The average of a set of floating-point and integer SPEC benchmark results.
While the old average SPECmark89 has been popular with the industry and the press, SPEC has intentionally *not* defined an average “SPECmark92” over all CPU benchmarks of the 1992 suites (CINT92 and CFP92), for the following reasons: With 6 integer (CINT92) and 14 floating-point (CFP92) benchmarks, the average would be biased too much toward floating-point. Customers’ workloads are different, some integer-only, some floating-point intensive, some mixed. Current processors have developed their strengths in a more diverse way (some more emphasizing integer performance, some more floating-point performance) than in 1989.
Some SPECmark results are available here (ftp://ftp.cdf.toronto.edu/pub/spectable).
See also SPECint92, SPECfp92, SPECrate_int92, SPECrate_fp92.
benchmark An old SPECmark benchmark result derived from a set of floating-point and integer benchmarks. It is the geometric mean of ten SPEC ratios of the outdated 1989 SPEC benchmark suite. The use of SPECmark89 is strongly discouraged, having been superseded by CINT92 and CFP92. (1994-11-29)
[“SPECOL – A Computer Enquiry Language for the Non-Programmer”, B.T. Smith, Computer J 11:121 (1968)].
- Spec rate
benchmark Results of the throughput measurement using SPEC benchmark suites CINT92 and CFP92. With the throughput measurement method, several copies of a given benchmark are executed. The method is particularly suitable for multiprocessor systems. The results, called SPEC rate, express how many jobs of a particular type (characterised by the individual benchmark) can be executed […]
- Spec ratio
benchmark Results for each individual benchmark of the SPEC benchmark suites, for example CINT92 and CFP92, expressed as the ratio of the wall clock time to execute one single copy of the benchmark, compared to a fixed “SPEC reference time”, which was chosen early-on as the execution time on a VAX 11/780. See also SPEC […]