User Tools

Site Tools



This shows you the differences between two versions of the page.

Link to this comparison view

base:mathematics_in_assembly_part_3 [2015-04-17 04:32]
base:mathematics_in_assembly_part_3 [2015-04-17 04:32] (current)
Line 1: Line 1:
 +====== Prologue ======
 +Hello dear sceners, this is the third chapter of my tutorial on maths in assembly,
 +first published in "GO64!" magazine. The first two chapters of this turorial were
 +re-published in "Attitude #4" and "Domination #17". Have a nice read!
 +====== Mathematics in Assembly, Part 3 ======
 +After increasing the precision of our numbers we're going to have a look at a
 +slightly more difficult topic, namely the multiplication. But take it easy, this
 +a bit more complex mathematic operation can be done in many different ways,
 +depending on the memory an execution time requirements. The simplest implementation
 +of the multiplication would of course be a simple loop which just increases a zero-
 +initialised counter by the first multiplication factor as often as the absolute
 +value of factor 2 demands. This extremely inelegant brute force method is naturally
 +very inefficient and slow, that's why it's just mentioned in passing here and not
 +going to be discussed in any more detail. let's take a look at some more elegant
 +====== Multiplication by Constants ======
 +A kind of multiplication relatively often needed is to multiply by constants, as it
 +just takes quite a small and fast algoritm. As an example constant we just take 11
 +($0B, %00001011). Using our knowledge about binary arithmetic, we can easily reduce
 +this number to powers of two:
 +11 =  8  +  2  +  1
 +   = 2^3 + 2^1 + 2^0
 +If we wanted to multiply this number with let's say 25 ($19, %00011001), it would
 +look like this if written on paper:
 +11 * 25 = 2^3 * 25 + 2^1 * 25 + 2^0 * 25
 +        =  8  * 25 +  2  * 25 +  1  * 25
 +What's all this needed for, you might ask. Very simple: the seperate summands of
 +the created expression are all shifted 25's. So it's as easy as this: the accu
 +contains an arbitrary number to be multiplied by the given constant. The number is
 +shifted by the needed positions an buffered. Afterwards, all summands are added.
 +Just a small pice of example code:
 +LDA #$19    ; arbitrary factor
 +STA BUFFER0 ; some byte in the memory, 2^0 * factor
 +ASL         ; doubling of the factor
 +STA BUFFER1 ; buffer 2^1 * factor
 +ASL         ; doubling
 +ASL         ; doubling, all in all 8 * factor
 +CLC         ; clear carry for addition (whithout overflow, like here, actually
 +            ; unnecessary)
 +ADC BUFFER0 ; 2^3 * factor + 2^0 * factor
 +ADC BUFFER1 ; 2^3 * factor + 2^0 * factor + 2^1 * factor
 +Simple. Oh, all examples will, just like this one, for the sake of simplicity,
 +refer to unsigned 8-bit integers to be multiplied. By the way, there are some
 +"factors" to be aware of with the multiplication.
 +====== Whats to be paid attention of ======
 +#1: After multilying two factors, the result has the double size, for instance
 +16 bits with two 8-bit factors. Quite obvious, as for example
 +255^2 = $FF * $FF = $FE01. Nevertheless, for the sake of simplicity, I'll take
 +appropriately small values in order not to cause any overflows.
 +#2: Also with computers, the exact result of multiplying two fractions has as
 +many decimals as both the factors have together. So the product of two fixed
 +point numbers with let's say one fraction byte each has two fraction bytes.
 +Together with the squaring of the number range this means, for instance: the
 +product of two 16-bit fixed point numbers with one fraction byte each is
 +32 bits in size, 16 bits left hand of the point, 16 bits right hand of it.
 +#3: in order to multiply signed numbers, the signs of both factors are
 +buffered and the absolute value of the result is calculated by multiplying
 +the absolute values of the factors. Afterwards, the buffered signs are
 +avaluated and the result is signed accordingly. Of course, the common rules
 +also apply here, so plus my minus is minus and minus by minus is plus.
 +Good. The multiplication in general and in particular with constants can also
 +done in another way, the magic word here is tables.
 +Tables? Correct. More exactly "Look Up Tables". Using these tables, it's
 +possible to calculate products extremely fast, or let's better say "to find
 +them out" extremely fast. For tha, a table has to be generated at the
 +beginning of the program, in this case containing the multiplies of the
 +constant in rising order. With the constant 3 ($03, %00000011), it would look
 +like this:
 +$00, $03, $06, $09, $0c, ...
 +This table is just read out, using the arbitrary factor as index. On
 +principle, this table is as large as the range of the factor is, i.e. 256
 +bytes with a 1-byte-sized factor. By the way, generating this table, the
 +above mentioned brute force method can be used, as with the initialisation
 +speed plays a minor role compared to memory efficiency.
 +That's it for the first couple of possibilities to multiply. The next three
 +chapters will be published in "Vandalism News Ruby Edition", discussing some
 +more ways to multibly assembly, starting with the most flexible kind of
 +multiplication, the bit-wise multiplication. See You!
base/mathematics_in_assembly_part_3.txt ยท Last modified: 2015-04-17 04:32 (external edit)