In the part number, the numeral that immediately follows the S (meaning static) refers to the process geometry. For example, the full part number for the 65C816 I'm using in POC V1.1 is W65C816S6PLG-14. The 6 means 0.6 µm process, PL means PLCC-44 package, G means the part is RoHS compliant and the -14 refers to the speed grade.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Thanks!

I have always though one of the main advantages of 6502 assembler is that it "fits in your brain" -- it was created in an age when humans still wrote the code. Now have all of these chips that are created with compilers in mind, so you can have these really powerful, but rather complicated instructions not fit for humans. Put simpler, ARM assembler doesn't seem fun. When every instruction can be conditional and you can shift everything at the end, it gets pretty complicated.

What surprises me more is that the 6809 hasn't stuck around. You'd think that if there is a market for the 6502 and the 65816, it could have found a home somewhere, too.

Having ported a compiler to ARM (a large part of which was writing small assembly-language snippets), I found ARM assembler to be great fun. Predicated code allows for some impressive tricks even without teaching the compiler to take direct advantage, and having the program counter in a register is good for not just position-independent data references, but also position-independent computed jumps, and then there's having a barrel shifter on one of the ALU inputs... My main complaints are the limited range on immediate values and the lack of registers (just having ONE more register would have saved me a great deal of trouble, for example, and several months after completion we're still trying to scare up another register in order to implement a major chunk of "optional" functionality).

Oh, and the stupidly huge number of FPU variations. FPA, iwMMXt or whatever it's called, I don't know how many variants of VFP, plus probably others. Ugh.