For less than the cost of recording his last album, Tom Robinson has managed to buy his own 24-track studio and the building it occupies.

He's known as the man behind hits such as 'War Baby' and 'Atmospherics', and he now owns a brand new 24-track studio, but despite all this, Tom Robinson insists that the most useful weapon in his studio armoury is the humble Portastudio, but why?

It's a question of writing. Originally I used to write all my songs on a 144 Portastudio, but since then I moved up to the Fostex 250. The Fostex gives a much cleaner sound, due to their choice of Dolby C which I much prefer to DBX and also the heads last much longer. I went through two sets of heads on my original Portastudio, so be very wary if you're offered a second hand one. It's worth carefully checking the condition of the heads.

Since the first Portastudios in 1980 and more recently, the introduction of good drum machines, there's been a distinct shift away from the 'garage band' of the seventies to the 'bedroom' band of the eighties. The garage band spends most of its time rehearsing for the occasional gig whilst the bedroom band may never play live but nevertheless are producing music which emerges as a finished tape. Their music is filtered through the limitations of their recording technique whilst the garage band's music is filtered through the limitations of their playing technique. I started out in a garage band and then moved over towards the bedroom band way of life. Live playing costs have escalated whilst recording equipment has plummeted in price. Take the Sony PCM701 PCM encoder, couple it to a video recorder and it will give you stereo recording quality beyond what would have been George Martin's wildest dreams in the days of the Beatles, but now you can buy one for a few hundred pounds. In 1963 terms I suppose that would be the equivalent of around £160 complete.

I can understand the use of a Portastudio to help write your songs, but why did you feel the need to buy a complete 24-track studio?

The bill for recording the last album came to something in the order of £120,000. That's a lot of money by any standards. The money is advanced by the record company in the hope that you'll write another 'War Baby' or whatever and they can either make a profit on that or lose a lot of money. Having talked to Dave Stewart about how cheaply they built DNA studio, I realised that it was a sensible move for me to do the same. RCA gave me a modest advance to go away and record a few tracks so I spent the money on this studio and presented them with a fait accompli. If I only ever make just the one album here the studio would easily pay for itself. The whole lot cost me just over £45,000 and that includes the building.

How did you cope with the transition from Portastudio to 24-track?

Having watched producers at work I picked up quite a lot, but the important learning was actually done on my Portastudio, such as learning the difference between track 1 and channel 1, and working out routing and bouncing. I didn't see any point in buying anything between the Portastudio and the 24-track. More gear does not necessarily mean better results. More tracks can actually be a distraction if you are writing songs rather than producing masters; the recording process gets in the way of creativity and in the end, you're supposed to be writing a song, not engineering a production. So the 24 track is here to produce masters, not to write or do demos.

Production versus Composition

Do you not find that there's a shift away from straight melodic songs to the point where the production is of equal importance in many cases. Take the old example of Frankie's 'Relax'. How would that sound demoed up with an old acoustic guitar and a vocal?

That's true, but if you gave Trevor Horn a Portastudio, a reasonable drum machine and a few budget bits and pieces, he could probably come up with something that was very recognisably 'Relax'. There's a lie that the manufacturers would have us believe; that if only I had the such and such digital reverb or sampler or whatever is currently the favourite I could come up with amazing music every time. Synth manufacturers are particularly guilty in this respect. It's sometimes actually helpful to limit yourself. Another related point is that everyone talks about 16- or 24-track as though they're so much better, but what they lose sight of is that music used to be recorded onto only two tracks and that MIDI sequencers are now making it possible to go back to that in some cases. What I'm interested in is this; all you need is a Sony PCM701 decoder and two video machines to enable you to bounce from digital to digital and you can overdub from one to the other with no noise or perceptible loss in sound quality. For around £2000 you can produce digital master quality recordings. If you make a mistake you just do it again. It's like the old fashioned sound on sound principle but without the loss of quality. If for any reason I lost the 24-track, then I'd probably stick with this system and put more down live. The only problem with the 701 is editing but I transfer the recording to my Tascam 52 and edit up a listening copy on that. If it's good enough for release, I can always take my 701 to a digital editing suite and have it done professionally once I've decided where the edits should be.

Decisions

How did you set about choosing the equipment for the studio, given that you had a specified amount to spend?

It was rather like a colourblind man shopping for matching clothes. Although I know about instruments, I'm not really up on what is available in the recording field. So I started by reading magazines such as this one to find out. It's a bit like buying your first computer; if you just go and buy the first one you see, you're an idiot. It's far better to buy all the computer magazines and see which names keep cropping up. It rapidly became apparent that the Soundcraft 24-track was the budget machine to go for, though I know Aces have now brought out something even cheaper. If on the other hand you buy second hand Studers, they're likely to be clapped out after a long and active life working round the clock.

"Take the Sony PCM701 PCM encoder, couple it to a video recorder and it will give you stereo recording quality beyond what would have been George Martin's wildest dreams..."

The choice of console came down to Soundcraft or Soundtracs and I chose the Soundtracs CM4400 after arranging an all-in deal with Don Larking at Luton. The advantage with buying a ready-made package is that you know that you can just plug it in and it'll work. The desk has computerised muting and routing and it's never caused any problems other than confusing a few outside engineers who weren't used to it. You can set up the mutes for a mix and then step through manually which is almost as good as automation. Another advantage apart from the good price is that you can use the monitor channels as extra line inputs on remix. We've had to get the desk modified for -10dB operation to make it compatible with everything else we've got. It's a good desk but it's not perfect. For example, as you buy it, the Solo level is much too loud and so someone has to go inside with a soldering iron and modify it. There's no switching for alternative sets of monitors and no way of patching in your cassette machine; we've had to use an extra patch bay to get around this. Another major beef was the amount of crosstalk the desk had when it came and when I complained about it, Soundtracs came down and did a standard mod, but why they didn't tell us about this in the first place I don't know. The worst problem was channels leaking on to the auxiliary busses and that gremlin is still present, albeit to a very small extent since the mod. I also use external parametrics to lift the top end as the internal EQ doesn't do quite what I want. You can't beat it for the money though; you get a computer and a free patchbay. You can't go wrong.

Effects Rack

Let's take a look at the effects rack, there seem to be a few familiar items there.

As far as noise gates are concerned, the Drawmer dual gates seem to be the clear market leader and they're British. I like to buy British where I can. They're ergonomically well set out. They're rather like Anglepoise lamps: perfect for their intended job.

Don Larking recommended TC parametrics and we've had no reason to complain about them at all, they just get on with the job.

As for compressors, DBX 160Xs are what we chose and we're perfectly happy with the way they work. I'm not one of those people who can tell you all about the characteristics of different compressors, I just like to compress the vocals and the odd guitar track. The DBXs don't give me any problems so I'm happy.

In the reverb department I have a Roland SRV2000. I reviewed it for Roland's Newslink and was so impressed that I asked Alan Townsend at Roland if I could buy it. I think that it's really underrated; if I had to choose just one reverb unit, I'd go for this one. You can even get reverse reverb effects by setting a negative decay time on Non-Linear though the manual doesn't tell you this. It's a fabulous machine. Underneath it is the REV7 which I consider a bit harsh as a pure reverb. However, it excels at ambient treatments and the flanged reverb is just out of this world.

The Yamaha D1500 I bought when they were being offered at silly prices and anyone who didn't buy one really missed out. At £300 for a programmable digital delay, I simply felt unable to turn it down and I've been pleased with the way it works. Oddly enough, though it seems to change the bass sounds, the top end is fine; I don't know why.

I wish though that we didn't have to put up with MIDI with everything because I'm sure it adds to the cost of the equipment and there's a rack full of MIDI sockets over there doing nothing. Some manufactures could make a fortune by selling a remote control unit that allowed you to change patches on all these machines via MIDI, even AMS doesn't have that. You could set up each unit on a different MIDI channel and just use a little box like a TV remote control or something similar to switch settings. It might even be cordless using infra red.

Of course I've got an SPX90 and I'm surprised just how few people have realised how good the harmoniser section is. Up till now you had to use the Eventide or AMS if you wanted reasonable results. The Yamaha glitches a bit if you set up big pitch changes, but then they all do. At small pitch shifts it fattens things up amazingly and I think it's worth double for the harmoniser alone. I haven't yet used the sampling facility, though that's obviously a useful inclusion. Tony Swaine of Swaine and Jolly likes the reverb very much. He'll hate me for telling you this, but his favourite setting is REV3 Vocal on a 400mS decay. That's supposed to be a trade secret by the way. There's nothing to touch this machine at the moment, yet there are still people complaining that you can only get one effect at a time. But what do they want for £600. The dealers are having problems because studios are coming back and ordering another five or so units and supplies are still not able to cope with demand. I could use another three right now.

The Bel sampler is a useful tool and whereas I might think of an Emulator as the equivalent of an artist's mixing palette, the Bel is the palette knife that lets you take a little piece of sound and put it here or there or wherever it is needed. If the tuning's slightly out, just tweak the pitch before you replay it. Likewise you only have to get a difficult phrase right once and you can trigger it whenever you need it. It's the equivalent of spinning in short sections from tape but without the problems of cuing up the start or waiting for the machine to get up to speed. Having said, that we do spin in from tape too.

"You notice very quickly if someone tries to make a song work by production alone when the basic musical ideas just aren't there in the first place."

Monitoring

I use a Quad 520 amp driving a pair of Tannoy Super Reds. They're good speakers and were the best value at the time. I'm not a Tannoy man as such but they offer the right performance at the right price. They also sound good in this room and other studio users have spoken well of them.

My desk top monitors are Yamaha NS10Ms with the obligatory tissue paper. The NS10 has become an industry standard so even though I'm not a great fan of the sound, I think it's a good choice simply for that reason. I would have liked a pair of AR18s but the silly sods have stopped making them now. My engineer Nick Godfrey, who I stole from Red Bus, is instrumental in getting top class results and is very exacting. He can even tell when I've left a pair of cans on in the same room as I'm recording a Marshall stack! So if he thinks the equipment is OK, then I'm happy.

This room started life as a double garage underneath your flat didn't it? How did you treat it acoustically?

I was lucky. Don put me in touch with a builder from Manchester called Eric 'O' Neil who has a good knowledge of acoustics. There's a lot of pine boarding at the front whilst the rear is covered in hessian with bass traps built in behind. The whole studio is a floating structure within the outer room. We seem to be a little false around 3kHz but you can correct for that once you get to know how the room sounds. In the studio area itself, I've got a pile of Ilsonic sound absorbing panels which can be stuck to the walls to deaden the acoustic. I'm considering using Velcro or something similar so that I can move them around easily.

Favourite Techniques

How do you go about producing a piece of music? Are there any special techniques that you use?

One little box that I must tell you about is the Archer Mini Amp/Speaker from Tandy's. It doesn't have a model number but it's important to get the right one as they do a bigger version which isn't so good. This one is about the size of a pack of playing cards and runs from a battery. Steve Lillywhite discovered it and used it extensively on the Peter Gabriel 3 album to process synths, drums and even backing vocals. It has an interesting way of distorting the sound very cheaply rather like a telephone. We then mic it up from a few inches away. It's also handy for tuning or setting up sounds.

The other useful piece of gear that Tandy do is the PZM, and what a bargain that is! Until I borrowed a couple of Neumann U87s, I used those for vocals and they are amazing. Even the popping is minimal if you stick on that little foam Hitler moustache thing that they give you. I find you have to sing close up to it if you aren't going to record the room sound but it's not a problem. They're also great for congas; I just stick a pair on the glass screen in front of the player.

Drum Machines versus Drummers

When I'm recording I use either a drum machine or a real drummer but in any case I like to augment drum machines with real percussion, even if it's only a tambourine. I'm lucky to have Steve Laurie as a drummer. He's often concerned about the lyrics and makes suggestions for improvements; the sort of drummer who sees a drum machine and takes it home to programme it. He doesn't take the negative attitude in thinking that it's doing him out of a job, he wants to be involved. Programming a drum machine or sequencer is a musical performance and the musicians union will have to recognise this whether they like it or not. Over eighty per cent of the charts use drum machines or a drummer working to a click track and I also like to work to that level of exactitude. If a beat's out of place, sample it, spot erase the original and then drop the new beat in in the right place. One thing that I have learned about production is attention to detail and not letting something through that you might accept live. If you build on sand, the whole thing is dodgy and shaky so if it takes a week to get a good drum track, then it takes a week. If you've spent three months recording something and you suddenly realise the basic track is not up to scratch, you must be prepared to scrap it and start again. People are used to hearing that metronomic beat so it's important to programme carefully so that it breathes. Playing the hi-hats live helps enormously here. Drum machine cymbals always sound the same so any naturally variety you can add helps. On the TR808 you could have open and closed hi-hat at the same time and you could also vary the accent to give a human feel. With the Linn 9000, you can manually vary the amount of openness of the hi-hats as it's playing to do the same thing. What I fail to understand is the failure of manufacturers to build in some real humanising features to let you lag or lead the beat slightly or add randomness. Some manufacturers have made token efforts in this direction but not nearly enough. The pitch should be slightly different from one beat to the next and the level and timing would benefit from minor variations. Often we replace the drum machine sounds by samples of our own drummer's kit or worse, maybe a Steve Gadd sound, and then I'll manually vary the pitch during playback to give that variety. The Bel does have a slight delay when triggered from tape so you need to perform the old trick of turning the tape over and re-recording the track to be replaced through a short delay onto a a spare track. When the tape is played the correct way, this new track comes just before the beat and compensates for the delay in the sampler. If you have a constant snare beat throughout the song, you can record an extra beat at the beginning and then delay the sampler by a whole beat, You can tell when the timing's right as you can hear your sampled sound phasing with the drum machine voice if they are played together.

As soon as you get real instruments playing around the drum pattern, it all starts to sound more convincing.

The same ideas on precision apply equally to vocals. I might record the vocal line onto eight different tracks and then select the parts that give me the best take. This would be mixed onto ¼" tape and then the whole thing spun in. If you can get the timing right at the start, the thing will stay in sync for the length of a song with no trouble.

Producing

How do you do you see the job of the producer with music such as yours which always seems to speak for itself without the obvious use of special effects or gimmicks?

The difference between playing live and recording is like the difference between a play and a film. You can't make a really good film just by setting up a camera in front of a play and letting it roll, even if the script is good. Similarly you don't make a good record by just getting a group to play a song through while you record it. You might know just what the bass drum should be playing but what bass drum sound should we choose to suit the song? As a producer you are inevitably put into the same position as a film director who tries to envisage the finished product by visualising camera angles, lighting and close-ups and so on. The actors are doing their job but the director has do decide on the best way to shoot it.

To take the analogy further, Steven Spielberg, the Trevor Horn of the film world works by combining spectacular effects with good storylines. His imitators might get all the effects right but they don't have the story or the cutting isn't quite right and it's the same in music. You notice very quickly if someone tries to make a song work by production alone when the basic musical ideas just aren't there in the first place. That's basically why I find that the Portastudio stage is still the most important part of the whole process. Unless you get the song right to begin with, no production is going to save it.