That's what the instructions do, but why do they do that? This is the same thing I spent hours earlier this year (unsuccessfully) pounding into your brain. :banghead:

Also, for the love of god next time you post a snippet of code, add a few line breaks. They don't bite.

Instead of making the generic observations of "branch if carry set" -- which isn't really an observation so much as stating what the instruction does -- try to look at what purpose that serves.

That is... when you come across BCS... look at what last changed the C flag, and use that in your comments instead

given the following:


Right now you're looking at that and saying  "Compare A to #$007E", "Branch if Carry Set".  Which isn't really that useful, since the instructions are already plainly saying that.

To know what purpose that BCS serves, you need to know what the C flag is doing.  So look at what set the C flag -- in this case, the CMP.  C will be set when A >= #$7E.  Therefore instead of saying "Branch when carry set", say "Branch when A >= #$7E".   Or "if A >= #$7E, jump to $xxxx".

That's the first step in understanding it... the next step would be to figure out what A is doing.  Is A a character in the string?  Then revise your comment:
"If the character is >= #$7E, jump to $xxxx"

Now you might start to get a clearer picture.  Perhaps characters over $7E are DTE characters, and that is jumping to a DTE routine or something.  Or perhaps they're control codes.  Or whatever.  But just using register names and dictating the bare-bones function of the instruction doesn't help you understand.

The next step would be to figure out what the code at $xxxx is doing, then adjust your comment yet again.  "If character >= #$7E, jump to DTE routine" or somesuch.

One small question:

Since a previous CMP (CMP #$004B) set the Carry flag, would that mean "if A >= #$4B, jump to $9E9E"? Because really, the second CMP does nothing, since the Carry flag was already set, right?

~DS

No.

CMP updates the carry flag.  That means it will change its state according to the new value -- meaning it will set or clear the carry flag according to the instruction result.  If you do two CMP instructions in a row, it's the first one that will be overwritten by the second one (so the first would be the one doing nothing)

Why would you switch it?

And you can access the ROM with several CPU addresses, depending what bank is selected...

Yeah, I was switching it to deal with the way SNES and NES deals with pointers.

And thanks, I'll try a different corruptor.

Edit: I took a better look at the EOR instruction (quoted above). So it's taking the value from $7E001D ($002C) and doing what? I'm confused, all it seems like it's doing is throwing that value, reversed, into the accumulator (taking A from $0088 to $2C88). Also, I'm not sure if it matters or not, but $2C is probably a control code (I have an entry in my table for $2B as a code to open a new window, and there's a thousand of them, so $2C is probably one that I haven't encountered yet).

I also took a better look at the two CMPs. The first CMP compares A to $004B. Anything that is $4A or below in the table is the DTE/Dictionary values. The second CMP is $7E, and directly after that is the normal kana ($4B-7D is the dakuten/handakuten section). So is it checking to see if the value is DTE/Dictionary in the first CMP, and seeing if it needs dakuten/handakuten in the second CMP?

~DS