dupuqub's Forum Posts

  • I found out. I'm using a 1600x900 display and the game works under 1920x1080, so there's the difference. When I set the game display to 1600x900 it works perfectly.

    My question now is, is this a bug?

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • This is what's happening.

    You can (barely) see in the top left that the Mouse.AbsoluteX and Mouse.AbsoluteY are precisely equal to the X and Y of the Square following (Aim). It has more definition at Gyphy. They are the last 2 items in the log.

    There is no Parallax in any of my layers and every scale is still 1:1.

    I'm stumped.

    Tagged:

  • You do not have permission to view this post

  • oosyrag Now... furthering the subject, I've been reading this link and I got to the part where he tiles the noise. I tried it myself, but my knowledge is still very limited. Basically I tried replicating his numbers on my noise, but it didn't go as planned.

    This is what I intended.

    This is the code I used.

    This is what I got...

    I don't have a math background at all, but I imagine there is something I can plug into the Classic3d() function that would output exactly what I need, have you ever had any interest in it and/or can you give some insight on the matter?

    as always, I appreciate any feedback, thanks in advance.

    ps: this is totally non-critical or even necessary to my project, but I thought it was very interesting. I'm not getting it at all right now, but I believe I can with some minor guidance.

  • I was able to fix it by using different noises for the X and Y values...

    But is there a way to use the same noise for both? I feel that using 1 less noise calculation could be very beneficial for performance in the long run.

  • As a side note, I'm getting these very straight lines on diagonals, I imagine it's beacuse of the Perlin noise being used in the same manner for the X and Y values on the Voronoi. Any ideas on how to get rid of it?

  • For future reference, if anyone uses this approach, imagine that when you don't use Perlin, you're giving a flat piece of paper to Voronoi draw onto, but when you ask for the Noise with Perlin distortion, then you're giving the Voronoi a crumpled piece of paper to write straight lines.

  • oosyrag thank you so much! your help was invaluable. In the end I did exactly what you said, first I got a Perlin noise going and then I added it to the Noise requests for the Voronoi. Worked like a charm, thanks again. great articles by Amit also.

  • I already have a map generated through Perlin in my game, so what I need now is being able to section this map in organic chunks, but Voronoi is too stiff.

    Another issue I encountered when mixing Perlin and Voronoi is that Perlin is usually good for layered sections that exist within one another, like height maps (mountains and beaches in my case), but they don't really work for "area-like" sections like Voronoi. I'm unable to see how they would interact to achieve the image above.

    Maybe you already gave me the answer, but I couldn't catch it, anyway, your link gave me a clear north though, where they talked about "domain warping". I'm 100% sure that it will solve my problems (when I understand it), so thanks a bunch for the reference.

  • Yeah, I thought it really looked like Voronoi, but Perlin will not work (at least based on my skills) because I need these different cells, just like Voronoi, except I need them distorted and I don't really know how to do or research that.

    As a point of reference, I'm looking at biomes in Minecraft, they behave just like the image above, but it's been hard finding any useful info on how to edit the noise as I don't really know how to blend noises together.

    My last idea was to try the billow effect on the voronoi cells... but hell if I know how to go about it.

  • I've been working with noise recently, but I don't really get how to approach this idea. I would appreciate any insight or reference or even a name that I could research.

    Cheers.

    EDIT: I think it's essentially a distorted Voronoi, but I can't figure out how to distort its edges.

    Tagged:

  • I did post the wrong screenshot, sorry about that. this is the real issue.

    thanks a bunch oosyrag for the catch.

  • I made this slightly modified advanced random example file to showcase how Voronoi and Cellular break if you move around (maybe others too, but less visibly).

    Just use the arrows to move around and very soon you'll see it breaking in straight lines like in the example below:

    Is this a bug?

    EDIT: it looks a lot like a bug, so I reported it, still, any insights are welcome.

    Tagged:

  • I used Inkscape tool "Text => Put on Path". I'm aware there are similar tools in Photoshop, Illustrator, Corel, etc for you to take a look, but Inkscape is free and open source, so I'd love if you could take a look there.

    This is a video showing how the tool works if you're interested. See how letter spacing works differently, like the text still works naturally, but curved on the path.

    Every font has a specific kerning setting which is virtually impossible to replicate if each symbol is its own object, unless you are able to read the kerning value between symbols, which is waaaaay above my skills. That's why I said what you did is probably the best solution possible.

    But either way, hope I added something to the conversation with this.

  • I intended to investigate further before commenting, and here it goes.

    Your tool is good, very good, but you can see how the text behaves weirdly because of the font spaces since every letter is an object (right?).

    I also don't think it is even possible to make something better than you did, so thank you for your effort, but I believe that for it to look properly good, the text should've been continuous because of letter-spacing and therefore impossible with the tools we currently have.

    I'll probably post this tool in the ideas page.

    Again, thank you a bunch!