tunepunk's Forum Posts

  • I think it also depends on how well the games are promoted by for example buildbox and Gamemaker. One problem is that the C2 showcase isn't very updated, and they could update the showcase with some recent released well made games made with C2, that would be both beneficial for Scirra and the gamesdevs.

  • Ashley do you have any windows phone to test this one?

    http://www.tunepunk.com/touchtest

    I just need to confirm if my device is faulty or it's a problem across all Windows phones. If the problem is a windows phone issue, i can contact them about it. And if it's a pure windows phone problem I need to put the windows phone version on hold and focus on Android/iphone until it's fixed.

  • I just tried this scene on a LG Nexus 5X, stock android. The Touch never failed even as I was approaching 80% CPU.

    The 50% mark is not an issue here. But on Windows Phone it is, even if I build an app with Visual Studio and XDK

    I havn't tried this test on any Apple device yet. But the same seemed to happen on Samsung galaxy S3 as well.

    I just also tried the touch test in another browser (UC Browser) on same windows phone device (Nokia Lumia 830). Touch doesn't fail, but as soon as I'm touching the screen the FPS drops about 20%.

    I don't know if this is a windows phone issue, or I'm just having a faulty device, but it seemed to happen on the Samsung Galaxy S3 as well.

  • Ashley I created this test scene.

    Capx:

    https://dl.dropboxusercontent.com/u/20560446/Scirra/touchfail.capx

    Test directly.

    http://www.tunepunk.com/touchtest

    press + to add more dots.

    press - to remove dots.

    hold to move blue dot.

    as you can see FPS is still decent, but touch is failing above 50% cpu.

  • You add a variable (elevation) to any objects that are supposed to be elevetad above something. Just sort like in newt example but set the ytoz value to Y+elevation. The height should be the distance to the ground, (as if the sprite was at ground level) then the sorting will be correct. I'm using something similar in my project.

  • *** EDIT: Updated original post with recent findings ***

    Currently I'm doing a lot of testing for my game on several devices. Ranging from upper low end phones, to high end phones. I've noticed that on some phones, especially windows phones touch will fail and cause a massive lag in the touch.

    On my Main testing device - 2 year old - mid range - Nokia Lumia 830, when the test runs at just about 40-55 CPU usage, and as soon as it goes above that magic 50% I'm noticing lag or failure in the touch input.

    Have anyone noticed anything similar when testing their game on low-mid range phones?

    I made this test to try it and confirm touch is failing as CPU get's higer.

    http://www.tunepunk.com/touchtest

    If you notice any touch lag please report in this thread.

    Phone model.

    CPU - Min, Max, Average

    FPS - Min, Max, Average

    UPDATE: Touch is very slow, not responding well in some cases, but in game when I lock the screen and unlock again with the touch starts to respond again. (Mortar Melons, Last of the Survivors, Made with C2, windows on phone store, has the same issue)

    So far it seems to be a Windows Phone issue, but curious if any other devices has the same issue.

    Any help appreciated.

  • Nice. Good job! Looking good. Hoping to try it out.

  • SoldjahBoy I managed to identify that the problem. Theres is something causing a very high spike in Draw Calls. Sometimes it goes away by itself but sometimes i have to quit and restart the game, for it to go away. But i don't know what is causing the spike. This spike can happen at anytime. Even when idle, nothing moving, no scrolling, nothing changing zOrder, no animations running.

    Ashley Thanks for the link. Do you know anything that could cause high unusual spikes in drawcalls? These idle spikes seem to be higher than normal, even higher than during gameplay when objects are moving, animations playing etc etc. Or is there any way to test it, so i can identify what is causing the spike.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • Does running in debug mode show anything different or abnormal? I had some similar issues when one of my loops would get stuck on for some reason - and it was causing a lot of collision checks that it shouldn't have been doing. Maybe you have something similar happening?

    ~Sol

    In the debugger the only things abnormally high in those cases are engine and draw calls. I've tried to disable groups of events one by one to try to identify if it's any event causing it, but no luck. Draw calls should go down to 0% if nothing is moving, changing, it does so in older versions of my game.

    I opened an older capx of my game and it doesn't have this issue. I'm starting to think it's some behaviour causing it.

  • Usually when I run my game it runs at approx 15-20% cpu, but sometimes hovering around 30-35 with even when idle (nothing moving) on my surface, and the only way to get it down to normal rate is by restarting the game a few times. This is especially critical on mobile where those values are at 25-40 normal rate, but every now and then 40-60% cpu usage.

    This is especially noticeable since touch starts to fail/lag above 50% cpu usage on mobile.

    Has anyone experienced anything similar or know what might be causing the game to run at varying performance every time? Plugins, Behaviours, etc etc not loading correctly on start of game? Any help appreciated.

    This seem to happen on both pc and mobile.

  • R0J0hound Brilliant, just brilliant. Those kind of solutions I love. You never stop to amaze, thanks!!

    I wish math was one of my skills. I'm going to try out your solution as well.

    AllanR I got it working in this way using the sine behaviour instead. (similar thing)

    capx: if anyone finds it useful.

    https://dl.dropboxusercontent.com/u/20560446/Scirra/snappyrun.capx

  • Bouncing is a very specific movement. Why not use the build in 'bouncers' ?

    https://drive.google.com/open?id=0B1SSu ... Th4aVdidGM

    Thanks. Just Checked out your example and it's a good solution, but unfortunatly i'm doing an isometric game so can't use the platform behaviour in this case.

  • AllanR it should be enough. 0 is on the ground 180 is on the ground, so everything in between is "in air". while progression is moving up to 180, left leg will be in front. (frame1), and while progression is moving down to 0 again after contact with the floor again at 180 the other foot will be in front (frame 3), so you solved it in one go i believe. There's the 2 bounces.

    progression =0 left foot touching ground(frame 0)

    progression 180, right foot touching ground (frame 2)

  • AllanR the bounce animation is not really a squash stretch animation, so I'm not really worried about it looking weird. I will try that out and see how it goes. I think it could work.

    It will be more of like, when the addition is positive the left foot will be in front, when the addition is negative the other foot will be in front.

    I'm using the sin to set the bounce height.

    I have a snappy 4 frame run animation very similiar to the one in the link, except that the character is not moving in Y axis. The sin controls the Y movement (bounce of the step)

    progression is between 0 and 20) set Frame 0 (feet together - step)

    progression is between 20-160 & addition is +1 - set Frame 1 (left foot in front - in air)

    progression is between 160-180 - set Frame 2 (feet together - step)

    progression is between 160-20 & addition is -1 - set Frame 3 (right foot in front - in air)

    Kind of something like that. I'm aiming to achieve a nice snappy & bouncy run animation using only 4 frames, where the sin controls the Y movement (the bounce of the step), having a smooth up and down bounce but a snappy four frame walk animation.

    The reason I'm doing it this way is that otherwize i would need a at least a 15 frame animation in every axis (8 direction isometric), to get a smooth up and down movement. That would add up to a lot of frames, for each character.

    I think your solution could work. I will try it out when i get back home.

  • I have an object that I want to bounce using the sin expression, as it will be linked to an animation. I kind of found a solution but it's not exactly what I'm after.

    I'm using an instance variable called progression. each tick I'm adding to this value until it reaches 180. Once it reaches 180 I then use a trigger once to set the value back to 0. This kind of worked to get the bounce, but my goal is to have the value decrease instead when i get to 180 to make it reverse again when I get to 180, and forward again when it get's to 0.

    I need to do this as a division of this value (progression/180*5) will also be driving which animation frame to show.

    I can't really wrap my head around how to make it reverse. I'm guessing I would need another variable to detect which end of the scale it last hit?

    Any ideas?