WinUAE 1.5.0 beta series

[UncleSpam]

Just tried Beta 8 and it crashes upon the initial Amiga boot... did not have a problems with WinUAE crashing prior to this version.

- Bogdan

Update: Once I turned the Filters OFF, things worked again.

[Toni Wilen]

Originally Posted by UncleSpam

Just tried Beta 8 and it crashes upon the initial Amiga boot... did not have a problems with WinUAE crashing prior to this version.

- Bogdan

Update: Once I turned the Filters OFF, things worked again.

Any possibility for more information? It shouldn't be anything too secret

What filter? All filters? Windowed? Fullscreen? Filter adjusting? etc..

EDIT: do you have some beta windows updates because user32.dll and other windows dll's don't have matching symbol file. (and makes debugging difficult)

[Steve]

Beta 8 seem to be pretty good. Sprites in SWOS are now working correctly again as are the hires sprites in super skidmarks. I've noticed a couple of small black lines which appear now and again in the top left corner of the screen during a game of swos. I'm sure they weren't there before. Well its nothing too major really. The speed seems to be back to normal again in full screen mode although the sound can get quite choppy on my system (though thats probably due to my slow computer).

I still get odd looking icons on my Workbench though. See the screenshot below.

[thomas]

WinUAE crashes if ROM file is not found.

[thomas]

... and the kom.key file is not searched in the same directory as the kick.rom file, but only in the directory configured in the paths section. Kick.rom without rom.key crashes, too.

[Ian]

Originally Posted by Toni Wilen

- timer calibration removed, uses now performance counters only, also checks periodically if counter frequency changes and adjusts to new rate automatically (possible fix for random power saving slowdown problems)

It's much better now (hires and double line enabled), but if I use when I use the scale2x filter (set to lores and single line) it's far worse.

Constant crackling noise. When I set the throttle of my CPU to 100% again the crackling disappears.

[UncleSpam]

Originally Posted by Toni Wilen

Any possibility for more information? It shouldn't be anything too secret

What filter? All filters? Windowed? Fullscreen? Filter adjusting? etc..

EDIT: do you have some beta windows updates because user32.dll and other windows dll's don't have matching symbol file. (and makes debugging difficult)

Toni,

I normally run the emulation in Fullscreen mode for Native and RTG.

The WB resolution is my native laptop/external LCD display, 1920x1200x32.

The emulation crashes if the Filter Settings have filtering enabled, using the above settings. If I disable the filtering, the emulation runs properly.

I tested with many filter settings, it makes no difference.

If I change the WB resolution (1280x1024x32 tested) to less than 1920x1200, then it works in full screen mode with filters set to enabled.

It also works in Windowed mode, obviously with a lower resolution than 1920x1200.

- Bogdan

[Ed Cruse]

B8 is working pretty good, one or two cores doesn’t seem to matter anymore, you almost have it. There are two minor problem that you should probably know about. Listed below are the conditions I used to create the problems, there may be others.

1. PC side – 1280x1024, 16/32 bit color, Default, Fullscreen, Filters disabled. Amiga side – 8 bit RTG. Hitting F12 causes the Amiga part to go black/red/green/blue, with the background being black. If I don’t do anything but hit OK everything goes back to normal. It appears if I do anything that causes the monitor to blank for a second then the colors remain wrong when back in the Amiga. In Display, setting the resolution, color depth, and vertical freq are settings that cause the monitor to blank. Filters enabled or 16/32 bit RTG don’t seem to have this problem.

2. PC side – 1280x1024, 16/32 bit color, Default, Fullscreen, Filters enabled/disabled, Scale2x filter. Amiga side – 8 bit RTG. The left half of the color wheel in Prefs->Palette is solid orange instead of what it’s supposed to be. 16/32 bit RTG don’t seem to have this problem.

[Toni Wilen]

Originally Posted by Ed Cruse

[FONT=Courier New][SIZE=2]1. PC side – 1280x1024, 16/32 bit color, Default, Fullscreen, Filters disabled. Amiga side – 8 bit RTG. Hitting F12 causes the Amiga part to go black/red/green/blue, with the background being black. If I don’t do anything but hit OK everything goes back to normal. It appears if I do anything that causes the monitor to blank for a second then the colors remain wrong when back in the Amiga. In Display, setting the resolution, color depth, and vertical freq are settings that cause the monitor to blank. Filters enabled or 16/32 bit RTG don’t seem to have this problem.

I guess you have "match RTG depth" setting enabled? (I think I said always mention this setting state when using fullscreen RTG modes.. AND always say if there is difference between windowed and fullscreen) GUI needs different palette. (not resetting to normal palette is a bug but GUI color change always happen)

2. PC side – 1280x1024, 16/32 bit color, Default, Fullscreen, Filters enabled/disabled, Scale2x filter. Amiga side – 8 bit RTG. The left half of the color wheel in Prefs->Palette is solid orange instead of what it’s supposed to be. 16/32 bit RTG don’t seem to have this problem.

Same in fullscreen and windowed?

[Toni Wilen]

Originally Posted by thomas

... and the kom.key file is not searched in the same directory as the kick.rom file, but only in the directory configured in the paths section. Kick.rom without rom.key crashes, too.

and fullscreen/windowed testing here would have helped to debug this quicker. This only happens in fullscreen because it tries to show dialog but fullscreen isn't yet fully enabled.

Please, ALWAYS test both fullscreen and windowed.

[Toni Wilen]

Originally Posted by UncleSpam

Toni,

I normally run the emulation in Fullscreen mode for Native and RTG.

The WB resolution is my native laptop/external LCD display, 1920x1200x32.

The emulation crashes if the Filter Settings have filtering enabled, using the above settings. If I disable the filtering, the emulation runs properly.

I tested with many filter settings, it makes no difference.

If I change the WB resolution (1280x1024x32 tested) to less than 1920x1200, then it works in full screen mode with filters set to enabled.

It also works in Windowed mode, obviously with a lower resolution than 1920x1200.

I can't crash it using 1920x1200x32 + filters. What do you mean by changing WB resolution? Do you mean RTG resolution (but RTG does not use filters..)
Does it crash when switching to RTG mode?

Does it crash in 1920x1200 windowed? (You can select and use it even if WinUAE window does not fit very well..)

[Toni Wilen]

RTG 8-bit colorwheel fixed. Both 8 and 16-bit optimized blitter routines blitted too much data when blit was >32 bytes wide and it was not divisible by 32.

[UncleSpam]

Originally Posted by Toni Wilen

I can't crash it using 1920x1200x32 + filters. What do you mean by changing WB resolution? Do you mean RTG resolution (but RTG does not use filters..)
Does it crash when switching to RTG mode?

Does it crash in 1920x1200 windowed? (You can select and use it even if WinUAE window does not fit very well..)

I mean RTG resolution.

My Workbench is set to 1920x1200x32 resolution and I normally enable filtering, to be used not for RTG resolutions, but for the native Amiga resolutions. It helps me better position the displays on my monitor.

If I set the resolution below the 1920x1200 resolution, in Fullscreen or Windowed mode, it works fine.

Just tried Windowed RTG in 1920x1200x32 and it crashed. In 1600x1200x32, it worked properly using both Windowed and Fullscreen RTG.

Hope this helps.

Bogdan

[Ed Cruse]

Originally Posted by Toni Wilen

I guess you have "match RTG depth" setting enabled? (I think I said always mention this setting state when using fullscreen RTG modes.. AND always say if there is difference between windowed and fullscreen) GUI needs different palette. (not resetting to normal palette is a bug but GUI color change always happen)



Same in fullscreen and windowed?

Sorry, I'll try doing better.

[Toni Wilen]

Attached in this post because winuae.net seems to have issues again.

Beta 9:

- crash when fullscreen enabled and error dialog was shown before display was changed to fullscreen
- CD32 state restore compatibility improved
- SSCAN2 'undocumented' feature emulated. "If SSCAN2 bit in FMODE is set, then disable SH10 horizontal coincidence detect". This really means both one AND zero SH10 match = single sprite can appear twice on screen! (Fantastic Dizzy CD32 background)
- AGA mode only: last line of sprite may have been redrawn if bitplane width was changed and sprite was reused later (introduced in 1.4.5)
- debugger sp command improvements
- AGA sprite outside bitplanes position bug fix (introduced in 1.4.5)
- Picasso96 8-bit mode wrong palette after mode switch
- Picasso96 8-bit and 16-bit optimized blitter bug fix, too much data was blitted if width was >32 bytes (not pixels) and it was not divisible by 32. 32-bit mode was working properly.
- emulated LPSTOP instruction (68060 only)
- MMU instructions didn't check supervisor mode. Note that there is no MMU emulation, I think I have to repeat this every time I talk about MMU
- added 68040/68060 MMU instructions to table68k (no more generic "MMUOP" in debugger's disassembler) 68030 MMU instructions are still shown as "MMUOP030", they don't really fit to table68k format because first instruction word does not specify the instruction type
- MOVEC weird disassembly updated (MOVEM is still weird)

[turrican3]

do you need p96 speed test ?

[Steve]

This is great stuff Toni. Fantastic Dizzy is working perfectly. Its incredibly slow on my machine but its fairly playable using frameskip. This is going to be one of the best versions of WinUAE yet! Well done.

[turrican3]

Originally Posted by turrican3

do you need p96 speed test ?

i asked this because i launched p96speed by curiosity and it was so slow !!!! I stopped it before the end.If you need results ? tell it.

[UncleSpam]

Toni,

Here's my latest testing. See attached file.

Bogdan

[Ed Cruse]

Originally Posted by Toni Wilen

Attached in this post because winuae.net seems to have issues again.

Beta 9:

- crash when fullscreen enabled and error dialog was shown before display was changed to fullscreen
- CD32 state restore compatibility improved
- SSCAN2 'undocumented' feature emulated. "If SSCAN2 bit in FMODE is set, then disable SH10 horizontal coincidence detect". This really means both one AND zero SH10 match = single sprite can appear twice on screen! (Fantastic Dizzy CD32 background)
- AGA mode only: last line of sprite may have been redrawn if bitplane width was changed and sprite was reused later (introduced in 1.4.5)
- debugger sp command improvements
- AGA sprite outside bitplanes position bug fix (introduced in 1.4.5)
- Picasso96 8-bit mode wrong palette after mode switch
- Picasso96 8-bit and 16-bit optimized blitter bug fix, too much data was blitted if width was >32 bytes (not pixels) and it was not divisible by 32. 32-bit mode was working properly.
- emulated LPSTOP instruction (68060 only)
- MMU instructions didn't check supervisor mode. Note that there is no MMU emulation, I think I have to repeat this every time I talk about MMU
- added 68040/68060 MMU instructions to table68k (no more generic "MMUOP" in debugger's disassembler) 68030 MMU instructions are still shown as "MMUOP030", they don't really fit to table68k format because first instruction word does not specify the instruction type
- MOVEC weird disassembly updated (MOVEM is still weird)

I've downloaded the attached file twice and both times I got zip file corrupted error. Any ideas? I tried unzipping b8 again and had no trouble.