mark@mips.UUCP (Mark G. Johnson) (10/10/87)
In article <1717@bath63.ux63.bath.ac.uk>, pes@ux63.bath.ac.uk (Smee) writes: > Really what it comes down to is that you can't meaningfully > compare RISC and non-RISC machines on the basis of MIPs, > because the two sorts of MIPs are totally different. Depends on the definition of "mips". Practically every commercial supplier of RISC products (e.g. Pyramid, Sun, MIPS, Fairchild, HP) uses a definition that goes something like this: ########################################################################## # If machine X runs program P in time Tx, and if the DEC VAX 11/780 runs # # the same program P in time Tv, then BY DEFINITION # # (a) The VAX 11/780 is rated at 1.00 mips for program P # # (b) Machine X is rated at (Tv / Tx) mips for program P # ########################################################################## To most RISC companies, "mips" is an abbreviation for "number of times faster than a VAX". And they usually say so quite frequently in their literature. Note that this definition gives a (possibly) different number of mips for each choice of benchmark program. So most RISC companies measure quite a few programs and quote them individually and in an aggregate rating. The only exception I am aware of is the Inmos Transputer, which appears to quote "number of native Transputer opcodes executed per second", i.e. exactly the sort of thing which Smee objects to in the quote above. -- -Mark Johnson *** DISCLAIMER: The opinions above are personal. *** UUCP: {decvax,ucbvax,ihnp4}!decwrl!mips!mark TEL: 408-720-1700 x208 US mail: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086