SkoolKid

I have, and the answer is no, it doesn't.

Can't say I have - time to update my JSW disassembly TODO list. :) By the way, I checked and this is the only instance of a non-conveyor tile being affected by the Cell-Graphics bug.

One advantage of generating an ASM file from the SkoolKit source files is that it retains all the annotations. You can also generate an ASM file that includes several bugfixes: skool2asm.py -f 2 jsw.skool > jsw-bugfixes.asm

Yes, you can generate an assembler-friendly ASM file from the SkoolKit source files for MM or JSW. In fact, that's one of the main points of SkoolKit - you can use the same source files to generate both the HTML disassembly and an ASM file. You will need to download SkoolKit - for which you'll also need Python, version 3.4 or later - and use skool2asm.py to convert jsw.skool into jsw.asm (for example): skool2asm.py jsw.skool > jsw.asm I've successfully tested the output of skool2asm.py with pasmo, SjASMPlus and z80asm (the assembler that comes with z88dk), but it might work with other assemblers too.

No, there's particular reason to use XOR instead of OR here. As you've noted, there's no difference in the resulting value in the A register, and there's also no difference in the effect on the flags (which are not checked anyway).

This bug just got more interesting! :) I think I know why Willy is relocated further down the rope on your first attempt: in the section of code that handles Willy's movement along the rope, the rope status indicator is reset to 12 if it's found to be 11 or less. How you then manage to circumvent that code in your second attempt, I have no idea yet.

Oh, I know that - what I meant was that they are erroneously skipped by my cavern-drawing code. I don't know if I've mentioned this before, but none of the images in my disassemblies are derived from in-game screenshots - they are all programmatically generated.

As far as I know, the printer buffer (5B00-5BFF) is used only by the stack - the stack pointer being 5C00 when the game starts. Regarding the cyan guardians in the Kong Beast caverns: well, well, well! My first thought is that because the third horizontal guardian in those caverns is unused, the fourth one (the cyan one) is erroneously skipped. I will look into it... Regarding the comment at 91EE: the comment at the beginning of that section of code (at 91A3) says "The Kong Beast is falling", but I suppose you might have forgotten that by the time you reach 91EE. :)

Just checking in to acknowledge the comments and see what new stuff I should be adding to my MM/JSW disassembly TODO lists! :) I've been preoccupied recently with SkoolKit 5.3 and the Spectrum ROM Disassembly, but I'm going to take a break from those and concentrate on MM and JSW for a bit. Now, where were we the last time I was here...

But this implies that the first pair of offsets at 8300/8380 are not used, and that the pair at 8356/83D6 are used. You see my confusion? I need to (re)examine the code closely at some point to figure out what's going on.

So why should the text read "from F to F+32" (33 offsets) instead of "from F to F+31" (32 offsets)?

I'm really confused now! Let me put the questions another way. 1. How many pairs of x- and y-offsets are used to draw the rope during each animation frame: 32 or 33? 2. Is the pair of offsets at 8300/8380 used? 3. Is the pair of offsets at 8356/83D6 used?

Isn't that a contradiction? If only segments 1-33 have x- and y-offsets applied to them, and the bytes at 8300/8380 are used for segment 1, then the bytes at 8356/83D6 are unused - and the text as it stands now is correct: entries from F to F+31 inclusive.

In which case the zero bytes at 8356 and 83D6 - currently marked as unused - are in fact used. Which also means that when the rope is the leftmost or rightmost extent of its swing, the last segment is not drawn in the correct place (the bytes at 8356 and 83D6 should probably be 3 and 4 respectively). Which is a bug. Is this actually the case? Investigation needed!

No, not at all - keep them coming! I want the disassemblies to be as accurate and comprehensive as possible, and I'm always happy to receive new trivia suggestions and bug reports.

My earlier answer covers snapshots, but for TAP files, there's always a loader, even if it doesn't look like it. Most emulators have the ability to 'fast load' TAP files, which means the USR command in the BASIC loader is executed almost immediately, without you having to wait for the tape to play in real time.

Are you asking how an emulator knows where to begin execution when loading a snapshot file? If so, the answer is that the emulator takes the value of the program counter (PC) that is stored in the snapshot file. In a SNA file, PC is stored at the top of the stack (and must be popped off before execution begins). In a Z80 file, PC is stored in the header (like all the other register values). In an SZX snapshot, PC is stored in the ZXSTZ80REGS block (along with all the other register values).

Phew! :) I've just read the 'Through the wall' description a few times to get it straight in my head, and I must admit I'm not seeing the discrepancy between the text and the pictures. When I say that there is a nasty tile below Willy's sprite in step 16, what I mean is that when the nasty tile check is made at the point when the jumping animation counter is 16, Willy's x-coordinate has not yet been updated from what it was in step 15 - at which point there is a nasty tile below Willy's sprite. (His x-coordinate is updated before he's drawn during step 16, though.) Perhaps the wording doesn't make that clear enough. Would it help if I stated the actual x- and y-coordinate values when the nasty tile check is made?

Actually, Willy gets an extra life only when the thousands digit is rolled over from 9 to 0, so that POKE won't give rise to a too-many-lives problem.

Regarding unused code: the table of unused blocks is generated automatically, and I can't add an entry for a section of code/data that's embedded in a block. I could add some intro text to the 'Unused addresses' page that lists such sections, though. Regarding the item below the portal in The Sixteenth Cavern: after looking at the code, I'm sceptical that Willy doesn't get 100 points for collecting the item - the points are added at the same time as the item is collected. You could try POKE 36739,41 to check this - it would award 100,000 points per item collected, which is more obvious than 100. (Actually, that's a good candidate for the POKEs page.)

And so the TODO list starts filling up again. :) I've started a new thread in this forum to announce the disassembly updates, but it's still in moderation. Once it's accepted, let the comments and suggestions begin...

Rather than announce this in the existing disassemblies thread, I thought I'd start a new one to rebase the feedback (as it were). I've just made a new cut of the MM/JSW disassemblies: Manic Miner disassembly (decimal) (download here) Jet Set Willy disassembly (decimal) (download here) Manic Miner disassembly (hexadecimal) (download here - click 'Download ZIP') Jet Set Willy disassembly (hexadecimal) (download here - click 'Download ZIP') To see what's new, check the changelogs: MM, JSW. As always, comments and bug reports are welcome!

Wouldn't POKE 35770,201 turn crumbly platforms off? (And POKE 35770,77 would turn them back on?) If we could bind this operation to a key in 10 POKEs or fewer, I'd consider including it!

I'm not sure about 'aura', because an aura is usually depicted as brighter than the surroundings (like a halo). If there is a generic name for a field of differing brightness around a body, I'm not aware of it. 'Guardian Brightness Bug' would be more accurate, but also less interesting.

'AND #C0' sets the zero flag if A holds any value from 0 to 63 (00-3F), and 'AND #E0' sets the zero flag if A holds any value from 0 to 31 (00-1F). Note that you can't use 'AND N-1' to set the zero flag on every Nth pass through a loop that increments A unless N is 1, 2, 4, 8, 16, 32, 64, 128 or 256.