Lesson on Accelerated Rendering

JJS jjs at krutt.org
Mon Apr 29 13:08:47 EDT 2019


This blank screen and the shifting of the image were a few of the bugs 
that were solved.

In the BUG quality list is a tip on how to solve it by sending it to me 
in a tick or something.

But this was before the bug was solved.

I previously turned on and off the accelerated when i needed a keyboard 
on android. This is not needed anymore, that bug is solved.

But maybe in your way of using it between stack, i might think that you 
need to turn it off before loading a new stack and then turn it on again.

It's worth a try.


Op 29-4-2019 om 03:50 schreef Sannyasin Brahmanathaswami via use-livecode:
> And here is Mark Waddingham's more technical explanation:
> <https://www.mail-archive.com/use-livecode@lists.runrev.com/msg84428.html>
>
> BR: After reading the above, in so far as my app is modular, made up of many stacks, all of  which have object that move, either with dynamic or scrolling. It is simple enough always to "set acceleratedRendering to true"
>
> Caveat he does not mention, that doing so in a preopenstack handler  of Stack B, if you have another stack open first (A), will cause a blank black screen, when opening Stack B, on Android. Whether this a bug or not, has yet to be determined. But if  we allow the stack to render first, and "set acceleratedRendering to true" in the top of the first handler run... no blank screen.
>
> That’s for moving between stacks, problem solved.
>
> Now within a stack, the open ACLR set to default-false. Inside one handler, I still get a blank black screen before I render our puzzle tiles after rearranging them with screen locked. I suspect that
>
> "set acceleratedRendering to true" while the screen is locked will be the problem.
>
> More testing needed. What is missing in Mark's reply is "how this all affects performance on Android."  I supposed it can be inferred... but  I can wait to hear Mark W. talk  on "Optimising" at the conference (I'll be tuning in with SimulCast)
>
> Stephen Barncard wrote:
>
>      It is frightening how much mumbo jumbo goes down every time we 'touch' a
>      field.
>      I had no idea.
>      No wonder loops could get glacial quickly if one is not careful with using
>      fields as a destination.
>
>
> _______________________________________________
> use-livecode mailing list
> use-livecode at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode




More information about the use-livecode mailing list