Viewing Issue Advanced Details
ID Category [?] Severity [?] Reproducibility Date Submitted Last Update
00378 Graphics Minor Always Jan 29, 2008, 06:39 6 days ago
Tester rcoltrane View Status Public Platform
Assigned To Resolution Open OS
Status [?] Confirmed Driver
Version 0.99 Fixed in Version Build Normal
Fixed in Git Commit Github Pull Request #
Summary 00378: all pitfight sets: During the TY intro sequence, it shows a fast 'flash' gfx glitch that seems to be a bad rom dump or something.
Description A graphic glitch occurs during the TY intro sequence. It shows a fast 'flash' gfx glitch that seems to be a bad rom dump.

UPDATE (12/18/2011): According to the video posted by Smitdogg from Pit-Fighter Rev. 5 original PCB running attract mode/gameplay, the MAME romset has a graphic issue from a bad dump. Here's the link to the original PCB video showing no issues during attract mode:
Steps To Reproduce Just run the game and let it go through the attract mode for a couple of loops. You will be able to see the graphic glitch at ty's training sequence.
Additional Information Verified against a video posted by Smitdogg running an original rev. 5 Pit-Fighter PCB that shows no graphic glitches during attract mode. Video posted at bug ID 04082.
Github Commit
Flags Verified with Original
Regression Version
Affected Sets / Systems all pitfight sets
Attached Files
There are no relationship linked to this issue.
User avatar
Mar 18, 2008, 01:16
I see this consistently on all pitfight sets. It looks like a legitimate game bug to me, but it needs verification.
User avatar
Senior Tester
Jul 3, 2010, 02:54
edited on: Nov 16, 2016, 18:15
My setup is still a WIP and I can only get grey scale for some games but I made a video of Pit Fighter. This one is even extra funky, colors come in and out in one part... It probably doesn't satisfy the question. I can make another video in a few weeks when I can get color on this game, but FWIW:
User avatar
Senior Tester
Oct 16, 2010, 20:05
I got this PCB working correctly with color. It looks like it's just rough transitions like lots of arcade games had. I'll upload a color reference video soon.
User avatar
Dec 26, 2011, 19:30
I'm curious why certain things were changed in this bug report rcoltrane. The version number is the version this bug was found in, not the latest version the bug exists in.

Also if it happens every time the attract gets to Ty's it shows the graphic glitch, that reproducability is set to 'always' not random. Lastly, no new information was atually added at all, except to add Smitdogg's info from a year ago to the description.

In short, this seems like a bug report that was just randomly altered to push it up onto the 'last modified' listing.
User avatar
Dec 27, 2011, 06:40
I reverted the version number and some insignificant fields.
User avatar
Senior Tester
Feb 22, 2019, 20:17
edited on: Feb 22, 2019, 20:17
it's not going to be a bad dump.

also videos of games that aren't recorded at 60fps are useless for verifying this kind of single frame glitch behaviour.
User avatar
Jun 11, 2020, 04:52
I need to say that perhaps not is a bad rom dump and also not is bug in atarig1 core.
That problem happen in others games exactly in a similar way randomly since before 2012.
The background tile map randomly flickering in others games for example Street Fighter 2 Hyper Fighting in E.Honda, Ken and Dhalsim stages.
I will try make video capture and after I post the result here.
User avatar
Jun 11, 2020, 21:30
edited on: Jun 11, 2020, 21:38
An reply before of my reply say that not is bad rom dump. Thus I had remembered that others games also has the same glitch.
Yes have others games where flickering the tile map in background similar how is saying in that issue.
"Compare" can show that is perhaps an bug in video core, other core or even atarig1 driver.
Have an exact same glitch happening in others games, but is random and not all time how is reported in Pit-Fighter.
I will try play SF2HJ and post here an video capture and thus being an random bug not will be soon to post;

Yes the atarig1 has problems.

I had posted in
the revision 1 is using an date after of revision 9 only to say that perhaps revision 1 not is exactly revision 1.
Your reply was right, but the report was closed.
Acknowledged status is better than close that report because is for try figure what is the exact revision number of revision 1.

Have an other report in
where have input delay in revisions 1, 6, 7 and 9.
You say that had happened changes in gameplay between revisions. You known the glicth where Kato does an 100 % combo ? If yes was removed after of revision 5 ? In MAME in revisions 1, 6, 7 and 9 not is possible to do Kato 100 % combo.
User avatar
Jun 13, 2020, 01:56
Yes has others games having similar bug how are being in Pit-Fighter.
My act was try add that information, but NOT say that atarig1 not has bugs.

About my report about revision 1 with strange BIOS date also was to add information because searching in not has any issue about it.
Thus being look how revision 1 is 10 how you are saying.
Any way that information in "revision 1" need to be fixed that was the because of my report.

Yes has more input delay in revision 1, 6, 7 and 9 than in others revisions.
I asked about if was removed Kato's infinite combo to see if has any detail related with input delay. Not exactly about gameplay.
Even thus using input twice forward or back has delay if comparing with first revisions.

Your post is correct about that game that is unhappily underrated for several gamers simply because is only 1 life and not refill energy.
User avatar
Jun 16, 2020, 00:28
edited on: Jun 16, 2020, 00:46
I use OSD values for debug.
Overclock 68000 to 101 % not happen the gfx glitch bug.
Verified in all sets not happening the bug.

Need to test if the vertical gaps bug also is "fixed" if overclocking the 68000.
User avatar
May 7, 2024, 14:43
edited on: 5 days ago
EDIT: I've replaced the old camcorder footage with a new footage I did using a video capture card straight from my Rev. 3 PCB @60fps. This demonstrates that the graphic glitch during TY intro sequence only happens in MAME and not in the real pcb.

The new video below shows the first 7 intro loops:

As we can see, there's absolutely no glitch during the Ty training sequence in the first 7 intro loops. And it never happens on the pcb.

UPDATED INFO: For some reason, in latest MAME versions, the glitch seems to be happening in the 3rd intro loop and not in the 1st loop. That's why I recorded a video from the first 3 intro loops. But in MAME this glitch happens randomly several times if you keep the emulation running for some more time. So it is definitely a graphic glitch caused by something related to the emulation itself. It's not a bad rom dump as I thought because in most part of the intro loops, the Ty intro sequence runs perfectly fine, with no glitches.

And last but not least: Today I tested the overclock trick suggested by Augusto with the latest MAME version 0.266 and after setting up the main cpu overclock to 101% the Ty intro glitch is gone! I really don't know what is the logic behind this but the fact is that it worked somehow. Really weird! But the glitch comes back if you set the overclock back to 100% (no overclock). Could it be wrong screen refresh rate? Because when I drop the screen refresh rate from 59.923 (default) to 58.923, the glitch also disappears!
User avatar
6 days ago
Updated info, please see my previous post above.