Zaga-33 seems to have gone down really well - I've had lots of positive comments about it, and I'm just really happy about that (and no really negative comments, which is even better). The iOS version's sold ~1200 copies so far, which I guess isn't super amazing or anything, but I'm still pretty pleased with it.
It's a bit of a weird one for me because the game isn't really "innovative", at least not in the way I usually try to work - it's more of a fresh combination of old ideas rather than having new ideas.
Just wanted to make a few more comments about it, address a couple of things that have come up in user reviews and such. Quickly written poorly-edited post.
Gamecenter. Before release of the iOS version Brandon recommended I add an online scoreboard thing because people like that. I decided no. A few user reviews have said they'd like it. I'm still leaning towards no.
I decided no after listening to Bennett Foddy's GDC talk this year, in which he attributed some of the popularity and social-mediability of QWOP to the lack of any built in social media elements which forced people to find their own ways of sharing things. Lots of videos posted to youtube of people inching their way along on their knees terribly slowly, which they were proud of because they thought they'd beat the system. If there's an online leaderboard, instantly everyone can compare themselves to the best in the world and feel inferior with their lesser achievements.
So in Zaga-33, I'm seeing people being proud of getting to level 17 say, and feeling like it's an accomplishment - and I totally support that, it *is* an accomplishment, level 16 is damn hard. If they knew that @rocketcatgames had scored 40pts within a couple of days of release it would deflate that whole sense of achievement. If you're familiar with playing roguelikes, it's not too difficult to complete it; the leaderboard will full up with COMPLETION scores in no time; without a board everyone can just compare to their own best and feel good as they improve.
Also, there's a limited range of scores. Score is "level reached", plus if you kill the end-boss 5 points for killing him and 1 for each item left in your inventory. I think leaderboards work best if there's a wide range of scores with more incremental improvements. I'm by no means ruling out doing it for a different game.
Lack of a "wait" button (or diagonal movement). This makes combat a bit weird because the game board is bipartite; monsters move whenever you move so if an enemy has the same parity as you you'll never be able to get next to it without it getting a hit on you first, unless you find some way to switch your parity (fighting a different enemy or using an item). It pushes it away from 'realism' to more of a puzzle. I think most roguelikes go through this phase early in development before a "wait" key is added; in keeping with minimalism I chose to keep it and make the weirdness that comes from it be part of the game. If you could wait a turn, an item like the nuke would be way overpowered given the deterministic combat system - it would reliably let you clear a level while taking no damage.
I don't want to; minimalism is most of what makes this game any good. I totally get that this is the kind of thing people like to see in updates, but just adding more stuff indiscriminately would break the game so I have to be careful.
Lot of people complained about the touch controls. Something I'm learning making iOS games - no matter what controls you pick someone will complain about them. I don't like adding lots of different control options (Hard Lines has ~8 and they all have weird names so you can't guess what they'll do without trying them out). But I added swipe to move as an option you can turn on from the titlescreen, that seems to have cheered people up. I find the original controls (touch somewhere to move towards there) feel really good on iPad, but I can see why they're not ideal for everyone on iPhone/Pod, especially if you're trying to play one-handed.
If you're using the original controls and having trouble exiting levels - just tap on the @ when it's by the exit to exit. There's not touch-sensitive material past the edge of the screen, so obviously you can't touch there to move there.
One review suggested: "some kind of benefit to killing monsters, as it would another dimension to the game".
Most people seem to appreciate this but I wanted to address it anyway: adding a benefit to kill monsters would actually *remove* a dimension from the game. As it is, you have a choice about how to get past enemies - fight them, use items, or try to stealthily sneak past them. Or some combination thereof. It's a strategic puzzle to find the best approach to each random level while conserving resources (items/hp). There *is* a benefit to killing monsters - it removes them as obstacles. Most roguelike (or generally RPG-style) games give direct benefits from killing, usually experience levels making your character more powerful. Generally this means you're almost required to kill them; otherwise you fall behind in power level and won't be able to keep up as the game gets more difficult. It removes a choice; bypassing encounters is suboptimal. Omitting this killing-bonus restores that choice.
(aside: in some tabletop RPGs experience points are awarded for completing encounters regardless of the method used - whether you fight or sneak past or sweet-talk your way out of trouble; this is cool.)
The area-damage items (laser, bomb) damage monsters if they move into an affected tile on the next turn. I'm really pleased with this. It came about by accident (remind me to post about creating in a way that admits serendipity next time I'm ill). Because the monsters move straight away once you move they were overlapping the explosion animations and it just looked off. Obvious solutions were to delay monster movement, to make the explosions animate faster, or to make monster AI just not go there, OR to do as I did and make it destroy them as they move into it. Turns out this made for a really interesting mechanic! It interacts really well with the game of understanding the monster AI and predicting where they're going to move. Generates a moment of excitement when you use these items, because there's an element of gambling, you don't know what the precise effect will be because the monster movement has random elements. (Also made the items a little stronger, which felt generous.)
Whenever I play another roguelike now I expect area-effect powers to act like this and it throws me for a moment when they don't.
I really like that "how you have to limit yourself to something quickly" overlaps so much with "good design". I've started making roguelike games before and they tend towards complexity - all these different item types, spells, experience levels, character classes, etc. Just sticking with single-use items, a linear sequence of levels, deterministic combat, all monsters having 2hp - these were decisions to make sure I could complete the original game in a week, but they turned out to be really good design decisions as well. There's a lesson there about both the value of jams and how to do design in general. Roguelike Radio did an interview with Glenn Wichman, one of the creators of the original Rogue recently; what interested me was how much of the design of the game came out of the hardware restrictions of the time - things that turned out to be really elegant design decisions were forced upon them by technology rather than deliberately chosen.