[PC-BSD Testing] FF3.5.2 runs slower on PCBSD than Windows?
Jeff
dejamuse at yahoo.com
Thu Aug 27 12:40:35 PDT 2009
I had run Geekbench before, and here are the results (I had FF3 running in two windows with multiple tabs, Krusader, and a doc open in OO). Not sure how accurate the results are since it's running as a Linux process. Note it incorrectly reports the OS as 32 bit and shows RAM of 2GB when I actually have 6GB, and some other system number don't look right either.
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Linux 2.4.2
i686
Model: Linux PC (Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz)
Motherboard: Unknown Motherboard
Processor: Intel(R) Core(TM) i7 CPU 920 @
2.67GHz
Processor ID: Genuine Intel Family 6 Model 26 Stepping 5
Logical Processors: 8
Physical Processors:
1
Processor Frequency: 673 MHz
L1 Instruction Cache: 0.00
B
L1 Data Cache: 0.00 B
L2 Cache: 256
KB
L3 Cache: 0.00 B
Bus Frequency: 0.00
Hz
Memory: 1.98 GB
Memory Type:
N/A
SIMD: 1
BIOS:
N/A
Processor Model: Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz
Integer
Blowfish
single-threaded scalar 1336 |||||
multi-threaded scalar 5470 |||||||||||||||||||||
Text
Compress
single-threaded scalar 1399 |||||
multi-threaded scalar 7011 ||||||||||||||||||||||||||||
Text Decompress
single-threaded scalar 1411
|||||
multi-threaded scalar 7604 ||||||||||||||||||||||||||||||
Image Compress
single-threaded scalar 1154 ||||
multi-threaded scalar 3798 |||||||||||||||
Image Decompress
single-threaded scalar 1074 ||||
multi-threaded scalar 4753 |||||||||||||||||||
Crafty
Chess
single-threaded scalar 1849 |||||||
multi-threaded scalar 3544 ||||||||||||||
Lua
single-threaded scalar 2182 ||||||||
multi-threaded scalar 9980 |||||||||||||||||||||||||||||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 1468 |||||
multi-threaded scalar 5709 ||||||||||||||||||||||
Dot
Product
single-threaded scalar 2219 ||||||||
multi-threaded scalar 8042 ||||||||||||||||||||||||||||||||
single-threaded vector 1048 ||||
multi-threaded vector 4282 |||||||||||||||||
LU
Decomposition
single-threaded scalar 1127 ||||
multi-threaded scalar 3266 |||||||||||||
Primality
Test
single-threaded scalar 2420 |||||||||
multi-threaded scalar 10348 |||||||||||||||||||||||||||||||||||||||||
Sharpen
Image
single-threaded scalar 3965 |||||||||||||||
multi-threaded scalar 16323 |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Blur
Image
single-threaded scalar 3639 ||||||||||||||
multi-threaded
scalar 16021 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 2176 ||||||||
Write Sequential
single-threaded scalar 4717 ||||||||||||||||||
Stdlib Allocate
single-threaded scalar 1630 ||||||
Stdlib
Write
single-threaded scalar 3000 ||||||||||||
Stdlib Copy
single-threaded scalar 3139 ||||||||||||
Stream
Stream Copy
single-threaded scalar 3051 ||||||||||||
single-threaded vector 3625 ||||||||||||||
Stream
Scale
single-threaded scalar 2777 |||||||||||
single-threaded vector 3811 |||||||||||||||
Stream Add
single-threaded scalar 2584 ||||||||||
single-threaded vector 3544 ||||||||||||||
Stream Triad
single-threaded scalar 2708 ||||||||||
single-threaded
vector 2527 ||||||||||
Integer Score: 3754 |||||||||||||||
Floating Point Score: 5705 ||||||||||||||||||||||
Memory Score: 2932 |||||||||||
Stream Score: 3078 ||||||||||||
Overall Geekbench Score: 4204 ||||||||||||||||
---------------------
I closed down all other apps and reran the test and it was, uh... lower: 3400. So not sure what credibility it has.
The Geekbench browser reports an average score for this processor running on Windows of 7500. Is PCBSD 64 bit not taking advantage of the multi-core processor or is this because of the Linux compat layer?
Again, I don't think I'm comparing apples to apples here.
I suggested making some of these benchmark suites into PBIs, but nothing's happened yet. Also tried running the Phoronix test suite, but couldn't get it to run.
...Jeff
--- On Thu, 8/27/09, Brodey Dover <doverosx at gmail.com> wrote:
From: Brodey Dover <doverosx at gmail.com>
Subject: Re: [PC-BSD Testing] FF3.5.2 runs slower on PCBSD than Windows?
To: "PC-BSD Testing list" <testing at lists.pcbsd.org>
Date: Thursday, August 27, 2009, 2:57 PM
Have you tried installing geekbench with runports? I can provide some performance metrics to compare against my system:AMD Athlon X2 5000+ BE @ 2.9GHz4GB RAM2x 7200RPM 320GB WD RAID 0Asus M2N-E
GeForce 7600GTPC-BSD 7.1.1 - AMD64
On Thu, Aug 27, 2009 at 2:54 PM, Jeff <dejamuse at yahoo.com> wrote:
I built what should have been a kick-ass system with:
Intel i7-920
6GB RAM
RAID-0 7200RPM drives
Asus P6T Motherboard - (Intel on-board RAID)
Radeon HD-4250 VGA
PCBSD 64Bit
But in various benchmarks it comes up short every time.
I just ran this browser test and I scored 1296 pts using FF3.5.2 VS double that running on Windows ?? It's not clear what OS the average results came from, but I presume it was Vista.
My benchmark results breakdown:
Rendering - 1247
Social networking - 1833
Complex graphics - 3134
Data - 1714
DOM operations - 1264
Text parsing - 741
I did some other OS benchmarks too recently, and was similarly underwhelmed. Anyone have a clue why? Anyone else done any benchmarking?
This machine is sucking up 175 Watts in electricity and I don't appear to be getting my money's worth. Bumming...
_______________________________________________
Testing mailing list
Testing at lists.pcbsd.org
http://lists.pcbsd.org/mailman/listinfo/testing
-----Inline Attachment Follows-----
_______________________________________________
Testing mailing list
Testing at lists.pcbsd.org
http://lists.pcbsd.org/mailman/listinfo/testing
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.pcbsd.org/pipermail/testing/attachments/20090827/3020fbd1/attachment-0001.html
More information about the Testing
mailing list