Better to compress TO tape, or POST?

[mixmasterdsr]

Hi everyone.

I realize this is one of those questions where there's not likely to be a catch-all answer, but I thought I'd ask it anyways. The question being, If you're going to compress a sound, say a vocal or a guitar, is it better to do it TO tape, or coming back out? More to the point, are there any tangible benefits to compressing a sound TO tape? Does it help the sound 'print' better to the medium of analog tape? Does it give the sound coming back off the tape a better signal to noise ratio? Or, are you better off NOT compressing anything to tape, since you'll obviously have a lot more control over your mix if you leave it dry until you mix?

I'm recording directly to a Tascam 8 track (yes Cassette!) 688 unit, which will be going out into the computer for mixing and post-production work. The program material on tape will be Guitars, Basses, and Vocals, to a click track.

Thanks for all your replies!

Sean

[Boogdish]

hey, I record on a 688 too! For me it depends on the source, but more often than not I find myself compressing to tape. I like putting things onto tape sounding just how I want it to (or as close as I can get), and if the sound calls for compression, then why wait?

[Roboburger]

Do you know how to work a compressor well? If you're good, you can compress a little to tape. Done right, you can get a bit more signal onto the tape and thicken things up a bit, and reduce the noise floor.

If you don't know how to work a compressor too well, then you're better off just waiting until post, and just dealing with the maginally dimished signal:noise ratio.

When I was four-tracking, I printed effects (as in compression, reverb, and delay) to tape all the time. it was a great exercise in learning how to make those things work.

[drumsound]

If you want the sound of compression on a source, print it that way.

[vibesof20hz]

I usually compress to tape if there is a lot of dynamic variances in the performance. If a guitarist is just strumming away on a rhythm part, I would let it go to tape. But maybe the vocals have a lot of dynamics, or the guitarist plays really hard, then really soft. In this case I would compress (lightly, maybe 3 to 5 db reduction at most) just to get a better source level to tape.

Sometimes ill play the acoustic and during little breaks ill hit the instrument like its a piece of hand percussion. That part has much more energy and will probably end up clipping, so a limiter would get put on my acoustic to catch to really loud transients to prevent distortion. Of course, the other logical answer is to turn the pre amp down but this gets a better level to tape.

[Professor]

Personally, I never compress on the way in, but I also don't record to analog tape. If I did, I probably would.
A lot of guys think of compressing to tape as a sonic choice, as expressed above and elsewhere when they say they want to get the final sound printed to tape early so there is less to do later. But the original reason why engineers would compress on the way to tape was because the tape machine was the noisiest part of the signal chain. Since the rooms were quiet, the mics & preamps weren't too noisy, but the tape machine introduced hiss and other low-level noise, and also lost signal strength over time and multiple passes, it made sense to push the most signal onto the tape. The trouble is that simply turning up the gain risks seeing the peaks going over the limit and distorting, so an automatic volume control (which is really all a compressor is supposed to be) would be placed in line to push the quiet stuff up while keeping the loud stuff from blowing through the roof.
In my studio, and with most modern gear, the recorder is among the quietest part of signal chain, and mics & preamps are about as quiet as they've ever been, so the noisiest part of the signal chain is usually the room (and mine is really quiet but still noisier than the recorder). So I don't compress on the way in because I want every bit of dynamic range available to me in mix down so that I can concentrate on the sound, play with it, tweak it, and perfect at whatever pace the music dictates. For me, I think the commitment of placing the source in the room and the commitment of microphone choice & placement is more than I care to worry about while I'm running around the room getting things set, dialing in monitor mixes, trying to encourage & coax out performances, and still moving the session along in a timely manner. Adding more subtle mix decisions to that part of the process is something I find unecessary as a concept and as an added stress. Especially since I can't 'undo' a bad decision the next day like I can if I overdo EQ or compression in a mix session.

But that's my situation, and not yours.

You are recording to analog tape... and more than that, you're recording to cassette tape split into 8-tracks.
Open reel analog at 15-ips was considered the noisy part of the system way back when. And 8-track cassette at 1.75-ips or whatever speed it's running, is going to be noisier still. So I don't think compression on the way in is a choise for you... it's a necessity.
But here's the other catch - you have to be careful to not overdo it.
The other consideration for you is that you're dealing with very narrow tracks on the physical tape, and very little tape rolling past the record head per second. That means the tape will saturate much more quickly (at a lower level) and because of the proximity of the tracks, you will have increased crosstalk as you push the level hotter. The tape saturation provides a little compression all by itself, and with a light bit of compression on the way in, you should be able to get better levels going in and put more distance between you average signal and the noise floor of the recorder.
You may want to be careful about your low end though since that will saturate your tracks very quickly. Be careful to place LF tracks like bass and kick next to each other and far away from stuff like vocals and lead guitar. Also, use low-cut filters on your mics for instruments that aren't going to be delivering frequencies below about 80-100Hz for you because rolling that off will allow you to push those tracks onto the tape a little harder.

So that was a long way to say, 'yeah, compress to tape' but you know I like to explain why too.

-Jeremy

[creature.of.habit]

i don't hear anyone complaining Prof.
thanks for the great post.

[TapeOpAndy]

Another great reason to compress to tape... if you end up wanting to compress more tracks than you own compressor channels for, come mixdown, then you'll be bummed that you didn't compress to tape. All-digital folks don't have to worry about that, since they have as many compressor plug-in instances as their CPUs will allow, and furthermore, they can freeze tracks to free up CPU resources.

I have a couple racks of great analog compressors, but I don't own multiple boxes of each (except for my five FATSO Jrs), so even when I record to digital, I compress to tape as I'm tracking and overdubbing, otherwise, I'd have to go out and rent a whole bunch of additional analog compressors to get the sound that I want (or bounce tracks one or two at a time thru the analog compressors I do own).

[mixmasterdsr]

Wow, great replies. This is exaclty what I was looking for. Thank you ALL for your thoughtful responses. I love this board.

[lancebug]

I don't have the rig necessary to try (no more tape), but I am curious: If you compress on the way in and then hit the tape hard, would you get a lot more overall saturation/distortion due to keeping more of the signal in the "hot" range, versus just having the uncompressed peaks running into the red?

[creature.of.habit]

i have a question too about this particular subject...

it's more of a cable thing..

this would be easy with something mono, or dual mono, but how would one do it with stereo comps? like for example, using my dbx 128 with my drum machine...

i guess it would be better to get out of the drum machine, into a Y cable, and then from the dbx from another Y cable into the recorder...in theory, the comp would be something in the signal path, much like a stomp or something....am i assuming this correctly?

or i guess i could simply insert the comp as i always do, since the tascam 244 channel strip is active while you're recording to it...maybe i'm overcomplicating this...

[allbaldo]

I rarely compress anything except bass guitar, vocals, and sometimes room mics on the way in. I'm usually hauling ass, and need something flexible for later. I think it's a good idea to have as "complete" a sound as possible, but I don't want to marry myself to a sound that may be innapropriate later.

[audiogeek1]

I am like Andy. I have a few great channels of compression that I like the character of. I choose sources carefully as I track. Then other tracks that need help in the mix get those compressors because the other tracks have the character I wanted all ready.

When I work at some of the bigger studios this is not a problem. They have more channels of better compression plus the 6 channels of good compression I own so it is not as much of a decision.

Mike

[wwittman]

I never compress drums (except room mics), but otherwise, I probably compress almost everything, in a pop rock band, going to tape.

I don't generally compress strings, horns, percussion.. that sort of thing.

but guitars, bass guitar, keyboards, vocals, etc. almost always.