Article about music recording in general. Published in AtariMagasinet (swedish magazine) #3-96 - October 1996. Translated into english by the author.

How to succeed with recording your music - by Claes Holmerup

Hello all musicians and other interested people. It's always hard to get started on an article (or series of articles if that would become the case), but I'll start by making a quick presentation of myself.
My name is Claes Holmerup and I work at Mr Data Syd in Malmo (if, by chance anyone didn't know). I've been working with computer repairs - and strangely enough also been a user of Atari computers for about nine years. Some of you met me at NAS, where I tried to make some demonstrations at a convent with the help of a Falcon that actually wasn't very cooperative...
For about half of my life (I'm soon to be 32), I've been much - not to say VERY much into music. A fact that leads me onto the right track - as this article is going to deal with music.

I'm going to tell a little about recording in general (and a little about the equipment needed) as a start and then we'll see where it leads to.
What does music recording have to do with Atari computers? - Tremendously much, I'd say, since almost all current music productions use partly or only midi-controlled instruments. Naturally there are many different methods to record your music and I'll try to explain some of them here, as well as trying to make the pros and cons understandable.

The most simple way to record is when you play and sing live and record it to a cassette taperecorder. This method has mostly cons, since you don't have any chance of changing the result after recording. If you record through a couple of microphones, there is also the risk that some instruments don't sound good when they're recording through a microphone and some instruments (i.e. the drums) might drown all the rest.

When you record through a mixer, you have a little more control of the sound and you can choose whether to use line or mike signals. When you start thinking about mixers and such things, it's important to think about how many inputs you're going to need. Otherwise you might need to exchange the mixer to a larger one very soon after the purchase - or buy another one that's connected to the first one. As an example, the drums can become very good if you use 2 or 3 mikes, but the best sound control is achieved if every drum has it's own mike, which might lead to about 10 mixer channels for the drums when you have an ordinary drum set. This gives the best possibilities to fix your own sound, but it also gets rather expensive (large mixer + many microphones).
Often, you'll want to add some effects in the music - reverb, echo etc - and to be able to use effects, you can use one of two methods in the mixer. Either by using the aux sends, typical for a cheaper mixer is about 2 sends, which gives the possibility for two different effects and you choose the amount of the effect on each channel's aux send fader. The other way is by using insert jacks, which means that the effect can't be used by any other instruments, but at the same time it doesn't use an aux send (which might be important sometimes). If there aren't any insert jacks on the mixer, you can also connect the instrument to the effect before it goes into the mixer's input.
Many synths have built-in effects, but too many of those aren't any good compared to the ones available as separate rack units - though many have very good effects. In general you don't need equalisers for the synths, since it's better to edit the sound instead. That's why there are line mixers available (which don't have any eq at all, but do have a couple of aux sends). Those mixers are much cheaper than similar ones with equalisers. The line mixers add a little less noise, since the signal doesn't have to go through filters that aren't used anyway (every part of the mixer is an amplifier which adds it's amount of noise).

Tape recorder
If you want to record, you need some sort of tape recorder (unfortunately more expenses). Naturally, you can use a standard cassette recorder and record live when the sound has been adjusted to a nice balance in the mixer and the effects. It's cheap, but some limitations come along:
1. You can't redo part of the recording, but have to start over again when someone plays something wrong.
2. You're stuck with the result and can't make changes afterwards if you notice that some detail could have been better.
3. If an instrument gets too strong or too weak, you have to record it all again.
4. The music that's played live in the room where you sing will leak into the vocal mikes, giving you less control over the sound.
5. If you want to add an extra guitarr solo or some other thing, you have to play the first recording through the mixer, at the same time as you make the additions live and record it all on another tape recorder. This method will give much noise when it's finished.

You get rid of some problems when you use a multitrack recorder, for example a portastudio. Then you can record four tracks (on some even six or eight tracks) instead of the two on the cassette recorder. The portastudio also contains a mixer, so you have what you need to be able to make rather good recordings. You don't have to record it all at the same time, but you can record the basics on one track, the vocal on one track and other additions on the two remaining. If you need to, you can even mix three tracks and record them on the fourth track - then you can use the three tracks for more additions. However, if you make a mistake during a mixdown, you'll have to redo the whole thing if you've recorded additions on the three original tracks.
If you don't make a mixdown, you're not stuck, but you can test various settings for the effects, the vocals and whatever, make a remix to a cassette recorder and listen - if something's wrong, you just make a new remix and listen again. In this case, you don't have to redo all the recordings, but the main problem with the basics' timing and lack of errors is still left.
Why do I talk about making the basic recording on only one track? Well, if you're limited to four tracks, you have to count on recording in mono, since otherwise you'll have less tracks available for additions, or you'll have to make lots of remixes and re-recordings. Naturally, you can also record the basics in stereo and only make two additions one on each remaining track, for example one for vocals, and one for backing vocals, solo guitarr, singing whales, computers being thrown onto the floor or other fun effects - only your imagination sets the limits and there are many different ways to use the equipment.
One way of having the music played perfectly once and for all is to let the computer manage the synths and drum machine with a sequencer program, record it and then make the additions on the portastudio. You won't get rid of all problems this way; you still have one of the live recording problems - if you notice that an instument was too intense or other faults are recognised when the vocals have been recorded, you'll have to make adjustments in the sequencer and re-record all of it (but with the advantage that the sequencer does what you tell it to do, as opposed to when a band member gets stuck with the same fault time after time).

Now we're getting into another interesting phenomena that is used in almost every recording on modern records - syncing of sequencer and tape recorder with the aid of a time code. This works as follows: on one track on the tape, a time code is recorded, which tells the sequencer exactly where in the song you are. If you wind or rewind and start replaying the tape again, the sequencer will start exactly right. This has the advantage that you don't have to record the synths on the tape. Instead you let the time code control the sequencer. This gives you the advantage that is necessary when recording professionally (and at amateur recordings to be true), which is the possibility to change everything - even after the live recordings have been made.
If you have a four channel portastudio and an extra linemixer for the synths, you're able to make very good mixes. The sync signal uses one track, but there are three left - maybe one for the lead vocals and the other two for backing vocals or however you want to use them. Since the sync controls the sequencer - which in turn controls the synths - the result will be a very good stereo mix (that is, if you're able to adjust the sound to something reasonable of course). Since the synths never are recorded on the tape, the sound quality on them will be the best possible and will only be limited by the sound quality on the master taperecorder (and naturally the quality of the mixer and effects too). With a multi-track recorder, you'll raise the sound quality a bit more (the tracks have more tape width and the tape speed is higher), but the principle is still the same. If you have an 8-track recorder, it's also easier to make all the additions you want to, with less compromises than with the porta. If you need to make lots of additional recordings so that the 8 tracks aren't enough, you can make mixdowns as I described earlier - but since the sound quality of the multitrack recorder is better to start with, you'll notice the loss of quality much less than on the portastudio.

Now over to the master recorder. Most people use their cassette recorder for the mastering, but if you want to give tapes away to your friends (or record companies), you'll get a lot of noise if you copy from one cassette to another. It isn't very fun to mix every tape separately either, but if you do you'll get less tape-noise on the copy. Unfortunately there are often real-time adjustments that have to be done during the mixdown, so it's rather tiresome to mix every copy. A DAT is by far the best master recorder, but also an investment that digs quite a hole in your wallet.
With the DAT, you'll get rid of all the tape-generated noise and you can copy from the DAT tape to cassettes when you need to - with only one generation of tape-noise. Also MD and DCC can be suitable as master recorders, but if you at a later point change you system to be a harddisk based system, they're less suitable than the DAT because of the frequency-band reductions that are made in the MD and DCC technology.

Another thing to keep in mind is that in most studios, the greatest source of noise are the synths (if you don't count noisy guitar amplifiers). There are several noise-eating rack units available, which can reduce the noise without affecting the original sound, so that the synths don't destroy an otherwise perfectly noise-free recording from harddisk.

What do you want to read about?
A natural step at this point is to let the article continue in the next issue - it may deal with harddisk recording with Cubase Audio (which I'll be planning on) or something else that's music related. I can also publish answers to music related questions if any reach me. If I've failed to mention something important in this article, please let me know and I'll get it into the next article. I'm open to suggestions, but don't wait too long - the magazine has a deadline that must be kept.I may not have the time to answer everyone who asks questions, but if any questions seem to have a general interest, I'll bring them up in the next issue.

Play on, Atarians!