04623: drivers that use SCREEN_VBLANK: change from SCREEN_EOF to SCREEN_VBLANK regressed timing

ID	Category [?]	Severity [?]	Reproducibility	Date Submitted	Last Update
04623	Timing	Minor	Always	Jan 20, 2012, 21:37	Jan 25, 2012, 16:13

Tester	hap	View Status	Public	Platform
Assigned To	hap	Resolution	Fixed	OS
Status [?]	Resolved			Driver
Version	0.144u6	Fixed in Version	0.144u7	Build
		Fixed in Git Commit		Github Pull Request #

Summary	04623: drivers that use SCREEN_VBLANK: change from SCREEN_EOF to SCREEN_VBLANK regressed timing
Description	change from SCREEN_EOF to SCREEN_VBLANK in 0.144u6 regressed timing in drivers due to the outcome of 2 separate timings (vblank start or vblank end) instead of one. I don't know how many drivers are affected, known ones will be listed.
Steps To Reproduce	In games that buffer spriteram, it usually causes spritelayer/bg layer to go out of sync.
Additional Information	List of users of SCREEN_VBLANK in src/mame/video, add note if known to have a problem or not, and if it's fixed. "not sure" means it needs to be rechecked: it doesn't have any visible change or might even get worse if trying to fix. - File airbustr.c: misc kan_pand sprite problems - FIXED - File armedf.c: sprite/bg sync problem - FIXED - File baraduke.c: sprite/bg sync problem - FIXED - File bionicc.c: sprite/bg sync problem - FIXED - File blktiger.c: sprite/bg sync problem - FIXED Y File bosco.c: starfield update timing - FIXED - File commando.c: sprite/bg sync problem - FIXED - File cps1.c: sprite/bg sync problem - FIXED - File crshrace.c: sprite/bg sync problem - FIXED - File deco_mlc.c: sprite offset glitches - FIXED - File djboy.c: misc kan_pand sprite problems - FIXED - File dooyong.c: sprite/bg sync problem - FIXED - File dynduke.c: sprite/bg sync problem - FIXED - File exedexes.c: sprite/bg sync problem - FIXED Y File galaga.c: starfield update timing - FIXED Y File gaplus.c: starfield update timing - FIXED - File gng.c: sprite/bg sync problem - FIXED Y File ladybug.c: starfield update timing - FIXED - File lastduel.c: sprite/bg sync problem - FIXED - File macrossp.c: sprite/bg sync problem - FIXED - File mcatadv.c: sprite/bg sync problem - FIXED Y File model1.c: glitches when frameskipping - FIXED Y File mw8080bw.c: starfield update timing - FIXED - File namcos86.c: sprite/bg sync problem - FIXED Y File n8080.c: line update timing - FIXED - File namcos1.c: sprite/bg sync problem - FIXED - File othldrby.c: sprite/bg sync problem - FIXED - File pgm.c: sprite/bg sync problem - FIXED - File psikyo.c: sprite/bg sync problem - FIXED - File psikyosh.c: sprite/bg sync problem - FIXED Y File redclash.c: starfield update timing - FIXED - File seta.c: misc seta001 sprite problems - FIXED (emulation still imperfect) - File sidearms.c: sprite/bg sync problem - FIXED - File srumbler.c: sprite/bg sync problem - FIXED - File taito_b.c: sprite/bg sync problem - FIXED - File taito_f2.c: sprite/bg sync problem - FIXED - File taito_f3.c: sprite/bg sync problem - FIXED (though broken gfx in u6) - File taito_l.c: sprite/bg sync problem - FIXED - File taotaido.c: sprite/bg sync problem - FIXED - File tceptor.c: sprite/bg sync problem - FIXED - File tigeroad.c: sprite/bg sync problem - FIXED Y File tmnt.c: missing sprites in detatwin - FIXED - File tnzs.c: misc seta001 sprite problems - FIXED (emulation still imperfect) - File toaplan1.c: sprite/bg sync problem - FIXED - File toki.c: sprite/bg sync problem - FIXED - File twin16.c: sprite/bg sync problem - FIXED - File twincobr.c: sprite/bg sync problem - FIXED - File tx1.c: misc gameplay bugs - FIXED File atarig1.c: good - OK File atarig42.c: good - OK File atarigt.c: good - OK File atarigx2.c: good - OK File bking.c: good - OK File carpolo.c: good - OK File deco32.c: (empty function) File lockon.c: good - OK File mermaid.c: good - OK File sprint2.c: good - OK File sprint4.c: good - OK File sprint8.c: good - OK File sspeedr.c: good - OK File starshp1.c: good - OK File suprnova.c: (empty function) File tank8.c: good - OK File ultratnk.c: good - OK File wolfpack.c: good - OK File xmen.c: good - OK File fuukifg3.c: not sure File gaelco2.c: not sure File goal92.c: not sure File hng64.c: missing 3d gfx File homedata.c: File lemmings.c: not sure File megasys1.c: File ninjakd2.c: not sure File nmk16.c: not sure File seta2.c: not sure File shadfrce.c: not sure File slapshot.c: not sure File ssrj.c: not sure File toaplan2.c: not sure List of users of SCREEN_VBLANK in src/mame/drivers: H File asuka.c: sprite/bg sync problem - FIXED - File bbusters.c: sprite/bg sync problem - FIXED - File champbwl.c: misc seta001 sprite problems - FIXED (emulation still imperfect) - File crystal.c: game locks up - FIXED - File dec8.c: sprite/bg sync problem - FIXED - File dreamwld.c: sprite/bg sync problem - FIXED - File galpanic.c: misc kan_pand sprite problems - FIXED - File hvyunit.c: misc kan_pand sprite problems - FIXED - File raiden.c: sprite/bg sync problem - FIXED - File sandscrp.c: misc kan_pand sprite problems - FIXED - File seta.c: misc seta001 sprite problems - FIXED (emulation still imperfect) - File slapfght.c: sprite/bg sync problem - FIXED - File snowbros.c: misc kan_pand sprite problems - FIXED - File wwfwfest.c: sprite/bg sync problem - FIXED File champbas.c: good - OK File ddz.c: (empty function) File mirage.c: good - OK File pgm2.c: (empty function) File photon2.c: good - OK File psattack.c: (empty function) File igs011.c: not sure File maygayv1.c: not sure (preliminary driver) File megaplay.c: File megatech.c: File sigmab98.c: not sure File wheelfir.c: missing fg layer
Github Commit

Flags
Regression Version	0.144u6
Affected Sets / Systems	drivers that use SCREEN_VBLANK

Attached Files

No.08068 hap Developer Jan 23, 2012, 09:41 edited on: Jan 24, 2012, 02:28	I'm done (and fed up:P) with this bug. Of the remaining drivers, hng64.c and wheelfir.c are broken the most. edit I think they're major enough to get their own separate reports
No.08084 hap Developer Jan 25, 2012, 07:42	oh man :P gotta go over them again after Aaron fixed a core issue
No.08085 hap Developer Jan 25, 2012, 16:13	ok, most of them could be reverted, and the remaining ones should be fixed with aaron's vblank timing core fix