Supermodel Forum

Hello - I have 2 questions regarding MTA.

1 - is there any way to change the region to a English? I can get in the service Menu but region is unchangable.

2 - During the attract mode screens after the first initial title screen, I get a black screen with just the instruction prompts onscreen. After that during the next attract mode level it shows the underwater level with strobing graphical glitches. Are there any settings that can fix this or is this a bug with the emu?

I’m using Win7x64 with an Nvidia 1050ti.

Thanks!

I think its possible a bad dump is causing the issues in attract mode

Ian wrote:I think its possible a bad dump is causing the issues in attract mode

Hi thanks for the reply. I'll look for another dump.

Could this be why the region can't be changed in the service menu as well?

So I have the newest MAME romset and this problem is still occurring. Tried a couple others I found and no dice.

Is anyone else able to replicate this issue? I don't think it's something with my video card either. I tried it on my work laptop with Intel video and the same thing happens.

This problem occurs for everyone. When Ian says it's possibly a bad dump, he doesn't mean that only you have a bad dump (if you did, it wouldn't pass Supermodel's CRC check); he suspects the ROM set itself is bad. We don't know what's causing it.

Would that mean that we would have to find a working board for this game in order to get a proper dump of it? Last time I ever saw this game was at a Chuck E Cheese fifteen years ago, so it probably won't be easy finding one :/

Yes, it would. But I'm not sure it's a bad dump. What's weird is that the ROM test only seems to check ICs 1-4. I think Ville has this board, maybe he can confirm the ROM test?

I think bad dump is possible. The bad values also leak into the viewpoint data and ram models. Either that or nan cpu bug

Ian wrote:I think bad dump is possible. The bad values also leak into the viewpoint data and ram models. Either that or nan cpu bug

I think a CPU bug is very likely. Or, this could be a timing issue. It's conceivable that e.g., IRQs happening in the wrong order or some time delta somewhere being too small or too large is causing a calculation to barf. One thing we could do is write some code to detect when a floating point register first becomes NaN and then try to trace back from there to see if there is something obviously weird going on with inputs to that calculation.

If it is a CPU bug mame has the same issue