Cool! Lot of work, not just the build but the assembler too, but the emulator too!
Poking asm without reading the manual? Brave! ...or dumb?
(Just kidding)
Yeah, becomes clear when contrasted with CLR Rx. Although it's interesting that MVx src are listed at all. Are these not just assembly aliases of MOV Rx, src? (Impossible to tell without encoding given alongside it.) Or, why enumerate them but not others?
Interesting choice to make Rb the... well, it's not the accumulator as such I guess, but the default (implicit) source. I'm not sure offhand if this is actually much (or any?) worse than the usual way (accumulator default destination).
And one register is named Acc, though it's not obvious that "ALU output" necessarily works the same as all the other registers? And if arithmetic and logic instructions return data to a selected register (not just Acc), is it really accumulating anything..?
Well, nevermind. Again, cool stuff, great work!
Tim
I know you were joking, but...
When I was seeing it for the first time, I didn't know of the existence of the assembly language manual, or where it was. Also, I was NOT expecting to write any programs for the unit. So, no need to read it anyway.
But, I was then fascinated, and playing with the prime number programs. But, they were taking ages, just to count from 1, to 2. Then eventually 3, Then some time later 4 etc etc. But I wanted to see much bigger (potentially) prime numbers.
Also, at the time, I didn't initially realise the buttons and controls/potentiometers, are actually user adjustable in real time (Wow, just wow!

). So, the (at max speed),
worlds slowest computer not the fastest computer, was defaulting, to a really slow clock speed setting.
So, I adjusted the assembly listing, (I didn't realise there was an alternative prime number program, which already lets you choose the selected number, via a constant near the top), to make it start at a much bigger number.
Then the bug (excitement, not fault) caught me, and I modified more and more stuff. Until I ended up writing complete programs.
My latest program (unfortunately NOT saved), calculated to 32 bits!
Using all 4, 8 bit registers.
By setting the clock speed to maximum (apparently around 6,500 instructions per second), I was (eventually, after much patience), able to count to 2^24 + 1, and beyond!.
At some point, I
cheated, optimised it, by putting in a huge bunch of increment instructions, set to be a binary multiple (e.g. exactly 16 Incs), so that the carry flag would still be set correctly, and only one (to make a loop) jnc instruction was needed for every 16 (later 32 or even 64) increment instructions.
It worked out very well. The 256 bytes is much more practicable, than its (I believe) original 16 bytes. Which wasn't very useful, beyond some early learning experiences for people.
The extra instructions, really helps as well.
I've NOT been able to find out much about the original spec "Ben Eater" computer anyway. I've just taken educated guesses, from watching one or more of his videos about how to make it, and what it does.
Another way of naming it, would be to have a register name, which means all of registers, A, B, C and D, such as K, E,
Z (EDIT: Z is no good, as could get confused with a zero flag) or something.
Then when you say CLRK, it doesn't get confused with register/Accumulator RA.
E.g. Some processors (if I remember correctly), have Accumulators A and B. But when they are used together (combined), they are called D.
I.e. like the 6809 (and others).
https://en.wikipedia.org/wiki/Motorola_6809in the 6809 are the use of two 8-bit accumulators (A and B, which can be combined into a single 16-bit register, D)
EDIT: Don't worry too much about it (like me). There are only 26 letters of the alphabet, and many/most instruction sets, end of with some confusing mnemonics in them. (E.g. Motorola (old company name) has a Sign Extend B into A instruction!, with a funny name).
I.e. the (please don't laugh), 'SEX instruction'. (Ignore the PUN, PLEASE!).
https://atjs.mbnet.fi/mc6809/Information/6809.htm