To be honest, I've learned a lot from C2, by thinking "how I would do this If I were to do so" about the behaviors.

Other than that, I was once a student on Sciences and technics of laboratory, mostly based on physics, and later more into how human can perform and transform things with devices, that leaded to how objects react to gravit, how fluid works, etc.. and also computer related stuff (like how do I control a devices ith a command, via a graphic based language)

So C2 wasn't really a problem since I've already had to understand the logics behind this concepts:

-Why coding the same y is good, not only for myself

-Why comments are great

-How the world works, more like how to find ways to interpret how it works

-What a computer understand, and how it translate things (the float-rounding errors? yeah, that too)

-Also some experience with a graphic-based language so I haven't any misjudgement about that too.

C2 helped me to understand how games truly works.

At the end, try to understand what you need to do a game, and what tools are at your disposal to help you, because I will never say it enough: keep it simple to read and understand, every coding language offers tools that can help you, you can even create your own, see what you need, then if it is present, or if you can do it, if so, try it (without things that doesn't makes sense if possible).

Exemple:

What do I need to apply gravity like earth?

-to move an object at a direction

-to increase this movement in a regular way (gravity is a constant acceleration)

-To stop my movement when there is something on my way.

So how do I know if there is something on my way? (etc..)

at the end (and also for tl;dr ones): making a game isn't about coding, it is about translating an information to the computer, but most people I help sometimes want to do this directly, It is more like:

What I want -> What logics behind -> translations of the logics -> Optimisations & Standarisation so it can be used easily in the future.

EDIT: going a little offtopic, but I think it is still a good way to process for every tool

just a little issue:

at th bottom it says:

Download

Stable Release (rstable)

Beta Release (r100)

other than that, no other bugs that the ones that have been posted before

Aphrodite

> I think you have 1/2 chances to miss the collision like that, it is more overlapping That'll work like this I think

>

i don't think there's a difference, if you only check once per 2 ticks, on collision or is overlapping will have the same result, nothing is checked between ticks, there is certainly more chance to miss the collision in both cases, but in my tests once per 2 ticks is unnoticeable, i have it set even higher, its a tradeoff

I was just refering to the fact that On collision triggers.. on collision, if the collision happens on one tick that you don't check, it won't register the tick after, but I could be wrong.

vtrix : "i do this, system compare tickcount%2 = 0 (checks every 2 ticks)

subevent on collision"

I think you have 1/2 chances to miss the collision like that, it is more overlapping That'll work like this I think

That is because the inspect tab checks a lot of things each ticks, it is meant to be like that, try to use the watch values when you can

kkletsgo ok, If you don't have enough rep to send me a PM, try reading some manual pages (it adds reputation points somehow).

Couple of quotes from my original post will wrap this up for me, as we seem to be going in circles...but thanks for all the input...

"I think most devs on here are blind to the real capabilities of C2 as they are running fast gaming spec PC's, many still struggle to push 60 FPS in a game that a Sega mega drive could breeze..."

"Maybe Java script is just not upto the job, or maybe we are just making excuses..."

I don't have a fast gaming PC, and still can achieve that, so javascript is potentially up to the job I think, the problem is somewhere else to be found, like why can't it be as well for everyone, why some just don't have all the advantages, that is more here the problem.

But I can understand your frustration, and also the frustration of people that can't enjoy our games :/

Agreed, I like to think it was a way for scirra to say "Don't use CocoonJS as we don't want to support it anymore", but I'm going into personal opinion

I wonder how this topic didn't get yet into HTML5 performances instead of C2's Performance, C2 is a great HTML5 engine, but it is an HTML5 engine, maybe that is part of the problem, how HTML5 works on older computer

Sprite collisions with themselves doesn't work well at all (A sprite is overlapping itself constantly)

Bad point Of C2 I think personnally, families can correct that I think, but not sure yet

Also you are checking the collisions for each individual object, let's see...

Number_of_object? collisions I think

EDIT:Indeed, checking like that means you check collision of an object with himself, and so the number of collisions to check will be huge, not a C2 problem, every engine would have trouble with this kind of logic, try instead to check collision between 2 different kinds of objects, or add some kind of other test

Crosswalk does only android, while CocoonJS does Android and iOS

also Node Webkit exports for Windows, Linux and MacOS, so .exe will not be accurate

You can always name a constant global variable "N" with value 0 in your example.

Instead of having to use the "0" index with less meaning, you can instead replace it with the "N" variable (if that makes more sense to you).

Example capx

Kyatric : when you call the function, you can set only

Parameter 0

Parameter 1

etc.., it is not like you can set

Parameter My_Constant_So_Awesome

I think that is what it is about too

EDIT: If that is too much, not a big deal I agree (+ninja'd).