Back in 94 three of us released Prince of Destruction -- a "big" game that was initially released as shareware and was later commercially published. (When I say "big" -- it takes a playtester who knows exactly how to solve every quest about 10-12h to play through it at top speed; it generally takes 24-48h of solid play for a player starting cold to finish it assuming they don't get stuck. That's more content than most commercial titles -- and you're not walking back and forward through mazes to flip switches.)
We'd initially estimated it would take us six months to complete; it ended up taking three years. I'd say that if we'd been working on it full time it would have taken 18 months.
1) We had an extraordinarily talented programmer: Andrew Barry. (Think of him as the Mark Sibly of Mac development tools; among other things he wrote RealBasic 1.x and 2.x and the Spotlight Debugger.)
2) We had two talented artists.
3) The standard of quality of computer games back in 94 was FAR lower than it is today. Today's games have FAR more content and far better architected engines.
Our game had 1400 map layouts, three internal scripting languages (items were scriptable, so were map objects -- e.g. monster spawns were invisible scripted map items, doors were visible scripted map items -- and so was npc behavior), a system for handling conversation data, and allowed NETWORK cooperative play (up to four players). We had real content development tools (in fact we made our content development tools available to our players). It won a number of awards, was reviewed in every high profile Macintosh magazine, was distributed on a LOT of cover disks (back when cover disks were fairly novel)...
I think we made a total of $30,000 from it (split three ways).
Typical triple-A titles today cost $5M. Ten years ago it was $1M. Ten years earlier it was probably $100k or less. Ten years before that it was something a good hacker could knock together in a couple of weeks given enough coffee (literally -- read the history of Sierra Online or Origin).
Most triple-A titles LOSE MONEY. Being a game programmer is the geek equivalent of forming a rock band: most people do it for love; a small proportion of rock bands ever get a paid gig; a tiny proportion make real money at it; the average professional musician doesn't earn enough to live on from music and has a "day job". This includes rock musicians and members of symphony orchestras. And guess what, the same is true for game programmers.
We'd initially estimated it would take us six months to complete; it ended up taking three years. I'd say that if we'd been working on it full time it would have taken 18 months.
1) We had an extraordinarily talented programmer: Andrew Barry. (Think of him as the Mark Sibly of Mac development tools; among other things he wrote RealBasic 1.x and 2.x and the Spotlight Debugger.)
2) We had two talented artists.
3) The standard of quality of computer games back in 94 was FAR lower than it is today. Today's games have FAR more content and far better architected engines.
Our game had 1400 map layouts, three internal scripting languages (items were scriptable, so were map objects -- e.g. monster spawns were invisible scripted map items, doors were visible scripted map items -- and so was npc behavior), a system for handling conversation data, and allowed NETWORK cooperative play (up to four players). We had real content development tools (in fact we made our content development tools available to our players). It won a number of awards, was reviewed in every high profile Macintosh magazine, was distributed on a LOT of cover disks (back when cover disks were fairly novel)...
I think we made a total of $30,000 from it (split three ways).
Typical triple-A titles today cost $5M. Ten years ago it was $1M. Ten years earlier it was probably $100k or less. Ten years before that it was something a good hacker could knock together in a couple of weeks given enough coffee (literally -- read the history of Sierra Online or Origin).
Most triple-A titles LOSE MONEY. Being a game programmer is the geek equivalent of forming a rock band: most people do it for love; a small proportion of rock bands ever get a paid gig; a tiny proportion make real money at it; the average professional musician doesn't earn enough to live on from music and has a "day job". This includes rock musicians and members of symphony orchestras. And guess what, the same is true for game programmers.