Zapster Solitaire is a game made for taking just a few minutes to play something light, and I’ve been dogfooding it a lot in that capacity. In all my playtime, I’ve found a strong point in the game’s design as well as several weaknesses, in addition to the “false coin flip” I discussed previously:
- 👍 Jacks are by far the most interesting element of the game’s challenge. Their varying badness depending on which numbers you have, as well as their ability to introduce duplicate cards in your cells, creates a lot of space in strategizing how to use your zaps.
- 👎 However, they’re made a little less interesting when they use an ace and add only one card.
- 👎 It too often feels like the RNG (random number generator) just hates you and doesn’t want you to win. I accept this as inevitable to an extent due to this being a highly luck-based game, but I think things are tilted a little too strong in that direction right now.
- 👎 This is especially true in the early game, when all ranks are equally represented in the draw pile and you have only six cells.
I want not just to address the weaknesses, but build on that strength. I’ve got a draft of a redesign that could do all those things: Cards of rank 10 will be put back in the game, aces will go out, and starting cells will be increased to 7. In theory, the early game will be more survivable and the false coin flip gone with the extra cell, and the higher 10-cell maximum will improve overall survivability. Jacks will keep the game interesting despite that by drawing at least two cards, and an average of one more (when they have numbered cards to use at all).
The only problem with the theory is… no evidence! I’d like to really study the effects of the changes, not just assume everything I said will be true. I made a plan to test-drive the new design before I implement it:
- Release the JavaScript code of the game rules implementation, so that other coders can join in implementing the rest of this plan if they so desire.
- Write bots that can play Zapster Solitaire. One won’t use zaps at all, so that I have a baseline to know what a difference the use of zaps will make. Other bots will play with varying levels of decision-making and risk-taking.
- Have the bots play lots of games with both old rules and draft rules. Study how both the ruleset and playstyle influence winrate, rate of losing due to jack powers, and anything else I can think to measure.
Once I’m satisfied the draft rules are making a substantial change for the better, I’ll publish a new version of Zapster Solitaire with them, and continue dogfooding and gathering feedback to see what real humans think!
One thing that’ll need to change under the draft rules – and one reason I want to be sure the new rules are good before I make a version with them – is the playfield layout. It’s currently done as a 3×4 grid, with the draw pile, just-drawn card, and trash pile each taking a space, and 10 cells is currently 1 more than will fit. The just-drawn card and trash pile are important for following what’s happening, and that will only get more true when I implement disabling animation, so neither of their spaces will be removed. I have a new layout in mind that could fill a 3×5 grid but won’t quite align to it – the starting 7 cells will be offset to make a honeycomb formation. I’ll post screenshots when that’s ready.
I hope you look forward to the new & improved Zapster Solitaire! Please support me on Sponsus if you want to help make it happen!