Well if you want to do any tests...name them, and ill run them.
My specs are.
Dell Inspiron Notebook 5100 P4 2.66 512MB Ram 40 gig hdd + 200 gig external USB2
$ df -T -h
Filesystem Type Size Used Avail Use% Mounted on
/dev/hdc3 reiserfs 37G 30G 6.8G 82% /
/dev/hdc1 reiserfs 71M 48M 24M 67% /boot
none tmpfs 251M 0 251M 0% /dev/shm
/dev/scsi/host0/bus0/target0/lun0/part1
xfs 187G 77G 110G 42% /mnt/removable32MB Mobility Radeon 7000 (i think)
I'm not happy though...i had GLXGears running up at 1350 consistently...and i dont know what i have changed in my graphics options...but now i only get about 730-740, thats a huge decrease!
Luke
O Plameras wrote:
Just a matter of comparison I did these calculations:
Item LukeS PeterR Comparative Nos.
(1) (2) (3) (3)/(2)
CPU-Speed MHZ 2658 450 0.17
Mem MB 512 517 0.99
Real-Time sec 0.577 2.564 4.44
User-Time 0.031 0.172 5.54
Sys-Time 0.026 0.190 7.31
Analysis:
Data-
LukeS has CPU-speed that's faster than PeterR by a factor of 5.88 (100/17) and
both has about the same amount of Real-Memory (0.99).
Findings- LukeS has Real-Time exec that's 4.44 times faster; User-Time exec that's 5.54 times faster; and Sys-Time exec that's 7.31 faster than PeterR.
Reasonable results if we look at User-Time vs CPU-Speed. They are close for both and they scale proportionately.
But Real-Time is only 4.44 times faster for LukeS and does not scale relative
to CPU-speed (5.88). Should be around 5.88.
System-Time is 7.31 faster and does log'th scale (vs CPU-speed of 5.88 times)
which is excellent. Way above 5.88 norm.
Would be interesting to compare both results by running other Browsers and/or desktops.
O Plameras
Luke Skywalker wrote:
Out of interest i performed the same test, here are my results...
i have a quite a few extensions, if that makes a difference, but your results are impressive.
$ time firefox
No running windows found
Component returned failure code: 0x80470002 (NS_BASE_STREAM_CLOSED) [nsIFileOutputStream.write]Component returned failure code: 0x80470002 (NS_BASE_STREAM_CLOSED) [nsIFileOutputStream.write]Component returned failure code: 0x80470002 (NS_BASE_STREAM_CLOSED) [nsIFileOutputStream.write]
real 0m2.376s
user 0m1.869s
sys 0m0.115s
but if i already have one open
$ time firefox
real 0m0.577s user 0m0.031s sys 0m0.026s
$ cat /proc/cpuinfo
processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 2 model name : Intel(R) Pentium(R) 4 CPU 2.66GHz stepping : 7 cpu MHz : 2658.076 cache size : 512 KB
$ cat /proc/meminfo
MemTotal: 512832 kB MemFree: 103692 kB Buffers: 34620 kB Cached: 258480 kB
Im pretty happy with firefoxs performance...
When i had XP on my machine, it took the same time to start!
Im Using KDE (dont know if that makes any difference)
Luke
Peter Rundle wrote:
Firefox/Mozilla startup on Linux/Unix/etc is a dog's breakfast to put it lightly.
$ time /usr/lib/firefox/firefox
real 0m2.564s user 0m0.172s sys 0m0.190s
That is the time to open a window on my desktop that I can click on to then close firefox.
$ cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 5 model name : Pentium II (Deschutes) stepping : 2 cpu MHz : 450.035 cache size : 512 KB
Not exactly the fastest PC on the planet.
$ cat /proc/meminfo MemTotal: 517164 kB MemFree: 4904 kB Buffers: 79380 kB Cached: 233092 kB
Ok so it's chock full to the gills with RAM and the kernel has cached some 230Mb of the disk. Here endeth the lesson
P.
-- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
