mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support
-
@akafox said in mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support:
thank you both.
Yeah it may just have something to with kodi itself or the porting the core from a real libreto core to a kodi add-on. Not a deal breaker just curious. I'll keep you posted from the other forum if you want.
I can launch kodi from ES sure but never could never figure out how to get into kodi and then exit it cleanly back into ES without a reset of the system. It's nice to have a "all-in-one" entertainment box but then again two PIs are better than one I guess? :P
Okay I am not going to hijack this thread so back on topic...
is there a one stop place I can look to find a list of improvements or/and "m3plus exclusives" that is available? And thank you for your HARD work (yeah we used to be called coding wizards...but honestly if we were wizards you think we would code instead of making gold out of air? xD )
https://github.com/libretro/mame2003-plus-libretro/blob/master/CHANGELOG.md
-
@arcadez2003 said in mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support:
@WeirdH said in mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support:
@grant2258 Is it possible that palette changes also affected colors in other games?
Because I might be imagining things, but since the update some effects in Storm Blade have been multi colored where they didn't use to be (shockwave before enemies explode and apparently some shadows). At least to the best of my recollection. Again, I might be remembering wrong, was just wondering about it, because I noticed it yesterday.
A year or so back we ported some video changes from later MAME for Storm Blade and Ultra X Weapons to fix the colours and graphical problems............
Are you saying I should not be seeing colors glitching then?
-
@WeirdH said in mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support:
@arcadez2003 said in mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support:
@WeirdH said in mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support:
@grant2258 Is it possible that palette changes also affected colors in other games?
Because I might be imagining things, but since the update some effects in Storm Blade have been multi colored where they didn't use to be (shockwave before enemies explode and apparently some shadows). At least to the best of my recollection. Again, I might be remembering wrong, was just wondering about it, because I noticed it yesterday.
A year or so back we ported some video changes from later MAME for Storm Blade and Ultra X Weapons to fix the colours and graphical problems............
Are you saying I should not be seeing colors glitching then?
Not that im aware of, show me a pic
-
@arcadez2003 Okay, some horrible phone pics, but the problem is visible. It appears when a shockwave of an exploding enemy overlaps with a shadow in the background. (these same shadows appear red when using mame2003, by the way)
-
Ok thanks for taking the time to upload some pics, i can confirm im not getting this in my test build, but as i always say it's different MAME72 standalone emulator
on a different platform although the driver for this game and the video code are identical in both builds, is anyone else getting these graphical problems
when playing Storm Blade in this core.?? -
I am seeing the same issue running RetroPie 4.5.1 and m3plus commit 0278635 on a Pi 3. Screenshot below:
-
Im starting to think it's this way in current MAME also, going by this pic from the arcade database and this YT vid
I guess that's how i didn't spot it as it's a case of blink and you'll miss it, two things it could be the graphical emulation is not yet 100% or it actually did this on the arcade pcb.
-
@arcadez2003 - I'm not completely sure if it's always been there. I have an image at least 6 months and another at least a year or older if it helps to track it down I'll load them up.
Current MAME 0.214 does look correct from what I can tell, just tested it now. With Retroarch you can frame advance so it's easy to catch but with current MAME you get about a half second or less to capture it during play so, not even double tapping the pause key helps. Maybe @WeirdH can add to it as it's a good catch, good eye! :) I would have just kept playing blissfully unaware.
-
Aye i see what ya mean i didn't spot that video was from 7 years ago :) your right it seems to be sorted in the latest builds, this might be the issue
im sure we had some troubles with the screen flipping a while back............https://mametesters.org/view.php?id=4669
If not then it'll likely be one of the many graphical updates to the SSV driver over the years, i dont really have the time to look at this in more detail just now
but i might take another peek at the video code down the road. -
@arcadez2003 @Riverstorm @WeirdH FBNeo also got some issues with some SSV games on arm/aarch64 (still trying to figure out the issue), but FWIW this one seems fine.
-
@Riverstorm Yeah, it's a curse. Friends joke that I could spot a dead pixel on a 4K screen from across the room. Sometimes blissfully unaware is the preferred way...
I only have a MAME 0.139 to test with on my htpc, don't know if that'll add anything new, now that more knowledgable people have already weighed in?
-
@WeirdH - Yeah it would have to be one of the programmers like @arcadez2003 @grant2258 @barbudreadmon to fix it as they are the true "Githubians of Libretronan" or "Libretronians of Githubio" "The Wizards of Retroly Street" ... :)
MAME 0.214 does still flag Storm Blade (SSV.cpp) with graphical issues on launch. Also ArcadeItalia lists cocktail issues. Still some good games in their though like Dyna Gear, Drift Out '94, Eagle Shot, Twin Eagle II, Vasara, Storm Blade, etc.
@barbudreadmon - That's great, thanks for the headsup! I'll be loading it up later this afternoon for testing. My build is mainly a hybrid of m3plus and fbneo with a bit of advmame for a handful of vector games.
I don't know if this was the fix back in 2005 but it does relate to shadows.
15th May 2005: Luca Elia - I finally got around to implementing proper shadows in SSV driver, replacing the hack that was there instead. The difference is most noticeable in Storm Blade. The way shadows work is this: The frame buffer is filled with 16 bit pens, that are indexes into the palette (i.e. the actual rgb colors), from the previously drawn, "normal" tiles. When a "shadowing" tile is drawn, it does not overwrite the contents of the frame buffer, but instead replaces the top bits of the pens there with the low bits of its pens. This results in a shift of the affected pixels from one section to another of the game's palette. The game has to just divide the palette into increasingly dark chunks of the same base colors. Some games do highlights this way, while Storm Blade again exploits the mechanism to cross-fade between graphics, which now works.
-
@Riverstorm said in mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support:
I don't know if this was the fix back in 2005 but it does relate to shadows.
15th May 2005: Luca Elia - I finally got around to implementing proper shadows in SSV driver, replacing the hack that was there instead. The difference is most noticeable in Storm Blade. The way shadows work is this: The frame buffer is filled with 16 bit pens, that are indexes into the palette (i.e. the actual rgb colors), from the previously drawn, "normal" tiles. When a "shadowing" tile is drawn, it does not overwrite the contents of the frame buffer, but instead replaces the top bits of the pens there with the low bits of its pens. This results in a shift of the affected pixels from one section to another of the game's palette. The game has to just divide the palette into increasingly dark chunks of the same base colors. Some games do highlights this way, while Storm Blade again exploits the mechanism to cross-fade between graphics, which now works.
I've already backported the following improvements........
0.141u1: Improvements to the SSV sprite position and flipscreen support [Robert Zandona].
0.139u3: SSV update [Robert Zandona]: Implemented visible area registers in the SSV driver (now the resolution for all the games is the max visible area, the values of the registers after the boot process). Implemented y tilemap offset register. Implemented x and y visible area register. Disabled tilemap draw when mode=0. Removed kludge for sprite x offset and implemented 1 bit of 7a-7b register. Implemented y sprite inversion CRT register. Added a kludge for Dyna Gear. Used CRT registers to get the sprite x offset. Removed offset values from the init; the offset now are handled by the CRT registers. Fixed bit #a of 7a-7b CRT register. Fixed background disable condition. Corrado Tomaselli confirmed visible area register with a Twin Eagle II PCB.
0.139u3: Roberto Zandona fixed graphical corruption on left side of screen beginining stage 3 in Storm Blade.
0.96u3: Luca Elia added shadow effects to the SSV driver. Changed palettesize to 32768.
However from checking the commits i've spotted a small fix for the shadows which never got a mention in the MAME WIP, i've added it to my test build
and it seems to sort the problems although my eyesight ain't what it was, i'll add in these changes to this core update to latest build then
try Storm Blade once again and lemmie know how it pans out.https://github.com/libretro/mame2003-plus-libretro/commit/ed9a8684e987e39e1981a1ec0a9412663f3f47ff
-
@barbudreadmon said in mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support:
@arcadez2003 @Riverstorm @WeirdH FBNeo also got some issues with some SSV games on arm/aarch64 (still trying to figure out the issue), but FWIW this one seems fine.
I hadn't played Storm Blade for quite some time but i do remember now sometimes on shooting down certain planes or finishing off certain bosses
you would as part of the explosions get a yellow or blue flash which i always thought didn't look quite right,But i was never 100% on this until the recent posts with the pics and reviewing the gameplay on YT, certainly that commit i just pushed to the git seems
to sort it as the flashes before the explosions or afterwards are all white now which looks correct i think, of course maybe on arm/aarch64
hardware the graphical problems are worse and show up more.?? -
@arcadez2003 - Thanks a bunch for the fix. It's looking darn right proper here unless my old peepers deceive me! :)
I mainly use the Pi 3 which I think is a quad core Cortex-A53 (ARMv8). Screenshot below running commit ed9a868. Shadow/shockwave overlap looks good.
-
@arcadez2003 Appears to be fixed on my 3B+ as well, awesome work!
-
Well it was Luca Elia who fixed this issue, i just backported the code into this core, kinda like copying the guy next to ya at school :) but still it's another happy ending at least.
-
@arcadez2003 You made me remember this quote from http://blog.wolfire.com:
We have opensource facial hair: if one guy has a big beard we all copy his code and get our own beards.
-
@arcadez2003 said in mame2003-plus: hundreds of new games, improved input, features, new bugs - now with runahead support:
ng the guy next to ya at school :) but still it's another happy ending at least.
opensource is a great thing when its truly libre :) including copying beards lol
-
Thanks a lot all for your help last week :)
Now i have my full set MAME 2003 Plus <3
I began to try some games and your work is so fantastic :)But i have one question for CHD games ...
Mine are CHD2 not v5, but i think it help because less compressed.
I try to use them on USB drive or directly on the MicroSD car.
I try Area51 and KI1 KI2 and all are slow, impossible to play them fine :(
On PI 3b+ ... is it normal ? Is there a trick to help ?
Because i remember on XBOX @arcadez2003 we can play KI1 and 2 without CHD and not with MAME, but at full speed on a low 700Mhz proc with 64Mo of RAM ... :(
Why not on PI who is a 4 core, 1400Mhz with 1Go of RAM !!?
Thanks
Contributions to the project are always appreciated, so if you would like to support us with a donation you can do so here.
Hosting provided by Mythic-Beasts. See the Hosting Information page for more information.