With a VM, a single byte can encode one of 256 subroutines/jump targets. On the 6502 processor, whether you're using JSR or JMP, your shortest option is a 3-byt

With a VM, a single byte can encode one of 256 subroutines/jump targets. On the 6502 processor, whether you're using JSR or JMP, your shortest option is a 3-byte instruction. That's a 3x code size reduction. Of course you pay a cost in dispatch overhead. Incidentally, there's also an intermediate option which is called a direct-threaded interpreter in the Forth world, where your VM code stream consists of jump targets rather than opcodes. That requires 2 bytes per operation rather than 3 while eliminating the table lookup to map the 1-byte opcode to the 2-byte jump target. The 3:1 reduction doesn't necessarily apply to operations that would normally map directly to CPU instructions. But especially for that generation of 8-bit microprocessors like the 6502, as soon as you're working with 16-bit data (which you definitely want for something like Zork) you end up spending a lot of code space for even simple operations. There's a famous example of a VM designed for the 6502 explicitly for reducing the code size (and inconvenience) of working with 16-bit operands: https://en.wikipedia.org/wiki/SWEET16 https://en.wikipedia.org/wiki/SWEET16