WinUAE 4.9.0 beta series (Was 4.5.0)

[falken]

Originally Posted by Toni Wilen

Right side had one extra pixel, fixed in next beta. Bottom should be correct, it is usually line that is black but can be have other colors too but not graphics. This was not fully emulated previously.

btw, don't scale screenshots..

Thanks for your clarifications Toni!

[Zarnal]

Hello,

there is again a problem with BCKID (with Quickstart A1200 this time). The screen seems to switch permanently (A500 and A600 are ok).

[Toni Wilen]

https://download.abime.net/winuae/fi...uae_4900b26.7z
https://download.abime.net/winuae/fi...e64_4900b26.7z

Beta 26:

- Write to most programmed mode registers (ECS/AGA only) caused display emulation reinitialization, even when modified register was not in use. Now reinitialization is only done if matching BEAMCON0 bit(s) is set. Fixes BC Kid screen flashing, game updates color registers but writes too much and modifies first few programmed mode registers.
- Remove interlace artifacts option works again but is not fully functional. (last line might flicker, copper modifications are not always accurate)
- On the fly switch to subpixel mode caused hang in some situations.
- Default overscan mode had one extra pixel row and line. Overscan+ size also fixed.
- Fixed display glitches in 68020+ memory cycle exact modes (memory cycle exact only).
- Shader files are again supported in Direct3D 11 mode. FX11 moved to separate static library.

[mark_k]

Originally Posted by Toni Wilen

Beta 25:
- On the fly PAL/NTSC switch (VPOSW trick or BEAMCON0) incorrectly adjusted screen width in some situations. Height is now adjusted relative to total vertical lines, some games use VPOSW trick to generate 56Hz mode which was previously scaled as NTSC.

Is the height adjustment supposed to continuously adjust the vertical stretch factor, so that however many scanlines there are, the whole image aspect ratio is (close to) 4:3?

[Toni Wilen]

Originally Posted by mark_k

Is the height adjustment supposed to continuously adjust the vertical stretch factor, so that however many scanlines there are, the whole image aspect ratio is (close to) 4:3?

It is supposed to always fill the whole display vertically, like real CRT monitor would do (at least in supported frequency range).

[hexaae]

Sent false positive report to Microsoft (Defender) again... sigh...
Should be safe in a couple of days.

[mark_k]

Most CRTs (incl. my old 1084) have an internal height switch. Here height really means distance between successive scanlines.

If you gradually decrease the number of lines per frame (so increase refresh rate), the height of e.g. a 640×200 CLI window, stays the same until refresh rate reaches the threshold (about 55Hz I guess). Then it switches to "NTSC height".

I remember using some tool to adjust the number of lines from Workbench back in the day. Could it have been 60HzEmulator? Or maybe Amicom's PowerUtility? Or maybe a program which came with the MacroSystem deinterlace card???

Some game using a 56Hz display could be explained by that behaviour. Developer adjusted number of scanlines until their monitor switched to "NTSC height" but didn't go any higher in refresh rate so as not to affect gameplay or music speed too much.

[hexaae]

Excellent beta this b26 (x64). New Picasso96 RTG screen grabbing (e.g. 24bit WB and Chaospro own screen) and sliding works very well now without glitches (there were some issues with pointer corruption and freezing in all previous WinUAE versions...).

The only problem I've found testing 20-30 games (WHD and non) was with Leander (WHD latest slave) sword sprite(s) having a vertical "hole":

[Toni Wilen]

Originally Posted by hexaae

Excellent beta this b26 (x64). New Picasso96 RTG screen grabbing (e.g. 24bit WB and Chaospro own screen) and sliding works very well now without glitches (there were some issues with pointer corruption and freezing in all previous WinUAE versions...).

Did you test 32-bit version? Sounds more like 64-bit accidentally does not trigger the bug.

The only problem I've found testing 20-30 games (WHD and non) was with Leander (WHD latest slave) sword sprite(s) having a vertical "hole":

I can't duplicate this.

But one sprite (sprite 2) is missing from your screenshot (see top left "bombs"/"coins" counters), it is also part of sword animation.

[Toni Wilen]

Originally Posted by mark_k

Most CRTs (incl. my old 1084) have an internal height switch. Here height really means distance between successive scanlines.

If you gradually decrease the number of lines per frame (so increase refresh rate), the height of e.g. a 640×200 CLI window, stays the same until refresh rate reaches the threshold (about 55Hz I guess). Then it switches to "NTSC height".

Ok, I probably confused it with some later "more intelligent" displays that behave differently.

I remember using some tool to adjust the number of lines from Workbench back in the day. Could it have been 60HzEmulator? Or maybe Amicom's PowerUtility? Or maybe a program which came with the MacroSystem deinterlace card???

AFAIK 60HzEmulator uses same hack as Dynablaster on OCS to "switch" to 60Hz. (Too lazy to test)

Some game using a 56Hz display could be explained by that behaviour. Developer adjusted number of scanlines until their monitor switched to "NTSC height" but didn't go any higher in refresh rate so as not to affect gameplay or music speed too much.

Yeah, this could explain weird 56Hz.

[hexaae]

Originally Posted by Toni Wilen

Did you test 32-bit version? Sounds more like 64-bit accidentally does not trigger the bug.

Tested also 32bit version, both work really fine with screen dragging compared to previous WinUAE 4.4.0+ versions where I had those random issues with pointer when dragging in P96RTG mode.

(Sent PM about Leander)

[Zarnal]

Hello,

while re-testing OCS demos I found a glitch ( pic 1 ) that always appears after any choice in the "Nevermind 14" compilation menu. Since b25. A500 Quickstart.

[amilo3438]

Blue Venture (Demo) by Anxious

A500 ECS 1mb chip ... ANXIOUS text glitches !


Crash Landing (Crack Intro) by Scoopex

A500 OCS / ECS ... some glitches !


Arc Ed v1.02 Intro (Intro) by United Forces

A500 OCS / ECS ... some glitches after exit ! (start the demo in CLI with "ufo-intro" and after a while press the LMB !


Also some glitches found in:

Hit the Road (Megademo) by Flash Production -> http://janeway.exotica.org.uk/release.php?id=2754
Legalize It 2 (Musicdisk) by Anarchy -> http://janeway.exotica.org.uk/release.php?id=9007

and probably some others caused by the same problem! ... so there is no point to continue !

[Toni Wilen]

I can't duplicate any of above. Which probably means some host config option affects this.

[Zarnal]

Toni, while checking the "Crash Landing by Scoopex" demo that Amilo3438 reported to you, I found the same problem of "residual" glitch after a menu of a packdisk (pic 1 and 2).
Note: as for my report from "Nevermind14", if you move the mouse pointer the glitch disappears + -.

[amilo3438]

Originally Posted by Toni Wilen

I can't duplicate any of above. Which probably means some host config option affects this.

Hmm, maybe because I tested on my Win7 machine ! (64-bit WinUAE)

So I will have to check it on my Win10 machine to see is it still the same!

[Toni Wilen]

I don't think it can be host specific but for example some Display panel option or similar.

Does it happen without default.uae in configurations?

EDIT: I can duplicate similar looking corruption if I load statefile and then another statefile. But it only happens with some statefiles.

[amilo3438]

Originally Posted by Toni Wilen

I don't think it can be host specific but for example some Display panel option or similar.

Does it happen without default.uae in configurations?

Yes, I checked also 32-bit WinUAE and combinations D3D9 and D3D11 and no change in Win7 ! (not sure will today manage to try it on Win10 machine, maybe tomorrow)


EDIT: No problem on Win7 with beta 24! (it starts from beta 25)

[Zarnal]

Originally Posted by Toni Wilen

I don't think it can be host specific but for example some Display panel option or similar.

Does it happen without default.uae in configurations?

Yes ( at least for my reports ).


Note : i'm on Windows 8.1 and Amilo on Windows 7. I'm going to try it on Windows 10.

[Zarnal]

I just tested in Windows 10 and it doesn't change anything.

There, i really do not understand what is going on.

Beta 24 is ok but not b25 and b26 in same conditions.