>>>TP knit circles around anything else I had on CP/M.
>>
>>My first machine was a used Apple //e with a Z80 card and a RAM card (with a WHOLE 1meg of RAM). I would boot to RAMDisk C:, load TP AND my working files to C: and I was cooking along at machine speed. Ran a (dog) pedigree program that I had written for quite a while. Then came assembler class and my instructor couldn't read the 8080 code. C'mon, there's not THAT much difference between JUM and JMP.
>
>Z80 had too many registers and knew how to combine them into 16-bit registers, plus those darn index registers. Above what can be understood from the 650x perspective.
Coming from a Z80 background and I found the 6502 rather lacking in registers (though it sort of made up for it by with zero-page memory instructions), and somewhat limited stack space (constrained to page 1). As for assembly language mnemonics, it felt weird seeing mnemonic always expressed as 3 characters. Transition from Z-80 to x86 assembly language was a bit easier due to somewhat their similarity in mnemonic style (Z-80 didn't differentiate between "load" and "store" in their mnemonics -- they simply used "LD" mnemonic, which did lead to confusion for some folks). As for the relatively large number of registers on the Z80 -- especially the "primed" set. I do remember fiddling with them on the TRS-80 (only thing you needed to remember is to always switch back to the "normal" set -- otherwise the system would hang). The only computer I remember seeing them used extensively was on the Sinclair ZX80 and ZX81 micros (on these you dare not fiddle with them -- they were tied with hardware interrupts, so even executing the switch instruction could cause the system to crash).