Page 8 of 15

Re: ADFFS 2.71 beta

Posted: Sun Feb 03, 2019 9:58 pm
by JonAbbott
richw wrote: Sat Feb 02, 2019 11:14 pm I noticed that exiting ADFFS causes a bit of a graphical mess in the desktop. The screen needs a redraw (F12 and enter fixes it).
Yes, I did spend some time looking at it a few days ago. GraphicsV doesn't handle switching graphics drivers very well, so I'll need to find some workarounds to reinstate the desktop.

Re: ADFFS 2.71 beta

Posted: Mon Feb 04, 2019 4:45 pm
by Vanfanel
@Jon: I was testing latest beta with Fireball (the first part) and after a while playing the game started to slowdown and the ship and balls partially dissapeared, as if only a part of the screen was being redrawn.

Re: ADFFS 2.71 beta

Posted: Mon Feb 04, 2019 6:26 pm
by JonAbbott
Vanfanel wrote: Mon Feb 04, 2019 4:45 pm @Jon: I was testing latest beta with Fireball (the first part) and after a while playing the game started to slowdown and the ship and balls partially dissapeared, as if only a part of the screen was being redrawn.
I played it for 10-15mins, but couldn't reproduce the issue. I did however notice that it may not by syncing to VSync, which sounds like the issue you're describing. I'll take a look at the code and see if there's any obvious bugs I can fix.

I've been through all the publicly available packages that were being installed to package ID folders, and have modified conflicting game !App folders so they're unique. Please uninstall and then reinstall the following games if you want to get rid of the package ID folders they're currently installed too:

Code: Select all

1049200 Adventure Collection, The (1993) (Alpine Software)
1003600 Ballarena (1994) (Uffenkamp Computer Systeme)
1023100 Ballarena (1990) (Sisteme)
1004400 Blaston (1994) (Uffenkamp Computer Systeme)
1064000 Blaston (1991) (Eterna)
1002200 Cartoon Line part one (1993) (Uffenkamp Computer Systeme)
1019500 Cartoon Line part one (1991) (Eterna)
1009100 COPS (1989) (Alpine Software)
1048400 Krisalis Collection, The (1993) (Krisalis Software)
1023300 Mad Professor Mariarti (1990) (Krisalis Software)
1028100 Pipe Mania (1989) (Krisalis Software)
1041000 Plague Planet (1988) (Alpine Software)
1059500 Revelation! [BUZZ version] (1993) (Krisalis Software)
1056600 S.W.I.V. [BUZZ version] (1993) (Krisalis Software)
1036100 Tactic (1996) (Uffenkamp Computer Systeme)
1036600 Terramex (1988) (Grandslam Entertainments)

Re: ADFFS 2.71 beta

Posted: Mon Feb 04, 2019 11:38 pm
by JonAbbott
JonAbbott wrote: Mon Feb 04, 2019 6:26 pm
Vanfanel wrote: Mon Feb 04, 2019 4:45 pm @Jon: I was testing latest beta with Fireball (the first part) and after a while playing the game started to slowdown and the ship and balls partially dissapeared, as if only a part of the screen was being redrawn.
I played it for 10-15mins, but couldn't reproduce the issue. I did however notice that it may not by syncing to VSync, which sounds like the issue you're describing. I'll take a look at the code and see if there's any obvious bugs I can fix.
Looking into this has opened a whole can of worms. The game was written for Arthur going by the date (1988) and is accessing an area of Page Zero that was moved in RISC OS 2. Once I've added support for Arthur to the JIT, I can resume looking into the original issue :roll:

From the quick look I did manage to take, I couldn't see any obvious means of syncing the screen updates to VSync. In fact I'm not entirely sure how it's locking it's frame rate, other than possibly counting VSync as it claims EventV and increases a counter every VSync. Its not double banked as it's using one screen bank for the display and the other to store what's behind the sprites, so it can remove them between updates.

It's probably a fair assumption that as the sprites flicker, the update isn't linked to VSync, but it is waiting n VSync between updates to regulate the speed. The wait is possibly in the wrong place and being done after the screen updates instead of before.

Re: ADFFS 2.71 beta

Posted: Tue Feb 05, 2019 5:42 pm
by Vanfanel
I'd like to report a couple of issues I found yesterday when testing:
-Joystick in Legend of the Lost Temple does not work
-Left mouse button in PipeMania stops working after some play

Re: ADFFS 2.71 beta

Posted: Tue Feb 05, 2019 8:20 pm
by JonAbbott
Vanfanel wrote: Tue Feb 05, 2019 5:42 pm -Joystick in Legend of the Lost Temple does not work
I'll refer you to the Joystick Support page, or the game manual:
- The 'Joystick 1' option refers to the Voltmace Mouse Eliminator.
- The 'Joystick 2' option refers to R.T.F.M. joystick board.
- The 'Fire Button' panel is used for selecting the button used for firing on mouse.
I suspect you selected Joystick 1, thinking it was Joystick 1!

Re: ADFFS 2.71 beta

Posted: Tue Feb 05, 2019 10:17 pm
by JonAbbott
v2.71w package now available. Lots of bug fixes, mostly to fix IOMD and low vector builds but there was a major issue where 26bit Module * commands weren't working.

It now attempts to refresh the desktop when ADFFS is quit, but RISC OS seems to be messing up some fonts. Any open command windows for example change to a small font :?
Vanfanel wrote: Tue Feb 05, 2019 5:42 pm -Left mouse button in PipeMania stops working after some play
Can you try mounting the floppy image, it may be protection kicking in. I've also noticed it doesn't always run first time, which may also be protection. I'll need to debug it to check.

Richard - if you want a title to test Voltmace emulation, PipeMania only supports Voltmace.

Re: ADFFS 2.71 beta

Posted: Wed Feb 06, 2019 11:23 am
by Vanfanel
@Jon:
-I can't reproduce the problem with PipeMania anymore. I was playing for twenty minutes and cleared two stages, with no problems. Will come back to that if I encounter the mouse problem again.
-Legend of the Lost Temple has workig joystick, yes! It was just a matter of selecting the Joystick 2 option... I should have RTFM before coming here to report, sorry about that!
-HeroQuest intro is workig again (love that music, eve if the game is not that good!)

Re: ADFFS 2.71 beta

Posted: Wed Feb 06, 2019 6:50 pm
by JonAbbott
Vanfanel wrote: Mon Feb 04, 2019 4:45 pm @Jon: I was testing latest beta with Fireball (the first part) and after a while playing the game started to slowdown and the ship and balls partially disappeared, as if only a part of the screen was being redrawn.
I know why this is occurring, but don't understand why - if that makes sense.

The game syncs to the 50Hz Counter, which is tied to VSync. I've switched it to wait for VSync, as its more power efficient but it makes no difference.

As soon as OS_Byte 19 returns to the game, it clears the bat/ball and redraws it so there's no way the blitter could ever capture that mid-redraw as the blit it performed before the soft VSync is raised and hence just prior to the OS returning to the game - and yet it somehow manages to perform the blit within a hundred instructions of OS_Byte 19 returning to the game...which makes no sense what-so-ever.

It implies there's a major issue somewhere.

Re: ADFFS 2.71 beta

Posted: Wed Feb 06, 2019 8:01 pm
by Vanfanel
@Jon: Sorry to have opened the can of worms, man. I just try to do my best reporting these issues, but really sorry that it means more investigation on your side :(