WinUAE 2.0.0 beta series

[Toni Wilen]

Originally Posted by Maren

Ok, here's the numbers for the CPU load:

Config above + 1.6.1 = max 92% (ok)
Config above + 1.6.1 minus PAL filter = max 31% (ok)
Config above + 1.6.2 = max 122% (slowdown)
Config above + 1.6.2 minus PAL filter = max 51% (ok)

It appears that WinUAE became more demanding and my system is no longer up to the challenge. Sorry about the confusion.

Only in NTSC-mode games? Or same difference in all games?

[Maren]

B.C Kid is the only game I know that's affected by this. I can even force NTSC + 60hz on regular 50hz games/demos with the config above and CPU load will still be like 70-something %, which means no slowdowns.

[Toni Wilen]

http://www.winuae.net/files/b/winuae_2000b12.zip


Beta 12:

- increased vertical filter display buffer size by 4 pixels (max PAL overscan fully visible now, again..)
- autovsync vertical resolution autoselection really works now
- more MMU fixes, Linux boots now. MMULib does not work but it most likely is a bug in MMULib which expects some undocumented behavior of 68040 (like how ATC caches work etc..) Commodore's 68040.library works. Note that native code (for example filesystem emulation) can't support MMU address translation (impossible to call bus error handler from non-68k code without some heavy black magic) -> any non-debugging MMU software most likely won't work without using IDE or SCSI emulation. MMU support is considered finished now. Enforcer and Linux working = good enough
- version number bumped to 2.0

[SailorSat]

Auto-VSync works like a charm now

Thank you very much

[gilgamesh]

Has the 1.6.1 beta 12 been uploaded somewhere already? I can't find it.

[TCD]

Originally Posted by gilgamesh

Has the 1.6.1 beta 12 been uploaded somewhere already? I can't find it.

Originally Posted by Toni Wilen

http://www.winuae.net/files/b/winuae_2000b12.zip

[gilgamesh]

Oh, ahem

[Steve]

Using the latest beta of WinUAE I am getting strange results with Oscar CD32 (original CD version). In 1.6.1 the game plays fine but in this beta series the intro works normally but the actual gameplay is in slow-motion.

I am using the standard CD32 config, no special features, no JIT. Thing is the sound is fine, CPU usage is around 3% and the frame rate is 49/50 FPS. Strange eh?

By the way. Oscar WHDLoad version works with no problems using the latest beta so it seems to be a CD32 specific emulation problem!

EDIT: Just played the WHDLoad version and CD32 versions again and noticed a couple of other things. When you collect the 'Gameboy' pickup (I played horror level 1) the screen should change to Gameboy colours temporarily. This happens on the WHDLoad version but not the CD32 version. Also the CD32 version plays the wrong sound effects during gameplay if JIT is turned off. Once JIT is enabled the correct sound effects are played. I think these two bugs were also in 1.6.1 but thought I would mention them here for completeness.

[Toni Wilen]

Originally Posted by Steve

Using the latest beta of WinUAE I am getting strange results with Oscar CD32 (original CD version). In 1.6.1 the game plays fine but in this beta series the intro works normally but the actual gameplay is in slow-motion.

I can't duplicate slow gameplay.

EDIT: Just played the WHDLoad version and CD32 versions again and noticed a couple of other things. When you collect the 'Gameboy' pickup (I played horror level 1) the screen should change to Gameboy colours temporarily.

I don't really care what whdload version do, especially if there is no source available. Only real CD32 test is acceptable

This happens on the WHDLoad version but not the CD32 version. Also the CD32 version plays the wrong sound effects during gameplay if JIT is turned off. Once JIT is enabled the correct sound effects are played. I think these two bugs were also in 1.6.1 but thought I would mention them here for completeness.

Standard problem when timing is bad and sound routine is stupid. (JIT mode has special hack to prevent this)

[Steve]

Originally Posted by Toni Wilen

I don't really care what whdload version do, especially if there is no source available. Only real CD32 test is acceptable

This is a real bug because it works fine on a real A1200 using original disks and real CD32. I ran Oscar in windowed mode if that helps?? I'll do some more testing then. Can anyone else confirm these bugs?

[Toni Wilen]

Originally Posted by Steve

This is a real bug because it works fine on a real A1200 using original disks and real CD32. I ran Oscar in windowed mode if that helps?? I'll do some more testing then. Can anyone else confirm these bugs?

as I said, I don't care what happens unless run on real CD32 (is this very common Oscar&Diggers version?)

Better question is: how do I confirm the bug? Exact steps please because I never waste even one second to learn some random game I have never played and/or never want to play

[Steve]

Originally Posted by Toni Wilen

Better question is: how do I confirm the bug? Exact steps please because I never waste even one second to learn some random game I have never played and/or never want to play

This is the common Diggers/Oscar version.

Ok...

1. Start the game
2. In cinema enter Screen 2 by pulling down on joypad. (horror level)
3. Once in the level run right and jump up onto the platform. Run and jump the gap till you reach the position in the screenshot.
4. The spinning film cannister you can see, will release a Gameboy. When you collect the Gameboy the background palette should turn monochrome like the old Gameboy. You can confirm this is true on a real CD32. Should take no more than a few minutes to confirm.

I don't really think that bug is beta related but oh well. I've also noted that if you leave the character for a few seconds during gameplay he starts his 'idle' animations. On WinUAE the sprite vibrates slightly up and down and causes total corruption of the screen for a couple of seconds during its 'speaking on the phone' animation. This only occurs on the CD32 version. The IPF and WHDLoad is fine just like the background palette bug.

As for the beta related problem I have found out some more.

Start the game in windowed mode = slow-motion
Start the game in fullscreen mode = slow-motion
Start the game in fullscreen + vsync mode = normal speed.

Switching from windowed to fullscreen + vsync doesn't fix the slow-motion bug. It remains in slow-motion. However start the game in fullscreen + vsync mode and its fine. Weird. I also noticed when the game is running normally that turning ON JIT causes the mixed up sound effects. When its running in slow-motion JIT has the opposite effect. Quite bizarre.

Bit more testing has just revealed that if I start in fullscreen + vsync the speed is normal. If I then switch to windowed mode the speed is still normal. If I switch to standard fullscreen mode the speed is still normal. The bug only occurs when starting in either windowed mode or normal fullscreen mode. Starting in fullscreen + vsync works!

One last bug which I don't think is beta related but I'll add anyway is when you enter the water in the CD32 version. The screen flickers and jumps when you're underwater. This shouldn't happen. You can see all these bugs by testing the first horror level on WinUAE.

[Toni Wilen]

Originally Posted by Steve

Start the game in windowed mode = slow-motion
Start the game in fullscreen mode = slow-motion
Start the game in fullscreen + vsync mode = normal speed.

I ALWAYS run in windowed mode = must be something specific in your host configuration.

4. The spinning film cannister you can see, will release a Gameboy. When you collect the Gameboy the background palette should turn monochrome like the old Gameboy. You can confirm this is true on a real CD32. Should take no more than a few minutes to confirm.

Colors changed to ugly greenish color. (not background, player sprite or scoreboard) Is that wrong? (too lazy to do real CD32 test today)

EDIT: perhaps I changed something, try the version linked in MMU thread..

[Steve]

Originally Posted by Toni Wilen

I ALWAYS run in windowed mode = must be something specific in your host configuration.

I've just tested it using the quickstart CD32 option and I get the same slow-motion problem. This didn't occur in the last version of WinUAE. Does no-one else experience this?

The vibrating sprite bug I mentioned is seemingly fixed by enabling JIT. Strange. Why is JIT selected on the quickstart for CD32? I thought JIT was mainly used to speed up emulation of OS operations and not games.

Another thing I noticed on the quickstart was that the compatibilty settings for my Sidewinder Plug & Play gamepad are set wrong and aren't like a real CD32 pad. The settings are set to this:

0 = Red, 1 = Blue, 2 = Yellow, 3 = Green, 4 = FFW, 5 = RWD

The correct setting should be this:

0 = Red, 1 = Blue, 2 = Green, 3 = Yellow, 4 = RWD, 5 = FFW

Originally Posted by Toni Wilen

Colors changed to ugly greenish color. (not background, player sprite or scoreboard) Is that wrong? (too lazy to do real CD32 test today)

Yes this is correct. The foreground (not background - sorry) should change to ugly greenish colour like the old Gameboy. The player sprite and scoreboard do not change colour!

EDIT:

Originally Posted by Toni Wilen

perhaps I changed something, try the version linked in MMU thread..

Just tried that version using the quickstart CD32 option. No difference. Did you get the palette change playing the CD32 version because its definitely not working here.

I've noticed the HOST settings remain as the previous custom settings if you load a quickstart setting. Is it supposed to do that?? I assume it does that by design. Had me puzzled for a while.

EDIT 2:

I've just tested all the betas to see when the first occurence of slow-motion in Oscar CD32 occurs. It seems that Oscar is broken and doesn't get further than the intro sequence in betas 1 to 7. In beta 8 the game loads properly but then slowdown is evident in this beta.

[JohnF]

More Quickstart A500 1.3 best compat testing:

Desert Dream (D1 3B9D369B, D2 DA6B9D52): Colour (copper?) bars disappear from inside cube vector on starfield in b12, b11, and b9. Worked okay in b10.

[prowler]

Try the beta 12 fixed version posted here.

[JohnF]

Originally Posted by JohnF

More Quickstart A500 1.3 best compat testing:

Desert Dream (D1 3B9D369B, D2 DA6B9D52): Colour (copper?) bars disappear from inside cube vector on starfield in b12, b11, and b9. Worked okay in b10.

Same in Beta 12 2009.09.01.

Also noticed a minor audio glitch: when the 'revolution' sample is played (accompanied by the word on screen), the sample repeats.

[Ian]

When I tried to run Oscar last night it didn't work at all, I didn't have time to check why but I suspect it's another "weird" issue in the new cycle exact mode. (tested both Digger and Oscar images in the TOSEC dats)

Can anyone confirm this as I am away from my PC for days and will not be able to check it.

[Toni Wilen]

Originally Posted by Ian

it didn't work at all

....

[Steve]

I think Toni wants more info. lol. Do you mean you just get a black screen and it doesn't boot? Does the CD32 intro and spinning CD appear? If it doesn't boot correctly first time I find a reset usually gets it going.