Comments on: Using BitmapData to Improve Performance (And One Small Annoyance) http://blog.nobien.net/2010/02/09/using-bitmapdata-to-improve-performance-and-one-small-annoyance/ A nerd blog about nerdy things by ... nerdy guys? Tue, 20 Jul 2010 20:39:29 +0000 hourly 1 http://wordpress.org/?v=3.0 By: zach http://blog.nobien.net/2010/02/09/using-bitmapdata-to-improve-performance-and-one-small-annoyance/comment-page-1/#comment-4692 zach Mon, 08 Mar 2010 18:47:31 +0000 http://blog.nobien.net/?p=199#comment-4692 What happens is that the height will get so big and exceed the max value allowed(8191, I think). It is with multiple filters and as I mentioned I just have to move something one pixel up or down and there are no problems. When it first happened I found the same issue happening to others and the resolution was the same (just move up or down), but no one understood what was actually happening. You seem like the kind of guy who would have figured out the how/why if you ran into this, so I was just interested if you could shed any further insight. But it seems as if you haven't seen so if you do you'll know how to work around it. What happens is that the height will get so big and exceed the max value allowed(8191, I think). It is with multiple filters and as I mentioned I just have to move something one pixel up or down and there are no problems. When it first happened I found the same issue happening to others and the resolution was the same (just move up or down), but no one understood what was actually happening. You seem like the kind of guy who would have figured out the how/why if you ran into this, so I was just interested if you could shed any further insight. But it seems as if you haven’t seen so if you do you’ll know how to work around it.

]]> By: Matt http://blog.nobien.net/2010/02/09/using-bitmapdata-to-improve-performance-and-one-small-annoyance/comment-page-1/#comment-4662 Matt Fri, 19 Feb 2010 16:34:30 +0000 http://blog.nobien.net/?p=199#comment-4662 Zach, when you say "can get giant" do you mean they can eat up a lot of memory or the display dimensions go out of whack? Zach, when you say “can get giant” do you mean they can eat up a lot of memory or the display dimensions go out of whack?

]]> By: Zach http://blog.nobien.net/2010/02/09/using-bitmapdata-to-improve-performance-and-one-small-annoyance/comment-page-1/#comment-4659 Zach Thu, 18 Feb 2010 22:30:18 +0000 http://blog.nobien.net/?p=199#comment-4659 Nice Matt, I've been enjoying your blog for some time. This is semi-related, but I noticed in flash player 10 that my bitmapData objects and associated filters can get giant even though they are specified as a number typically less than 10. I think the error code was 2015, but I don't remember for sure and I don't have any code that is currently offending. I've always been able to resolve it by moving the object up or down one pixel, it's as if something in the rendering engine explodes my height or width. When I first researched this others were having this issue, but no one had a conclusive reason as to why. Basically, I was wondering if you have run into this and if you can add any insight as to why this happens. Anyway, thanks for the great blog. Nice Matt,

I’ve been enjoying your blog for some time. This is semi-related, but I noticed in flash player 10 that my bitmapData objects and associated filters can get giant even though they are specified as a number typically less than 10. I think the error code was 2015, but I don’t remember for sure and I don’t have any code that is currently offending. I’ve always been able to resolve it by moving the object up or down one pixel, it’s as if something in the rendering engine explodes my height or width. When I first researched this others were having this issue, but no one had a conclusive reason as to why. Basically, I was wondering if you have run into this and if you can add any insight as to why this happens.

Anyway, thanks for the great blog.

]]> By: Matt http://blog.nobien.net/2010/02/09/using-bitmapdata-to-improve-performance-and-one-small-annoyance/comment-page-1/#comment-4646 Matt Wed, 10 Feb 2010 15:12:28 +0000 http://blog.nobien.net/?p=199#comment-4646 Will do, Mike. It didn't even cross my mind to log it as a bug since I found a little "fix". Will do, Mike. It didn’t even cross my mind to log it as a bug since I found a little “fix”.

]]> By: Elliot Rock http://blog.nobien.net/2010/02/09/using-bitmapdata-to-improve-performance-and-one-small-annoyance/comment-page-1/#comment-4643 Elliot Rock Wed, 10 Feb 2010 08:27:25 +0000 http://blog.nobien.net/?p=199#comment-4643 Nice work mate. Clear coding as well. Nice work mate. Clear coding as well.

]]> By: Li http://blog.nobien.net/2010/02/09/using-bitmapdata-to-improve-performance-and-one-small-annoyance/comment-page-1/#comment-4642 Li Wed, 10 Feb 2010 06:27:10 +0000 http://blog.nobien.net/?p=199#comment-4642 I once wrote an animation that consisted of a bunch of 2D points that moved around with some physics + arbitrary behaviour and on each frame, I would draw a lineTo from each point's previous position to its current one. As you I found that every so often I needed to do a graphics.clear() on the sprite in charge of rendering the graphics just before I redrew everything to a bitmap. That seemed to give the renderer a "break" and optimize performance. Also, regarding your weird findings with filters, I think this problem is called screen tearing, and in my case I found that it was related with the wmode of the html wrapper holding the swf, the FP version and the browser version. In my case the problems dissapeared when I upgraded my firefox to the latest one. There seems to be a bit of weakness with the flash player's compatibility with the sync to the graphics card lately. I once wrote an animation that consisted of a bunch of 2D points that moved around with some physics + arbitrary behaviour and on each frame, I would draw a lineTo from each point’s previous position to its current one. As you I found that every so often I needed to do a graphics.clear() on the sprite in charge of rendering the graphics just before I redrew everything to a bitmap. That seemed to give the renderer a “break” and optimize performance.
Also, regarding your weird findings with filters, I think this problem is called screen tearing, and in my case I found that it was related with the wmode of the html wrapper holding the swf, the FP version and the browser version. In my case the problems dissapeared when I upgraded my firefox to the latest one. There seems to be a bit of weakness with the flash player’s compatibility with the sync to the graphics card lately.

]]>