Jump to...
Columns:
Advanced Software and Technologies
BYTE Media Lab
Chaos Manor
Conference Reports
Features
Free Features
Gigglebytes
Letters to BYTE.com
Mr. Computer Language Person
New Products
Op/Ed
Portable Computing
Serving with Linux
The Upgrade Advisor
April 1996
/
Letters
/ Benchmark Confusion
In
"Intel Beats the Clock...Again"
(News & Views, January), the SPECint benchmarks reveal that the 150-MHz PowerPC 604 processor is slower than the 167-MHz UltraSparc. However, in the review
"UltraFast UltraSparcs"
in the sa
me issue, the BYTEmark test reveals the contrary: A slower, 133-MHz PowerPC 604 beats the 167-MHz UltraSparc. How come?
Hendrik Verroken
hob@eunet.be
The SPECint benchmarks that we published in the News & Views section, including those from Sun and IBM Microelectronics, were estimates provided by the companies. Like BYTEmark results, SPECint results depend not only on hardware factors, such as caching and memory architecture, but also on the OS and compiler performance.
The Sun Ultra I that we reviewed, a preprodu
ction unit, was tested with a beta OS and compiler, which certainly influenced the BYTEmark scores. Unfortunately, we neglected to point this out in the review. We have arranged to retest a production system and will post updated BYTEmark results on the BYTE Web Site (
http://www.byte.com
). For more information, including Sun's latest BYTEmark results for the UltraSparc, see "Benchmark Update" in News & Views, this issue.--Eds.
Copyright
Matthew Wilson more...
BYTE Digest editors every month analyze and evaluate the best articles from Information Week , EE Times , Dr. Dobb's Journal , Network Computing , Sys Admin ,
and dozens of other CMP publications—bringing
you critical news and information about wireless communication,
computer security, software development, embedded systems,
and more!
Find out more
Benchmark Confusion
