Subject: bytebench comparison 1.4.1 vs. 1.5_ALPHA2
To: None <tech-kern@netbsd.org>
From: Hubert Feyrer <feyrer@rfhs8012.fh-regensburg.de>
List: tech-kern
Date: 09/01/2000 02:33:23
1.4.1: 	    http://www.feyrer.de/Misc/bytebench.NetBSD-i386-1.4.1.miyu
1.5_ALPHA2: http://www.feyrer.de/Misc/bytebench.NetBSD-i386-1.5_ALPHA2.miyu

Please ignore the increase in File writes, the 1.5 system has more buffer
cache than the 1.4.1 system has. The test is from
pkgsrc/benchmarks/bytebench, each set of tests was compiled using the
compiler distributed with that system.

That aside, it seems that 1.5 is ~10% slower than 1.4.1 in many areas. :(

Esp.:
1.4.1:		
System Call Overhead Test                 57000.9 lps   (10 secs, 6 samples)

1.5_ALPHA2:	
System Call Overhead Test                 50161.6 lps   (10 secs, 6 samples)

How does this come?


 - Hubert

-- 
NetBSD - because Unix isn't just #include <linux.h>, i386, ILP32, ELF, ...!