Re: Improvement of mobile scrollings

2014-10-08 Thread jude
I think it should be the same as the framerate unless the developer overrides it. So, public function get minUpdateInterval ():int { if (_minUpdateInterval==null) { _minUpdateInterval = stage.framerate; } return _minUpdateInterval; } So minUpdateInterval would be null by default. On Wednesday,

Re: Improvement of mobile scrollings

2014-10-07 Thread after24
Hi Justin, Every device capable (for a specific scroll case) to go beyond 30 fps should take advantage of this change. One more time, the improvement is observable during a user "dragged scroll" (application framerate must be greater than 30 fps, I set it to 60 on my devices). Vincent. -- Vie

Re: Improvement of mobile scrollings

2014-10-07 Thread Justin Mclean
Hi, > Anyone mind if I increase the default from 30 to 90 and see if Mustella has > issues with that? I think it’s worth it given the perceived performance > improvement and my guess is that the possible side effects are small, given > that the field we’re messing with was marked private. Do we k

Re: Improvement of mobile scrollings

2014-10-07 Thread OmPrakash Muppirala
On Oct 7, 2014 11:04 PM, "Erik de Bruin" wrote: > > Hi, > > I’ve implemented after24’s patch and there were no objections from > Mustella. The way I implemented the patch doesn’t automatically give the > performance improvements he intended it to have. That is because I didn’t > increase the defau

Re: Improvement of mobile scrollings

2014-10-07 Thread Erik de Bruin
Hi, I’ve implemented after24’s patch and there were no objections from Mustella. The way I implemented the patch doesn’t automatically give the performance improvements he intended it to have. That is because I didn’t increase the default rate, but kept it at 30. After24 suggests that the performa

Re: Improvement of mobile scrollings

2014-10-06 Thread Erik de Bruin
I applied the patch to ‘flex-sdk’. I left the default on 30, instead of the suggested 90, in order to reduce the chance of side effects and to maintain backwards compatibility. We can always increase the default if the Mustella test don’t object to the current change ;-) EdB On Mon, Oct 6, 20

Re: Improvement of mobile scrollings

2014-10-06 Thread after24
Hi Erik, Just added a proper patch. Have a nice day. Vincent. -- View this message in context: http://apache-flex-development.247.n4.nabble.com/Improvement-of-mobile-scrollings-tp41029p41107.html Sent from the Apache Flex Development mailing list archive at Nabble.com.

Re: Improvement of mobile scrollings

2014-10-05 Thread Erik de Bruin
Hi Vincent, Thank you for submitting your patch! I took a quick look, and the first thing I noticed is that it is not really a patch, but rather a patched version of the entire class file (Scroller.as). As the whitespace and some other stuff seems to be very different between the version you subm

Re: Improvement of mobile scrollings

2014-10-03 Thread flex
flex.apache.org Sent: Fri, 03 Oct 2014 7:17 AM Subject: Re: Improvement of mobile scrollings Hi David, In fact the changes of the value doesn't affect the absolute performance of the scrolling but it removes a very limiting factor during the drag of the content of a scroller. Combined with a hi

Re: Improvement of mobile scrollings

2014-10-03 Thread after24
I just added a patch that replaces the private static const MAX_DRAG_RATE by the public static property maxDragRate with a default value of 90 (former value of the constant was 30). link to JIRA -- View this message in context: http://apach

Re: Improvement of mobile scrollings

2014-10-03 Thread Harbs
The question is whether it should be a class property of an instance property. I’d lean towards a class property. On Oct 3, 2014, at 3:59 PM, Erik de Bruin wrote: > I would say that you can best implement this by adding property that you > give the initial value of the constant you mention. Ins

Re: Improvement of mobile scrollings

2014-10-03 Thread Erik de Bruin
I would say that you can best implement this by adding property that you give the initial value of the constant you mention. Instead of assigning the constant (elsewhere in the code, I haven’t looked), you would assign the property. The property setter would have to take care of any additional assi

Re: Improvement of mobile scrollings

2014-10-03 Thread after24
Hi David, In fact the changes of the value doesn't affect the absolute performance of the scrolling but it removes a very limiting factor during the drag of the content of a scroller. Combined with a high framerate (60 fps) and a stage quality set to low during "critical animations" I managed to

Re: Improvement of mobile scrollings

2014-10-03 Thread flex
Wow so after all that complaining about "performance" on mobile it could end up being nothing more than a constant being set to a static value ... David -Original Message- From: after24 To: dev@flex.apache.org Sent: Fri, 03 Oct 2014 5:00 AM Subject: Re: Improvement

Re: Improvement of mobile scrollings

2014-10-03 Thread after24
In fact the performance improves even for values greater than the application framerate (90 seems to be the limit on my phone). -- View this message in context: http://apache-flex-development.247.n4.nabble.com/Improvement-of-mobile-scrollings-tp41029p41032.html Sent from the Apache Flex Dev