Search the Community

Hey, I'm currently working on decreasing my game's size. Recently I tried swapping png files for something with smaller size (image files take 60% of my game's size) and tried using jpg. Unfortunately, jpgs are not transparent, so most of files after conversion don't work the way intended. I tried converting tem to SVG which do support transparency, but after conversion their quality decreased to noticeable levels. Is there any slick way to avoid remaking those same images in SVG? Or maybe there's a way to make jpgs transparent with masks?

Howdy Babylonians! When I used the unity3d to babylon exporter it currently takes my cubemap and exports it out to a series of JPGs for the different faces of the cubemap. I would like to have use PNG files instead. I can fix this on the exporter side fairly easily by switching the command to from File.WriteAllBytes(faceTexturePath, tempTexture.EncodeToJPG()); to File.WriteAllBytes(faceTexturePath, tempTexture.EncodeToPNG()); and then also changing the output naming convetion from case CubemapFace.NegativeZ: faceTexturePath += "_nz.jpg"; to case CubemapFace.NegativeZ: faceTexturePath += "_nz.png"; When I change these it makes the PNG and names them correctly, but the exported BJS scene doesn't read in the reflection textures correctly. Here is what it looks like when I export the JPG version: - What it looks like when I switch everything over to PNGs - When I load the scene into the babylon sandbox it says " [15:15:49]: Error while trying to load texture: file:ReflectionProbe-0_pz.jpg " So it's still looking for the jpgs insted of the PNGs. Here are the settings for the shader: { "name" : "babylonjs_sample_material_reflection", "id" : "fb8dec12-8352-4e22-8ec5-83008b1780c1", "backFaceCulling" : true, "wireframe" : false, "ambient" : [1, 1, 1], "diffuse" : [0, 0, 0, 1], "specular" : [1, 1, 1, 1], "emissive" : [0, 0, 0, 1], "specularPower" : 53.588665, "alpha" : 1, "diffuseTexture" : null, "diffuseFresnelParameters" : null, "ambientTexture" : null, "opacityTexture" : null, "opacityFresnelParameters" : null, "reflectionTexture" : { "name" : "ReflectionProbe-0", "level" : 1, "hasAlpha" : false, "getAlphaFromRGB" : false, "coordinatesMode" : 3, "isCube" : true, "uOffset" : 0, "vOffset" : 0, "uScale" : 1, "vScale" : 1, "uAng" : 0, "vAng" : 0, "wAng" : 0, "wrapU" : 1, "wrapV" : 1, "coordinatesIndex" : 0, "isRenderTarget" : false, "renderTargetSize" : 0, "mirrorPlane" : null, "renderList" : null, "animations" : null }, "reflectionFresnelParameters" : null, "emissiveTexture" : null, "emissiveFresnelParameters" : null, "specularTexture" : null, "bumpTexture" : null } Is there a way I can adjust the unity exporter code to have it look for the PNGs that were created instead? Or if not is there a way to alter the shader settings after the fact to have it look for the PNGs?

Hi, I am trying to "hack" the spritesheets limitations by using JPG images with alpha channel. The process is simple: - I have a starting PNG with alpha channel - from this PNG I save a JPG (with varying fill colour for originally transparent areas) - from the PNG I extract the alpha channel and save it as an opaque indexed colour grayscale image - I load the new JPG and the grayscale PNG, and use the PNG as alpha channel for the JPG. These are the two images, the JPG and the grayscale PNG animBitmapData is BitmapData from the JPG animBitmapData_ALPHA is BitmapData from the grayscale PNG I did everything, but the last step doesn't work, animBitmapData is still opaque animBitmapData.copyChannel(animBitmapData_ALPHA, animBitmapData_ALPHA.rect, new Point(0, 0), BitmapDataChannel.ALPHA, BitmapDataChannel.RED);animBitmapData.transparent returns false, could this be the problem? Or maybe it is not correct to use a grayscale image as alpha channel? - Coding in Haxe + OpenFl -