For the most recent issue, you can solve it by using a system like this:

https://www.dropbox.com/s/8d2co8ain01ol ... .capx?dl=0

For the previous issue, it's hard to say. Are you still experiencing that? I have theories but if you have it fixed I'll save it lol

Hey,

it's a little hard for me to see exactly what's going on in this event sheet but I think your problem stems from the decor menu being used as a trigger to set your decorative sprites as visible and movable. When you re-open the menu, the cursor changes in the cupcake viewer to the left again. Perhaps you were attempting to get around these triggers by having the sprites in 2 families but I'm not sure that will work. If for some reason re-working the decor trigger proves untenable, you can always have identical sprites spawned from your decorative ones when drag and drop is released, having the originals destroyed.

Okay! After really looking around I’ve found tons of posts about crazy cpu utilization and memory issues in Edge with some posts being as recent as 3 days ago https://www.drivethelife.com/windows-10 ... pdate.html . I’ll keep looking into this and trying things but if anyone has any Macguyver ideas I’m game to try anything!

Split screens will always hafe performance, and my experience is if you think you might have too many objects then you probably do.

Hmmm I hear you , I just don’t know why that wouldn’t be the case across all browsers. A jump from 20-30% to the 90s is no where near comperable. Also these are pretty tight controlled testing levels, at the moment, whose object counts have been dwarfed and outperformed by stress test projects (though those are not stellar on Edge either).

On the Xbox itself the fps drops off to the mid teens, essentially a 75-80% drop off. No animations or props aside from pickups. We’re talking bare rectangular sprites running around.

newt EDIT add: ...albeit, I've tried creating another project with tons of (read: waaaay more) objects and collision/poly checks with paster splitscreen to stress test the xbox and it should perform way worse but it actually does much better (though the CPU usage is still quite high and fps hangs around the mid 40s).This was actually my initial response a few days before posting this topic, because I assumed a 3rd party plugin must have been causing the hit (I used identical events to create the splitscreen, building and exporting piece by piece to locate the limit). With the ability to actually play the test at an acceptable (by comparison) fps might indicate something else is at play?

Absolutely, please let me know if you come across any snags

Ashley Thank you for responding! Webgl is definitely on. I just tried running things out of the debugging mode and performance is still quite bad. Also, that huge increase in overhead is occurring when I test from using Edge from C2 as well (albeit, while the game runs well enough to play on PC in from the C2 editor, the CPU usage hangs around 90%). For some reason Chrome runs everything perfectly fine (60fps, and CPU usage around the high 20s and low 30s).

newt Yeah, I couldn't recall at the time whether paster could run without webgl on so I tried when getting home to find that indeed it won't. I'm trying to get a splitscreen multiplayer working...perhaps I should try using canvas to do the heavy lifting and have smaller paster objects? I'm not too sure what performance differences between the two would look like.

Hey, I believe you can do something like:

--If condition met (I'm assuming you don't always want the player stuck in a box)

--every tick

-----------Player -> set x to clamp(player.x,camera.x-(window.width/2),camera.x+(window.width/2))

I'm away from home so I can't test it but I think that should work. If not I'll to make an example when I get a moment.

First I would check if Edge has it enabled using the renderer expression.

If its canvas2d then its not getting the benefits of hardware acceleration, there's not much you can do other than try to update video drivers.

Great! I’ll look into this when I get home tonight! *fingers crossed!*

newt Yeah, webgl is on, indeed. I'm using the paster plugin so perhaps it's something going on with that and Edge...

Sounds more like a Webgl issue.

Thank you for responding!

Yikes!! This leads to new questions! newt have you run across this before? What makes it feel like a webgl issue? And perhaps most importantly, can I do anything about this?

So I've been tinkering around with exporting my current project as a UWP 10 app and testing on the Xbox One. In Chrome/Nw.js the game runs at a smooth 60fps and has very manageable cpu utilization. However, when running in Edge on PC the cpu utilization is over 300% greater (growing as high as 94%), with a dip in fps from 60 to the 40s (at times). Even worse, when testing on the Xbox the game becomes completely unplayable - or at least, enjoyable due to the severity of these issues. I have to believe this is an issue with the Edge browser and the way it handles javascript (at least on PC, I assume the UWP wrapper is contributing to problems on the Xbox). I'm wondering if others have run into these performance issues and what they've done to try and counteract them??

What I've tried regarding rectification:

1) lowering polygon points

2) lowering amount of every tick specifiers to make triggers more dynamic (which did benefit performance few percentage points)

3) Lowering the amount of objects on screen (This works in this project, but is still too constraining to develop cross platform)

My general feeling is that object count should be the culprit, as objects (with collision/poly checks) appear to hit Edge performance in a separate project, but after testing this does not seem to consistently answer the question as to where the problem lies. Perhaps it is more global? Any personal experience developing on Edge and Xbox One, regardless of outcome, is welcome and appreciated.

Congratulations! The game looks so fun!! It's obvious a lot of hard work went into it .