WinUAE 1.5.1 beta series

[Toni Wilen]

Originally Posted by MandiATO

heavily flicks when I move mouse pointer

Do you mean it is either 100% correct looking cursor or it has pink rectangle depending on location?

Do you see any "logic" in cursor behaviour?

For example, does it stay correct if you move mouse down only? Or is it totally random looking? Set mouse speed to very slow and try moving it pixel by pixel.. (Perhaps the problem only happens in specific positions and my test is using "correct" positions..)

[/QUOTE]

Originally Posted by Toni Wilen

Do you mean it is either 100% correct looking cursor or it has pink rectangle depending on location?

Pointer itself is 100% correct, but uncer pointer whole pointer area is filled by non-transparent pink rectangle.

Originally Posted by Toni Wilen

Do you see any "logic" in cursor behaviour?

For example, does it stay correct if you move mouse down only? Or is it totally random looking? Set mouse speed to very slow and try moving it pixel by pixel.. (Perhaps the problem only happens in specific positions and my test is using "correct" positions..)

There's nothing "logic" in that. This doesn't depend from cursor speed. And it only occurs when I boot to 32 bit GFX modes. When I boot into any other bit-depth GFX mode there's no problem. When I change GFX mode after booting it's also ok, but booting directly to 32 bit mode results in pink rectangle under pointer. I can live with that, I can set 16-bit mode, even 4-bit mode is enought to my work:-). Maybe easiest way will be adds new option in Display option - with description that use it only when problems with pointer occurs. And don't automate dertecting that, this is not worth effort for one user right now.

Regards
MandiATO

[Toni Wilen]

http://www.winuae.net/files/b/winuae_1510b5.zip

Beta 5:

- avi recording crash if only audio recording was enabled
- native dll support fixed
- do not enable second Z3 board if RAM limit reached (=decrease Z3 size) and Z3 size is smaller than 256M (4M+2M combination for example is quite stupid..)
- adjusted max RAM limit calculation (max RAM + max page file instead of currently available RAM + max page file)
- sound adjustments, previously it was possible to lock non-safe portion of directsound sound buffer causing glitches in some cases (which may or may have never happened)
- SuperEagle, Super2xSaI and 2xSaI 32-bit color support added
- disable filter if filter size multiplier is too big for configured display size (instead of overflowing buffers..)
- resolution and doubling setting change on the fly updates filter centering
- added -ddsoftwarecolorkey command line option. Workaround for buggy Intel integrated laptop chipset Vista graphics driver causing non- transparent (pink rectangle) RTG hardware cursor
- (new AHI driver will be only available after 1.5.1)

Originally Posted by Toni Wilen

- added -ddsoftwarecolorkey command line option. Workaround for buggy Intel integrated laptop chipset Vista graphics driver causing non- transparent (pink rectangle) RTG hardware cursor

Thanks! This solve my issue! Now everything works here as expected! Again many thanks!

Regards
MandiATO

[hexaae]

Since v 1.5.x WUAE shows a strange sprite bug with Apidya.
Here is a shot: the dropping bombs (you have to play for 30-40 secs) have those vertical lines following the bomb trajectory:

With WUAE 1.4.6 and previous doesn't appear...
I'm using Apidya with the very latest WHD installer.

[thomas]

uaescsi.device (with CD0 mounted) + Check4GB crashes WinUAE. Screen becomes black and crashdump appears.

[Toni Wilen]

Originally Posted by thomas

uaescsi.device (with CD0 mounted) + Check4GB crashes WinUAE. Screen becomes black and crashdump appears.

It seems there is no check if media is not inserted when calling uaescsi.device CMD_READ and friends.. (HD_SCSICMD works)

What does real scsi.device return in this case? Error 29? (trackdisk "no disk" error)

[Toni Wilen]

Originally Posted by hexaae

With WUAE 1.4.6 and previous doesn't appear...
I'm using Apidya with the very latest WHD installer.

Does it happen with disk version using a500 config? I don't really care about whdload specific problems. (if it is the usual "too fast" everything)

[Steve]

Originally Posted by Toni Wilen

Does it happen with disk version using a500 config?

No it doesn't. I tried the IPF version using a standard A500 config. I didn't find any sprite problems.

[thomas]

Originally Posted by Toni Wilen

It seems there is no check if media is not inserted when calling uaescsi.device CMD_READ and friends.. (HD_SCSICMD works)

The only read command Check4GB uses is TD_READ64 with io_Data/Length/Offset/Actual all 0. This should return either IOERR_NOCMD (if TD64 is not supported) or something like invalid length or so. That's the documented way to check for TD64 support. It should not even try to read anything IMHO. Especially because io_Data is not valid.

Check4GB uses these commands (in this order):

NSCMD_DEVICEQUERY
TD_READ64
HD_SCSICMD Read Capacity
HD_SCSICMD Read(10)

What does real scsi.device return in this case? Error 29? (trackdisk "no disk" error)

Don't know (no real Amiga at hand). Trackdisk.device returns 29.

BTW, it cannot be CMD_READ because this is what TSGUI uses and TSGUI does not crash. In TSGUI uaescsi.device returns -5 on an empty drive. TSGUI does not check for TD64 if the drive is not bigger than 4GB. Check4GB always does the check.

[Toni Wilen]

Originally Posted by thomas

The only read command Check4GB uses is TD_READ64 with io_Data/Length/Offset/Actual all 0. This should return either IOERR_NOCMD (if TD64 is not supported) or something like invalid length or so. That's the documented way to check for TD64 support. It should not even try to read anything IMHO. Especially because io_Data is not valid.

It was the length check that passed when length was zero
Now it returns IOERR_BADLENGTH (same fix also added to uaehf.device)

It still can return weird error codes if no disk in drive and data, length and offset are valid.

[hexaae]

Originally Posted by Steve

No it doesn't. I tried the IPF version using a standard A500 config. I didn't find any sprite problems.

You have to play it for a while. The dropping bomb trash bug happens after the first mushroom...
I tried with another WHD installer (written by another author in 2001) and it does the same...
Even slowing down emulation (CPU, Match A500 speed) the bug is still there. With WUAE 1.4.6 there is no trashed sprite instead. That's strange...
If I try to run the Quickstart config A500 the emulator quits saying the ROM setting is incompatible with the Apidya1.ADF (???)...

[Retro-Nerd]

Originally Posted by hexaae

If I try to run the Quickstart config A500 the emulator quits saying the ROM setting is incompatible with the Apidya1.ADF (???)...

Try the IPF images. There should be no bugs and it boots flawless here.

[StingRay]

Originally Posted by hexaae

If I try to run the Quickstart config A500 the emulator quits saying the ROM setting is incompatible with the Apidya1.ADF (???)...

Sounds as if UAE can't find a Kickstart ROM. Check the path in the "ROM" settings.

[Photon]

When I read 'window sizing': Is it possible to get a 1280x512 ECS resolution on a 1280x1024 capable PC in windowed mode?

To be honest, I haven't experimented, but it seems it's 320x256, 640x256 doubled, or 640x512 with interlace.

Looking forward to next version, will try this one - please advise me as to the above quandary.

[Toni Wilen]

Originally Posted by hexaae

You have to play it for a while. The dropping bomb trash bug happens after the first mushroom...
I tried with another WHD installer (written by another author in 2001) and it does the same...
Even slowing down emulation (CPU, Match A500 speed) the bug is still there. With WUAE 1.4.6 there is no trashed sprite instead. That's strange...

JIT enabled? 1.5.0+ had some JIT adjustments that increase speed but can cause even more weird timings to happen with custom chipset and blitter. (it still works as long as blitter waits are correct. at least in theory..)

Does it start working if you run winuae with -jitevent 3000? ("winuae.exe -jitevent 3000")

If I try to run the Quickstart config A500 the emulator quits saying the ROM setting is incompatible with the Apidya1.ADF (???)...

It means disk has FFS bootblock = non 1.3 compatible. (weird disk..)

EDIT: I was wrong, it really means no KS ROM found. (message means "kickstart replacement" could not boot it)

[Toni Wilen]

Originally Posted by Photon

When I read 'window sizing': Is it possible to get a 1280x512 ECS resolution on a 1280x1024 capable PC in windowed mode?

To be honest, I haven't experimented, but it seems it's 320x256, 640x256 doubled, or 640x512 with interlace.

Looking forward to next version, will try this one - please advise me as to the above quandary.

What do you mean? Scale laced superhires to 1280x1024 with 1:2 pixel mapping? Selecting superhires in resolution selectbox and enabling null filter and setting horiz=1x and vert=2x should do it.

[hexaae]

Originally Posted by Toni Wilen

JIT enabled? 1.5.0+ had some JIT adjustments that increase speed but can cause even more weird timings to happen with custom chipset and blitter. (it still works as long as blitter waits are correct. at least in theory..)

Does it start working if you run winuae with -jitevent 3000? ("winuae.exe -jitevent 3000")



It means disk has FFS bootblock = non 1.3 compatible. (weird disk..)

EDIT: I was wrong, it really means no KS ROM found. (message means "kickstart replacement" could not boot it)

Tried with -jitevent 3000 and tried also to run the game without JIT at all. Still the same gfx bugs.
Happens also with disk images from the original game, I've got it with full box, and shows no sprite problems with WUAE 1.4.6. That's why I thought to a sprite bug in latest WUAE... Isn't strange that in the same conditions (same WUAE settings, same disks etc.) with WUAE 1.4.6 there was no garbage around those sprites 'till WUAE 1.5 came out?

[hexaae]

Originally Posted by Retro-Nerd

Try the IPF images. There should be no bugs and it boots flawless here.

I got the original game.
I WHD-zied it.

PS
The ADF files I tried to use were from a quick game crack downloaded from the net... just to test...

[Photon]

Originally Posted by Toni Wilen

What do you mean? Scale laced superhires to 1280x1024 with 1:2 pixel mapping? Selecting superhires in resolution selectbox and enabling null filter and setting horiz=1x and vert=2x should do it.

Yup, that's it exactly. Thx.