[openMSX-bugs] [ openmsx-Bugs-3427184 ] ASO test 7: screen flickers enormously

[SourceForge.net <no...@so...>]

Bugs item #3427184, was opened at 2011-10-22 02:38
Message generated for change (Comment added) made by manuelbi
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=421861&aid=3427184&group_id=38274

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Graphics
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Manuel Bilderbeek (manuelbi)
Assigned to: Nobody/Anonymous (nobody)
Summary: ASO test 7: screen flickers enormously 

Initial Comment:
Tried this game: http://sharksym.egloos.com/5042966 on openMSX and it's not playable. It seems like every other frame is black. Tried it on real MSX and there it runs smoothly, no flickering is visible at all. (Apparently it also works fine on blueMSX.)
Note that the game runs on turboR only, so it could be turboR specific timing. I get the same broken effect in openMSX 0.8.1 release.

----------------------------------------------------------------------

>Comment By: Manuel Bilderbeek (manuelbi)
Date: 2012-02-10 13:57

Message:
Some info BiFi found out:
Oct 22 2011 16:38:08 <BiFiMSX>  hey, it uses overscan
<BiFiMSX> it was the first time I've seen overscan using R#18

Oct 22 2011 16:51:32 <BiFiMSX>  WOW... it misses interrupts
Oct 22 2011 16:53:06 <BiFiMSX>  try this:
Oct 22 2011 16:53:14 <BiFiMSX>  - start the thing
Oct 22 2011 16:53:19 <BiFiMSX>  - press ESC to exit
Oct 22 2011 16:53:31 <BiFiMSX>  - start it again with ASO /I
Oct 22 2011 16:53:44 <BiFiMSX>  you'll see it running with idle borders

Oct 22 2011 17:17:59 <BiFiMSX>  ah, nice trick... using overscan to do a
page change

Oct 22 2011 18:23:50 <BiFiMSX>  it would be useful though to have some
display now where the FH and F status flags are set and acknowledged
Oct 22 2011 18:24:10 <mth>      BiFiMSX: it will likely require experiments
on the real MSX to figure out exactly when certain things are initialized
Oct 22 2011 18:25:28 <BiFiMSX>  that way you'll have line interrupt and
vblank information and also if an interrupt is missed
Oct 22 2011 18:25:48 <mth>      yep
Oct 22 2011 18:26:03 <mth>      I usually use printfs for that
Oct 22 2011 18:26:23 <mth>      print the line number in frame + clock
ticks within line

Nov 29 2011 21:19:55 <Quibus>   mth: try that ASO test 9... It shows pixels
in places I didn't know were legal on MSX

Nov 29 2011 22:16:02 <Quibus>   I ran that ASO demo 9
Nov 29 2011 22:16:12 <Quibus>   it still bugs like crap on openMSX
Nov 29 2011 22:16:21 <Quibus>   But it looks like it does some odd type of
overscan
Nov 29 2011 22:16:41 <Quibus>   one frame have the screen totally up, next
screen have it totally down (vertical adjust reg)
Nov 29 2011 22:16:50 <Quibus>   and this seems to go wrong timing wise
Nov 29 2011 22:17:09 <Quibus>   next frame I meant
Nov 29 2011 22:17:26 <Quibus>   So, I guess the intention is to use this
with some screensplit to get a rather full screen
Nov 29 2011 22:19:53 <Quibus>   Anyway, it's of course just a wild guess




----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=421861&aid=3427184&group_id=38274