jobel's Forum Posts

  • I know the website has many issues, but the search function for searching the forums is now terrible. It used to be decent on the old scirra.com - Now I can never find anything. And when I have 'Relevant' selected I always get posts that are from 10 years ago and completely irrelevant.

    I can never find any match on things in my search string.. even when I know the topic or its something I posted on. I search and it never comes up. I have to look in my post history to get to it.

    how about some options? like 'only these exact words' or except 'these words' and some better matching? It used to be good before why is it so poor now?

    for example, I am looking up if someone has talked about key bindings and remapping. I get a very high number of hits all having nothing to do with key bindings or remapping, not even close.

  • I would mock something up in a test project.. a very basic example of what you are trying to do. You will definitely get more help that way.

    you don't want to go the collision route for detecting matches. I would use instance variables 

    sqaure.top = "A"
square.bot = "B"
square.lft = "C"
square.rgt = "D"

    use collisions for when the two squares collide, and I assume you have some sort of grid to line up the squares. The grid could be numbered so you know the location of each square.

    then once that's done compare the variables to tell you if there's a match.

  • looks great congrats!

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • like this..

    drive.google.com/file/d/1i7h9mL7QMHtcPqmn_k9FmIlFOVKH_-zb/view

    I have some overlap in my coordinates for moving.. so you should probably fix that otherwise it goes extra fast when you are in the corners...

  • I've never done it, but for a smooth camera movement I would track mouse.x/mouse.y and when they get to the edge (Compare X/Y vs the LayoutWidth/Height) and I'd scroll lerp to a new position.

    Also you can change the mouse cursor when it gets to the edge of the screen to show the player they are going to start moving the screen.

  • the only issue with Construct compared to GMS2 or Unity (2D) is yes the export is not native but rather a web wrapper around your game. It's basically a version of Chrome for your game. Does the player know? Doubtful. Performance is probably the only thing that would make you (the dev) notice it. I doubt a player would ever know. But I've built things side by side with GMS2 and Construct with the same assets and I can tell the difference of GMS2's export, its just better, I don't know how to describe it.

    However, all that said, Construct is an amazing prototype tool. It's ridiculously fast. So many corners are cut for you that any engine I work with I'm always wishing for the ease of Construct. When I first got into to GMS2 I couldn't believe how much I had to do myself..although GMS2 has other pluses, but C3 is all about speed.

    I hate the subscription model, it's a big turn off for everyone I talk to about it. But I use it a lot so it's worth it for me. When people ask me if it's worth it I always ask them: will you use it as much as you use Hulu? because its the same price. If yes, then there's your answer!

  • newt LOS would be interesting to try, you'd have to create obstacles though and still it seems like you need a "ruleset".

  • oosyrag haha, that's funny because I kept working on it as well.

    I ended up leveraging your Overlapping at Offset and made a "ruleset" with a priority.

    When hitting UP ARROW one space up is top priority, 2 spaces up next etc.. 3rd priority it doesn't matter you can check right or left first. I'm not entirely sure it works as expected. At first I had immediate right/left the 2nd priority and sometimes I would get stuck and not be able to get to a piece. That seems to have been fixed by giving two spaces away 2nd priority. but still needs more testing!

    I only check 9 spaces since there should never bee that many empty spaces - but it makes it limited since you can't leverage the code for a bigger board or less pieces.

    here's the code..really long and not very efficient.. there's probably a more stream line way of doing it!

    drive.google.com/file/d/1xr1b8XgVFO22Y-mVgXyaHmmBciuBQjQh/view

  • can you mock up a simplified version in a separate capx what you trying to do?

    you should be able to save any data you want to an array.

  • oosyrag it's all moot anyhow since I don't think I am going this direction anymore, it's not really worth it - and I'm sure there are scenarios I haven't thought of yet. But thanks for you your help on this! I very much appreciate it.

  • oosyrag either F5 or F7 would be okay looking. but that's what I'm saying the rules are somewhat complex.

    if you started at B4 and pressed RIGHT, if there is no piece immediately to the right, then you would search outward in a radius from there for the next closest piece.

    You would not want to move all the way to G4 instead you'd probably want to go to D3

    EDIT: actually not a radius since you wouldn't want to go up to B3...maybe a half radius?

  • oosyrag I've also never used Stop Loop, will that break out of a sub event?

    i.e. where do I put code that never passed and executed the stop loop so I know to check the next column?

    however it does seem like it could get complex if I had something like this:

    if I had G6 selected and hit LEFT ARROW I'd want to pop to D4. you'd have to write code that kept checking columns.

  • oosyrag oh that's really cool.. I've never used Overlap at Offset.. I like that approach thanks!

  • oosyrag oh I didn't mean lots of checks as in performance issue.. I just meant writing complex code in Construct Event script.

    yeah I might be over thinking it.. having the player just navigating to an empty grid tile might be okay looking... I think popping right to the " filled piece" might be more work than it's worth.

  • oosyrag

    If you run into a filled grid object, stop loop and move the selection to that object. Otherwise, if no valid spots were encountered, nothing happens.

    yeah that way is the "lots of checks" I was originally thinking. but it gets even more complicated in the case where nothing happens. You'd always want something to happen, so if there were only 1 in a particular column and you hit UP, you'd want to go the nearest up piece to the left or right.