When to Release?

[dephbokks]

I am approaching the time when I will be able to release my current roguelike in development. It is the first true rl project I have undertaken. Now that I am at this point, I have been thinking about release schedules and the big question:

When to release?

One school of thought is to release early and often. This has the auspicious effect of building a community around the game who follow its progress and provide feedback. Another school of thought is to wait until it's finished or at least in the home stretch so that any people checking out the game would be impressed, blown away, etc. You 'capture' their imagination via that first impression.

The problem with the former approach is that some may check out the project and say, "This sucks," since maybe the game in its nascent stage will not be as impressive (ie the wow factor). Maybe that impression will be burned in their mind and it may take some persuasion to re-check out the project in its current stage, in a stage that may 'wow' them. It seems this approach is more often, but not always, utilized by indie studios, indie soloists, hobbyists, first-timers, etc.

The problem with the latter approach is that you may have an awesome, polished, wow-inducing game that nobody knows about. It seems this approach is utilized by AAA studios since they can advertise the hell out of it to television-droolers.

So what are your opinions? It can be in the general sense or specific to roguelike development. I've seen both extremes: games where the author(s) say, "All you can do is move around.... Yeah, it's just placeholder art, for now.... No, we haven't finalized the exact game mechanics...." and games that come out and blow you away.

Which is better? Does it even matter? (with respect to the final release)?

[Krice]

When it's ready.

[Darren Grey]

Release when it's enjoyable to play.  Roguelike players can put up with a lack of polish, and early development feedback can be priceless.  If you keep releasing often and list the updates you've made then any that were turned off by early releases will eventually come round to trying it again.  If the game is good enough it will get enough attention.  And even a supposedly polished release may turn off many, as everyone is fussy and there's absolutely no way your game would be perfect.  In general people are less fussy about what is clearly a beta release and more likely to provide helpful rather than negative feedback.

I'm not personally aware of any roguelike that has suddenly appeared on the scene full and polished.  They've all appeared in early stages and built themselves up, whilst also building up a fanbase that supports it and provides ideas and code-fixes and amendments (one of the advantages of open source stuff is that you can get a lot of useful help, especially in such a knowledgeable community).  The release early, release often model is simply the *only* successful model of release amongst roguelikes.  There are many failures as well of course, but not 100% failures.

Keeping a game hidden also runs the risk of vapourware - a rather severe risk amongst roguelikes.  If you decide to go down this route then try your best not to spend years bragging about how great your game is, how great you are for coding it, and how every other game is inferior, whilst generally being a bitter and sarcastic asshole to everyone, without ever releasing your supposedly brilliant game.

[Z]

Some people claim that roguelikes are never finished. Corollary: if you release a roguelike only when it is finished, you'll never release it.

(I don't know how much truth is in that claim, ADOM feels unfinished, but most other major roguelikes do. Actually, I release a new version of Hydra Slayer whenever it is "finished" in a sense that at the moment I see no feature that I think would be worthwhile to add.)

[dephbokks]

Thanks for the replies.

I like the line "release it when it's enjoyable to play," since about one week ago, when  my game was in a playable state, which, for me, meant a fun and interesting challenge from the developer's perspective, I found a moment of enjoyment. That quoted line ostensibly is an elaboration of the more succinct, "When it's ready."

Good post Darren, very helpful. I know you have some rl's under your belt, and I've checked out your work before (your last two entries to 7drl), so it was nice hearing an experienced perspective discuss what you discussed.

And, Z, your corollary is especially cautionary to me since I could definitely see myself just keep refining and refining, tightening and tightening, until the whole process spirals deeper.... I started this project after watching the movie Kill Bill and I thought that could inspire a cool roguelike. I said I'll make a 5drl and bang it out. Now about 120 days later it is now a 120+drl challenge.

[Pteriforever]

If there are bugs in the version, spend 1-2 days making a fix version. If there aren't any bugs, spend 5-6 days adding new features. If you can't think of any features, spend 3-4 days making a content patch. There's always more updating to do!

[wire_hall_medic]

Release it once it's fun.

I doubt anyone who posts on a rougelike dev forum is really interested in the wow factor; you get here by pledging your allegiance to function over form.  It's nice to have pretty graphics, but they don't make a game fun; they make it pretty.

Where projects die is in the last 10%; bughunting, endless tweaking, getting this feature just right . . .  It seems like that'd be the time to really have a fan base pushing you.  You can always add more, but at some point your changes stop being meaningful, and actual players are a good indicator for that.