I noticed a change from the last version I played. It's harder to rotate the pieces without adding more moves. I know exactly what the puzzle solutions are but I can't get 3 stars because when I try to rotate the pieces I accidentally move them a few pixels and add a move. I came across this exact same problem too and the solution was to add a threshold of a few pixels (depending on screen resolution) that made the difference between moving and turning.

I don't think level design is a exact science, so one idea might be to just have a lot of people test the levels and notice which ones takes more time to solve.

Yeah, I think that would be the best way to go about it, but also the most time consuming. Personally I just galloped through my levels because I was pressured by time, tested them myself a few times and again while recording hints and hope really hard I found all the blunders in design. But I think it's safe to say the players will find a few more and I'm pretty sure my levels are no smooth progression - next time I need to plan my time better.

Now, game have helper about rotate blocks. And smoosh rotating blocks. Also fixed some bugs. Achievments now is working. And now I will make web version.

Good stuff, the helper does the trick. I need to learn from the simplicity of your interface. May I ask what compiler you use? The app is so small in size I'd guess it's cocoon.io & Canvas+, right?

Hello everyone,

I've just finished the production version of my first game on Google Play.

It's called Laser Links and it's a block-sliding, brain twisting puzzle game with colorful lasers.

Google Play link - > https://play.google.com/store/apps/deta ... odes&hl=en

The game is my first completed creation after a few short projects that helped me learn C2 and it took over a year to make (half of it was correcting mistakes). I compiled it with cocoon.io and Canvas+ and designed all the graphics in Adobe Illustrator. There are 120 levels split in 3 campaigns and I intend to make more packs in the near future. Also, every level is playable in 3 game modes designed for different kinds of players.

The laser beam mechanics are quite flexible and dynamic and were probably the hardest part to code, but somehow not the most time-consuming. Here's some gameplay screenshots:

I am now working on a light version to upload on the arcade and maybe Kongregate and Newgrounds and in the next months I plan to get it on the App Store as well.

If you like what you see give it a try and let me know what you think! I've only had friends test it and I'm starving for some feedback. If anyone has a device higher than Full HD resolution I'd really appreciate a test - the game has graphic assets for multiple sizes all the way to 4K and I have only been able to simulate the larger ones on my PC.

In any case, I hope you enjoy!

Yes, it was unclear that it is possible to rotate the blocks.

Oooooh Ok, solved. I like the rotation part but clearly it needs a visible hint. I'm sure it's probably in the works. I love that the game uses very little text and can convey everything you need to know through icons and design. The level progression doesn't have to be that linear. I guess it depends on what type of player you're after. Some like brain-busting puzzles and others just like to relax moving blocks around. Roll The Ball for example is more oriented towards the latter and it's levels aren't very smooth in difficulty progression.

Anyway, now I can actually play the game. I'm curios about that 10th level, thanks!

I gave it a try but I can't complete the first level. Either the 2-square block was supposed to be vertical instead of horizontal or I'm being really stupid. I love the graphics and simple interface and the themes look great. Let me know how to get past that first level (show me a screenshot or something), I'd love to play more.

I think with all the extremely cool C2 action games in development it's gonna be hard to catch the eye of people here with a puzzle game. Which has me a bit worried because I too have a puzzle game approaching publishing and I was hoping to give it a small boost by showing it on the forums. In any case, your puzzle game looks top notch, keep it up.

What do you mean by C2 slow on mobiles? What kind of phones are your lowest target? Are you using WebGL effects? Because those can drastically slow your frame-rate on some devices, even new and powerful ones. I use Canvas and dropped all WebGL effects and the game runs like a charm even though it's gotten pretty complex.

Well, sorry, i'm mistaken, but only regarding PC's, which runs any C2 game really well. But for mobiles i still have my doubts...

Yes, you probably can't make a graphics-heavy game like that for mobile using C2 because WebGL effects can really take their toll on many phones. But you can make perfectly good 2D games like Candy Crush puzzle games or platformers or RPGs without the need to use almost any effects and make them run at a steady 60 fps on devices that are pretty low end by today's standards. I'm working on a puzzle game with large-scale graphics and pretty complex mechanics and testing it on my 2 year old 4-core cheap Samsung with 1 GB of ram and it runs perfectly well.

I've switched to Cocoon with Canvas+ for a few months now and I don't miss Intel XDK. I'm just doing the finishing touches on my banner and interstitials and for now they seem to work fine using the CocoonAds plugin. I've only tried AdMob and Android so far. I hope the in-app purchases will go as smoothly with the Cocoon plugins because I'll be implementing them next.

As far as performance is concerned, Canvas+ is really the only difference between XDK and Cocoon and it has it's tricks. You must not use regular vectorial text with Canvas+ as that will severely impact performance so you have to switch your text for sprite fonts. Also some Browser object commands may not work with it, like exit app for example, but that's no problem because they provide you with the CocoonCanvas plugin that can do that among other things. One downside of Cocoon I can think of is that you have a splash screen that's very expensive to get rid of and depending on game size there may be a black screen gap of a few seconds between splash screens (you can also ad a custom one in addition to the Cocoon screen) and your loaded first layout. Supposedly using a loader layout should fix this but it doesn't. It may just be a bug and fixed in the future.

One other thing you should know if you're used to developing for PC and want to expand to mobile is that WebGL effects should be used carefully because some phones, for some reason choke hard on these effects. I couldn't tell which exactly but I know that with WebGL effects my game was running at 5fps on a brand new HTC octocore and under 30 fps on my 2 year old samsung with 4 cores... To be safe I removed all WebGL effects and all of a sudden it was running at 60 fps on anything. It even managed a steady 30 fps on my old Huawei dual core with 512 ram. So if you must use effects, try to test each one on as many devices as you can before you implement it.

I think if you go with Cocoon you will probably have less headaches than with Intel. If you choose Canvas as well the number of bumps in the road will increase a bit but it should be well worth it because the performance is clearly above Webview, especially if you're making games that require reaction speed and agility. Oh yeah, and Canvas+ also means a smaller app package because it's wrapper is about 10mb smaller than Webview.

I don't know if anybody's gonna see this but yeah, I second that. Look what the lack of 9 patch scaling has done to my pre-loaded objects layer. Also it would be amazing if we could control the size of the 9patch's slices by script or load a new image from URL.

> Make sure you didn't accidentally set Player.PressedKeys to a number type variable

>

I did :/

Feel so stupid

Hehehe coding can do that to you. I was thinking it's impossible within my script because the variable was already assigned some text values and that makes Construct forbid you to change it's type but I forgot you were replicating a custom version of my script.

Sorry, was away for a few days. This seems strange. You could do this:

tokencount(str(Player.PressedKeys), " | ") - 1[/code:k7jq0g9z] The "str" turns whatever is in brackets into a text string but it's weird to have to use it because Player.PressedKeys is already a text string. Make sure you didn't accidentally set Player.PressedKeys to a number type variable but I think that's impossible. The problem may be that I set the default values of Player.PressedKeys to "-1", so maybe the event sees it as a number. You can also try changing that value to "a" or "start" by selecting the Player family and editing the variable. That way it starts with text.

Yes, something like that. Basically either make the portraits new objects but add them to the player family and name their Class variable the same as their corresponding actual players, or simply add new animations to each player to use only in the selection menu. The most correct way to do it would be to make a new portrait object that doesn't share the Player family and give it multiple animations for each class portrait. An object that really doesn't function at all like the Players but with the same variables and family may cause some head aches further on.

Then you can give it a text variable that stores the class just like the Player objects have. It can even be named Class as well. Then you replace the references to Player in this here segment with references to the new object like, instead of "On Click on Player", "On Click on Portrait" and then setting the second Array value to Portrait.Class instead of Player.Class.

Wow, better and better...

Now I need the entire week to understand all, and apply it to my project. Or should I start again from beginning ? I have succeeded in doing interesting things, like adding operational live bars for each active character, behaviors and animations not repeated, etc.

Anyway, thanks again for all your work, I'm pretty sure it will be usefull not only for me but for anyone.

The more the merrier. I figured you might have to restart your project but if you've already restarted once know that you can copy paste chunks of script form one project to another if you open them both. The key is to first copy the Variables, Families and Objects the chunks of script use, otherwise it won't work. Functions, token strings and arrays may be a bit difficult at first but believe me they are worth learning as soon as possible.