This is a game inspired by Level Devil. My version is top down instead of a platformer. I originally went with tiles to check for collisions, but now I’m using kinematic for walls and AABB for things that scale.
I would have liked to use a collision object for checking for traps, but scaling doesn’t work so well with those. I found a solution with AABB instead. It checks the size of the pivoted sprite vs the player’s pos. This works great and is probably less expensive.
I also would like to share each version instead of the latest, but this isn’t the easiest thing to do on Poki or Itch. I could do it myself if I self host, but it’s not that important.
So, here’s how the latest version looks and plays.
While I have 15 levels done, I’m only testing the first 5 currently in Poki. This is helping me focus on the most important moments for the game — the first experience. I will continue to improve these initial moments until the majority of players finish the first 5 levels.
It’s looking promising so far.
In one of my other professions, copywriting, a popular rule is that the headline is where 80% of the ad’s effectiveness comes from. In other words, the 80/20 rule.
Average playtime is picking up, so I’m adding 5 more levels with the theme Spikes.
Focusing on the first 5 was key as they take an average of 2 minutes to complete. So if I can get 25-50% of the testers to complete the first 5 levels, that’s a good sign I’m heading in the right direction.
Now I want to see how many players continue beyond the first 5 levels.
This is a challenge. Sometimes the playtests look promising and other times they aren’t. It’s hard to identify what exactly the issue is. It’s a lot of guessing, testing… and luck.
I feel like the mobile controls could be better. I’m seeing longer playthroughs on average from users on desktop. I think this is because the mobile controls can feel confusing or not as accurate, which frustrates the player.
There are smaller things I can do such as add sounds or effects, but I think the improvements from that would be minor. I still make them from time to time, but they’re not the focus.
It also seems like making the game easier helps. I think the formula for this type of game is easy, but with a subtle trick. The trick/trap shouldn’t be a huge surprise, like a random tile disappearing, but something that can be avoided if you’re paying attention.
I’m going to try and remove the camera following the player. I think it may be causing some confusing with how movement works. It makes it more difficult to follow what’s going on. I strongly believe one of these many “minor” changes will eventually make a big difference.
Hi! I tryed your game on mobile, and yes I think the stick control is too small and too sensitive.
i am thinking i just touch my finger in the middle of the stick, but the character move up because i am not right in the middle!
→ you should make the disc bigger and have a “dead” zone in the center where there is no movement taken into account.
So we can put the finger in midle, then move if we want.
I add that the place of the stick should be configurable! I want to use my right hand, not left. And the stick is too much on bottom of screen for my bigg hands…
I think if you change these 2 things (sensitivity, size, dead zone, configurable position) it will be really nice!
I stopped playing because of the controls! But the game is really nice
There’s definitely more work to do with the controls. The played time durations are higher on desktop.
I’m not sure how to “fix” the mobile controls so they feel right. I can see players struggling in the video.
Maybe show 8 direction nodes that show where the player can move and the nodes light up based on the current direction. Some visual feedback like this might help.
Nice that’s much better!
But there is a bug with the stick position, it appears above from where I touch the screen.
Maybe that’s a scaling issue or something…
That’s actually intentional – the stick appears above the finger so you can still see it. I read about that being the standard in game/app design. Maybe it’s too high above the finger.
This has led to the longest durations on mobile, but it’s hit or miss. They either play for very long or not at all. More testing is needed.
Based on feedback, I will make the levels even easier. Seems like it’s better to shorter levels for this type of game. When I get stuck, it helps to go back to Level Devil and understand the psychology behind the level design.
Short level
Something unexpected happens
A secondary unexpected thing may occur based on the previous mechanic
Latest player fit test hit a new record of 1:59 average time played. I think adding more levels would help get the average time played increased based on the 10-player test review.
Making 25% of the players hit 3+ minutes is harder. Maybe if I can identify a certain audience that plays this type of game more that would help. I will also continue to make small adjustments to the levels to make them more interesting and fun.
Seems like moving to a d-pad helped with mobile. The stick, as I had it set up, was too sensitive and frustrating for a game like this.
As for more levels, I’m adding 5-10 more soon. The next 5 will have the theme “Pits”.
I’m now reworking some of the levels. The “trick” should be based on thinking, not controls. I would rather have the player fail because of a bad decision, not because they didn’t move their finger correctly.
Hopefully this helps push the play testing durations closer to 3 minutes. The challenge continues.
I think adding more levels will increase the playtimes of those who play until the end, thus increasing the average time played.
The 10-player tests are useful, but also too small of a sample size to make major decisions about the overall game. Here’s how I use them.
For durations <1 min, I check to make sure if there’s anything confusing about the gameplay. Some players are just not into this type of game and that’s okay, but others may want to play but can’t figure out the controls.
For longer durations, I check to see if they got stuck on a level. It’s okay to get stuck and die a few times, but ideally once the tricks are figured out the player can beat the stage, as long as they remember what to do.
I’m thinking of changing the name and maybe even the art style too. This game is inspired by Level Devil but I don’t want to make it too similar. Maybe a name like Tricky Dungeons would be better.
The current art style is okay, but maybe I’ll switch to vector art style (Inkscape) instead of sprites.
I discovered a game from 2007 that has a similar “troll” like appeal.
After watching some videos, I think adding jump and shoot actions would help with the level design and re-playability by adding a time completed to each level. Maybe a leaderboard too. The actions would be unlocked and players can replay previous levels with the actions, trying to improve their time.
Also, seems like adding more levels helped. I still have players who drop off early, but I guess that’s to be expected. In some of the play tests I’m seeing players stick around for longer.
Adding sounds seems to have made a difference. Also selecting “new players” helps with increasing duration played, I think. These secondary updates feel like cheating because I’m technically not updating the game with more content, but… they work. Seeing the stats increase is also motivation for me to create more content.
I’m going to experiment next with seeing if adding “platformer” as an audience helps. That’s the same category as Levil Devil, so it might make a difference in attracting a similar type of player.
Some significant changes lately. First, the name change. To avoid confusion with Level Devil, I’m changing the name to Tricky Temples. This new name also matches the new aesthetic and style of the game.
While the overall goal is to improve time played, these changes will improve the branding and style which I think will help. It’s also more fun to be creative and apply my own twist.
The new style simplifies the art, which allows me to focus on the level design and gameplay. Each level uses two main colors only now – a decision was inspired by Tomb of Mask.
I have to update all the tiles, but the good thing about a simple game style like this is that won’t take as long as it would if the times were more complex.
