WinUAE 3.1.0 beta series

[huepper]

just incredible

[amilo3438]

Quickstart A1200 4Mb Fast

PB-Twisted-A.DMS -> http://janeway.exotica.org.uk/release.php?id=10549

WinUAE Beta 17 (2015.05.02) -> it starts to happen from this version!

[Noth]

Just wanted to add that I can't get OpenBSD 3.2 to get past the "probe HD for installation" phase with 3.0 but it works fine in 3.1-beta. The le0 (A2065) in SLiRP mode is damn slow though... Much slower than in AMIX 2.1.

[Toni Wilen]

Originally Posted by amilo3438

Quickstart A1200 4Mb Fast

PB-Twisted-A.DMS -> http://janeway.exotica.org.uk/release.php?id=10549

WinUAE Beta 17 (2015.05.02) -> it starts to happen from this version!

Will be fixed.

Originally Posted by Noth

Just wanted to add that I can't get OpenBSD 3.2 to get past the "probe HD for installation" phase with 3.0 but it works fine in 3.1-beta. The le0 (A2065) in SLiRP mode is damn slow though... Much slower than in AMIX 2.1.

If 68040 config, then it probably was fixed in b1.

Sorry but I am not going to debug any emulation problem that is not simple "not working at all/crashes" when OS is "alien". Too much work x 100. (Unless OpenBSD sources reveal the problem easily)

[Noth]

Originally Posted by Toni Wilen

Will be fixed.



If 68040 config, then it probably was fixed in b1.

Sorry but I am not going to debug any emulation problem that is not simple "not working at all/crashes" when OS is "alien". Too much work x 100. (Unless OpenBSD sources reveal the problem easily)

It all works (TM) so I'm not complaining. Not a dev so pinpointing that stuff is way beyond my skill set. It hasn't been supported since 2003 anyway.. The hanging in 3.0 is moot, although I'm surprised it's a 68040 issue, not an i/o driver issue.

[Toni Wilen]

http://www.winuae.net/files/b/winuae_3100b21.zip

Beta 21: (RC1)

- b17 DSKSYNC update should not have started DMA in WORDSYNC mode.
- Moved A590/A2091 ram autoconfig board after HD board. A2091 install disk RAM tester expects it. (Some other boards may also have wrong HD/Fast RAM board order)
- Debugger deep trainer "not modified" condition fixed.
- AROS ROM updated.

[amilo3438]

TLC-powertrax
http://eab.abime.net/showpost.php?p=...&postcount=389
http://eab.abime.net/showpost.php?p=...&postcount=393

Above glitch start to happen again from Beta 5! (worked fine in WinUAE 3.1.0 from Beta 1 to Beta 4)

p.s.
Also, regarding the demo, something with plasma was broken in official WinUAE 3.0.0 ?! (just for info)

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

UPFRONT-PlasticPassion

http://janeway.exotica.org.uk/release.php?id=348

White screen with music on plasma part in OCS! (if change to ECS seems better)

It starts to happen from Beta 3!

[Toni Wilen]

Originally Posted by amilo3438

TLC-powertrax

More undocumented features remaining when copper writes to BPLxPT and next cycle is DMA. Apparently bitplanes 5+ work differently or something. (This is not exactly same problem as previously, similar result but different reason..)

I probably try to fix this after 3.1 because full fix needs lots of real hardware testing. Any attempt to guess will break other programs. (As you must have noticed!)

EDIT: Fixed because solution appears to be very simple.

White screen with music on plasma part in OCS! (if change to ECS seems better)

You found really rare bug edge condition. Fixed.

Scroller is actually broken on ECS too and works only accidentally.. It sets DDFSTRT/DDFSTOP value pair so that it causes DMA conflict but because DDFSTOP is >maxhpos, ddfstop condition does not match. Here comes the OCS/ECS difference: At the start of next line OCS Agnus will always clear "horizontal window open" flag but ECS leaves it alone (only ddfstop match will clear it) -> after first line (that caused the conflict), next line will start horizontally from smallest possible position (0x18) which does not cause conflicts anymore! (Technically same as DDFSTRT=0x2c becoming DDFSTRT=0x18)

[Mclane]

Slightly OT but just wanted to thank Amilo3438 for his stunning long term demo testing, helping make WinUAE even better by pointing out so many emulation errors for Toni to fix..

Truly amazing dedication there..People rarely understand just how mind numbing testing is, you have shown real determination..

Thank you...

[amilo3438]

Thank you... but I'm afraid it's like this ...

until my "outdated" WinXP laptop be able to run a new version of WinUAE I'll be able to help.

p.s. I do not intend to purchase a new laptop until the current works... ie while covers all my needs.

Cheers!

p.s.2
Also, as long as it is fun I will try as much as I can (get), despite recently, the lack of free time and consequently difficulties in finding motivation.

And yes, debugging really is not easy, so the occasional "thank you" feels good because you know that there is someone who appreciates the effort.

Perhaps a little too far with emotions, but I felt like... I hope it will not turn out a cry-baby.

[Toni Wilen]

Twisted by Polka Brothers moved to problematic AGA demos. (and thanks for Upfront last minute report )

XP support is not going away as soon as I originally expected because next visual studio version still supports it. Some other "legacy" features may go away but it will still run on XP.

[Toni Wilen]

http://www.winuae.net/files/b/winuae_3100b22.zip

Beta 22: (RC2)

- In some rare situatations both "HARD_DDF_STOP" and "plf_passed_stop" test conditions incorrectly passed in single fetch block. Only happened when DDFSTRT/DDFSTOP value pair caused bitplane dma overrun so result was different corrupted display. (Plastic Passion / Upfront plasma part with OCS Agnus)
- BPL DMA pointer modified one cycle before same channel's BPLxDAT gets accessed behavior adjusted. (Powertrax / The Light Circle)
- Archos ADD-500 is also confirmed having fast ram autoconfig board after HD controller.

[Fieldday]

I ripped the kommos eproms to disk as .bin file (2 27c128 eproms). When I scan the rom folder they are not found. Is there any special to do?

[Toni Wilen]

Originally Posted by Fieldday

I ripped the kommos eproms to disk as .bin file (2 27c128 eproms). When I scan the rom folder they are not found. Is there any special to do?

Nothing special is needed. Perhaps it is different version? (EDIT: Attach or upload them to the zone and I'll check)

[amilo3438]

Originally Posted by Toni Wilen

XP support is not going away as soon as I originally expected because next visual studio version still supports it. Some other "legacy" features may go away but it will still run on XP.

That's nice. I also did not expect that the new version of VS could have support for XP. (really great)

[Leandro Jardim]

When I run a program that uses the Tocatta emulation, I need to disable "sound -> automatic switching" to hear any sound.

[Toni Wilen]

Originally Posted by Leandro Jardim

When I run a program that uses the Tocatta emulation, I need to disable "sound -> automatic switching" to hear any sound.

Argh (and interesting that no one noticed until now, it should have been same in 3.0 too). http://www.winuae.net/files/b/winuae.zip should fix this.

[dlfrsilver]

Toni, i have noticed a problem introduced in b21. Rawread freeze when i try to image any original disk. the problem was not there in b20.

I noticed you changed this :

- b17 DSKSYNC update should not have started DMA in WORDSYNC mode.

Can you have a look and see what happens ? BTW It's not related to my amiga configuration under winuae, Thanks

[Toni Wilen]

"Not enough information!" No command line, no used disk. No nothing.

EDIT: Duplication instructions needed because change you mentioned is related to disk dma and rawread does not even start any disk dma operations.

[Leandro Jardim]

Originally Posted by Toni Wilen

Argh (and interesting that no one noticed until now, it should have been same in 3.0 too). http://www.winuae.net/files/b/winuae.zip should fix this.

Thanks it is fixed.

I noticed that since the beginning, I tried to report it but I am sure I was not clear when I spoke about the issue, well you know, english is not my native language, it is far from perfect...