WinUAE 1.5.0 beta series

[Toni Wilen]

Originally Posted by laser

but I think that froggerNG is a bit faster now when play videos on a window..however Im not sure of that

Sorry but generally I don't care about application programs that have better Windows native version. (and this kind of program probably expect timing that won't happen on emulation, you can never be sure if you are benchmarking just video output or something else..)

[Mad-Matt]

Could it possile be more a hardware issue ? pc gfx chipsets probably never saw reason to have more then 2 colours max for a single hardware sprite used for mouse pointers which was likely more important in the old days so never updated for modern machines as a software pointer is more then sufficiant, unless windows is busy so end up with a stuttery pointer

[Toni Wilen]

Originally Posted by Mad-Matt

Could it possile be more a hardware issue ? pc gfx chipsets probably never saw reason to have more then 2 colours max for a single hardware sprite used for mouse pointers which was likely more important in the old days so never updated for modern machines as a software pointer is more then sufficiant, unless windows is busy so end up with a stuttery pointer

Picasso96 hardware sprite has 3 colors (+ transparent). Perhaps hires-mode uses different scheme or uaegfx.card has incorrect settings (which is very possible because I don't have sources or sdk yet..)

[laser]

Originally Posted by Toni Wilen

Sorry but generally I don't care about application programs that have better Windows native version. (and this kind of program probably expect timing that won't happen on emulation, you can never be sure if you are benchmarking just video output or something else..)

OK...but all programs made for a gfx card runs on winuae very well and fast....you can relax.....ppl should be satisfied...there is no need for speed improvements (I think)

BTW,the pointer never was smooth.....I have a 100hz screen and a line on my user-startup......p96refresh 100

but p96refresh not works fine....the pointer is not smooth on workbench like on windows ....

Im doing something wrong? or ....it is an old issue not solved yet?

[AmigaSurfer]

I use MultiCX 2.80 to blank the mouse pointer at keypress and blank it after approx. 30 seconds when no mouse movement has occured.

If I dissable SoftSprite, then the mouse pointer doesn't blank anymore at keypresses or after 30 seconds.

Inside e.g. the texteditor Ed, MUI TextEditor or MUI String Gadgets the pointer, rather the part of the pointer within the screen lines of the cursor, are blanked by pressing any character key. Even if the pointer is outside the Ed or TextEditor window or outside the String Gadget, but within the same screenlines of the cursor.
By hitting Return all parts of the pointer within the screen lines of the whole Ed or TextEditor window are blanked.
Sometimes the cursor blanks too inside this area with a short delay after pressing one charakter key or while pressing some character keys. But I can't find a scheme.
The cursor does never blank by pressing Shift, Ctrl or Alt.

If I have dissable SoftSprite at startup, enable SoftSprite and do a reset with Ctrl-Amiga-Amiga, then the hardware pointer is still at the left top border and can't be moved, in addition to the software pointer, which can be moved. If I switch the resolution of the pointer from HiRes to LoRes and vice versa, the color of the hardware pointer change as mentioned in this thread above. So the hardware pointer keeps active in addition to the software pointer, but don't move. If I quit WinUAE and start it again, then the hardware pointer is gone. But not with a simple reset with Ctrl-Amiga-Amiga.

[PaulyQ]

I'm getting a few strange artefacts on the 2d graphics. Icons now have a strange horizontal line hovering above them, not all the way across, around a 5th of the icon size.

Currently using the p96 ClassicWB setup with new uaegfx.card file installed etc...

Also p96speed crashes halfway through a test with a guru. Will do a bit more testing and get back to you.

[Toni Wilen]

Originally Posted by AmigaSurfer

If I dissable SoftSprite, then the mouse pointer doesn't blank anymore at keypresses or after 30 seconds.

Probably another uaegfx.card missing feature. I am getting SDK soon

I'm getting a few strange artefacts on the 2d graphics. Icons now have a strange horizontal line hovering above them, not all the way across, around a 5th of the icon size.

Windows 2000?

[PaulyQ]

Sorry...

XP SP2 + all updates, nVidia 7300GT card.


UPDATE:

Just had a "Create Device Failed 8876086c s=1 f=0876 c=086c (2156)0 " error in a dialogue box when I tried to start up then realised I'd forgotten to switch off Filtering.

UPDATE 2:

Possibly the fact I was using 16bit graphics. Upped it to 32bit graphics and the line(s) had gone. Went down to 256 colour graphics and the screen became garbled.

[Toni Wilen]

Originally Posted by PaulyQ

Possibly the fact I was using 16bit graphics. Upped it to 32bit graphics and the line(s) had gone. Went down to 256 colour graphics and the screen became garbled.

Screenshots please + test in both windowed and fullscreen modes.

[Toni Wilen]

Originally Posted by Toni Wilen

Picasso96 hardware sprite has 3 colors (+ transparent). Perhaps hires-mode uses different scheme or uaegfx.card has incorrect settings (which is very possible because I don't have sources or sdk yet..)

Ok, I have the card "sdk" now (can't distribute it yet, there are some license things Cloanto needs to solve first) and hires/lores hardware sprite is supposed to work identically.

I also noticed in clean OS3.9 installation hardware sprite works correctly, colors are exactly same as lores. I guess it is bug in some patch..

[Mad-Matt]

Edit: Forget previously written.

p96 seems to randomly work or not work. some issue with initialising memory at boot.

seems to be linked to amount of rtg mem set. 8mb is working, but higher(16+) isnt.

[PaulyQ]

Tested in both windowed and full screen.

Results as below in the screens. Please note, exactly the same happens in full screen mode as compared to windowed mode in both 256 colour and 16 bit (i.e. artifcacts above icons), no filters used and no vsync either.

[Toni Wilen]

Originally Posted by PaulyQ

Tested in both windowed and full screen.

Results as below in the screens. Please note, exactly the same happens in full screen mode as compared to windowed mode in both 256 colour and 16 bit (i.e. artifcacts above icons), no filters used and no vsync either.

Does the garbage disappear if you move software cursor over it?

Does 8-bit mode wrong colors only happen in fullscreen mode?

[PaulyQ]

Does the garbage disappear if you move software cursor over it? No it doesn't Toni

Does 8-bit mode wrong colors only happen in fullscreen mode? BOTH, but in full screen mode the colours change slightly, exactly same effect though.

Anything else I can help with?

[Toni Wilen]

Originally Posted by PaulyQ

Does the garbage disappear if you move software cursor over it? No it doesn't Toni

Does the garbage follow the icon if you drag it around?

[PaulyQ]

Does the garbage follow the icon if you drag it around? Nope

[Toni Wilen]

http://www.winuae.net/files/b/winuae_1500b7.zip

Beta 7:

NOTE: _DO_ use filters (except OGL/D3D). Vsync still broken.

ADDED: GUI disappears if adjusting filters in fullscreen mode.

- Gayle and Gary CIA select lines work slightly differently, Gayle does not select any CIAs if both CIA select bits are zero (address bits 12 and 13). Gary enables both. Gary-style selection was emulated long time ago, Gayle-style emulated now. (I think 'select both CIAs' was only used by some games accidentally and perhaps also used by some copy protections)
- car color flickering in hires mode super skid marks (ancient bug)
- real PCMCIA SRAM card detection really works now
- uaegfx.card updated, now works even if "UAE boot rom" is not in "standard" 0xf00000 location (=CDTV ROM enabled) or disabled. Previos version simply crashed.
- uae boot rom backup location moved from 0xe70000 to 0xef0000 (0xe70000 can conflict with KS mirror)
- jit memory allocation update, should prevent random RTG memory allocation failures
- Picasso96 emulation updates and tweaks
- CPU panel "After RTG" radiobutton removed, not needed anymore
- big filter rewrite (NOTE: OpenGL/Direct3D not yet updated) Settings needs to be readjusted!
+ works more intuitively now
+ scaling does not affect centering anymore.
+ position setting (0,0) = centered.
+ 1/2x scaling factor added
+ "FS" scaling factor added = full screen scaling (obsoletes short lived "full screen filter" setting)
+ user scaling factor (FS,1/2x,1x,2x..) never include filter's possible internal scaling
+ hq3x and hq4x added
+ onscreen leds are now rendered after filter calculations
+ (some kind of "keep aspect ratio" option needed?)

[Steve]

Hi Toni. Wonderful work! Super Skidmarks hi-res sprites work perfectly now. Thanks a lot!

Only problem now is that the speed of games has been reduced dramatically on my computer. Super Skidmarks was playable on 1.4.6 but now is so slow its practically unplayable. Its maybe running at 50% of the speed of what is was previously.

I tried Sensible World of Soccer and found the exact same slowdown problem on a game which previously ran at 100% perfect speed. There is also a very bizarre problem with the player sprites on SWOS. The players flicker and change colour during play. How very strange.

[Toni Wilen]

Originally Posted by Steve

Hi Toni. Wonderful work! Super Skidmarks hi-res sprites work perfectly now. Thanks a lot!

Only problem now is that the speed of games has been reduced dramatically on my computer. Super Skidmarks was playable on 1.4.6 but now is so slow its practically unplayable. Its maybe running at 50% of the speed of what is was previously.

You know that CPU% meter is included for a reason
And even more information please, like is only b7 slow or previous betas too? Windowed and/or fullscreen? etc.. Your report is unusable

[Ed Cruse]

I downloaded B7 and updated uaegfx.card. I’m using 32 bit color on the Windows side and RTG 1024x768x8 on the Amiga side. When I open a 1024x768 screen with less then 8 bit planes and then open a 640x768 CLI window centered horizontally, the CLI window’s cursor shows up on the screen instead of the window. The cursor is located all they way to the left side of the screen at the vertical level where the cursor should be on the CLI window. Once I start typing the cursor now appears on the CLI window in the correct place with the original cursor still showing on the screen at the original location. If I bring up a program like my text editor, which opens a window on the screen, and then close the editor the same thing happens again. If I type a Dos command like "Dir" the cursor continues to be on the CLI window in the correct place. If I use 8 bit planes there’s no problem. The screen uses standard colors, and is a public screen. The problem is new with B7.