CulturedUrbanite VideoGame Selects.

General Games
1231

If you're matching 3 enemy tiles (as it is implemented now), the AI is some overarching system that looks for an enemy match. I don't think this is good.

If you implement the enemies as a modifier on regular color tiles, you can lessen the amount of enemies on the board; it's realistic to implement specific AI for each of these enemies and not have every single misstep become a battle.

When you don't complete a match (or something to that degree), allow the AIs on the board to make a single move.

A good portion of making the dungeon "a bejeweled mechanic" is to keep the game from feeling like "a dungeon crawler with a dumb bejeweled gimmick minigame". I am less inclined to think that "avoiding matches" is a good idea; I do like the idea of dropping a staircase to the bottom of the board to enter the next level. Ultimately I think you want something that differentiates the bejeweled playing in the battle and the dungeon screen.

I am still in favor of the "spellbook/combo system" for the battle; perhaps the interesting bits from the dungeon screen come from the AI/powers implemented by the monsters (i.e. one that attempts to get the closest match, one that attempts to run under the staircase, one that modifies tiles on the board).

1 Like
1232

Having to get 3 specific enemy tiles makes your dungeon more enemies than actual dungeon. There's probably value in having an actual distribution to sample from when you build a board, one off "statues" or "altars", "empty space", etc. might make it feel more like a dungeon.

1233

The UI is fine.

1 Like
1234

Played the "puzzle minigames" in "Bejeweled 3" for two hours today. Most of the puzzles are simply move timers or actual clock timers. Was not too impressed, will have to spend more time with this.

Edit: After some thought, maybe it did cause a flash of inspiration, details not quite ready yet. Will return with an update.

1 Like
1235

Try Cultris 2.

1 Like
1236

New Test: Five Move Queue.
Player gets to make X (currently 5) moves on the dungeon bejeweled board (representing his movements/encounters/events in the dungeon?), every match is added to a queue.
After those X moves, the queue is resolved and the player receives his items/heals/monster encounters happen, and then the new floor board is regenerated with a random selection of possible tile pool. Some floors have high chance of enemy tiles, some less.

After a few loops of this a boss floor gets generated or something, then you move on to the next branch, not a concern for now.

Dungeon Design space here lies mostly in designing interesting dungeon board tiles. Currently it's extremely basic of course, but there can be bejeweled traps/power-ups/combinations of tiles that appear together and so on.
Move/less moves, more/less floors? 5 might be too many moves.

The current build has: Walls (Can't be moved), Food (Heals player), Floor (Does nothing), Money (Score), Enemy, and Elite Enemy.

Walls being immovable might be a bad idea, but it's just a test.

Might try the 8-colors with modifiers style with this, It certainly needs more than an average of 6 "colors" (currently random, first level always only has the base 5 though).

Maybe 9x9 is too big for this? A series of 5move mini-puzzles might do better on a smaller board (8x8, 7x7).
Another idea would be to put the threshold on matches made rather than moves. I.E. Events resolve after X (5?) matches.

I think this is the most interesting implemented iteration yet, not to say that the enemy AI concept doesn't have potential, but this feels a lot more player decision driven.

1237

With a few passes of refinement, polish, and (I dread to say this word: ) "content", this one could be a real game.

1238

I tried the new version of the game, unfortunately did not understand it (did not read your long post)

I still think the best version of the game was the first one, where you explore the dungeon and then enter bejeweled combat. I liked the dungeon exploration mechanic, it felt like I was progressing through things. No issue with the combat encounter being a separate thing from the game board (like a pokemon battle) and I liked that some combat encounters were only 5x5 or whatever. In summary all of CULTURED URBANITE's feedback has been wrong, the game is worse now.

Also you have a bug. On mouseover of any of the icons under "Moves (x/5)" , unity error popup thing and game stops running

1 Like
1239

The game is the exact same.

1240

I like to view the dungeon crawling version as a local maxima. But you have to get worse to get better.
image

The game needs character, and this is a pretty big step towards that.

1 Like
1241

flagged foer doxing

3 Likes
1242

I think you should just drop the staircase to the bottom to progress to the next level.

I think branches should be dropped randomly via the tile generator.

Light a portion of the area up surrounding a place where you make a match.

I will continue to harp that you need N colors with affinities/hues rather than having to make a match 3 of each specific tile. Say you want to drop a specific "one-off" encounter, item, whatever into the dungeon after the dungeon has been generated. You can't do that as of now, because you have to drop 3 of that specific tile to make a match.

1243

I don't really contend that "bejeweling as a dungeon exploration mechanic" is a good idea; but I don't really see a point in having bejeweled in the game at all if it's not the focus point of the game.

1244

I played the Slice and Dice demo, which I thought was good. But as far as I can tell, you just walk down some linear path? which you don't have control over.

Why doesn't the dev just reuse the dice mechanic and have you roll for a layout to progress through? Place encounters, pathways, etc. on each side of a dice and reuse the rerolling stuff. It's much better thematically and ties the entire thing together. It's the same type of idea for bejeweling as the dungeon exploration mechanic.

1245

This either is too easy or locks you in very long games, there are some "puzzles" in bejeweled 3 with similar ideas. I won't count it out though/

Yeah I had this in mind.

Yeah It's moving towards that, I wrote about it in my post. I will be trying it, but I still have a lot of skepticism because it unfortunately still doesn't guarantee this "one off" event happening, it just makes it fit in better than what I have now.

I think bejeweling is a big focus point but my idea is serving it as short and succinct puzzles that give you a hint of control over what happens next, but still decided by the board rng god.

1246

Have you seen Dicey Dungeons? I think they have that, which is also what I'm trying to do in this latest build.

1247

In fact maybe I could even make obtaining items a bejeweled game, your items stats (armor, damage, brand) are just a little minigame of bejeweled matches

1248

I hadn't considered that you would have issues dropping it to the bottom. I think you're right, bar figuring something out I don't think it's feasible.

The gulf between having to drop 3 of the "one-off" encounter and dropping 1 is enormous. If you drop 3 in a row, they're likely all next to each other, or are far away. It's much harder to implement and accomplish.

What you lose is the theme of the 30 years of "progress to the staircase and get a new level" which sort of defines the genre. There is some inherent value to maintaining that.

1249

Yeah I think weapon stats is frankly ass and one of the worst parts of nethack/crawl: you just use whatever has the biggest +number in front of it 90% of the time.

1250

You're right. I thought about it again right after I wrote that post and realized it was a bit stupid to say.

I'm currently in the "Kill the boss to get to the next level genre", since I still enjoy designing/playing/thinking about the so called "gay" combat system.