My strategy was to upload the latest version of the game at the end of every day, which I’ve been doing for a week or so. That way, even if I got kidnapped by aliens before the deadline there’d be a version of the game in the competition. That last day went brutality fast, though. There were a few tweaks I’d like to have made but didn’t have time to.
45 days full of fun, entertainment and mysteries.
Secrets i shouldn’t say;
I always leave my cheatcodes in the final releases
Because no one knows and last minute changes always break it. Always.
Oh yeah last minute changes are extremely risky indeed. I have an extra command in my game which I should have taken out but I felt it safer just to leave it in. Not mentioned in help etc and if people stumble on it they cant break anything anyways since its disabled… just a bit careless to leave such things in I guess.
Congrats to all the submitters!
You did it~!
Hope you’ll all be treating yourself this weekend for this feat!~
Having an acceptable version one week in advance? Well, yes, I did it and I only had to miss IFcomp 2022 to have the time.
Well what shall I say… last week I thought had my act together and was all ready with my entry for IFComp.
All ready that is, until ClubFloyd came along and taught me quite a few lessons on how to improve my game. Lessons learned indeed! Extremely helpful, but the backlog I had in my git was suddenly erm ballooning :D. Interesting learning experience.
One of my last changes was to change the “bypass a puzzle for debugging” command from XYZZY to not gonna reveal it yet. Nobody will ever guess it.
Rumpelstiltskin.
Is it just
>BYPASS
?
It’s wild to see how different people allocate time for programming and testing. For IFComp games I like to spend an equal amount of time on both, with a few months for programming and a few months for testing and not making any changes in the last couple of weeks.
But a lot of great games were thrown together in surprisingly short amounts of time! I’m always shocked to remember Jon Ingold saying he wrote All Roads in a day (although testing was longer), and a bunch of other games I personally like were made in short time. I’m very impressed by people who can handle that!
A challenge! Sounds like a bonus puzzle.
Well let me just whip out my handy English-Hittite dictionary and I bet I can figure it out!
What’s the Hittite word for “not gonna reveal it yet”?
It’s not actually hard to guess; I just don’t want to tempt people before the games are even out. (But I left it in for a reason: if you play the game to satisfaction and want to come back and see all the other endings, it lets you bypass all the puzzles and just explore the ship and figure out what you can do.)
I still wasn’t able to digest every ClubFloyd suggestion but I think I got the main ones.
I have a lot of very minor issues in my Github repo, which I will open once all the entries are released. I found 2-3 on the final day which were of the “too high risk low reward” variety.
One of my testers pointed out something that I think was a great final day fix. The THINK command mentions when you tried the right command but hadn’t solved other puzzles to prepare fully for it yet. But it could/should have mentioned where to use that command.
So “You can XX YY ZZ (once/now conditional) you have the right friends” became “You can XX YY ZZ in (here-in of xyz-location) (once/now conditional) you have the right friends” pretty easily with this simple code I already had:
to say here-in of (rm - a room): say "[if player is in rm]here[else]in [rm][end if]"
The thing I’d really have liked to have was Invisiclues-style HTML, since my entry is, for better or worse, far less linear than the previous ones. So a walkthrough could spoil things if people don’t do it in the order I thought! I wish I had realized this in July, once the game was technically complete.
But I still have 2 days to whip something up before people see all the entries!
I’ve found I’m able to sustain that for short stretches if it’s something I’ve thought about for a while, then I suddenly find the right idea I can riff on. Certainly my progress with games close to the z-machine ceiling generally have about 3 or 4 stretches where I create the main puzzles and get in a flow.
If you want to borrow my HTML-InvisiClues generator, you’re welcome to it. It’s not amazing and the CSS and JavaScript could really do with a redesign, but it works.
Thanks! I have one of my own. I wrote it in PERL, and it transcribes slightly marked up text to HTML. I pulled some code from something Juhana Leinonen had, IIRC.
It might be interesting to compare ours. Maybe someone could look at them and combine the best features.
(Just now, I seem to remember your Scroll Thief invisiclues were quite nice and used it.)
Absolutely. I should go make a new thread for that; my code was written ages ago for Scroll Thief and uses basically bare HTML, so I’m sure someone using proper CSS can make something fancier.
Crikey - I couldn’t possibly enter every year!