The Philco 212

Some time ago, I had learned that the Philco 2000 computers used their index registers as pointer registers, rather than as misnamed base registers, because the selector bit changed the part of the instruction word that contained a three-bit index register specification and a 12-bit displacement to a simple 15-bit absolute address.

Since the machine had at most 32 K words of memory, it didn't need the index registers for use as base registers, and could instead use them as pointers. That meant the 12 bit displacements were only occasionally useful.

Today I read an old DATAMATION article which said that adding just one bit, a control bit, to the index registers when going from the Philco 211 to the Philco 212 allowed the index registers to be incremented or decremented "an arbitrary amount" after use.

Hmm. How could a single bit contain so much information? It seems to me that the explanation is obvious - if an indexed instruction uses an index register with that bit set, the displacement field instead serves as a signed increment.

Of course, this guess could well be wrong. Does it jog any memories?

John Savard Usenet Zone Free Binaries Usenet Server More than 120,000 groups Unlimited download