Image groups images randomly flashing

Solar2D General Questions/Discussion
#1

I’ve been working with image groups all day trying to find out why when moving objects created from a image sheet the images will sometimes just randomly flash in place it should not be, sometimes it looks more like double copies of same image.

Before I go and post a bug, has any one else experienced this, I’ve kinda narrowed down to when a image is removed from a imagegroup. Its always ok if the group is getting larger but when its reduced strange things happen. [import]uid: 118379 topic_id: 25524 reply_id: 325524[/import]

#2

Hey Beloudest, I have heard of a similar issue before. If you could make up a test case which exibits the problem… just a really simple example that has only the code and image assets needed to reproduce this issue and put that into a zip file and submit it with as much information as possible here :

http://bugs.anscamobile.com

I will get this looked at personally asap [import]uid: 84637 topic_id: 25524 reply_id: 103189[/import]

#3

Its been done. :slight_smile: Wow im bug finding which kinda makes me feel pleased. [import]uid: 118379 topic_id: 25524 reply_id: 103346[/import]

#4

I’ve had this bug - it was a minor symptom of a bug that also leads to the simulator and devices crashing. In fact today I had it confirmed finally that it is a bug within corona itself, and it has been passed to the appropriate people. [import]uid: 46639 topic_id: 25524 reply_id: 103447[/import]

#5

I’ve had this bug - it was a minor symptom of a bug that also leads to the simulator and devices crashing. In fact today I had it confirmed finally that it is a bug within corona itself, and it has been passed to the appropriate people. [import]uid: 46639 topic_id: 25524 reply_id: 103448[/import]

#6

I replied to your message, did you get it. Or do i have to resubmit the bug. thx? [import]uid: 118379 topic_id: 25524 reply_id: 104175[/import]

#7

I don’t see the point really in resubmitting the bug - once it is in the system it is fine, and an Ansca engineer confirmed he saw and could repeat the bugs using code I sent him, so it is definitely on the list to be looked at. How long it will take is anyone’s guess, but the main point is that it *will* be looked at. After many years of using Director (the software, not the module here!) I can’t begin to tell you how refreshing that is! [import]uid: 46639 topic_id: 25524 reply_id: 104334[/import]

#8

Yeah mines in too so they should get it fixed. There is a workaround and thats to use a blank image as the last element to be added into the imagegroup. This stops the flashing from appearing. [import]uid: 118379 topic_id: 25524 reply_id: 104416[/import]

#9

i’m having same problem with this issue, i can see objects create using display.newImageRect and graphics.newImageSheet twice. do you know when can be this fixed? [import]uid: 74255 topic_id: 25524 reply_id: 112301[/import]

#10

@Peter are you using build 828? [import]uid: 84637 topic_id: 25524 reply_id: 112302[/import]

#11

danny: i’ve this problem in 828 and also in 833 [import]uid: 74255 topic_id: 25524 reply_id: 112303[/import]

#12

@Peter. Just to clarify, do you mean objects are doubling up or ghosting? [import]uid: 84637 topic_id: 25524 reply_id: 112304[/import]

#13

danny: doubling, i can see 2 same objects at same time [import]uid: 74255 topic_id: 25524 reply_id: 112306[/import]

#14

@peter, I thought that bug was fixed a few builds ago…

If your seeing this behavior then I would really appreciate it if you could make up a really simple example (as few lines of code as possible please, along with images needed to run the project) zipped up and submitted along with filling out the following form: http://developer.anscamobile.com/content/bug-submission

If you reference my name in the description It will get assigned to me to check out.

Thanks! [import]uid: 84637 topic_id: 25524 reply_id: 112316[/import]

#15

Danny, I’m beginning to get a bit frustrated with this bug.

This is now the third time (I think) that the bug has been claimed to be fixed, yet it still displays easily in the demo I sent you guys ages ago - it is like you aren’t even bothering to try it, when just running it will show within 5 seconds or so the problem.

I accept the code isn’t the most basic example, and possibly there-in lies the difficulty in tracking down the problem (although the issue is reasonably clearly .isVisible in new sprites in image groups), but really, you can still easily check whether what you are doing fixes it.

I do hope we’ve moved past the point where Ansca claims the bug is in MY code, as I’ve provided numerous ways of proving that is not the case :frowning: [import]uid: 46639 topic_id: 25524 reply_id: 112368[/import]

#16

@Rakoonic, I can understand you are frustrated. Just to confirm I nor anyone else at corona believes that these bugs are issues with your code. We know there are still issues with the new sprite apis.

The isVisible thing is definitely a bug and we know about it. This bug has been a case of “fix one thing and other breaks” we have all being there and we are aiming to resolve these issues as soon as possible. [import]uid: 84637 topic_id: 25524 reply_id: 112370[/import]

#17

@rakoonic, I’m confused. Are you referring to bug 13834? We fixed this in build 820. And I just tried your test case in 833 (the one with the coins and star sprites) and it works.

@peter/rakoonic, we do have an outstanding issue when setting the isVisible property of child objects inside an imagegroup. In that case, the child object will still be visible.

This will be addressed as part of some additional culling performance improvements in a daily build post-public release, as the code changes were not sufficiently isolated. [import]uid: 26 topic_id: 25524 reply_id: 112381[/import]

#18

I downloaded the latest daily build today (833 I think) and it still had the graphics going wrong.
I’ll double and triple check later on and post back, but I definitely did see the bug happening. You can’t just run my sample code and take a quick peek, it takes a few seconds for the bug to start happening. [import]uid: 46639 topic_id: 25524 reply_id: 112429[/import]