How do you choose the hardest and most interesting? By playing them yourself?
It’s automated, at most I’ll play a couple of games after making a change to verify I didn’t break anything.
The solver gives a bunch of metrics on the puzzle, and then there’s a scoring formula based on how deep and wide the solution tree was, and what kind of deduction rules were needed. (The formula varies by difficulty level, along with the size of the board). E.g. if there are 10 possible moves to make at the start of the game, which is actually pretty common for a freshly generated puzzle before it gets optimized, it’s unlikely to be very interesting.
There’s probably an endless amount of work one could do on making the metrics more human-like though. E.g. I suspect that in real life it’s a lot easier to notice a 5-length piece having to be vertical if it has no space at all horizontally, rather than having four squares of space. But right now I consider both of those moves to be equally hard.
Very cool!
Hey this is a pretty fun game! I’ve been playing easy - on puzzle 12. Is there an end on each difficulty?
How can you determine that? Are there leaderboards somewhere?
Yes, it’s 100 levels for each difficulty.
Will it remember my progress in a cookie if I close my browser?
Nice job on the game!
Yes.
Thanks!
Did about 20 hard puzzles and then jumped to hard-99 and then expert-99.
Really cool game, although I feel you want some extra restriction to come up after some point, the difficulty curve flattens somewhat in the later levels I sampled.
The difficulty just varies between the difficulty levels, not within them. There are a couple of different deduction rules you might need on expert that aren’t needed on hard.
Ah, I see!
It’s really good work, man.
Cross-posting from the roguelike topic…
It’s Seven Day Roguelike (#7DRL) time for 2019. Anybody interested in giving it a shot?
I might try to whip something up with Godot.
I didn’t even realize there was a thread here for this. Nice!
I’m traveling over the time frame for the rogue jam, shorter jams are tough for me, and I’m still very much a newb at all this, but I’ll keep an eye on the thread going forward.
I did a longer game jam last summer and had a blast, so I’m very interested in the topic from a hobby/fun standpoint, and interested in game design on the whole.
3.1 reached stable, just so you know.
Awesome! I’ll check it out tonight or tomorrow. I played around with one of the betas but they started churning them out so quickly that I decided to just wait.
Annnnnnnnnnd… finished!
For those of you who don’t know, I teach Game Design at a small college in Vermont. For our Intro to Game Design class, the main project is completing a game from top to bottom. For the entire semester, they slowly work toward this goal by following a series of projects and milestones that allow them to work from concept to completion. This semester I wondered why they should have all the fun, so I worked on my own project using the same tools and resources they had.
When thinking up what kind of game to make, I remembered reading one thread where someone mentioned Twin-Stick Farming in reference to some game. I thought that was pretty funny, so I used that as my jumping off point.
Behold! Twin-Stick Farmer!
Made in GameMaker Studio 1.4. Happy Farming.
I finally got around to writing up how the procedural generator for Linjat works.
And as always happens, while doing the write-up I found a bunch of things to improve. Most importantly it turns out there were way more levels than I’d expected where the knowledge of the solution being unique was helpful. So I ended up changing the generator such that it detects the most common such situations and either corrects the board on the fly (by adding dots) or penalizes them in the scoring function.
Hey peeps, if you are still into and supporting the Godot engine, they are currently $100/month on their patreon away from hiring a third full time developer (goal: $12,100/month). It’s going around reddit right now so maybe it will be closer once you read this message, but consider supporting them if you’re still using the tech:
Is this the appropriate thread to talk about games one is making, whether or not one did it in ten minutes? My game is taking more like ten years.