Programming history of JP Set 3 - US Set 1

[mrchrister]

Hey guys,

I have some questions about the differences between US and JP DK roms and am hoping some of you here have some theories of their own around this.

So apparently the original DK was programmed by Ikegami who did an amazing job at putting this game together for Nintendo in Assembler code. One thing I don't understand is why the structure of the levels in the most common rom "US set 1" is pretty screwed up. The first few levels are out of order of what Miyamoto apparently had in mind. Level progression is:
1 - 4
1 - 3 - 4
1 - 2 - 3 - 4

I don't think this was an intentional move on either Miyamoyos or Ikegamis side. If I was Migamoto back then I would be pretty upset that my story has been screwed up by buggy program code. Not only does the info screen in between levels (How high can you get?) Not make sense to go from 50m to 200m but you also rescue Pauline pretty easily before the game starts over.
I read that back in the day when people saw the pie level for the first time it was quite exciting that the game had surprises after so many played stages but I wonder if this was ever an intentional move? Also why do you win when you reach the top of the pie factory platform? It looks like the game was meant to continue and you had to climb the ladder to get past moving Kong to rescue Pauline. You also never get to play the easy pie level on US set 1 in contrast to JP set 1 because the first time you see it you already beat DK twice.

I find it interesting that the Japanese code has the correct order while the US version doesn't. I know that this game was primarily developed for the US market to convert the hard to sell Radarscope cabinets. Wouldn't you think Nintendo paid extra close attention to programming bugs for the US market? I'm also surprised that the ladder cheat, where you can hide on top of a ladder without a barrel ever hitting you was never fixed on the 3 Japanese ROM sets. Also why would there be different roms at all between US and JP?

[ChrisP]

It doesn't make sense story-wise, but it definitely makes sense in terms of creating intrigue and getting more quarters into the machine.

If you complete one loop, then go back to the barrel board, you might think that's all there is. But then you complete that barrel board, and all of a sudden you're on a new board (the springs). You start to wonder if there's more. So you keep playing, putting more quarters in, and eventually reach the third loop where you discover yet another new board (the pie factory). Is there more? You don't know. Want to see that mysterious pie board again? Better keep playing.

Even though the original Japanese game just cycles 1-2-3-4, the code actually strongly implies that shuffling around the screen order was intended to (at the very least) be possible.

There's a table at #3A65 with 20 "slots" for the screen order. In the US release it's 1,4,1,3,4,1,2,3,4,1,2,1,3,4,1,2,1,3,1,4. In the Japanese release, it's simply 1,2,3,4 repeated five times. If the game were only intended to go 1,2,3,4 and never any other way, then there would have been no reason to set aside 20 bytes for this table, since 16 of them are redundant.

In addition, the text for "125m" and "150m" are options explicitly built-in for the "how high?" screen, and the part of the code that determines how many stacked monkeys to draw does check for 6 monkeys, which you never see in the Japanese version since all levels top out at 100m, but these options are in the original Japanese release. Why would they add 125m and 150m text blocks and a monkey-stack limit of 6 if they were never intended to be used? So I'm thinking that they actually DID code the board-slot scheme with flexibility in mind, in order to keep options open for easy tweaking later.

NOA took advantage of that, and again I'm pretty sure it just came down to adding replay value.