[Evan]

http://www.snescentral.com/article.php?id=1000

It doesn't work, but it obviously is pre-release. Hopefully some crafty rom hacker might be able to extract the graphics and level layout data.

[MathUser2929]

I hope they somehow get the rom working. It'd be awesome to see the blob enemy in action, and maybe more stuff.

[BMF54123]

The unused font and "blob monster" are actually leftover data from Super Metroid. You can find them at 0x15D800.

There are some other leftover SM tiles at 0x31E000. All of them have been identical to the final so far.

I have my doubts about one of the ROMs being from a different version. I suspect ROM #6 may actually be bad, judging by the strange lines of dotted/corrupt tiles cutting right through parts of the graphics data (e.g. the one near the Funky's Flights graphics), and the fact that the block of data at 0x3502C0 is perfectly aligned to a bank boundary in the final (0x350000). Is it possible that the chip was outputting bad data in a few spots, shifting the remainder of the good data down a few hundred bytes?

[BMF54123]

I hate to say it, but I think this is just a seriously broken copy of the DKC Competition Cart.



Those yellow numbers are only present in this prototype and the Competition Cart; they were used for the timer and score display. The Super Metroid graphics are also present in both ROMs, oddly enough.

Both ROMs skip the title screen and drop you straight into the first stage. This is why its music starts playing in ZSNES.

A lot of code/data blocks are identical between the prototype, Competition Cart, and v1.2 of the final game (the text is in the same location in all three).

I dunno what's up with the rest of the ROM, though. I thought maybe it was just compiled from an earlier version of the source, but that doesn't explain why there are so many corrupted graphics, duplicated strings of bytes, and blocks of data that are clearly in the wrong place. I'd blame it on a bad EPROM, but it affects all eight to some degree. Really strange.

I'm assuming this didn't work before you dumped it, either, Evan?

[Evan]

d4s also believed that the sixth EPROM was bad, so it it good to know that this is confirmed. The actual cart doesn't work, even with the chips in the correct order, so I wouldn't be surprised if the chip is bad.

I actually never thought to check the game against the Competition Cart or later versions of the game. I guess they would have to prototype those ones too. Too bad if it is just a corrupt version of the competition cart, though I guess it would make sense (they did have the competition cart in prototype form for the 1994 Powerfest finals, which happened prior to the release of the game).

[Evan]

I updated the article. The game has badly dumped sectors throughout, so hopefully a redump will solve the problems.

I am convinced that a perfect dump will not be 100% binary identical to the final Donkey Kong Country Competition Cart, mainly because the internal checksum is different.

[Tongueman]

Beware also of certain types of EPROMs whose pin configuration is slightly nonstandard. For example, Famicom dev carts that come in FDS shells (a very common type for later Famicom protos) accept only (IIRC) 27C010 EPROMs that have pin A16 and /OE (output enable) pins swapped. So if one removes these EPROMs and dumps them like a regular 27C010 EPROM, the 2nd half of the ROM data comes out as garbage.

If your dumper is using just a cheapo Willem EPROM reader, and it's a Nintendo devcart, there's always a possibility that the EPROMs could be a nonstandard, finicky type.

[BMF54123]

[Evan]

The Competition Cart also had decidedly more chunks of Super Metroid in it than the v1.0. The European version also has things laid out differently than the US v1.0.

Also, I doubt that it is non-standard EPROM problem, because the data corruption happens in random spots. Most of the data are fine.