you gotta be kidding me.... why rectangle changing shape after colliding?? build#758 onwards

Solar2D Daily Builds and Public Releases
#1

I’m using build#761 now,
I was playing around with rag doll code, then I found this

when a rectangle collides with other physics object, it changing shape. colliding side of rectangle is vanished

here is the code

function createWalls()  
 --\> Create Walls  
  
 local walls = display.newGroup()  
  
 local leftWall = display.newRect (0, 0, 1, display.contentHeight)  
 local rightWall = display.newRect (display.contentWidth, 0, 1, display.contentHeight)  
 local ceiling = display.newRect (0, 0, display.contentWidth, 1)  
 local floor = display.newRect (0, display.contentHeight, display.contentWidth, 1)  
  
 physics.addBody (leftWall, "static", {bounce = 0.5, friction = 10})  
 physics.addBody (rightWall, "static", {bounce = 0.5, friction = 10})  
 physics.addBody (ceiling, "static", {bounce = 0.5, friction = 10})  
 physics.addBody (floor, "static", {bounce = 0.5, friction = 10})  
  
 walls:insert(leftWall)  
 walls:insert(rightWall)  
 walls:insert(ceiling)  
 walls:insert(floor)  
  
 return walls  
end  
local box = display.newRect(0, 0, 64, 256)  
box:setFillColor(255, 0, 0)  
box.strokeWidth = 3  
box:setStrokeColor(128, 128, 128)  
  
box.x = 160  
box.y = display.contentHeight\*.5  
physics.addBody(box, { density = 0.5, friction = 0.3, bounce = 0.2 })  
  
createWalls()

I wanted to try new sprites. but now I changed my mind, going back to build#745 which seems reasonably stable though some facebook bugs are there in #745 also

edit:
Circle also losing its face, but you don’t have to add circle to physics. just draw circle at some x,y coordinates
local crcl= display.newCircle( 160, 160, 60 )
crcl:setFillColor(255, 0, 0)
[import]uid: 97420 topic_id: 22936 reply_id: 322936[/import]

#2

We are aware of some issues relating to shapes in the latest daily builds. What is your bug # for this?

Please be aware that daily builds are not guaranteed to work work perfectly, or at all - that is why they are daily builds and not stable releases. [import]uid: 52491 topic_id: 22936 reply_id: 91661[/import]

#3

Thanks Peach for enlightening me :slight_smile:

so, I can’t take advantage of new sprites and image sheets [import]uid: 97420 topic_id: 22936 reply_id: 91694[/import]

#4

You may want to wait until the shape issues are solved which should be very soon, yes - although you could still start playing with sprites so you are not simply sitting on your hands in the meantime.

Did you have a bug # for me? :slight_smile: [import]uid: 52491 topic_id: 22936 reply_id: 91865[/import]

#5

ya, you are right. I really can’t sit idle. more over I stopped writing in forum and raising defects. from my past exp forum post getting answered and bug is attended is very unlikely ( may be I sound sarcastic here). felt I rather spend time on finding alternate way of doing things. if there is no alternate then I tell my employer (I’m a freelancer) that it’s not possible at this point of time with this version of Corona.

but I couldn’t stop myself asking about this.
No, I don’t have bug# for you Peach :slight_smile: [import]uid: 97420 topic_id: 22936 reply_id: 91893[/import]

#6

There are some bugs that are lower priority than others, particularly those with workarounds - however this is not one of those and is a high priority.

That said, when you have a bug you should file a bug report - posting in the forum is good to let others know about it, share their experience with it, etc. - but really we need the bug report too.

Hope that makes sense :slight_smile: [import]uid: 52491 topic_id: 22936 reply_id: 92118[/import]

#7

priority is something depends on individual.

I raised this bug http://bugs.anscamobile.com/default.asp?12313 almost 3 weeks back. which is very important for me to continue.
I still kept this part of game pending hoping Ansca will resolve it, and finishing remaining of the game.

It may be low priority to Ansca but for me it is high priority.

time and again Corona’s physics engine is letting me down

don’t get me wrong, it’s no blame game :). at the EOD I want my work finished

for this defect I really not able to find work around. code is attached in bug case. it’ll be a big help if you can suggest any possible workaround

[import]uid: 97420 topic_id: 22936 reply_id: 92313[/import]

#8

I’ve looked at the bug report right now - can you try adding this line after starting physics? It appears to have a positive impact in build 704, which is what I’m running currently.
[lua]physics.setPositionIterations( 16 )[/lua]

Let me know.

Peach :slight_smile: [import]uid: 52491 topic_id: 22936 reply_id: 92332[/import]

#9

Thanks Peach for looking into this bug.

I’m using build 745, should I use 704?

I’ve already set it 200 for making rope bit stiff.
I’m still having this problem.

[import]uid: 97420 topic_id: 22936 reply_id: 92337[/import]

#10

Is the sample you’re running the one you submitted or something else? For me, the gems do not seem to be pushing through the cart walls at all with that change, whereas they were a little bit prior to it.

I downloaded and did the same test with 745 and it was the same as in 704.

If you’re using something other than submitted sample would be useful to either test with sample or update bug report if possible. [import]uid: 52491 topic_id: 22936 reply_id: 92343[/import]

#11

sample is part of a game I’m currently working on.
making it to 16 is working fine in sample,

here is the screen shot of problem in actual game, even after making change

http://tinypic.com/r/2mcyo8o/5

in picture, I’ve marked with circle

last cart’s movement is stopped when gems in the cart infront of it trying to pass through the walls [import]uid: 97420 topic_id: 22936 reply_id: 92356[/import]

#12

Ah I see, the tilt into the floor a little.

I can’t seem to replicate that in the sample although if I can make some extra time tonight I’ll try to run through it again and modify it a bit.

Were you experiencing the floor issues in the project you submitted or only the gems? [import]uid: 52491 topic_id: 22936 reply_id: 92508[/import]

#13

in sample code, floor is static. so we don’t see any tilt issue.
I’ll try to give you modified code in which we can probably replicate this issue

edit: tilt issue reproduces when we have dynamic objects jointed like bridge, we can use bridge example. [import]uid: 97420 topic_id: 22936 reply_id: 92522[/import]

#14

I found a work around for this problem, it is working fairly good
finally it seems to me corona unable to handle collisions properly when we have good number of collisions.

thanks for your time, Peach. your effort is much appreciated

one question bugs me, what stopping Corona to have latest physics engine, where moai sdk (which is fairly new sdk) is able to have latest Box2D physics engine. [import]uid: 97420 topic_id: 22936 reply_id: 93165[/import]

#15

@y.sravankumar: In reply to your comment:

It may be low priority to Ansca but for me it is high priority.

I understand your frustration when it comes to bugs and issues, especially when you feel like our priorities aren’t aligning well with yours. Of course, it’s impossible for us to cater to everyone’s individual priorities when it comes to bug fixes, features, etc. but we try our best to keep on top of issues as much as possible, and sort our priorities based on what makes sense for our users as a whole, and the Corona SDK in general.

I’m happy to hear that you found a workaround for your issue, but for future reference, I want to bring to your attention that we do offer standard and premium support if there is a show-stopping issue or feature you need addressed as soon as possible. [import]uid: 52430 topic_id: 22936 reply_id: 93189[/import]

#16

@jonathanbeebe, thanks for dropping by.
I can understand nobody accommodate individual priority. but, I can see lot of problems and forum posts regarding physics engine. I’m just puzzled if still it is not priority on whole. [import]uid: 97420 topic_id: 22936 reply_id: 93298[/import]