Hello! I started this last week and I now have a prototype ready.
Feedback welcomed and encouraged.
Desktop version only at the moment.
Every day you’ll be presented with a new puzzle based on the day of the day, day (number), and month. Solve the puzzle by adding all of the pieces without them overlapping.
Press a piece to select it. Once selected, move your mouse to position the piece. You can press Q and E to rotate.
This prototype got accepted for Poki play testing. I’m now setting the game up to work for both mobile and desk. While the game is simple, there’s a lot to be done to get it ready for release. The best advice I could give to a game dev who’s struggling to release a game is this: start with the most simplest game possible (even if it’s Pong) and get it released.
That is great news! But why are you not targeting both mobile and desktop?
I plan to!
That was a typo. I meant “now”.
I developed desktop first because it was easier to set up the controls. Mobile requires some creative thinking for rotating and flipping the pieces. I’m already working on a mobile version.
Making progress with getting the game ready for play testing. For a simple game there’s a lot of not-so-simple mechanics to figure out. Perhaps making an MMORPG would’ve been easier.
The main thing to figure out was rotating the pieces. I ended up with making it so a tap rotates the pieces and dragging a piece moves it. Seems simple enough. If I had flipping, I make it so a double tap flips a piece.
Some of the small, but important, things to consider:
Before
After
I switched from depending on collisions to determine if a block is on a valid square. I am using code instead. This is more reliable and also makes things easier later.
With all the work I’m doing with blocks, my next game will definitely be block/grid related.
Interactions are now smooth and intuitive. Getting closer to launching the play test version. A little more polish need, but not too much. Goal is to get it live asap. I’m sure there will be changes based on player feedback.
Play the playtest instead. It’s ready!
Loads very fast.
Within minutes of requesting videos, I got 10 ranging from a few seconds to a few minutes. The initial feedback was valuable. I saw that some players couldn’t figure out the controls. I added some basic instructions to make it easier to understand.
Hope you add a hint system. Even if its just for 1 block to know its correct position haha
It’s possible, but I’d have to write the code to figure it out. There are also 1-20 possible solutions depending on the day. The code would have to choose one and lead the player toward it with hints / rewarded ads.
Added a better tutorial. Most players couldn’t figure out you can rotate blocks by clicking/tapping.
Getting closer to an official launch instead of play test. I rewrote a lot of code to make the game more modular. It was also important to rewrite the code to determine the hint system.
When a hint is requested (by watching an ad), the game will determine a solution and provide where one of the blocks goes. The player can watch more ads to get more solutions. This is the idea and it’s in the works. I hope to have it ready within the next few days.
Average duration of gameplays are improving the more I update the game. The idea I had to add a hint didn’t work as expected. I wanted to have the game find a solution when requested, but due to how the logic works, it could take several seconds to minutes.
I came across several solution codebases for this type of puzzle. None of them were in Lua. Even if I did convert them, the delay in waiting for a solution would not be feasible. Therefore I will likely run the solutions myself and store them in a module. Then I can provide them to the player when requested.
Alternatively, I could break the daily puzzle into multiple puzzles that gradually increase in difficulty. Another to-do for this game is to allow rewarded ads to unlock different skins.
But for now, I’m going to try and take this game to the next step in Poki’s launch process.
I’m ready to move onto the next game. It was an amazing learning experience.
Web fit test results were slightly below average in some areas and above average in others. With a few tweaks, the game should be eligible.
One important thing to add is the hint feature. This is challenging because it requires a tool that solves the puzzle using code. I found some of these tools on git and after working with several of them all day long, ended up finding one that works with my game.
I initially wanted to run the tool from Defold, but this was too difficult. Perhaps it’s possible now that I think about it — but for now, I am running the tool offline and building a database of solutions.
This will require me to update the game every so often to make sure it has enough solutions for the next few months or year. Getting all of the solutions for a month takes a few minutes.
I already set up the game to read the solutions and update the squares. Next step is to work on the visuals and showcase the solution one block at a time — i.e., watch an ad, get 1/10th of the puzzle solved for you.
Screenshot shows how the squares populate based on block number. I will do something better to show where the blocks go, like change the square color or maybe even place the block for the player.
Second shot is how the solutions are compiled offline.
While waiting for another web fit test result I started another game. But after a few days, I decided to get back to work on this one.
One of the issues I think right now on the latest version – https://poki.com/en/g/daily-block-puzzle — is that it’s too difficult. My solution is to keep the daily puzzle element while adding a “classic mode”.
80% of the code is there. Some tweaks are needed to make different levels and grid variations. The refactoring will be worth it. Because once it’s done, there’s no limit to the amount of puzzles I can add.
Here’s the super simple and almost-finished first level.
I’ve also switched to using tilemaps instead of an x,y system. This will make it easier to build the levels.
One of the motivating factors behind sticking with this one game is this game called Hexasort.
It’s a simple concept with a lot of levels and have 10m+ players. I’ve also tuned into a podcast above mobile dev that talked about this game making $500k USD a day. Not sure how much of that is revenue, but I would estimate $50k-$100k+.
I’ve created another version of this game that has a system of progression. Today I added 40 levels. All of the levels are fairly easy.
I uploaded the game to Poki to run some play tests. I’m interested to see if players are more engaged with easier levels.
Creating the levels is much easier now that I’m using tiles, but the process of creating game objects and updating the tilemap numbers, factories, and so on is still tedious. I wonder if something like Tiled would be useful in this situation.
The tilemap system I created works very well. I map out the level using tiles and when the level loads it reads the tiles and adds game objects (squares) accordingly.
If the average playtime is higher with “Block Puzzle Classic”, I’ll launch this version instead and have the Daily Block Puzzle as an optional challenge.
The second play test of 10,000 users finished and results were unfortunately similar to the first. I think it’s because the puzzle is too hard for the Poki player base.
The second game, that I branched from the first, is under play testing. I expect to get results next week.
While the process for play testing on Poki is easy, it takes time. They say 2-5 days, but in my experiences it’s more like 5-7 business days.
The third web fit test finished and results weren’t good.
Need to revamp the game to get the play time increased.
95% chance I will scrap this type of game.
Today I worked on a version that mixes puzzles with combat, but not sure if I’ll finish this idea.
This type of gameplay would progress you through the stage, give you puzzles to solve, and reward you gold based on how fast you complete the puzzle. You then use gold to buy skills. The better skills cost more gold, of course.
I don’t now if the puzzle element is a good fit for this type of action game. Or if this type of puzzle game is a good fit for Poki.
