Hello everyone,
Sorry if my post is disturbing but I really need your help with this one.
A good a friend o my need to make 3 programs in assebly z80 language for a school project.

1st program:
Categorize the content of memory from FC00H until FFFF at ascending numerical order

2nd program:
registers in the memory from address FC00 until FFFF the content 00-01-02-… FF

3nd program
Multiply two 8bit numbers that are found in the memory and the result going to the memory [exp] (FC00) * (FC01) = (FC04) (FC03)
msb lsb

Any help would make us really happy
Thank you so much for your time

1) This is 4*256=1024 memory locations, so there are duplicates (not a problem, just an observation). Do you know what kind of sort you want to use?

2) Are you saying you want to store $00 in $FC00, $01 in $FC01, etc? Note that $FF will be stored in $FCFF, so what goes in $FD00?

3) This one's much easier. A quick Google search reveals a page I've used several times to help with Doors CS, http://baze.au.com/misc/z80bits.html. Their code for multiplying two 8-bit numbers:

Input: H = Multiplier, E = Multiplicand, L = 0, D = 0
Output: HL = Product

Code:


Notes on this: You'll want a ld l,0 \ ld d,0 before, obviously, as per the inputs. At the end, you'll want:


Code:


Which stores MSB, LSB to $FC04 and $FC03 respectively (this is little-endian, is that what you're trying to do?)

So now our complete code is:


Code:


Now, as for the unrolling part. That means that you need to repeat those three lines 6 more times for 7 total times. The speediest method is to copy and paste six times. However, I'm going to assume the negligible speed hit of saving space is better in this case, so we'll put it in a loop. The register b is used as an iterator:


Code:


The djnz decrements b, then jumps to MultLoop if it's still not zero. Otherwise, it continues past the djnz. The final step you need to add is load in H and E:


Code:


I'm curious, what kind of school project is this for?

First of all thank you KermMartian for your reply. We appreciate the fact you spend time to help us. We wish your kindness will be returned to you a 1000 times.

The first program checkes the memory numbers and put them in order the result must be something like this (FC00H)-> 00, (FC01H)-> 01, (FC02H)-> 02,...,(FFFFH)-> FF.
There have to be duplicates cause the memory is too big.

For the 2nd program I gues the result will something like this (FC00H)-> 00, (FC01H)-> 01,...,(FCFFH)-> FF,
(FD00H)-> 00, (FD01H)-> 01,...,(FDFFH)-> FF,
...etc

For the 3rd (FC04H)-> MSB, (FC03H)-> LSB

I'm sending you an exaple from how the program should look:

Code:

This is project for microcomputers lesson in greek technical school
Thats all that I can tell from the regarding what my friend told me.
We are so thankful for your help

Regarding the code you just posted, that's fairly unoptimized. How about:


Code:


I'm not sure exactly why you're resetting to 38h; I assume it's something with your particular device. Anyway, as far as your Part 3 goes, the code I posted should work. Part 2 is trivial (I'll type it the way your code looks):


Code:


For part 1, you have to chose a type of sorting first. May I recommend glancing at http://en.wikipedia.org/wiki/Sorting_algorithm?

38h is (always) an ISR (fine, only if you use interrupts in your device), why would you rst to that, anyway? If you really need that to fire, just 'halt'.

I thought the same thing, but now I'm assuming it's part of the assignment. I suppose it has to do with however their testbed is set up?

To tell you the truth I have no idea why we use that command. That's what we've been taught to. We are using this assebler



KermMartian
I cant tell which sort should I use at first program. I was thinking something which would compare 2 numbers by subtraction and carry check. Or whatever you think should work fine

Hmm, this is the MPF-1B:
http://www.sbprojects.com/projects/mpf1/mpf1.htm

Before we get to the sorting one, do Parts 2 and 3 make sense to you? I think those are both done; have you tested them?