggest concerns with compression are with the drums.  I rarely hit the close mics with compression even when recording a loud rock band.  I never compress kick or toms, never OH mics, usually snare(-3) usually room(-1 to -3).  Even though it would be useful I don't usually trust a drummer to keep the dynamics straight especially with the more low end sections of the kit.  However Vocals(-5) in rock get crushed along with bass(-5-to -8) and acoustics(-3).  Everything else takes the compression in the mix.  With electrics it usually depends on the amp.  If its solid-state like a jazz chorus I might hit it with just a db or 2 on the way to the box, if its a marshall or fender I let the amp do its own thing.  It also really depends on how your production looks.  If there are tons of overdubbed gtr tracks with different parts I'll squash some of the more melodic parts while leaving the rhythms alone.

I do know some guys that ascribe to compressing through all stages of a production, a little in the track, a little more in the mix, a little more in the masters.  I've never been happy with the end results I have gotten from this style.  However Ive never not been happy with the results that my friends get with this style.

In the end you just have to ask if it sounds better with the comp bypassed or not, Desol has a great point on making the mix rock from the get go. This is what makes the talent happy when your tracking.  If it rocks so will they.

If it's a full rhythm section where most of the track is being cut simultaneously, and I can therefore hear how things fit together, I'm comfortable using lots of compression when tracking.  If it is a 'drums first, then bass, then one guitar at a time, etc. kind of stacking approach, then I'll tend toward milder tracking compression just to tame the peaks and save the rest for the mix.  I love compression as a way of shaping the character of a signal (I use tons of it when I mix), but if I am not confident how the track will fit within the final mix then I'll take more of a tame-the-peaks approach and save the heavy stuff for later.

FWIW, when mixing, more often than not all that is needed to average/stabilize the perceived level of a given track is to knock of an absolute max of -6dB using a 2:1 ratio and the 'auto' time constant setting of the processor pretty much does the job just fine.  If it needs to feel more 'up front', then I slow the attack to let more transients through and if I need it to feel further back I increase the attack speed to limit the transients.

M2C

JC

rascalseven said:

If it's a full rhythm section where most of the track is being cut simultaneously, and I can therefore hear how things fit together, I'm comfortable using lots of compression when tracking.  If it is a 'drums first, then bass, then one guitar at a time, etc. kind of stacking approach, then I'll tend toward milder tracking compression just to tame the peaks and save the rest for the mix.  I love compression as a way of shaping the character of a signal (I use tons of it when I mix), but if I am not confident how the track will fit within the final mix then I'll take more of a tame-the-peaks approach and save the heavy stuff for later.

Click to expand...

This is a good point too...although i dislike, and try to avoid tracking people individually.

although i dislike, and try to avoid tracking people individually.

Click to expand...

+1

It has become a habit for me of plugging in a compressor whenever tracking. I'd love to just be able to ride the fader when recording vocals, but its too much work for me having to learn the song. The only time I do not use a compressor most of the time is when tracking synths and samplers, but for live instrument, I end up compressing when tracking. I would hate to miss out on a good take because it clipped which I have done every so often even with a compressor in.

canidoit said:

I would hate to miss out on a good take because it clipped which I have done every so often even with a compressor in.

Click to expand...

This reminds me of something that may be worth mentioning:  As a rule I set my gain structure with no compressor in the signal path (not even plugged in and bypassed... nothing physically patched in).  I leave some headroom for unexpected peaks (which always come during tracking), and then, and only then, do I patch in the compression.  I adjust for the GR I want, and then adjust the output, still leaving some headroom for unexpected events.  With this method I pretty much never have unexpected and unwelcome clipping.

This seems obvious, but I've seen people violate this many times, often to their detriment (they get to a loud part in the song and have to readjust everything, which alters the character of the signal).  I have a friend, who is an excellent musician and pretty knowledgeable guy, and he once called me concerned about the distortion he was getting with a Neve-style preamp I had built.  We talked for a bit, and I could not, for the life of me, understand how he was getting audible distortion on loud peaks if everything looked good when he was setting his gain.... then it hit me... I asked him if his compressor was patched in, and he said, 'yes,' so I told him to unpatch it, reset the gain (which he then lowered significantly) and repatch and readjust his compressor.  No more problems.

The distortion was in the signal path, not a result of overloaded A/D's, and that was what was tripping him out (his DAW's input meter wasn't peaking).  So set gain without compression, and if you happen to be using an all-in-one type device with preamp and built-in compression, make sure that not only is the compressor bypassed, but also that the gain makeup (which remains active on some units even when the compressor is technically bypassed) is set for unity, so you know you're dealing only with gain when adjusting the preamp.

(Same is true for eq as well... some eq settings can jack with your gain internally causing problems, so set gain first, then patch in the eq/compressor/whatever adjusting each one at a time before inserting another device.  It'll prevent lots of frustration.)

Carry on.

JC

What happens when every track is compressed with "just a little GR on the peaks to help it sit in the mix"?

I can imagine alternate philosophies.

One is to not compress any more than is useful to capture the take without exceeding the dynamic range of the recording media.

It is always easier to add compression, than remove too much later.

Another philosophy, especially if you only have limited signal processing available for final mixdown is to track as close to the final print as possible, to free up processing for final mixdown.

Back in the not so good old days, there were other rules of thumb, like adding HF boost preferentially while tracking and HF cut later during  mix down to get some free tape hiss reduction.

With modern gear and plentiful processing this may be 6 of one, half dozen another..

JR

One word.

Commit.

I like to get my project sounding as close to the final going into the box. It helps make the mix easier and helps the musicians to play against something that sounds like a record. I like to get my tracks pumping and exciting.

That mean's generous amounts of compression and EQ.

Compression in the box is flat and listless compared to the real deal.

It get's easier to see into the future when you've done it a few times.

Yes, it's old school.



Mark

JohnRoberts said:

Back in the not so good old days, there were other rules of thumb, like adding HF boost preferentially while tracking and HF cut later during  mix down to get some free tape hiss reduction.

Click to expand...

Haha.. I designed and used a "very special", in fact meticulously tuned bass preamp precisely for that purpose: to get the then very widely used, glossy slap sound without annoying hi-freq-boosted tape hiss on mix.

Biasrocks said:

One word.

Commit.

I like to get my project sounding as close to the final going into the box. It helps make the mix easier and helps the musicians to play against something that sounds like a record. I like to get my tracks pumping and exciting.

That mean's generous amounts of compression and EQ.

Compression in the box is flat and listless compared to the real deal.

It get's easier to see into the future when you've done it a few times.

Yes, it's old school.



Mark

Click to expand...

You know it!