An Interview with Neil Brennan (Melbourne House/Beam Software)

Remembering that I grew up in the ‘70s, computers weren’t really a thing when I was a teenager, except for the very small group of Apple II fanboys at my high school. One of those fanboys was Stephen Taylor, who eventually introduced me to the crew at Melbourne House many years later.

The first time I remember doing anything remotely computing-oriented was wasting hours on a programmable robot game (Robot Wars?) on Steve’s Apple II at his house on the weekend when we were in high school together. It was fascinating and addictive, which I think pretty much describes all of computing for me ever since.

As a teenager, I was a poorly-practised classical pianist who picked up a bit of guitar and singing on the side. I had a little talent with improvisation and worked hard with a jazz teacher to improve my knowledge, but loved the performance aspect of singing and playing guitar. It’s so much easier to stand up and sing with a guitar rather than crouch over a keyboard! I worked pretty hard on my singing and the harmonica. The other instruments were mostly to support my voice.

At about the age of 16 I started busking with the guitar around Melbourne, playing small coffee shops and support gigs for friends in bands. I played in a few Christian bands as a keyboard player before getting much more seriously involved with the Melbourne bands The Cracks in the Wall and Stick Figures, as well as picking up a gig singing and playing in a piano bar in the city called Adam’s Apple.

It was a sleazy dive, where drunk Greek businessmen would go to pick up teenage Greek girls. I loved it, of course. I would play for 3-4 hours, earn $50 cash-in-hand (this was the ‘80s, so that wasn’t bad), eat a free meal, and spend a third of what I earned to get back home to Doncaster by 2am, ready for another day of University in the morning. Sometimes I would hitch or walk home just to save the money, but it was around 20kms so I didn’t make a habit of it.

I got kicked out of Stick Figures at around this time (I was never that clear on why!), but ended up joining Macguffins for a Sydney tour, again as keyboard player, until finally deciding that what I really wanted to do was sing and play guitar in my own band. At Melbourne Uni I had become really good friends with Gerard Neil (Gid), a trained jazz guitarist. He and I did a bit of busking together then finally formed Redfish Bluefish, which stayed together, playing around Melbourne for about 5-6 years, through multiple line-up changes (but always including me and Gid). We incorporated players from both Macguffins and Stick Figures over the trip as well as friends we met along the way, even to the point of incorporating a brass section.

In 1994-95 I took about half a year off to travel around the world. When I got back I started my own solo project, Linoleum, which turned into a band that played from 1995 to roughly 2004. You can still find our stuff on Bandcamp: linoleum.bandcamp.com . Linoleum had small-scale success, with airplay on the popular youth radio stations and a short-lived deal with Polygram’s Mercury label until its untimely demise – the label, not the band. As I am sure is common with these things, I always felt like we were good enough to get lucky but maybe not enough to really break into the scene in a big way on just our own steam.

It’s funny but that’s not so much how it happened. On reflection, I think it may all have been Steve Taylor’s fault.

The degree in Computer Science wasn’t my first choice; my life was changed, as the cliché goes, by a book. The book was Gödel, Escher, Bach, loaned to me by Steve. I was entranced enough by the picture of real artificial intelligence that it painted to switch my major from Biochemistry to Comp Sci. The computer music only really kicked in after I started at Melbourne House. Remember, these were the days before MIDI was even a thing.

I was still playing in the piano bar at that stage, but Steve and his friend Doug Palmer had started doing part-time work at Melbourne House in South Melbourne. I can’t remember if it was Doug or Steve, but one of them suggested that I come down and meet Fred Milgrom as they were looking for a musician to work with their game designers. I went down to their tiny little office and met the very strange people working there (Gregg Barnett and Phil Mitchell come to mind, odd guys – Gregg used to like to call people treacle beak, just for instance). It was a great-paying gig for a nineteen year-old, and along with my piano bar nights I was quickly earning enough to move out of home and keep studying.

So I may not be the person to ask this (really no expert), but my impression was that Melbourne House was a book publishing company that initially put out a few computing books with microcomputer games in the back, that readers would laboriously transcribe manually into their tiny plasticky devices.

These games were so popular that it occurred to Fred that selling the games individually, without the attached books, might be a thing. And it was. They were actually more popular than the books, which were doing fine by themselves. Beam was the company that Melbourne House spun off to develop and market the games, separate from the book publishing. They shared offices, but Beam quickly took over pretty much all the territory available.

(But I could be wrong 😊 )

I think it was a combination of things. The Spectrum (as I recall, long time ago) didn’t have much room in its memory for music. But it was also a horrible device for sound, with a nasty little speaker we could only click on and off. We called it a clickatron. With all the tools at our disposal, we could only get 2 simultaneous voices out of the thing at the best of times, and that by clever PWM maths.

The C64 on the other hand had the wonderful SID chip, with multiple channels, different oscillators, noise, filters and ring modulation. It was like a tiny, little, poorly quality-controlled, synth in a box. Or beepatron, if you will. I loved it. It let me actually write interesting pieces with approximations of real instruments and had that bit more memory for me to play in, so it was always my first choice if I had one. The fact that the filtering was different on every machine was a constant bugbear, though.

I think Fred felt that there just wasn’t as much added to the Spectrum games by music as there was for C64 games, so he was reluctant to squeeze game-play or graphics for what he saw as little benefit and would drop the music, despite the whole aesthetics of constraint thing.

The ZX and C64, I think I’ve covered above. I’m not sure about the NES. I really don’t remember doing that much with it. I think I must have submitted pieces that were encoded onto it but not worked on it directly? Mists of time here.

Well that certainly made me laugh! Okay, yes, you've spurred my memory. I definitely wrote that letter, in my inimitable, charming style.

When I left Beam Software, it was to work in more mainstream computing for Library software. Books and the like. The company I moved to was in Port Melbourne, a few kilometres from South Melbourne where Beam was situated. Fred and I had a good relationship and he still needed someone to help with the music for the games, as well as the embedded code that played the music and sound effects, so I moonlighted for quite a few months afterwards, schlepping across in my new business shirts and shiny shoes. All my work for Beam was outside of normal working hours and the money was great, but I recall getting pretty frustrated if I had to spend a lot of time on setting up my workstation in the evening.

The biggest piece of work I took on in that time was porting the C64 sound module, the sequencer and synth, to the Nintendo as best I could. The intention was to support the music language I had created on both platforms so that we would be able to reuse any tracks we had created for C64 on the Nintendo as painlessly as possible. I think the music chips of the 2 systems were more similar than different, so this supported multi-platform releases of games. I remember having to arrange and transcribe music for the Nintendo games, but I don't think I wrote a lot of new music at that time. That also explains my lack of memory of some pieces I have seen attributed to me from the period; I will likely have arranged music from other composers and of course, those tracks didn't stick in my mind the way my own creations have. Accreditations were really never Beam's strong suit and over time we all become unreliable witnesses, I guess.

I started off as just the house musician, but as I acquired computing knowledge and skills from my university course, by the time I graduated I was actually useful enough to pick up programming work at Beam. As well as coding the 6502 assembly sound module and music language for the C64 synth, I got involved with the in-house downloader and cross-development suite written in C on IBM PC and took over the sysadmin work for the Unix III Pillar machine (used mostly as a file server at the time, I think, again hazy).

I think I was blissfully unaware of the changes. Beam was forever moving offices or engaged in other upheavals, so this one didn’t seem notably different from the others.

For me Beam was a happy accident. I wasn’t a big game player even after they became popular, so it was primarily a music gig that paid good regular money and allowed me to get a foothold in the computing world. I think if it hadn’t happened, I would have just tried to pick up more piano bar work and looked for computing work after uni. But they were good times, and I took the first steps of my eventual career there. I’m grateful for it all. I wasn’t aware of any other local game houses at the time, to be honest.

My next job was in library software for a huge pay rise (game devs never did get paid much), the one after that at a telco. Apart from those early days, I’ve very rarely made much of a living out of music, despite my very best efforts. I think the world isn’t really set up to support or respect 99% of artists, but you can be a pretty poor computer programmer and still get a great job. It’s not the fairest of things.

We all knew Rob Hubbard. He was amazing. Wonderful arrangements and the best sounding instruments on the C64 at the time. We would listen to his stuff and just try to figure out how they were doing it. As a musician, there’s always someone better than you and always someone worse, but honestly I always felt he and his wonderful sounding instruments were always a cut above. I didn’t really know much about the others, to be honest.

Haha. It was a pretty good place to work. Very much the no dress code, tinny cassette music always playing, pub for lunch and pizzas, beer and games on Friday kind of place you might hope for as a young developer. Starting times were not strictly enforced, but everyone worked late. Numerous people didn’t wear shoes.

The devs were smart and dedicated to their work, Fred was a gentle madman in a wrinkled linen suit and the graphic artists were very cool and let me sit with them. I was introduced to Johnny Cash, The Fall, Patsy Cline, Nancy Sinatra and the sublime Galveston by Glen Campbell, and learned how hard you can work on projects that both succeed and fail commercially, with sincerely no whisper of telling in advance which will be which.

From where I was sitting, the developers felt the pressure much more than the artists. The art usually came pretty early for most games, with late rewrites more the exception than the rule (and then usually for technical reasons), but the devs had to ride the game right down to the wire. My impression is that’s still the case in game development, but I haven’t been in that world for decades, I think fortunately, from the stories I hear.

I would normally talk through the key concepts any game with Fred, usually with an overview of the characters and artwork, and he would try to give me the kind of musical feel that he was after. I would take that as a brief and head off to my little 4-track to put down as many themes as I could muster, using whatever instruments I could throw together.

Normally this would be 8-10 pieces, which I would mix down and circle back to Fred with to see which ones we thought would make it. I would fully instrument the winners and explore related themes, then start trying to pull them down to the stripped-down 3-voice bare bones that the SID chip could support. The ideas were always the fast part, the painstaking precise work came at the end, especially trying to get feel into the mechanically rigid boundaries of our sequencing modules.