591 reputation
17
bio website gamealchemist.wordpress.com
location France
age 38
visits member for 1 year, 11 months
seen 12 hours ago

Hi,

I'm a french independent developer coding in Javascript.

I hire my services as Javascript Consultant, i already solved problems ranging from small debug issues to more important architectural design advices.

Do not hesitate to contact me for more informations.

I Have a small blog talking about Javascript and video games here :

http://gamealchemist.wordpress.com/

And i published some of my libs here :

https://github.com/gamealchemist/


19h
comment Parallelism and Functional programming
So basically you're tigthing your hand with an outdated approach you don't even know the basics of, and you think this is a good gamedev question. Original.
1d
comment Calculate required resolution
Your computation depends also on a model M and the distance d from model to camera (imagine a textured monster very far/very near from the camera). To avoid headache, you have to decide of some 'average case' i guess.
Sep
8
answered How to handle a negative coordinate 2d map with a growing map size
Aug
18
comment Algorithm for procedureral 2D map with connected paths
Nice ! I was interested by the id so i made a small attempt. You need to be careful when using random otherwise too-strange results will follow. And i wonder if the corridors shouldn't be handled right during the recursive split, because i don't see easy way to build corridors out of the tree. Anyway for anyone interested fiddle is here : jsfiddle.net/gamealchemist/xt57zwb8
Aug
9
comment How do I randomly position objects without overlap?
Interesting, however i don't agree with your fear of too much CPU computations: When creating a new platform, one must ask : where should i put the next tile in a 'possible' location ? So it's a computation involving many parameters ( horizontal and vertical distance between tiles, ... and the parameters you explained well), but there's no 'complexity' involved : compute the area(s) where a tile should be there, then depending on difficulty, put it randomly in the area, the more difficulty we have, the more the highest position in the possible area is likely.
Aug
9
comment A star pathfinding and discrete/smooth positions
In fact, about radius, a precision : if one is using axis-aligned Bounding Box collision tests, then the 'radius' should be max(width/2, height/2). For one using Bounding circle collision tests, the formula is the one above (computed once for each object creation).
Aug
9
comment A star pathfinding and discrete/smooth positions
Thx for your update. Since it is corridor width = corridor width - hero width, then on each side you should consider only (hero width / 2 ), or even better consider radius = sqrt ( sq(width/2)+sq(height/2)), in case the hero can freely turn. (in your scheme the corridors are too narrow ) ( Rq : I'm assuming here we consider the center of the hero -the only way to avoid nightmares -)
Aug
9
comment A star pathfinding and discrete/smooth positions
Very nice. Do you have by any chance a simple way to handle the 'hero' size ? i wonder if considering the corridors have width = width - hero's radius (while keeping same corridor center coordinates) wouldn't do the trick.
Aug
7
comment How to interrupt movement where the model can lag behind
no, you can have, for instance, a 'pauseLerp' and a 'resumeLerp' function that allows you not to change all your logic.
Aug
7
comment How to interrupt movement where the model can lag behind
It seems that if 'the character is interrupted in its movement', then lerping should stop where it is, no ?? After the shooting is done, re-launch another lerp to target point.
Jul
18
comment Detect which object conceals another one
don't think of 'z' as indicating the 'depth' or something : just think of it as way of sorting your items. In a second time you might use the z to change the scale / project y for a quite nice effect. The math are a bit tough though.
Jul
18
comment Detect which object conceals another one
isn't a z-sort of your elements the way to go ? (you'll have to think about intersection vs z, since character, when behind the house, intersects with the house but not the door).
Jul
16
comment How to check for collisions in “diagonal” tiles?
You should use the 'right' tile size, in this case your tiles should be half as big, this way you'll have far less special cases to handle (4 tiles would be enough for your slope i guess).
Jun
4
comment How can I find the right UV coordinates for interpolating a bezier curve?
By 'not valid' i mean that the texture will be stretched in the outer part of the curve if using only one quad per segment. -that will be case also when using one equal lenght segment like Anko suggests- Now if you use enough segments (more than in the O.P.'s image) the artifact might be negligible and the solution perfectly OK.
Jun
3
comment How can I find the right UV coordinates for interpolating a bezier curve?
I don't know why @KromStern deleted its post, which was the most valuable, not to say it was just the right answer. I deleted my post since, after some small experiments, i proved it completely wrong : approximating one of the trapezoid by a rectangle is impossible when the difference in upper/lower width can be >40% even for a moderate slope with a good amount of polygons. Opsenas and API-Beast 's answers are, for the same reason, not valid either.
Jun
3
comment How to implement soft edge areas with particles
Do you want the particles to move, or is just a still background ? i'm afraid i might have misunderstood... :-/
Jun
2
comment How to get scripted programs governing game entities run in parallel with a game loop?
Short answer is : using a parser to 'understand' the codes and one finite-state machine per robot to make it run. You might look also about what is an interpreter. (on wikipedia, a tutorial, ... )
Jun
1
revised collision detection algorithm issue
added 21 characters in body
Jun
1
answered collision detection algorithm issue
Jun
1
comment collision detection algorithm issue
@TimothyGroote : O.P. does the collision test the right way. Think of each inner condition as a rejection condition : too far on the left, too far on the right, ... It is faster since result is computed as soon as ONE reject condition is true. If you code it the other way ALL inclusion tests must be performed (&&).