Software compression questions?!*%$!

C

crazyhorse

New member
Someone recently gave me the run down on using a compressor to enhance tracks when mixing a song and I would like some other opinions.
My friend said that he compresses each track so that the wave looks flat (squarewavish) instead of peaky (sawtoothish) before normalization and mixdown. He says that by doing this and normalizing everything to just below clipping you can get the maximum 'hot' signal in the final mix.
How much is too much compression? What instruments should you use the most on? How about for techno/dance music vs. standard rock quartet setting? What are some good software programs being used?
He said that he talked to someone who switched from analog compressors to digital for their ease of use and transparency. I thought that highend tube compressors were one of the great secrets to getting that warm 'pro studio' type sound, especially on vocals. I am thinking of using my Johnson Millenium 150 amp which combines a tube pre-amp with S-disc fx processing (as well as compression) to run a SM-57 (all I have) to get that type of sound. The amp has direct XLRs out so I wouldn't be relying on the inefficient amp cones for sound, just whatever the processor could pull off.
I don't own a bass so I'm also thinking of pitch shifting my guitar through the Johnson to get a bass sound. It sounds crappy through the cones but I haven't checked the XLR outs yet to see how much low freq. the processor is actually pushing. But that's a whole other deal...
Please help me out on the compression questions! I am so close to getting a good mix!
 
Compression is used mostly on vocals and bass and most instruments that have a wide dynamic range.

How much is too much?.
It's a essential tool but you don't want to sqash the life out of your sound

As to your friends suggestion. althoug comnpression helps getting a hotter sound It would seem to me that compressing EVERY single track could lead to a very sterile sound with some loss of natural dynamics of a musician.

As for software- check out a shootdown on software at http://www.prorec.com/prorec/articles.nsf/articles/9FED7E1E97DC4C9A862566E70067B116
 
Shailat has a very good point about compression being a essential tool. I will try to elaborate on that a bit here.

What is compression? If you cannot answer that question then you shouldn't use it... :) Sorry, that is just the truth.

Compression is a form of a VCA circuit. Voltage Control Amplifier. It's job is toin some way control the Voltage. In the case of a compressor, it will do this automatically based upon pre set settings.

Why use a compressor? Obviously, to control the dynamics of a track, or source. Often, a vocal track may be a bit too dynamic to work effectively in a mix. By using compression, you can make the vocal sit in the mix a lot better. Lower volumes will be increased, higher volumes will be decreased. In effect, the compressor will tend to make the dynamic range of the track less.

Other uses for compression. Compression can actually be used to in effect shape the sound of a track also. In the case of like a snare drum, the compressor, when set right, can actually make the snare sound fuller and fatter by limiting the initial hit, but bringing up the resonance of the drum. Because a compressor also imparts a bit of distortion on the signal that it is processing, you can actually smooth out high frequency transients on a track. Of course by the time you compress that much, you have lost ALL dynamic range, so, compression of more than say 10db should be avoided unless the effects of it are what you are after. This is purely an artistic decision to be made by whoever is producing the mix.

As far as what your buddy says about compression, well, I have never heard his mixes, but I well imagine what they sound like with that kind of dynamic limiting going on. Probably quite stale, and no real "life" to the mix. But then again, I have done mixes that had compression on many of the tracks and have had great results, so really compression can be used in this way with great results also.

Some of the things you have to ask yourself before you do anything to a track is 1-why am I doing this. Believe it or not, many people just do things to a track because they were told to, but in a lot of cases it was totally not needed. 2-What do I want to achieve by doing this. If you don't have a clear idea of what you are hoping to acomplish by doing something then you really will not have a place to start setting things. 3-Did it make a difference in the end when I used this. Often, you will put something on a track and find that it really didn't do anything to enhance the part in anyway that embellishes the song. If you are applying anything to a track and it don't contribute to a better outcome in the mix, well, you are wasting time on something.

Here is my thought on compression.

If your mixes a pretty darn good without it, then compression can make it a bit better. It is the saying "Noise reduction works best on tracks that have hardly any noise". This can work on most any effect of dynamic processing. If you barely need it, then it will work really well. If you need it really bad, then it will not work as well because anything you do to a track has it's drawbacks sonically speaking.

You may try other solutions to compression before relying upon it to solve the problem. If you are using software to mix, then you have fader automation. You will probably have better luck using an occasional fader move to adjust the track in the few places that you need to control the dynamics than to insert a compressor over the whole track. Also, sometimes the right eq tweek can take away a frequency that really jumps out at you when hit hard. The point is is to really "LISTEN" to the music and make decisions about what to do to make it better rather than relying upon what others tell you to do to it.

Of course, having an understanding of how all of your tools work is the logical first step in determining which one to use on a problem track. Really, mixing is just a matter of solving a bunch of little problems with the sound. If approached that way, you will find that you will make better decisions on what to use on what track to fix it.

Now, it has taken me many years to learn this one thing about recording. No matter how many times I read and/or heard it, I still didn't do it. "Record the track as closely as possible to the way you want it to sound at mix". Eq, compression, delays, reverbs etc.... work best when used in slight ways. Using them heavily should only be used for very specific effects.

So, the only way you are going to know what works well is to experiment a whole lot. I have been recording for over 10 years now and I really wasn't until the last 3 years that I really started to get it down. It is going to take you years to even have the knowledge to start making decisions that are good for the music like a professional engineer does. Next, you will discover that you will always have a comprimised sound when using consumer quality equipment. If you don't have say at least $30k to invest in a studio setup, forget about getting anything better than just demo quality recordings, no matter how much you know about production techniques. A little computer setup with a moderate sound card and low end mic pre amps, and Shure microphones is just not going to produce the same quality of sound as a Studer 2" machine with dolbySR, a Neve mic pre, and a Nuemann U-87 will!!! That is a fact that you will have to live with in a home studio setup. So if you approach recording knowing your limitations of your equipment you will have a whole lot more fun doing it.

Good luck.

Ed Rei
Echo Star Studio www.echostarstudio.com
 
Hey Homeboy,
You bring up a point that is worth looking at. First of all... I am drunk... and I am having trouble typing.
Now that we are all using Software FX for mastering and recording, I wish someone would explain the relationship between the waveform that we see and the waveform that we hear... I see spikes in my waveform that I don't hear, and I hear loud sounds that don't appear to jump out of the waveform.
I have loaded a few professionally recorded songs into my software and looked at the waveform and it appears to be relatively flat, without all of those spastic spikes that my waveform mixdowns have. I will deduce that compression is the tool that pro-folk use to achieve this good-sounding flat-line. The question is: does amatuer software compression sound as good as professional hardware compression??? I think we should all run a test... Record a multitrack session and compress all of the tracks untill they are not cactus-looking, and then mixdown... Then remix the same session without compressing the individual tracks, but compress the mixdown waveform untill it looks smooth. Then lets compare results. I bet we will find out something about the nature of waveforms that we didn't already know.
I'll run the test and post the results... Will anyone else???
 
Thanks for the posts guys. According to what I've read here, it looks like my friend and I were both partly right. Thanks for clearing some stuff up.
sonusman, I'm always talking about your posts with my buddy and the way they seem to reflect a 'purists' approach to audio handeling. ie.. get it right the first time, less is more, & don't drive a go-cart in the grand-prix. My problem is that I have no money for good equipment so I will be forever struggling I'm afraid. I have talked my wife into building a DAW (Celeron 300a, 24 bit, etc.) but I'm going to have to buy it one piece at a time.
S8-N, what's up you runaway hillbilly? I'd be willing to help out on your compression tests if you're serious. I've had a bit of trouble getting direct answers to similar "How's it work/ What's it doing questions" regarding mp3s, dc offset. My question is after we run the tests, who's gonna explain the results?
 
S8-N
The leveled out peaks are most probably from limiting and compression and not just compression alone so check that out when you do your test's.

Many high end mastering studio's use a multiband and choose carefully the Freq to compress and therefore
manage to boost lows with out damaging highs ( for example ) and don't color the sound like many amatuer Comps.
 
I think one of my problems is that I don't know how to program the software compression that i am using. (That's right, PROGRAM... There's no GUI of any kind with it.)
I can dick around with the presets and get a nice compression for vocals but I cant get that fat waveform of a pro studio. I worry about weird artifacts popping up if I hard limit too much.
I'm gonna just have to devote some time to experimentation. When I am in the middle of mixing a session I dont really feel like experimenting on a single waveform all day.
 
Hey everyone, I'm new here and this is a pretty cool board. All this advice is much appreciated. However, I can't say I agree with Ed on his $30K estimate. I'd be lowering that to about $10K or even less if you're willing to overdub and spend a long time on your mix instead of a lot of money on gear. I'm sure Ed could do a better job and faster than me with his studio, but I'd also guess that someone who was as knowledgable could also get a professional caliber master on a streamlined computer system for a lot less money, if they were willing to spend five to ten times longer doing it. This may seem like a waste of time to an engineer with a studio, but to a home recordist (the subject of the board) who has all the time in the world, but little money, I believe very clean mixes can be achieved for a very low budget, by bringing audio as cleanly as possible into the digital realm, and keeping it there. I have heard very-pro sounding results from this or similar chains:

- good mics: $1000 to $2000
- tube compressor: $1000
- pre-amp or direct box: $500
- computer with pro sound card: $3000
- software for everything else, anywhere from
$2000 to $4000,or more to get good results.
- power amp and monitors: $2000

Mix down can be done on disk, and any of these prices can be severly lowered by sharing gear and buying used. Also, the high end mics can just be rented as needed, as can DAT's for distribution. Mix-downs on hard-disk and source tracks can cheaply be archived at high resolution with a CD-burner at $200 bucks and 650 Megs/disk, or a tape back-up. The catch is that you need to learn how to use everything properly, which a qualified engineer could do very quickly. I am still a ways from getting what I want out of my set-up, but I have heard incredible stuff from others with almost identical studios. Seems to me that the whole point of the studio now is for the know how. A good engineer is worth way more than a gold-wired SSL.
Ed was talking elsewhere about testing sound quality of what he could get out of high-end studio in a short period of time billing $100/hr, compared to home recordists here. A fair comparison would be what a pro-engineer could do in a high-end studio in a couple of days, compared to what the same engineer could do on a streamlined guerilla set up like mine, given as long as he wanted.
At any rate, I'm spending my money on engineering courses instead of hourly rates at the local gold plant.
What I'd really like to know from Ed is which of the expensive tools he would least want to part with, or what he thinks is the right price for best results for individual components in the set-up I outlined. Just how good a compressor, mic, etc would you use?
Any feedback is welcome.

Iain

[This message has been edited by Iain Duncan (edited 01-08-2000).]
 
yes experiment onit i been doing a lot of
mix test on each song this time with compression on this track,next mix with this and that...
many months passed i comeback to less is more!
and just trying to capture the mood
and using as less an fx as i can
i guess i'm matured now!
 
Thanks for all the replys! I've learned a lot since I started this thread. The compression article ot Prorec (thanks dmc!) has shown me that that heavy sqr wav looking compression is more like limiting. My friend who was doing that was writing hard techno/metal stuff so that makes sense. Now I understand how to use compression to shape 'notes', bringing out performance characteristics. I've been doing so much studying lately, I haven't had time to practice any of this much. I'm in heaven.
 
You must log in or register to reply here.
Back
Top