On 26-Apr-05, at 11:47 AM, Jonathan Johnson wrote:
On Apr 24, 2005, at 9:05 PM, Asher Dunn wrote:
lol, ok. It's fine on OSX,
Not quite. On OS X, it appears you're calling
CheckEventQueueForUserCancel.
Why does that happen automatically in a tight loop? I think we've been
over this before but I can't recall why this is still the case (not a
big problem, just curious).
That will prevent the SPOD from appearing, but it still makes your app
behave unlike any other app when it's in windowed mode. You should use
the thread on all platforms, which will prevent the OS from tagging
your application as unresponsive. It won't slow your framerate down
much, and you'll be a better citizen.
Unfortunately RB Threads make OpenGL views go whacky on Mac OS X (frame
swaps don't always "take" - with very nasty results - this affects RB3D
as well). This problem also got MUCH worse in a certain version of RB
I'm not allowed to talk about here.
I've whipped up a project that will hopefully shed some light on this
issue, and I've attached it to a bug report:
<http://www.realsoftware.com/feedback/viewreport.php?
reportid=boxjgviz>. I consider this to be a very serious problem and
hope you RS guys take it seriously too! I've eliminated all the causes
suggested in the past (e.g. partial frames, lower framerates than a
tight loop) and I've created a simple 3D scene that visualizes the
problem to the point of being blatantly obvious.
I've also made the RB project publicly available here:
<http://www.chaoticbox.com/frametest.zip> so the rest of ya can see
what I'm talking about and sign on if yer so inclined.
P.S. The project requires Quesa, and is Mac CFM Carbon only (I declare
into AGL to prove a point).
Thanks,
Frank.
------------
Open Source RB Plugins and Classes
<http://developer.chaoticbox.com>
_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>
Search the archives of this list here:
<http://support.realsoftware.com/listarchives/lists.html>
|