scarter@gryphon.COM (Scott Carter) (11/30/89)
In article <5275@omepd.UUCP> mcg@ishark.Berkeley.EDU (Steven McGeady) writes: >In article <31329@winchester.mips.COM>, hawkes@mips.COM (John Hawkes) writes: > >> The Atlantic Research Corporation, an independent group, has done some >> comparisons between the MIPS R3000 (25-MHz) and a 20-MHz 80960 executing Ada >> programs (the "Common Avionics Processor Ada Benchmark Suite"), and they >> discovered that the R3000 was usually more than twice as fast on hand- >> coded programs, and overall was more than five times faster on >compiled > > programs. > >The 20MHz 960 referred to here is the Military 80960MC part, *not* the >960CA. The 960MC hit silicon in 1985 and has not been upgraded since >then. ARC did not measure the 960CA, even though that would have been >a more representative measurement. More representative of what? I agree, I'd love see the 960CA numbers on the integer JIAWG benchmarks - not to mention a few of our own - but the about half the benchmark suite is floating point. Also, contrary to the"typical" embedded system, the JIAWG avionics systems _do_ need MMUs to implement the multilevel (B3/A1 - ouch!) secure OS requirements. The capability addressing of the 960XA was one of it's big selling points. >The part measured was running in a >PC/AT plug-in board. The MIPS system it is being compared to is a full >system with a significantly-sized off-chip cache. We got numbers of a BiiN 6220, which is [was] a full-sized system. If the caches on that system were smaller than on an M/2000 - and they were - why? Certainly not to make a cheap system. >The 960CA would perform approximately 2x *faster* than the MIPS R3000 >on the handcoded versions of the benchmarks. For compiled code, if >the code were written in C, we would also perform approximately 2x ^^^^^^^^^^^^ >faster. Why this caveat, i.e. why not Ada? Or is it just the lack of a suitable Ada for the 960CA? Using a set of Pascal versions of some of the JIAWG benchmarks [developed by an independent contractor for another program], we got much better performance out of the Mips than in the Ada version. >The code in question was compiled with a beta-release Ada >compiler available last spring. <...> >If people are more >interested in these tests, I will see how much information JIAWG will >allow to be released, and release it here. > >S. McGeady >Intel Corp. Scott Carter