Home Page Forums MAME MAME TMS-09xx/1xxx thread (was New Dumps)

https://archive.org/details/GameWatchMarioBros scans for the pcb I sent

I ID'ed the segments on ML102. Because of the broken corner, I couldn't turn on all of the segments, but by driving all the commons in the other 3 corners together, the ones that I couldn't turn on showed up faintly, so I could at least tell what segment (An/Bn) they are on. And I could positively ID some because I had already ID'ed the other 3 commons for that segment.

www.seanriddle.com/gnw_ml102_skrsvg.7z

I cut out the middle of my LCD PCB so I could light the LCD from below and milled a piece of aluminum to hold the panel in place. I wound up using my laptop's screen as a polarized backlight. I didn't work too hard on the all-segments-on pic since this panel is broken and you already have a good SVG, but I need to clean everything better and use a tripod so I can get better focus with the macro lens I'm using.

www.seanriddle.com/ml102segs.jpg

I identified the remaining segments with help from ml102lcd10khc.jpg:
http://tsk-tsk.net/net/temp/gnw_ml102_v3.svg

The game looks fine in MAME, but there are problems when you move the player on the elevator. Do you have ml102metalrom.jpg? to make sure none of the bits were 'washed' away.

Here you go: http://www.seanriddle.com/ml102metalrom.jpg

Thanks, the bits look same as on the acidrom.
As a last try, I compared the visual dump against your partial electronic dump. Of the correctly dumped pages, one bit was different. I changed it and now the game works. The wrong bit looks like a "1" on both the metal and acid rom, very weird. See section at x=4900, y=4100:
http://tsk-tsk.net/net/temp/ml102_evilfakebit.jpg (blue overlay=1, empty=0)

The specific bit In ml102.bin:
offset $12C change $2F to $2B

That is weird! I looked at the original pics to make sure it wasn't a stitching glitch, but they look the same.

I decapped and visually dumped the ROM from Mattel Computer Gin's NEC D650: http://www.seanriddle.com/necucom4.html

The bit layout on the D650 is the same as on the D553 and D557 (and probably the D546).

I compared the visual dump to Kevin's electronic dump, and noticed his dump had bit 5 stuck high. But pages 1, 5 and 7 are completely different.

Tigrou from Belgium sent me a Tandy Pirate to de-glob. The PCB had a few hints; buttons were marked K1-K4 and S1-S4, and it used a single 1.5v coin battery. Looking through Bitsavers' data books, the OKI MSM6051 looked very promising; the chip was wired up almost exactly like the example application diagram.

But after I cleaned the die, it turned out not to be the MSM6051. It doesn't even have a manufacturer's name or logo, just the label 3C5800-919. It turns out that Travis Goodspeed had posted a 3C5800 die shot from an RSA SecurID. Unfortunately, I haven't found anything else about it. Worse, the bits aren't visually discernible.

That's great, kevtris sent you the Mattel Computer Gin's NEC D650?
So far, the patent information looks correct. So I can probably make it playable (with internal artwork).

No, I picked up another unit. I decapped the HLCD0530, too, but broke off one edge.

Got it working: https://imgur.com/a/e70G4