ah!

thank you very much for clearing that up <img src="smileys/smiley20.gif" border="0" align="middle" />

now i know what to do. and yes, im using power of two in pretty much everything i create. now if i could only find out how to do gradients of different shapes. like the one in light behavior but only different shapes.

anyway, thank you, newt!

hi people. i just got a one simple question and i dont know if this is completely stupid and silly one, so i apologize <img src="smileys/smiley9.gif" border="0" align="middle" />

does transparency(of sprites) affect performance in any way? or excessive use of alpha and stuff?

the reason i ask is im wondering if i should make my environments using tiled sprites, or use more of a free hand method with "fading" sprites together.

again, sorry if this is a stupid question, i just honestly dont know and couldnt find an answer.

hi and first of all, sorry for im about to ask a really noobish question, but i cant seem to find my way around this obstacle.

situation:

player sprite. ground sprite. effect sprite.

player sprite overlaps ground sprite.

--ground sprite spawns one effect sprite.

player sprite does not overlap ground sprite, the effect sprite that was spawned gets destroyed.

now, the object picking. how to make it so that it doesnt destroy ALL those spawned effect sprites, but only the one that was spawned on a ground sprite that just stopped overlapping the player sprite?

i tried for each condition (and trigger once), but somehow i managed to make an infinite loop where the effect sprite kept spawning, thus quickly jamming the whole game.

to say it simply, imagine the player running through piano keys.every key is that ground sprite and something happens only once when player overlaps it.

i know this is amateur stuff, but i just got so confused with object picking and now i cant think straight.

thanks for reading!

hi guys!

i had some big issues understanding the water behavior in the example .cap posted by rojohound in this thread:

http://69.24.73.172/scirra/forum/viewtopic.php?f=3&t=7752#p60271

the effect is brilliant,but after spending alot of time trying to understand how it works i just got frustrated. is there anyone who would like to break it down for me and explain the basic idea and behavior of those few lines of code?

is it possible to have multiple instances of those water "puddles" ? meaning that i could scatter small puddles all over the level (layout)? i didnt have any luck with that and now i feel completely stupid <img src="smileys/smiley5.gif" border="0" align="middle">

arrays are pretty alien to me still. also, i get stupid everytime i see "loopindex" lol , it scares me. <img src="smileys/smiley36.gif" border="0" align="middle">

hi guys!

i have a feeling this is a complete noob question, but i really didnt find anything about it:

i want to create object randomly on a layer, this works fine enough, no problems.

but i also want to randomly set the size of the object while keeping its proportions. now, i cant find a simple set size action anywhere, i can only find set width and set height, but if i try to randomly set those upon creation, it doesnt keep the sprites proportions, instead, it can make for example width 50pixels, height 90pixels. and id want to keep them both the same amount. is there any simple enough solution i just cant figure out, being noob and all? <img src="smileys/smiley9.gif" border="0" align="middle" />

also, to make things more interesting, how can i make a random value between two values?

for example if i want to give a sprite a random size between the sizes 50pixels and 90pixels?

small bonus question:

ive been tinkering around with physics hinges, and already made some pretty neat stuff with it, but i still cant figure the basic logic behind hinges.

for example, if i create hinge from sprite a to sprite b, how can i make it so that the sprite b is also stiff? meanin that it would keep the angle relative to what it is upon creating hinge?

do i have to make hinges for both, sprite a and b?

thats all right, but it could get rather tedious when doing alot of hinges together.

little example:

if you make a car, 3 parts:

one body, two wheels. you have to make hinges FROM the body to the wheels, and that kind of confuses me. somehow i think it would be the complete opposite.

the mass is also something that puzzles me. if i made a say, simple ragdoll, with torso having the biggest mass, so i could apply force to hands, without it affecting the movement of the torso too much, what happens is that when moving the torso, the limbs that have smaller mass start to sort of act erratic, the hinges wont keep their position locked and they act "rubbery" even though having 100 stiffness.

anyone care to educate me? <img src="smileys/smiley36.gif" border="0" align="middle" />

hey guys.

been playing some modern games lately, since i got a new pc with all the dx11 eye candy, and started thinking:

would it be possible to implement something like parallax mapping(or its variations) in the construct?

because, if im not completely wrong, its still just a 2D texture effect used to make an illusion of depth, correct?

and since there already is bump mapping in construct, i was thinking why not parallax..

i think it could make a 2d game look really stunning when used subtly and carefully.

which lead me to wonder why it hasnt been used in 2d platformers, for example, very much. i really like the way it looks and while it could potentially go to waste in a 2D game where you cant rotate around textures, it would still add some depth to visuals.

any thoughts?

thanks tulamide! i knew there was a simple way to do this, even i can understand that piece of code lol!.

i'll try this when i get home, BUT (yeah yeah..)

once again i started wondering if:

there was a way to cut the amount of frames in half, so i could use more frames without sacrificing VRAM that much. because, correct me if im wrong, theres the option to mirror the sprite angle when moving the opposite direction..? so, if i had a frameset consisting of angles between 270 and 90 going clockwise, in other words a half circle of frames, i could somehow mirror the car sprite, thus drastically saving VRAM and i could make more frames and get smoother rotation.. but that would need some extra eventing and/or math, no?

hi guys.

my racing game is coming along pretty nicely, i have the base gameplay so that i can adjust stuff easily, and im overall pleased. BUT (yeah you knew it was coming lol)

i started thinking about artistic stuff and presentation. started toying with the idea of trying to make the perspective isometric. i could very well use the existing handling that ive made using the physics object(the actual car sprite is just set on the physics object every tick/always), but the car sprite's presentation would need some work,if its not top down view,but isometric.

so i ask you wise people, is it possible to to set animation frame to a specific angle?

for example, if i was to use 360 frames (oh god i really wouldnt, dont be alarmed) for my car's angles, how would it be done? meaning frame 0 is angle 0 and frame 359 is angle 359. i wouldnt want to make actions for every separate angle even if i used less frames, im sure theres simpler, easier way, eh?

am i too far off if i guess its somehow done by clamping? though i dont really know the correct action or event for it <img src="smileys/smiley5.gif" border="0" align="middle" />

continuing my monolog, i managed to get a somewhat working prototype for my racer, got a little distracted by lighting and shadows, so i decided to make a small example, if anyone is interested. spent almost the whole evening with learning how to do bump mapping and shadows. theres also a little bit of terrain effects, though just a brutally simple idea still.had to make a workaround for the velocity issue cause i still dont know how to determine it. also note that theres alot of plugins in use, so you may have to DL those. the event sheet is also a complete mess and im not surprised if it doesnt run well at all <img src="smileys/smiley36.gif" border="0" align="middle"> anyway, here it is:

http://dl.dropbox.com/u/11157144/night-rally.cap

im still open for hints, tips and suggestions of all kinds!

taking baby steps, but i managed to get a simple prototype real quick, if anyones interested:

http://dl.dropbox.com/u/11157144/rally-ish.cap

the canvas works just as i hoped and i even have a way to check collisions with the tire tracks, although everything is still very barebones and primitive.

i got a few questions though:

1. is there any way i could use the erase effect with a smaller magnitude. for example, while touching virgin track surface, only erase with 50% strength. when touching the same section again,erase all?

and further, is there way to use gradient erase,just to get smoother edges for the tire tracks?

2. as you can see from the .cap, there's still the issue with a little bit too much slide and being able to turn while not moving.

i guess this would probably be simple to fix with a velocity check, but problem is, how do i determine the condition, when car is moving in any direction? do i have to check all the directions at once?

for example

car Xvelocity is greater than 1

          or

car Yvelocity is greater than 1

          or

car Xvelocity is less than -1

          or

car Yvelocity is less than -1

is this correct? i mean im having problems to determine the fact that if the car is moving at all. but that condition list seems a little too complex, and i couldnt get it to work like that. when youre checking negative velocity, is it supposed to be less than -1 or greater than? i got a little confused by that <img src="smileys/smiley36.gif" border="0" align="middle">

aside from that, i like the use of a physics object as a car, right now it has a great sense of weigth and even a bit of oversteering and with a little thought and fine tuning it could work really well, and the collision feels good.when i get that velocity check figured out, i can add over and understeering related to speed and all those fine details.

so anyone got any helpful ideas? mainly that velocity thing is bothering me as of now.

hi rojo and thanks for replying!

i think im gonna try te canvas stuff. do you have any guidelines as to how big of a canvas would be "optimal" or how big is going to be overkill? i have a pretty beasty computer so its always kind of difficult to keep track on how demanding my project will be lol.

the array object is completely alien to me, havent really studied how it works, but i kind of get what you mean.i really have to study it and learn how to implement it,that would make it possible to have hills and bumps and stuff, good idea. but first i was thinking id focus on how the skidmarks affect your car,taking into account the angle,speed and stuff,when colliding with skidmarks.

i did some tests with physics object, but theres already some stuff that makes my brains fart <img src="smileys/smiley36.gif" border="0" align="middle" />

for example, i made it so that phys object has another object always set to an image point,so that when i push down up arrow i add force and the "car" mowes forward. turning and velocity is the problem. now the velocity works as if its a rocket. meaning that when im not pushing up arrow and turn, it doesnt move towards the angle anymore, because force is not added, instead you can just turn in place.so, the first obstacle lol. i could probably make a behavior,that when the car moves, it adds 1unit of force forwards,and if the the up arrow isnt pressed,it adds 2units in the opposite direction, to slow down while still having the momentum towards the right angle..? <img src="smileys/smiley4.gif" border="0" align="middle" /> or am i making any sense?

i had another idea of making a compromise and using the car behav, but upon collision, i could somehow destroy the car and spawn a corresponding physics object that would handle the collision. i would need a way to determine the speed upon impact and store the value in order to use it for the physics object though.

car physics are really difficult to make from scratch, considering all the physics calculations and math involved. sucks for me since i completely lock out when someone says "math"

<img src="smileys/smiley36.gif" border="0" align="middle" />

EDIT: thanks for the cap btw. just to make sure, is the scaling simply done by using the canvas:set height/width action? that seem pretty useful.

hi everyone! new user here. well not entirely new, ive been using construct classic for a few months now and downloaded a lot of tutorials and even made some simple stuff already.

i wanted to start a little discussion about a few issues im having. first of all, im really fascinated by all kinds of different, not very used technologies in games, and mainly im talking about terrain deformation in this post. anyone who has played sega rally revo or mx vs atv: reflex knows what im implying.

in sega rally it was subtly used, but it did impact handling in some situations, and frankly, the game would have been pretty standard/mediocre without that tech. but it really made the game a little more enjoyable.

reflex took this a bit further, because you were able to really mold the terrain with your tires. you could make little hills and bumps even and i remember spending alot of time just fooling around and marvelling the tech behind it.

now, im not hoping to make anything near that quality in construct, BUT.

it got me thinking if something similar would be possible in a top down racing game. maybe some prebaked, smoke and mirrors style effects.

okay, so ive been testing stuff and the immediate first problem arises when considering skidmarks, or tire tracks, whatever you call them.

im not really sure what would be the best method creating skidmarks? creating skidmark sprite(s) every x seconds while moving quickly drains resources, after a lap or two. i started thinking (from that destructible terraing example)that if i used a reasonably sized track and made it a canvas object, then erase effect, so that the skidmark sprite erases the canvas, shaping the track? i havent properly tried this but i wanted to ask if its even worth a try? or is it completely insane as an idea to begin with? ive heard the canvas object takes a lot of juice, so big racetracks are out of the question with that? maybe some small track like slicks n slides or generally- games for example?

now the second issue of this could be a little simpler actually, just time consuming. simulating the skidmark, bump, traction effect. i thought originally that it would be something like so:

-positioning 4 detector sprites at the place where the tires would be in the car sprites.

-check collision when the detector collides/overlaps the skidmark or tire track.

-make small subtle adjustments to car sprite's velocity/angle. (nudges,loss of skidding,traction boost etc.)

one big challenge(as if those werent enough already) would be the actual car behavior. im not sure if i would use the car behav itself, because the collision behavior is, to say it bluntly, kind of primitive.

so maybe i could use a physics object with custom made behavior using add force etc. that would make it easier to use the above mentioned adjustments to handling when traversing the terrain though.

so i guess i wanted to hear you guys opinion, hints, maybe even get some discussion and prototypes for this project

<img src="smileys/smiley1.gif" border="0" align="middle" />

OR you could always just say to stop even trying because it has been tried 1000 times over <img src="smileys/smiley36.gif" border="0" align="middle" />

anyway, thanks for atleast reading this poorly excuted novel of a thread, hope theres atleast someone who could help <img src="smileys/smiley1.gif" border="0" align="middle" />