OISC:2
OISC:2 (pronounced "whiskey 2") Obfuscated Indirect Subleq with Coprocessor: 2 word instructions
OISC:2 is a derivative and expansion of Subleq and is an OISC. However, it has been both extended and obfuscated by using the trichotomy of numbers to have different effects based on the positive, negative, or zero value of each of the two words per instruction.
OISC:2 accepts a mandatory positive and an optional negative memory file. The coprocessor is activated by sending data to the Mode address (-7). Positive memory is loaded as integers only. Jumps to negative memory Halts. Negative memory can be addressed by indirection, but instructions are only in positive memory.
The sample code (posmem.o2c and negmem.o2c) prints the sign of a memory word, inputs and prints a keyboard character, and prints "Hello, world!" using direct and then indirect addressing.
What's the point?
The purpose of the OISC: project, besides entertainment, is to see just how far the Subleq metaphor can be pushed. Subleq is an elegant one instruction set computer (OISC), but it uses direct memory addressing as a form of meta-programming to perform side effects like input, output and halting. OISC: takes the meta-programming to its illogical extreme to enhance the power of the language through obfuscation. And the coprocessor is just ridiculously powerful for an esolang.
Summary
Two word instructions: A B
If A & B are both positive: *B = *B - *A If A & B are both negative: **B = **B - **A If A is positive and B is negative: IF *A <= 0 Jump to |B| If A is negative and B is positive: IF **A <= 0 Jump to B If A is 0 and B is positive: STDIN --> *B If A is 0 and B is negative: STDIN --> **B If A is positive and B is 0: *A --> STDOUT If A is negative and B is 0: **A --> STDOUT If A & B are both 0: HALT
Negative Memory
Address Function -1 IP (initially 0) -2 NEXT (always IP+2) -3 RETURN (initially 0, set to NEXT before any jump.) -4 Register a -5 Register b -6 Register c -7 Mode (activates a, b, c; then resets to 0) -8 MaxPos memory size -9 MaxNeg memory size -10... Data (user address computation: start from 0, add 10, negate)
Coprocessor
Mode Function 0 NOP 1 c = ~b (bitwise not) 2 c = b & a (bitwise and) 3 c = b | a (bitwise or) 4 c = b ^ a (bitwise xor) 5 c = b << a (shift b left by a bits) 6 c = b >> a (shift b right by a bits) 7 sign of b (+1, 0, -1) 8 floor of b (integer) 9 truncate b (round down, integer) 10 c = b - a 11 c = b + a 12 c = b * a 13 c = b // a (integer division) 14 c = b % a (mod, integer remainder) 15 c = b / a 16 c = b to power of a 17 c = a root of b 18 c = log base b of a 19 c = sin(b) 20 c = cos(b) 21 c = tan(b) 22 c = asin(b) 23 c = acos(b) 24 c = atan(b) 25 c = sinh(b) 26 c = cosh(b) 27 c = tanh(b) 28 c = asinh(b) 29 c = acosh(b) 30 c = atanh(b) 31 c = sqrt(b**2 + a**2) (hypotenuse) 32 a = pi, b = e, c = phi (golden ration) 33 c radians <-- b degrees 34 c degrees <-- b radians 35 c = greatest common divisor of b and a 36 c = a permutations of b items (unique) 37 c = a combinations of b items (not unique) 38 c = b! (integer factorial) 39 c = sum of 0..b (integers, works with negatives)
Philosophy
What's the point of all this? What is an instruction?
Subleq is, to me, the premier one instruction set computer (OISC). Three memory addresses, A B C, where B=B-A, if B<=0, jump to C. That's it. Simple, elegant, usable. An operating system (Dawn, sadly no longer available) was once written using it. Since there is only one instruction, it is assumed. But to be useful, any language has to perform I/O. So Subleq uses memory mappings for input, output, and halt. Each of these attaches to a different part of the instruction (A, B, or C). It works well. Where is the instruction?
Subleq+ uses negative numbers to indicate indirect addressing. It is a touch more complicated, but works well and enhances the power of the language. Where is the instruction?
OISC: uses the trichotomy of numbers (positive, negative and zero), along with floating points and integers, to accomplish several different things using the same two (or three) words, which are memory addresses. There is even a parser which uses recognizable commands. But these commands are seen nowhere in the actual executable code. It's nothing but memory addresses expressed as numbers in different ways. Where is the instruction?
Complexity arises from relationships. Power arises from complexity.
Computational class
OISC:2, being a derivative of Subleq, is Turing-complete.
See also
- OISC One instruction set computing.
- Subleq The original, great OISC.
- Subleq+ Subleq with indirection signed by negative memory addresses. Subleq+ is the primary motivator for the OISC: project.
- OISC:2bis New and improved, now with a stack and a working parser.
- OISC:3 Obfuscated indirect subleq with coprocessor: 3 word instructions.
External resources
OISC:2 Github repository. There is not yet a working assembler, but the python interpreter works with the provided files.