moskowit@paul.rutgers.edu (Len Moskowitz) (07/17/90)
Does anyone have relative performance figures for the various transputers and the 68030/68040, 88100/88200, and i860/i960 processors? I'm more interested in integer math, logical operations, and memory access performance than floating point math or array operations.
iann@specialix.co.uk (Ian Nandhra) (07/17/90)
moskowit@paul.rutgers.edu (Len Moskowitz) writes: >Does anyone have relative performance figures for the various >transputers and the 68030/68040, 88100/88200, and i860/i960 >processors? I'm more interested in integer math, logical operations, >and memory access performance than floating point math or array >operations. The trouble is that the instruction sets will not allow a meaningful comparison. What type are you after. VAX MIP's comparisons are often quoted (OK, not by Inmos!!). Ian. -- Ian Nandhra, Specialix, 3 Wintersells Road, Byfleet, Surrey, KT14 7LF, UK {backbone}!mcsun!ukc!slxsys!iann iann@specialix.co.uk Tel: +44 (0) 9323 54254 Fax: +44 (0) 9323 52781 Telex: 918110 SPECIX G >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
moskowit@paul.rutgers.edu (Len Moskowitz) (07/17/90)
I'd be happy to hear of any reasonable yeardstick with which I could compare them. If anyone has run a task on these machines using commonly used compilers and without substantial optimization, I'd appreciate hearing the relative performance results. If I had to boil my question down it'd go like this: "Can a single transputer (any model) run my typical C application significantly (say four or more times) faster than my 20 megaHertz 68030?"