WinUAE 4.9.0 beta series (Was 4.5.0)

[Zarnal]

@amilo3438

I can't download from your link.
I've found another link :

http://janeway.exotica.org.uk/release.php?id=8706

[hexaae]

Originally Posted by hexaae

Unlikely, but I'll try...

@Toni
Possibly a bug in a PPaint plugin (personal_icon_io.library) I was using and I was able to reproduce working with Coloricons...
Re-downloaded from Aminet, replaced with "pio_icon/src/personal_icon_io.library" binary from the archive (which is suspiciously different from same file from: "pio_icon/libs/personal_icon_io.library" I was using Have no idea why the binary in src/ is different from the file in libs/ ) and now seems stable...
Ignore this.

[Toni Wilen]

Originally Posted by amilo3438

Critical Mass by Parallax (ADF from arabuusimiehet) has problem with this winuae.7z!
It stops with loading but the music is still playing! (was fine in official beta28)

Tested config: quickstart A500 1.3 1Mb Chip RAM

This (as everyone probably already knows) has broken first disk (and no one has found working one) but it should still work in 1M Chip config. Except intro animation. I'll check later.

[Toni Wilen]

https://download.abime.net/winuae/fi...uae_4900b29.7z
https://download.abime.net/winuae/fi...e64_4900b29.7z

Chipset updates are almost done. Really. At least I hope. If there was some other updates I delayed until chipset update is done, now it is time to remind me

(FloppyDriveBridge thing does not need reminding. It will probably happen.)

Beta 29:

- b28 copper update was wrong. Second attempt. (Hotbleeps etc. This time without breaking others.)
- Sprite fix, sprite DMA does not start if frame's initial SPRxPOS/SPRxCTL fill loads vertical start position that matches current vertical position.
- Added bsdsocket IoctlSocket SIOCIFADDR, SIOCIFFLAGS, SIOCIFBRDADDR, SIOCIFCONF, SIOCIFNETMASK, SIOCIFMETRIC and SIOCIFMTU support. Read-only.
- Fixed uaegfx indirect mode BlitTemplate() harmless one byte buffer out of bounds read.
- VPOSW register access always clears LOL bit (if it was set = NTSC long line). It can't be set to one by writing to VPOSW.
- NTSC long/short line order was wrong after horizontal wrap around.

[Zarnal]

Originally Posted by Toni Wilen

Chipset updates are almost done. Really. At least I hope. If there was some other updates I delayed until chipset update is done, now it is time to remind me

You wrote (# 642) that you would watch for this demo at the end of the new blitter emulation.



UpToFront-Coolfridge / Smooth Copper.

Thank you.

[Viceroy]

Just 2 outstanding bugs for me Toni ;-)

1) Sound issues still with pops / crack in some games i.e. Aladdin and Alien Storm cracktro + trainer.

2) On far right of video there's a blinking vertical line; can see this is Quake II (Not sure if you still have Quake II from last time)

Cheers Viceroy

[amilo3438]

Agony (1992)(Psygnosis)[cr CSL](Disk 1 of 3).adf

Upper pixel row in gameplay window (below scores) is behaving strange! (moving left then OK again)

A500 quickstart (most common) WinUAE b29 (fine in 4400)(didnt test previous betas)

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

Goldrunner (1987)(Microdeal)[cr HQC][t +1 HQC].adf

Gaming window is blinking during the gameplay!
It found it does not happen on this video: [ Show youtube player ] (original amiga?)

A500 quickstart 1.2KS 512Kb WinUAE b29 (seems it happens in 4400 but less frequently)

[hexaae]

Originally Posted by Zarnal

You wrote (# 642) that you would watch for this demo at the end of the new blitter emulation.



UpToFront-Coolfridge / Smooth Copper.

Thank you.

Nasty one! Fails with all versions I tested of WinUAE back to 3.3.0...

[hexaae]

Originally Posted by Toni Wilen

https://download.abime.net/winuae/fi...uae_4900b29.7z
https://download.abime.net/winuae/fi...e64_4900b29.7z

Chipset updates are almost done. Really. At least I hope. If there was some other updates I delayed until chipset update is done, now it is time to remind me

(FloppyDriveBridge thing does not need reminding. It will probably happen.)

Beta 29:

- b28 copper update was wrong. Second attempt. (Hotbleeps etc. This time without breaking others.)
- Sprite fix, sprite DMA does not start if frame's initial SPRxPOS/SPRxCTL fill loads vertical start position that matches current vertical position.
- Added bsdsocket IoctlSocket SIOCIFADDR, SIOCIFFLAGS, SIOCIFBRDADDR, SIOCIFCONF, SIOCIFNETMASK, SIOCIFMETRIC and SIOCIFMTU support. Read-only.
- Fixed uaegfx indirect mode BlitTemplate() harmless one byte buffer out of bounds read.
- VPOSW register access always clears LOL bit (if it was set = NTSC long line). It can't be set to one by writing to VPOSW.
- NTSC long/short line order was wrong after horizontal wrap around.

Just FYI, this one solved the strange bug with my Leander data files
http://eab.abime.net/showpost.php?p=...&postcount=788

[Toni Wilen]

Originally Posted by amilo3438

Agony (1992)(Psygnosis)[cr CSL](Disk 1 of 3).adf

Upper pixel row in gameplay window (below scores) is behaving strange! (moving left then OK again)

Fixed.

Goldrunner (1987)(Microdeal)[cr HQC][t +1 HQC].adf

Gaming window is blinking during the gameplay!
It found it does not happen on this video: [ Show youtube player ] (original amiga?)

Actually blinking or just horrible interference blinking-like effect caused by background pixel pattern? Most likely it is invisible with blurrier display.

Originally Posted by Viceroy

Just 2 outstanding bugs for me Toni ;-)

1) Sound issues still with pops / crack in some games i.e. Aladdin and Alien Storm cracktro + trainer.

PC hardware? Recently one similar issue was solved by creating new power plan, supposedly "optimal" wasn't that optimal after all.

EDIT: the biggest problem is the same as always, normally UAE CPU usage is quite bursty (do something, then wait few milliseconds, do it again) and some power saving features seem to assume this means: it is safe to enter lower power mode. Results are bad.

2) On far right of video there's a blinking vertical line; can see this is Quake II (Not sure if you still have Quake II from last time)

No idea but screenshot would be nice. Which screen mode?

[amilo3438]

Originally Posted by Toni Wilen

Actually blinking or just horrible interference blinking-like effect caused by background pixel pattern? Most likely it is invisible with blurrier display.

Have you tested it or just guessing?
There are moments when it is not blinking and background is same pixel pattern. (e.g. change the ship to going down)

[Toni Wilen]

Originally Posted by amilo3438

Have you tested it or just guessing?
There are moments when is not blinking and background is same pixel pattern.

Of course I tested it. It looked horrible

If display refresh rate does not match 50Hz, it probably looks even more worse and depends on which frames are missed and which are not. And speed of scrolling.

[Viceroy]

Sorry Toni I forgot to upload SS for Quake II :-(
Screen mode; it's RTG 400 x 300

[amilo3438]

Originally Posted by Toni Wilen

Of course I tested it. It looked horrible

If display refresh rate does not match 50Hz, it probably looks even more worse and depends on which frames are missed and which are not. And speed of scrolling.

Ok, there is solution! Just change the "Line mode" to "Double, fields" in Display option. (looks much nicer)

Originally Posted by Toni Wilen

This (as everyone probably already knows) has broken first disk (and no one has found working one) but it should still work in 1M Chip config. Except intro animation. I'll check later.

Critical Mass by Parallax -> got HALT2 in todays winuae.7z test version on some point! (I know about broken animation, but it should work on ECS 1MB Chip config. as it works on 4400)

Note: If I remember from "Problematic demos" thread, you said something about how you make it to work!? (need to check the topic)

EDIT: Seems it works in todays test version! I was not present 1st time when it HALT-ed but checking again it works till some point near the end when it crash! (it also crashed on 4400 so the disk must be damaged)
It does not work (tested again). Or I must be tired and start talking nonsense or something strange is going on!
EDIT2: It is working till HALT2 (tested 2x) after deleting my previous *.ini file! Must be that something disturbs config files randomly or by my mistake.
EDIT3: Case closed - http://eab.abime.net/showthread.php?t=73639 (confirmed)

[Viceroy]

The Sound info you req Toni.
My laptop specs are these: https://www.swsd.sk/img.asp?attid=288137

I've never had sound issues; and I've been using all versions of WinUAE on this laptop that I purchased in December 2011.

It started in these beta, probably around B5 or something.

My power plan has always been the same and never adjusted. It's "Balanced" but I can try changing to "High Performance".

[Viceroy]

Changing to "High Performance" fixes "Alien Storm (1991)(U.S. Gold)[cr VF][t +6 LGD].adf" cracktro + trainer pops / crackles.

Unfortunately same deal with Aladin.

[Toni Wilen]

Originally Posted by Viceroy

Changing to "High Performance" fixes "Alien Storm (1991)(U.S. Gold)[cr VF][t +6 LGD].adf" cracktro + trainer pops / crackles.

Unfortunately same deal with Aladin.

Where exactly in Aladdin? First game screen? Even when not moving joystick?

[amilo3438]

I noticed some strange things going on while mixing 32-bit and 64-bit WinUAE versions!

I will try to explain by testing Critical Mass by Parallax demo:

Start the 32-bit version of the WinUAE 4900b29,
go to the paths and click Rescan ROMs, (Note: I have enabled only "Scan subfolders" and "Use relative paths".)
go to quickstart and choose A500 1.3 ROM, ECS Agnus, 1MB Chip RAM and check "Set configuration",
go to chipset and change ECS to OCS,
go to configurations, give the name and save the configuration.

Now load the saved configuration and test the mentioned demo. (it will work till the end and than reset)

Now exit the WinUAE 32-bit version and start the 64-bit version.
Load the above saved config into 64-bit version and test the same demo!
(it will pass the animations and stop somewhere after the "critical mass" screen)

Now exit the 64-bit version and start the 32-bit version.
Load the above saved config into 32-bit version and test the demo!
(it will pass the animations and stop somewhere after the "critical mass" screen)
But if now press F12, go to paths and Rescan ROMs and into config to load the saved config, start the emu and the demo will again work fine!?

Hmm, is it possible that the 64-bit version somehow messing up something? (or 32-bit and 64-bit cant use same config files)

[Toni Wilen]

Sounds like the usual (with extra-mega-complex way to duplicate it!): first boot with disk inserted has always same timing. Every time.

Next boot after any kind of reset has new "seed" (because disk is "analog" and "random", floppy motor position relative to index position, gap glitches)

Sometimes this causes problems, if some program is "unstable", there is always small chance that UAE default boot always triggers "bad timing" that crashes the program. Even if in real world it only happens every 1/100th time or so..

This is by design. Without it debugging anything would be extremely painful.

btw, if you want to run both 32-bit and 64-bit at the same (or any multiple versions): it might be good idea to force every winuae.exe to have its own winuae.ini instead of sharing same registry. Especially if they are located in different directories.

if you think it is something else, get rid of any extra steps: simply disable quickstart mode, load pre-created config, start emulation.

[amilo3438]

Yes, that's what I thought. For now, I will stick to the 32-bit version. Thanks.

BTW. I wonder if there would be possible to exist two .ini files in the same folder: winuae.ini - for 32-bit version and winuae64.ini - for 64-bit version (in portable mode)