Lollipop Ninja is probably the hardest game to post a score on, as it seems to take around 30' for most people to beat the first time. After finishing it twice I finally feel like I have the hang of it. So, a couple tips:

• The cooldown for Smoke bombs is super low, so use them A LOT. Since they can both be used on enemies and on yourself, you can basically advance through levels hiding behind smoke. The more confident you'll get, the less you'll have to use that.

• Going invisible is mostly useful to advance tricky sections with lots of jumps/wall jumps, until you can finally reach either a safe spot, or a flat place to hide behind smoke again. The cooldown is quite high here, so if you run into trouble try to hide in smoke waiting for the skill to ready. Again, with more confidence you'll realize some enemies can be passed without using the skill at all, by just rushing through.

• Sometimes it's hard to figure out where to go next, leading you to get lost a bit or end up in a hole. Some guy did a video speedrun of the game, so if you're getting too frustrated with the trial and error feel free to check that.

• The very final jump of the game requires a specific wall jump technique. It took me a lot of practice so just smoke the guy below the jump before each attempt and you'll be able to try for as long as you need. Basically the idea is to hold Left after the ninja has jumped, and then immediately switch to Right to reach the opposite wall. Again when at the right wall, keep holding Right when you jump, then immediately switch to Left. After a couple back and forths you'll be on that sweet lollipop.

EDIT: Some more tips courtesy of @Raindrinker

• You can walljump on top of a non-smoked enemy for a looong time because the sight line collides with the wall or something.
• Also vertical shafts are super safe, so waiting there for your invisibility cooldown to restore is quite useful.
• Going invisible resets your alert rate, so sometimes just running towards an enemy and pressing the skill in the last moment possible is enought to escape his gaze.

---

Bonus: On Disappainted, use white paint to leave a slightly visible mark on the canvas. It can help you put down the contour of your drawing before you fill it with actual paint.

There's just an hour left to vote! If you haven't already voted, you really should. If you have, though, why not try to improve your score on Prison Breakpoint? :D

To get you motivated, here are the top 5 scores for each level so far ;)

Level 1

Player	Score
Wan	100
Thomas	100
Rain	100
vede	100
kdrnic	100

Level 2

Player	Score
Bubba	100
Wan	98
vede	98
Thomas	96
Ferri	96

Level 3

Player	Score
Bubba	100
Thomas	97
Wan	96
vede	93
Aurel	82

Level 4

Player	Score
Wan	215
Thomas	210
Ferri	206
vede	192
Bubba	142

Level 5

Player	Score
Wan	100
Bubba	100
Ferri	99
Thomas	94
vede	92

Level 6

Player	Score
Bubba	139
Wan	138
Thomas	134

Level 7

Player	Score
Wan	200
Thomas	186
Bubba	144

A remake nobody asked for

I'm planning to submit my MiniLD 74 entry, Cooperoids for the first AKJ tournament. Not happy with the way the game currently looks, though, I'm releasing a brand new edition. Introducing Cooperoids: Tournament Edition (oooOOooh).

In brief, expect (hopefully) fixed audio, better graphics and more intense gameplay (full description here). Enjoy! :)

I just realised the music was not there on the version of BDSM that was uploaded. There is now! It's music generated by a web that uses AI to create background music, so it's as fitting with the theme as it can possibly be.

Maybe you find it useful for your games -> https://www.jukedeck.com/make/tracks/browse

Just wanted to share! :D

And so I have jettisoned my fragile game, Prison Breakpoint, into the internet wilderness, ready to be devoured by a howling mob of people with probably totally valid criticisms.

I'll write a longer reflective post after the voting stage, but these are my initial thoughts on the game.

• I had to crunch pretty hard last night to turn what was a congealed blob of JavaScript spaghetti into something resembling a game, but am happy with the end result. It's definitely basic in some ways (the art styles are mixed and incoherent; there's no SFX or music), but it does have a reasonable level of polish.

• Because I lost a big chunk of time, there's not a whole lot of AI. The whole game is about writing some AI, which is enough to get into the jam, but it still feels like a cop-out. I would have preferred more AI in this game.

• The elephant in the room: I suspect this game is a bit too inaccessible. It's not particularly hard, but… you have to read a manual. Seriously. What the heck was I thinking?

I finished my Kajam entry… and I managed to do it before the sun rose.

I'll figure out how to upload it—probably to Itch—tomorrow (still a few hours left before the deadline), but for now here's a screenshot. I hope this piques your curiosity :)

A couple busy weeks put me a bit late with my 1v1 PvE platformer, but I'm back on track!

Throughout the month I have managed to complete all the basics & structure required to make the game playable, leaving this week-end to actually develop an interesting AI. All the iterations of that AI are actually going to be in-game as various levels. I hope to get at least 5 levels done:

• Level 1 (DONE): Almost idle AI
• Level 2 (DONE): Jumping around the map
• Level 3 (DONE): State machine-powered objectives
• Level 4 (IN PROGRESS DONE): Pathfinding
• Level 5 (TODO DONE): Tactics (health management, human movement anticipation)

I'm currently hacking together a waypoint system… and hoping it won't be too hard to build movement on top of it, so that the robot can actually jump on platforms correctly.

EDIT: And the game is finished at last! One day early for once :)

The GPU skinning I've been working on is going well—but not that well. I decided a couple of days ago that I need to prioritise finishing something for the Kajam, and then continue working on the OpenGL stuff. So I've started working on a new game in Phaser, but trying to reuse as much as possible from the other game (mostly the artwork).

I re-rendered the model from before, and did some tidying up in GIMP to produce this instead:

So I'm making a traffic light programming game for Kajam 4. I thought it would be fun to look at the cars zoom by and respond to the lights you've so cunningly set up.

The first thing I wanted to set up are the roads. Because I plan to make a bunch of different levels, it would be neat to have a little road drawing algorithm, so I easily have various layouts.

First I started with a basic horizontal and vertical road, like this:

Now of course, the next thing was to try to have an intersection, because that's where the traffic lights come in. So there we go:

Hmm… that doesn't look entirely right, does it. So I added some special intersect drawing code:

Pretty sweet, if I may say so myself :D It doesn't just support 2 roads either:

And it also supports more than 2 lanes :D

And that's it for now. Next up: adding cars!

My project will be called BDSM (Bomb Defusal Squad Machines)
A level (which is mostly the one I used while testing to make sure everything is bugless) is operative!

Is the UI understandable, even without tutorialisation? Please, if you want and can, try the level and tell me if you managed to beat it (deactivated the bomb) and if you understood everything or misunderstood anything. Thanks! :)

https://raindrinker.itch.io/bdsm