toddpw@nntp-server.caltech.edu (Todd P. Whitesel) (12/12/90)
Don't trust that figure. A 1 mhz 6502 is more like .3 _Native_ MIPS. Native MIPS is how many of the machines' own instructions can be executed in a second. This rating is a good way to rate the average number of clock cycles an instruction takes, but as a cross-CPU performance index it is absolutely useless. Workstations are often rated versus VAX MIPS, i.e. their performance is compared to that of a VAX 11/780 running at (yow) 1 mhz. The VAX had 32 bit registers, so a 6502 would quite likely do really poor in VAX equivalent MIPS (officially VUPS, for VAX Unit Of Processing or something like that). Todd Whitesel toddpw @ tybalt.caltech.edu
fadden@cory.Berkeley.EDU (Andy McFadden) (12/13/90)
In article <1990Dec12.064925.28609@nntp-server.caltech.edu> toddpw@nntp-server.caltech.edu (Todd P. Whitesel) writes: [snip] > Workstations are often rated versus VAX MIPS, i.e. their >performance is compared to that of a VAX 11/780 running at (yow) 1 mhz. The >VAX had 32 bit registers, so a 6502 would quite likely do really poor in >VAX equivalent MIPS (officially VUPS, for VAX Unit Of Processing or something >like that). If you use 16 direct page addresses, the question becomes: how many registers are available to the VAX microcode engine? If 6502 is nearly equivalent to what the VAX uses internally, things would be pretty even. I don't have my VAX tech ref with me, so I don't know; if anybody is terribly curious, ask again in about a week, and I'll check it out. >Todd Whitesel -- fadden@cory.berkeley.edu (Andy McFadden) ..!ucbvax!cory!fadden fadden@hermes.berkeley.edu (when cory throws up)
rhyde@ucrmath.ucr.edu (randy hyde) (12/13/90)
I'm not sure what you're getting at. A VAX is a 32-bit computer. It can do a 32-bit memory-to-memory addition with a single instruction. I dunno how long this takes, but it is much faster than the 13 6502 instruction (average- 3 cycles each) to do the same job. Perhaps you could clarify your comment. *** Randy Hyde O-)
brianw@microsoft.UUCP (Brian WILLOUGHBY) (01/03/91)
In article <10564@ucrmath.ucr.edu> rhyde@ucrmath.ucr.edu (randy hyde) writes: >I'm not sure what you're getting at. A VAX is a 32-bit computer. >It can do a 32-bit memory-to-memory addition with a single instruction. >I dunno how long this takes, but it is much faster than the 13 6502 >instruction (average- 3 cycles each) to do the same job. >Perhaps you could clarify your comment. >*** Randy Hyde O-) If you "dunno" how long it takes on a Vax, then how can you possibly say it takes less time than a 6502? If you were to actually look up the number of cycles needed for a 32 bit memory to memory addition, you would no doubt be surprised at how many cycles it does take. The 65x02 has some of the lowest cycle counts for processors in its price rance (I'm not counting pipelined, multi-register RISC chips, even though the 6502 was one of the first processors to use pipelining). Brian Willoughby UUCP: ...!{tikal, sun, uunet, elwood}!microsoft!brianw InterNet: microsoft!brianw@uunet.UU.NET or: microsoft!brianw@Sun.COM Bitnet brianw@microsoft.UUCP
rhyde@ucrmath.ucr.edu (randy hyde) (01/04/91)
>>> If you "dunno" how long it take on a Vax, then how can you possibly say it
takes less time than a 6502?
Because I have used a VAX. It is 100 times faster to me! Are you going to
start claiming a 6502 can outrun a VAX? Get real! I have an LSI-11/03
(Heathkit h-11) which is faster than a 6502 @ 1Mhz. Come on, this
response borders on true nonsense. Do you just want to argue or do you
want to discuss possible truths around here? Next you'll be saying my
Apple //gs is slower than my Mac IIfx!
*** Randy Hyde