I don't have plans to add support for modeling style features as they're not game oriented, and would require considerable effort for a feature very few would use. With that said i may include simple geometry controls in a future update, but i don't think it'd be enough to make a full modeler.

Doom uses BSP's , and im not sure about how you implemented raycasting, but wolfenstein/effecient raycasters use a hashing scheme that can very quickly determine collisions in a uniform grid.

Not yet through a simple api, you can only constantly create models on your server, edit the model file, then reload them. The reason is that such operations are generally very slow and require cpu->gpu transfers, rebuilding of buffers, etc, so i didn't think they'd be of much use. What use case do you have in mind?

hes using raycasting, basically shooting rays for each vertical bar, saving the distance to each bar, using that to scale the bar vertically, and using the position along the face of the tile to choose what to use as the texture for that bar.

codah, It only shoots right away if greater than cooldown time has passed, you wont shoot right away if you shot 0 seconds ago, you'll have to wait the length of the cooldown, but if you havent shot for 10 seconds, you wouldn't expect to wait possible the entire duration of a cooldown for your first shot.

imagine you had a gun that could only shoot every minute, if you used every "60" seconds, then if the timer was at 61 seconds, and you clicked, you wouldn't shoot a bullet for a whole minute, and would have to be holding the key down for 59 seconds until you shoot, even if you haven't shot once. by using a variable, if your weapon is "cooled down" for the required length of time, then you don't have to arbitrarily hold down the key till the timer hits some value, you'll fire, and then youll have to wait, and youll fire again, etc. Reloading takes a set amount of time after firing your first shot in a series of shots, it doesn't make you wait a random amount of time based on the game timer before being able to fire.

the problem with using every X seconds is that you wont shoot right away when you click sometimes, which feel weird especially for longer delays (imagine holding down the fire key for 6 seconds then the bullet comes out).

a better way to do it is to have a timer variable.

every tick

: add to variable object.shootTimer -> this.dt

// this value gives you the time since the last shot fired.

is fire button down

is object.shootTimer > cooldown

: shoot

: set variable object.shootTimer -> 0

// shoot the object, reset the cooldown

it's dependent on your rate of deceleration, current speed, and distance to target.

you haven't given enough information as to how exactly you want this to work, is the deceleration something that can change with time? that makes things much more complicated, and would necessitate some kind of PID loop to make an estimation of how to alter the velocity. If your deceleration is constant its simple algebra.

for the pid loop you'd have to alter velocity based on the current speed and distance to target and some value to control sensitivity. You'd need to carefully tune things not to overshoot. If you know deceleration will be constant for the entirety of the stopping motion you just need to solve using an appropriate kinematic equation, depending on how far you want to move before stopping / what constant rate you wish to decelerate. in each of these cases you'd need to do different things. If you had a specified stopping rate you'd need to find at what distance to begin decelerating. if you wanted to stop at a particular point from any distance to target you'd need to solve at that instant what constant rate to decelerate at until velocity is zero and the target has been reached.

set its position to scrollx and scrolly

your best bet is to write some kind of macro using software that can take control of the mouse/keyboard (Autohotkey would maybe be a good start, though im not sure to what extent it matches the flexibility you require). Construct 2 isn't geared towards intensely automated version control so there really isn't a way to "compile" (at least officially supported) via command line like you could do with some programming languages, but it shouldn't be hard to write a macro that launches Construct 2, clicks through the menus, exports and does everything you need at the execution of a single command. The only shortfall of this method would maybe be it failing to work on different screen sizes or something but if you're doing it from one machine it wouldn't be a problem.

I think you could even create a simple gui with autohotkey to tie everything together into some "compile program".

not sure if this is exactly what you want but usually least squares refers to fitting a line/function to a dataset, which doesn't really match your request.

I think what you're after is more like the center of mass (c.o.m.), which is given by the relationship (positions are vectors, so you'd need to do this for x and y seperately)

(total mass) * (position vector of c.o.m.) = mass1*position1+mass2*position2+....massN*positionN

you can solve for the center of mass (c.o.m.) position by dividing the right hand side by the total mass. masses act as your "weighting" and theres no reason you can't give them a negative value, although the effect may be different than what you're looking for, i'm not sure how else you'd define the effect of a negative weighting in something like this.

Ribis,

There are some limitations with using some behaviours, but for the most part they work, except obviously since theyre coded for 2D they only work in the xy plane. I havent built a 3D mark style test yet, but thats a good idea. For now you can look at some three.js examples to understand the performance the rendering engine gets though (in a way my modifications to the engine handle unique objects better than how vanilla three.js does in terms of memory and cpu use per frame, since matrix updates are triggered as needed rather than constantly, and i have an efficient geometry sharing system). In terms of this being a JavaScript engine and each object being unique, the performance is as good as it can be. Ive implemented many optimizations, and users can choose to share materials/textures between types for added performance. Users can also set the scale of the collision hashing grid to tweak it to the needs of their game.

Effects cannot be used on the models, but they have their own custom blend modes that can be specified (available blending options). Custom fragment shaders/vertex shaders will come eventually. Multiplayer works but you may have to use variables since the positioning of objects is designed for 2D with that plugin.

kmsravindra

Timescale works based on how you write your expressions, it has nothing to do with the plugin. There is something wrong with your events. Read the article and try to understand how to apply timedelta to your expressions. It wont automatically work just because you changed timescale without changing anything else.