kayin's Forum Posts

  • If you've made 5,000 off of your game, I think the extra 60 bucks won't really be a deal breaker.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • Looks good. I'll probably pick a license up eveutally, though I do want to see more progress first. Progress seems pretty rapid though. I'm hoping my next game project will line up just in time for the EXE export. I know it's a ways off, but whats the vague time line on that?

  • It's probably a lot to do with other stuff and not file size. If I made a game with lots of music and big graphics, it could easily be large while not having much sophistication nor benefit much from these changes.

    Complex projects with lots of code -- where the code is the bottleneck for FPS, not the display -- are what would be necessary to benefit. If you're using lots of effects and only a little code, no amount of code optimization would give you a better frame rate. Also whats sets this problem off is lots of objects. I'm working on an NES style game where tiles are 16x16 and everything is super small and barely uses any vram. But I have LOTS of tiles, thus making my relatively small game benefit from these changes.

    Now I have no idea what you're working on, but these are just some of the reasons you might now have seen any benefit.

  • I suspected a lot of the massive lost of FPS I experienced to be a result of what Ashley said too. Working on my current project, FPS loss from project growth seemed to slow down quite a bit once I was in the 200fps range. Losing the first 1000 FPS was quite easy though! Of course the amount of reduction seemed unsettlingly fast early on and the new files give me an FPS that seems more appropriate.

    Hopefully this means Construct Classic will get some big name releases in the next 6 months.

  • My old project that I bitched and bitched and bitch and bitched about? Went from 900fps to 2000. The other, developed project I'm working on is all the way up from 120 to 250. SO BASICALLY MORE THAN DOUBLE.

    I love you guys.

    Out of curiosity, what caused this issue? Due to the suffering I've gone through, I'm invested in the answer.

  • I already tried to set the client height and width with the same results With a zoom I get a chrisp image and with the window plugin I get a fuzzy image. The fuzziness I would assume is an intentional feature (Only so much you can do when stretching an image), but there is still no way to get a crisp pixels (point/nearest neighbor) without resorting to annoying zoom shenangians that messes other things up.

    I have photographic evidence if you want, but it's pretty much what you'd expect. A a game window with bilinear upscaling and a window that resembles nearest neighbor upscaling.

  • I've noticed that the resized windows functions in the windows plugin kinda create crappy little sub-pixel crap that mucks up the display, as opposed to zooming in with point sampling which is nice and clean. Anything I or you guys could do about that? The windows plugin is so much less infuriating when it comes to dealing with multiple layers and it's a shame that it makes things ugly, especially when someone is working with retro resolutions.

  • I'm just going to be sad that, I can still gain hundreds of frames of FPS by deleting objects in completely separate layouts, even if I delete every script I had available first.

    But glad to see you guys are still supporting this, casually. I gotta still be grumpy though if I have any hope of continuing with that project.

  • Sure thing! I'll PM you everything that I put in the bug tracker so you can get a better idea whats up. I definitely appreciate it, this has been the bane of my existence for awhile now and I'd much rather use Construct than GM.

  • It would eventually be debilitating. Anyways the cap has gone around before and no ones been able to come up with anything besides "bug", so I'm just going to wait to hear from David or Ashley.

  • I really don't think so, as I don't use global objects for anything. Even if I did, that doesn't line up I don't think. If that were the case, I believe FPS would drop as I move from layout to layout? In my case, if I test a layout, my FPS will be the same. If i go to another layout, delete some seemingly random junk and return to test again, my FPS will be higher. This is true for compiled files too (So it's not the editor keeping things in memory).

    Am I misunderstanding what you're implying?

  • It involves a cap where objects in one layout cause framerate drops across the whole cap. Basically the more I add to the file (not just the layout) the slower the game gets. And not just a few lost FPS. David looked into this and acknowledge the bug but couldn't figure it out. It's a private entry on the tracker right now since it involves project game files. Sadly since Ashley and David tend to be busy, I haven't heard about the issue for like 6 months or so and I'm reaaaaaaaaally itching to work.

    So I'm just trying to remind them once and awhile in hopes they can either deal with it or tell me I'm shit outta luck (or even tell me just what not to do to avoid the problem). I can deal with a lot of quirks and bugs, but in this situation it's absolutely crippling.

  • I'm just going to be a pain in the ass again and beg for my bug on the tracker to be looked into so I can get some closure. I'm pondering just moving to GM for the next IWBTG game (though I'd be checking back here for C2 for future games).

    I swear, if a solution is found, Construct will have a big release title under it's belt. I'd be back to work instantly. Even if the bug isn't fixable, I'd just like some closure on the issue of some sort so at least my hopes aren't kept up.

  • Application projects.

    Also I'm against removing the image editor because we'll always have to worry about hot spots and action points and stuff. Its better to exist and be crappy than to not exist at all since we're going to need to interact with sprites anyways. In theory it sounds good but in practice I think it would be a mess.

  • Only thing I want for 1.0 is the my single bug on the tracker -- the one that sucks the performance out of my game even when I add stuff in OTHER layouts. That seems kinda major, so I'd either like that fixed or at least for it to be figured out WHY it happens, so you and David can tell me what to do/not do to avoid it. Then I could actually pump out a game for Construct.

    Closure on the issue would be appreciated, even if it's "We won't fix it, but this is why it happened"