eli0s's Recent Forum Activity

  • This topic has become way too techy for my understanding!

    I'm getting weird results from Colludium 's test in Firefox and both Chrome 38 & Canary. Only IE 11 renders a curve that resembles the tests that you guys posted above...

    Chrome 38

    Chrome Canary (40)

    Firefox 33

    IE 11

    Am I the only one that gets this? What's going on? I fear now that my PC might be the culprit for the jitter and not the browsers... For example, many members posted that the latest Canary has no jittering problems, I still get almost the same results as with normal Chrome: Every few seconds the sprites that move pause for an instant and then jump a few pixels ahead...

  • Ashley ,

    Is it possible to have the option to color the layer with a gradient? Having linear, radial and mirrored gradient modes would be awesome! This would be useful for creating skies, colored filter ramps, colored vignettes and all sort of "theatrical" overlays and shading over the layers below.

    I know that all this can be achieved with sprites, but using layers would be simpler object managing-wise, and the sprites would have to have high resolutions to avoid noticeable stepping value issues. The other option is R0J0s canvas Plugin, but, again, using layers would just be easier.

    This is low priority of course, I just want to know if is feasible to go in the todo list

  • Try Construct 3

    Develop games in your browser. Powerful, performant & highly capable.

    Try Now Construct 3 users don't see these ads
  • DUTOIT ,

    Subscribe to Construct videos now
  • rozpustelnik , nice! I hadn't seen any other lens flare examples before, this is great!

    I think that you can't see the online example because you live in Poland. For some ^&%*$ reason my domain is been blocked in some countries. You can always try the capx however!

  • C-7 ,

    ,

    DatapawWolf ,

    volkiller730 ,

    rekjl ,

    Thanks guys!

    By the way, does anybody know a better way to use one Sprite in order to mask in an other, but in the manner that destination out works? What I mean is, when we have a large sprite and on top of that an other smaller sprite with the blend mode set Destination out, the small sprite cut's a whole into the Sprite bellow. For instance, we only need a small gradient circle with a solid center that fades in opacity towards the edge, and with that we can cut a whole in to a solid colored layer (the way lighting effects in a dark environments is done in the ghost shooter tutorial).

    The small circular sprite cuts a whole in to the solid black layer, enabling the content below to be shown.

    However, the opposite doesn't happen. If we wish to use the same small Sprite in order to show just the part that it covers (in respect with it's opacity), we realize that outside the physical borders of the Sprite, both Destination In and aTop blending modes fail: The pixel information of the sprite bellow (or of the colored layer) is been displayed outside of the small Sprite's edges.

    The same circular Sprite does reveal the black layer bellow, taking in to account it's opacity values. However, the effect breaks outside the Sprite's physical borders and the black layer is visible.

    In my Lens Flare example, the Sprite named "HotSpot" is used to show the lens dirt spots. In order to do that, I had to stretch it into a great size, something that is not very practical, has limitations and either need a huge Sprite if details are needed, or renders blurred results when stretched.

  • I am sorry for the newb question, but I see in the forums posts that have text active as hyperlinks that look like this. I'm trying to figure out how to do that...?

    I remember that some time ago, when I wanted to insert a hyperlink, the forum popped out a window that had two fields: one for the text that would be displayed on the post and one for the actual www address. I can't find anything like this...

    Thanks for any help on the subject

  • For any one interested, I've made an experimentation on recreating a dynamic Lens Flare effect. Ideally some genius could do it with a behavior or something, but here it is, in my own layman way <img src="{SMILIES_PATH}/icon_e_biggrin.gif" alt=":D" title="Very Happy">

    Here is an online preview: http://www.eli0s.com/Tests/LensFlare/

    And here is the capx file:[attachment=0:cpkhnibs]LensFlare.capx[/attachment:cpkhnibs]

    Please feel free to build upon and improve this example! I'll be happy to see any suggestions and ideas! Just don't turn your game into a JJ Abrams movie <img src="{SMILIES_PATH}/icon_e_wink.gif" alt=";)" title="Wink">

    note: with 32 lens objects I get 3% cpu usage (on the debugger running in Chrome). I feel like it's fare enough but I have no idea hoe this scales on non desktop devices.

    note2: Ashley , I don't know if it's my system's problem but the same capx gives me an assertion error with version 186. Is this a bug perhaps? Also, I've noticed in the last few versions of C2 that the Tilled Background object produces an artifact, at least on the preview mode. It's hard for me to explain in English, please have a look on the TB object in this capx, notice that I've intentionally left a 1 px line empty on the bottom of the image. Otherwise, when the height of the object is 256 px (the default image's height) there is a 1 px bleed on the top.

  • You can always lerp the positions. Use an variable to store the target x position for the box (lets say "goToX") and when you press the left/right arrow keys subtract/add to this variable the 130 pix offset (you'll need to find a way to limit me minimum/maximum x position). Then, add an event that does the following:

    every tick-----> Box | set X position to lerp(self.X, goToX, t)

    where t is a number between 0 and 1. The closest to 1, the "snappiest" it will get.

  • zenox98 , hehe, thanks! I don't know about hundreds, but I will try to dig a bit deeper into existing shaders In reality the only thing I did is to follow QuaziGNRLnose's step by step instructions, I'm not that confident that I'll do any good by my own :b

  • QuaziGNRLnose ,

    I'll be damned! After I read your thorough explanation I came up with this: [attachment=1:2v5tqtap]RadialBlurPlus.zip[/attachment:2v5tqtap]

    I don't know what to say! A simple thank you is not enough! But here it is anyway, thank you very, very much!!!! <img src="{SMILIES_PATH}/icon_e_biggrin.gif" alt=":D" title="Very Happy">

    -note for anyone interested, I've also changed the way the images are dispersed. Instead of 4 images in front of the virtual radius and 4 images behind it, I made it so all images are rendered behind the radius. I feel it looks more like a camera zoom effect this way.

    <img src="{SMILIES_PATH}/icon_lol.gif" alt=":lol:" title="Laughing">

    Edit: I've updated the html test to use the new effect (http://www.eli0s.com/Tests/Warp/)

    Edit2: And here is the Radial Pixellate effect <img src="{SMILIES_PATH}/icon_e_wink.gif" alt=";)" title="Wink">[attachment=1:2v5tqtap]RadialBlurPlus.zip[/attachment:2v5tqtap]

  • QuaziGNRLnose , what I wrote to Ashley applies to you also: you guys can't even begin to understand how frustrating coding looks to someone that wasn't exposed to it like you did, when you did.

    I am not claiming that I know you or something, don't get me wrong, I admire you guys for what you do! It's just that I wouldn't believe if you told me that you had no prior experience in programing what so ever up until your thirties, and suddenly you decided to grab visual studio or something and you came up with Construct or Q3D. Somehow you laid foundations in knowing how the logic works, what the syntax means and so on.

    On the other hand, to put it with an analogy, someone that was blind until his mid thirties and miraculously found his light, does not have the necessary neurons to interpret the visual stimuli.

    What is great in C2 it's that it bridges this gap with the event system's logic. However, making a platformer or a top-down shooter in C2 doesn't substitute for proper programming education.

    So, to sum it up, I do understand the logic behind your advice but I don't get where those parameters live inside the code! For example, this is how I see the code:

    //some stuff that are necessary for the shader.
    precision mediump float;
    varying vec2 vTex;
    uniform sampler2D samplerFront;
    
    //I guess that these are declared variables, like global variables in C2
    uniform float pixelWidth;
    uniform float pixelHeight;
    uniform float intensity;
    uniform float radius;
    
    void main(void) //this is the function (?) that does the trick
    {
    	vec2 dir = 0.5 - vTex; 
    	float dist = sqrt(dir.x*dir.x + dir.y*dir.y); 
    	dir = dir/dist; 
    	vec4 front = texture2D(samplerFront, vTex); 
    // I guess that the position of the center is been determined somewhere in the above 4 lines, but I don't get where or how.
    
    	vec4 sum = front; //No idea what so ever.
    
    	sum += texture2D(samplerFront, vTex + dir * -0.08 * radius);
    	sum += texture2D(samplerFront, vTex + dir * -0.05 * radius);
    	sum += texture2D(samplerFront, vTex + dir * -0.03 * radius);
    	sum += texture2D(samplerFront, vTex + dir * -0.02 * radius);
    	sum += texture2D(samplerFront, vTex + dir * -0.01 * radius);
    	sum += texture2D(samplerFront, vTex + dir * 0.01 * radius);
    	sum += texture2D(samplerFront, vTex + dir * 0.02 * radius);
    	sum += texture2D(samplerFront, vTex + dir * 0.03 * radius);
    	sum += texture2D(samplerFront, vTex + dir * 0.05 * radius);
    	sum += texture2D(samplerFront, vTex + dir * 0.08 * radius);
    //All of the above must create the extra images that blend together to create the blur. There should be 4 images between the sprite and the center and 4 behind the sprite and towards the center . 
    
    	sum /= 11.0; //something that sets up the way the images blend together?
    
    	float t = dist * 2.2; //how much the images are separated from each other in relation with the distance from the center...?
    	t = clamp(t, 0.0, 1.0); //no idea
    
    	gl_FragColor = mix(front, mix(front, sum, t), intensity); // No idea...
    }
    [/code:3y85ayee]
    
    @shinkan , I am glad you have your way around just by looking the code and experimenting! I 'm just not able to do that. I need help to cross the road on those stuff. I have my way on other things though. For example, I like to think of my self as a half decent artist and composer. I've been told that I am a good teacher. I get along with cats and dogs! I just can't learn code!!!
  • Ashley , if I was able to learn something like that I would have been a happier and more intelligent person. Unfortunately I am not. I know that this is probably something trivial for someone with your knowledge but I am illiterate in terms of programming languages.

    Perhaps if I invest some months, I'll learn the basics of GLSL (although I strongly doubt about that), but I estimate it will take much-much less time for someone that knows what he/she is doing, to add these features. That's the reason I ask.

    Since you are the one who made the radial blur and warp effects, you know better than anyone how and if it can be done. That is if you I agree with me about those options and if your developing schedule allows it.

eli0s's avatar

eli0s

Member since 24 Apr, 2013

None one is following eli0s yet!

Trophy Case

  • 11-Year Club
  • Coach One of your tutorials has over 1,000 readers
  • Educator One of your tutorials has over 10,000 readers

Progress

13/44
How to earn trophies