This page lists validlab technical correctness/performance
reports available online, in most cases in the original tbl | troff -ms,
as well as in ASCII and postscript.
- Compiler Correctness and Performance on LAPACK - 96/10/23
SPARCcompilers 4.0 and 4.2 with maximum optimizations provide superior run-
time performance on LAPACK, compared to Apogee and FSF compilers, due to
greater robustness and optimization, but at the cost of greater compilation
times.
- UltraSPARC and SPARCcompilers 4.0 Improve Scientific Performance - 96/02/13
Over many publicly available scientific applications in Fortran-77 and C,
UltraSPARC systems provide significant performance improvements over
previous Sun systems, due to synergistic interaction of UltraSPARC hardware and
SPARCcompilers 4.0.
-
UCBTEST vs. Pentium - 95/03/05
How much testing is enough for a product that will be obsolete in two years?
In PDF.
- Effects of Pentium Division Flaw and its Software Workaround - 95/02/22
The infamous Intel Pentium floating-point division flaw is seldom visible
in the results of realistic technical applications, nor does it perceptibly
affect performance.
Intel and Cygnus published a recommended compiler workaround that reduces
the effect of the Pentium division flaw to at most one unit per division, in
the least significant bit of extended precision.
The workarounds in the compiler and libm avoid any severe effects
of the Pentium flaw, but sometimes cause harmlessly different results in realistic technical applications. The modifications degrade performance of flawed
CPU's by a median of 1%, and SPECfp92 ratios by about 9%.
- Searching for a Solaris Workstation - 94/03/07
Obtaining and configuring Solaris Unix on a system of commodity PC components
can be surprisingly complicated. Three possible approaches include:
-
Buy a collection of components, put it together, and try to get it to
work with Solaris.
-
Buy an integrated Windows 3.1 system and try to get it to work with
Solaris.
-
Buy an integrated Solaris system and try to get it to work.
I report my experiences with the latter two approaches, which sometimes
turn out much like the first approach.
I also report -g compilation time and -O execution time results for SPEC
and Perfect codes, comparing Solaris PC's to each other and to SPARCstations.
For another interesting point of view on the differences between PC's
and more expensive computers, see
Crash-Proof Computing
from Byte 4/98
-
Suggestions for SPEC 3.0 - 91/06/27
What should be changed in SPEC 3.0 [later known as SPEC 92]
besides adding new benchmark programs?
Visitor count for this page starting 26 Oct 97:
