recording to analogue tape
My band will shortly be recording a live demo on a Teac 80-8 half inch recorder. Track 1 is kick, 2 is snare, 3 is toms, 4 is all cymbals, 5 is guitar, 6 is lead vocal, 7 is backing vocals, 8 is bass guitar.
Is it better to record everything flat and then compress on mixdown?
I think it depends on experiance.
Over compression plus tape saturation can leave you with unusable audio. However, peaks that push the tape too far into saturation can be even worse. No compression, and low levels will leave you with tape his. So you have to find a happy medium, and that takes experiance.
With your track list, you'll probably have a happier experience if you compress the vocals.
More than likely, they are the most dynamic. (assuming that you're not doing jazz or some avant-garde drumming)
Beyond that, as GeckoMusic pointed out, experience dictates what happens next.
80-8, a good old workhorse. But you didn't tell us if you were going to use any noise reduction? It's not built-in. My recording process would be different with and without noise reduction. And it would be different if I was using Dolby versus DBX. You really can't get any saturation effects with DBX and shouldn't try to "over" record with that. Plus, you need a very flat response record to play back. You won't find that on a TEAC/TASCAM as they are not flat. Any variations in frequency response is multiplied by a factor of two with DBX which can color your sound appreciably. Whereas Dolby, in any flavor, is not so much depended upon frequency response as it is on level matching of record to play back.
If on the other hand no noise reduction is being used? Like my other colleagues suggested, I'd definitely compress and/or limit the vocal. I'd also limit the bass. Meanwhile, I'd even consider utilizing some compression and/or limiting on the drums. Just to keep my signal as high above the tapes noise floor as possible. But too much compression and/or limiting can create an ultra flat recording. Not flat as a frequency response but flat like yesterday's beer you didn't finish until today. So just a little goes a long way. You can add more upon mix.
Without any noise reduction, upon mix down, any type of boost equalization in any of the upper mid range & high frequencies will produce a recording of nearly unlistenable quality. So not only do I advocate utilizing some compression and/or limiting during recording, I also recommend adding some upper midrange and/or some high-frequency boost. If you like it, it's good. If you find it too bright? Well, upon mix down, you can cut and/or lower the upper midrange and/or high frequencies. This will give you some additional noise reduction by utilizing an additional amount of preemphasis with an equal amount of deemphasis on mix down. That in combination with some downward expansion and/or gating will further help to improve signal to noise (ratio).
Utilizing an 8 track analog machine also doesn't mean you are restricted to 8 tracks. You can actually record many more than that. On most later generation multitrack recorders, like yours, sync playback for the record head, is virtually as good as through the playback head. This means you can bounce one channel, in combination with other channels, to an available open track that is not directly adjacent to a playback channel unless the level is slightly lowered. Otherwise, track one could cause feedback through the record head to channel two. But if the track that you are bouncing to is separated by a track, you can't balance at higher than normal levels. This allows you to really stack things up. But you essentially have to utilize clairvoyant premixing throughout the process. I had to record more than 24 tracks on a 8 track machine when I produced fully orchestrated jingles for a multimillion dollar international advertising agency 30 years ago! Of course, I didn't use a TEAC. I used a 1 inch MCI JH 110-A, barefoot, no noise reduction, at 30 IPS. The final mixes included elements that were up to 6 and even more generations down. So I utilized everything I've mentioned to you. There was never any objectionable noise when I completed the project. I'm still quite proud of it to this day, 30 years later. In fact, I think I'll post some of these as examples, with explanations, for all to hear. Millions of people have heard these as they were used repeatedly for 20 years, throughout the United States & Canada. If anybody is interested, I'll start a new thread to explain the techniques utilized that made everybody think I was lying about not having a 24 track machine. I'm just an engineer that sees the sound in my head. And then places it into the recorder. Believe me, clairvoyant mixing isn't easy.
So let us know how your session goes?
Ms. Remy Ann David
frosty55 wrote: Track 1 is kick, 2 is snare, 3 is toms, 4 is all cymbals, 5 is guitar, 6 is lead vocal, 7 is backing vocals, 8 is bass guitar.
You realize, as Remy says, that you could have more fun with this. What you'll end up with is a mono demo, or an early-Beatles this-side or that-side left and right "stereo" demo. Also, I'd put the kick and bass guitar next to each other on tracks 7 and 8, then the rest of drums 6-left, and run instruments/vocals from 1 on over.
You may consider:
WHAT?!!! What about the VOCALS???
What...you can't play the song through without vocals? :shock: If you can, and you do this, simply bounce this all down, while mixing, to a two-track recorder...tape, computer, hi-fi VCR (which actually works nicely, fidelity-wise, except for the cueing..which is a pain) ...whatever. Record that pre-mixed two-track recording back into tracks 1 and 2, at a good level, of the 80-8. That gives you 6 more tracks to do your lead vocal, and stereo back up vocals. That took 3. You have 3 left. Double the lead vocal. Put on another guitar. Do another 2 tracks of backup vocals for added texture. Whatever.
Mix all of those down to stereo. Maybe run effects on the way out to the mixdown at any stage.
In the end, you may have a nice real stereo demo where the instruments are placed better than grouped all by themselves...as in an early-Beatles tune.
BTW, WHY did I want to put the kick and bass next to each other? Because they pretty much work together anyway, and any track leakage from each to the other will be simpler to deal with than if you had the kick bleeding into the vocals, etc. Any bleed you have, when you manipulate the main signal, you are also messing with the bleed signal, which may change it's relationship to the mix and have you chasing your tail. "I brought the vocals up...now there's too much kick." Etc.
nd, the way the cymbals are right next to the guitar means there is not much common tonality betwixt the two, so if there is bleed, you are more likely able to just eq out the other stuff.
Just something to consider.