System benchmark

Hi, i was wondering if there is any program to perform some system speed testings.
I mean, back when i was using windows, i used to perform tests with some program called 3dmark or with everest, so you can test graphics acceleration, cpu speed, memory read and write speed, hd speed... and then compare it to have some idea about how "outdated" my system is.
Is there anything symilar for linux?
Thanks!

Hi,
Have you tried this one? :
http://www.archlinux.org/packages/commu … est-suite/

Similar Messages

[Athlon64] Noob needs help optimizing new system

Sorry to post yet another thread, but I've been lurking for the past few days and I only get bits and pieces of what I need. I'm pretty clueless about this stuff as it's my first home-built system
See my sig for system info...
I'm needing help with bios settings. The pertinent stuff (I think - hopefully I got it all down):
CELL MENU
CPU Clock speed: 2200
DDR Memory Freq: 100.0 MHz
High performance mode: Manual
Cool n quiet: Disable
HT Freq: 800 MHz
Dynamic overclocking: Disable
Adjust CPU Ratio: Auto
AGP Freq: By Default
HT Voltage: Auto
Mem voltage: Auto
CPU voltage: Auto
AGP Voltage: Auto
Spread sectrum: Disable
DRAM
DRAM Clock Mode: Manual
Mem clock value: DDR200
CAS Latency: SPD
Burst Length: 8 Beat
Bank Interleaving: Auto
Active to CMD (Trcd): SPD
Active to precharge (Tras): SPD
Precharge to active (Tvp): SPD
DRAM IT Timing: Disable
ADVANCED CHIPSET FEATURES
AGP Mode: Auto (this is "greyed out")
AGP Fast write: Enabled
AGP Aperture size: 128
VLink 8X supported: Enabled
My questions mainly concern memory. I bought all this stuff at the Tiger Direct outlet store in Naperville, Illinois. I had intended to buy the K8T Neo-FSR (754 pin), but they were supposedly out of the 754 pin 3400+, and this board/CPU combo had a $100 rebate. Yeah, it's one of those deals with the odd 939 pin 3400+ that everyone has been talking about.
In any case, I already had the Corsair CMX512-3200C2PT picked out, as Corsair said it was compatible with the board. I wasn't aware of the MSI compitibility chart until I came here later that night. I STILL can't find that chart off of the MSI web site.
Anyway, I got it all hooked up and got the standard 4 red LEDs. Not knowing what to do, I took it back to Tiger Direct and the tech played around with it until he stuck an el-cheapo stick (Corsair VS512MB400) in there and it worked. But it only works in slot 2. So here I am.
I started to run memtest86, but it took too long to run and I aborted it for now. For what it's worth, it said the RAM CAS value is 2.5-2-2-5. I don't know what the hell that means, but I know it's important.
My questions:
1. I assume I need manually to set my memory clock value to DDR400?
2. Now that I can get the thing to post, is there any chance I could get the "good" Corsair memory to work by setting the correct latency/CAS values? Again, I don't know what I'm talking about, so forgive my ignorance. What I'm wondering is if I can set everything to what Corsair recommends while I have the cheap memory in there and then replace it with the good stuff?
3. With this board, am I better off with double-sided or single-sided sticks? As I said, I went into the store intending to buy the Neo 754-pin board, but they were out of the Athlon 3400+ in 754 and only had the 939 pin.
4. Anything else in my bios settings that stand out and need to be addressed? I want to make sure everything is optimized.
EDIT: I almost forgot: My CPU temp is runing around 42-49 degrees. Am I in the danger zone? I just bought a generic cooler for now and intend to get a Thermalright XP-120, but I can't find it locally, so it will have to wait.
AND FINALLY...
I don't know if this is related to anything above, but I have been having a hell of a time getting the 6600GT to run DirectX9+ games. The latest drivers for Windows 98SE was 66.94. Yes, XP will be installed soon, but for now... It was failing on dxdiag test for Direct3d 9 and also the third DirectDraw test. I was about to give up and take it back when I discovered that a new beta driver came out today (71.84). I installed it and it passed all of the dxdiag tests. I thought I was in hawg-heaven, but when I run Call of Duty (the only DX9 game I have besides HL2), it slows to an absolute clawl. The level starts out somewhat normal at first, but quickly slows down to the point that performance can only be measured in seconds per frame! All of the patches 1.5 and 1.51 for UO have been installed, but that shouldn't make a bit of difference.
If I don't get my CoD fix soon I'm gonna reinstall that Geforce3.
Thanks much and sorry again for such a long post.

Thanks everyone for the recent replies, but I started this thread almost a month ago (March 3rd), and much has changed since then.
New PSU (see sig).
Replaced the PNY Geforce6600GT with a eVGA 6800
Added another stick of the Corsair Value Select 512 RAM. I didn't even bother trying to get the "primo" CMX stick to work. I just returned it.
Went from Windows 98SE to XP Home.
The PSU did nothing for me at all, but I suppose it's better to have some headroom in the amperage.
The biggest performance booster was the additional 512MB of RAM. It wouldn't have done me any good in 98SE, but I was shocked at how much it helps XP. Quitting a game is almost instant now, where before it would take a good two minutes or so for XP to recover itself.
As for games running slow, the 6800 fixed it right up. CoD absolutely screams now, as does everything else I've thrown at it (HL2, Doom 3, Far Cry). I think the 6600GT was just bad or there is a serious driver problem with the 6600GT's. Went from 3DMark03 score of around 5000 (when it didn't crash) to above 9000 now. And this was all before adding the additional RAM.
Oh yeah, I bought a 74GB Raptor. No RAID for now though. So far I'm not real impressed with the performance. I'm running off of the VIA controller, since my board doesn't have the Promise controller. I don't see much improvement in loading times as compared to the 40GB 7200rpm Maxtor I have running on IDE1. The only time it beats the pants off the Maxtor is defragging. But there's only so much entertainment value in that. The Sandra "file system benchmark" gave me 57 MB/s, which seems pathetic, but I've read some posts here where others are getting an identical rate.
And I'm not too happy to discover, too late, that my overclocking capabilities are severely limited while using SATA.
I do have one more question for now:
When I installed the second 512MB stick, I originally had both sticks in slot 1 and 2. I ran Sandra's memory bandwidth benchmark and got a much lower score than with the single stick (around 3500 MB/s). WHen I had the sticks in 1 and 3, it jumped up to close to 6000 MB/s. Why is that? Just curious.

KT3 Ultra ARU Slow HD Transfer Rates

Hello,
First, please see my system information at the bottom of this message.
I have a MSI KT3 Ultra ARU motherboard and have installed a Maxtor ATA133 60GB hard drive with 2 MB cache. The hard drive is jumpered as Master and is connected all by itself on the IDE1 ribbon cable. On IDE2 I have a CDRW (Master) and a CD-ROM (Slave). My other hard drive (WD 120GB ATA100 8 MB cache) is on IDE3 and is set up in RAID 1+0 stripe.
I recently ran the Sandra file system benchmark on my hard drives with what I think are very discouraging results. My Maxtor drive on IDE1 scored 32 MB/s and my WD drive on IDE3 came in at 23 MB/s. Those are some stinky results aren’t they?
I checked my Device Manager under IDE ATA/ATAPI Controllers and it indicates that for my Primary IDE Channel the current transfer mode for device 0 is Ultra DMA Mode 6. I don’t know how to check the transfer mode on my IDE3 hard drive.
Question #1: Are my 32 MB/s and 23 MB/s results slow for my hardware?
Question #2: If these results are slow, how can I increase the transfer rates?
Please help!
Thank you!!!!
My System
AMD Athlon XP 3000+
MSI KT3 Ultra ARU MOBO
2 x 256MB DDR PC2700 SDRAM
ATI Radeon 9600XT 256MB
Western Digital WD1200JB 120GB Caviar HD ATA100 8MB Cache 1+0 Stripe/RAID0
Maxtor 6Y060L0 60GB HD ATA133
Pioneer DR-A04S CD-ROM 32X
Lite-On LTR-48125W 48x12x48 CDRW
TTGI TT-350SS (350W) Power Supply
PSU Ratings: +3.3V/28A +5V/35A +12V/16A
Windows XP Home SP2

These tranfer rates seem fine for sequential read rates. The maxtor drive is at mode 6 and the WD should be at mode 5 if its ATA100.

W520 power supply rating vs. performance question

I have had a W520 with the i2820 processor option and the quadro 1000 video for over 3 weeks now. I bought several spare 170 watt power bricks for it, but in one location I kept the 135 watt brick from my W510.
When I moved from the w500 to the w510, when I used a 90 watt w500 brick, the w510 would boot with a BIOS level message that said that the power supply was not adequate and that the machine would be stepped down to a slower speed. The reduction in performance was dramatic and noticable. (The w510 had the 45 watt consumption i920 chip and the more power hungry graphics)
When I use or boot the w520 with the 135 watt supply, there is no message. My benchmarks do not change from those registered when on the 170 watt power supply, starting with the Win 7 system benchmarks for processor and video. Further, the 135 watt brick is cooler to the touch than the 170 watt one under sustained usage (3 hours plus).
The question: given that the w520 in my configuration seems to draw less power than the hot and hungry w510, is the 135 watt supply going to work "as well as" the 170 watt one? As a briefcase lugger, any way to shed weight is valued.
Other considerations: I have no WWAN card installed, and have the CD/DVD removed and substituted with an SSD. The main drive is also SSD, bringing down somewhat the power requirements.
David Gleason. Every T from the T20 to the T61p, W500, W510. Current system: W520 4270-CTO i2820 8GBx4 (corsair), FHD(1920x1080), NVIDIA Quadro 1000M and 2 Samsung 830 512gb SATA III SSD.

I remember another post, that I can't seem to find now, which refers to a software utility that displays wattage being consumed by the system. This would be quite helpful.
Couldn't that be used to study what is being used, and how much of a power adapter is needed to operate the system?
I am trying to come up with a solution for vehicle usage, and I am afraid, it will have to be a 12V to 120v power inverter with a brick plugged into it.
Another factor will be the amount of power consumed when recharging the internal 9cell battery and the external 9cell slice. Doing this while operating the system with WiFi / WWAN active will increase the demand.
= = = = = = = = = = = = = = = = = = = = = = = =
W520 4276-2QU i7-2720QM quad core, 4GBx1, FHD(1920x1080), NVIDIA Quadro 2000M, 500GB-7200, 6205WiFi, 720pCamera, Bluetooth, vPro, 57++9cell, 27++9cellSlice, Win7Pro64
IBM R50 WinXP for backup.

Speed up time of AI Create Channel.vi task containing thousands of channels

I am creating an AI task containing a total of 8000 channels of different channels with each channel being acquired many times before the next channel is acquired. The channel list is something like 0,0,0,...,1,1,1,...,2,2,2...3,3,3... where each channel number would be acquired 2000 times. The purpose is to perform averaging. I am seeing that DAQmx Create Channel.vi takes about 100 msec to create 1000 channels for a total of 800 msec to create all channels (DAQmx 9.4, LV 11, WIN 7, CoreDuo P8700 2.5 GHz). I would like to speed this up by a factor of 10 if possible. Test vi attached.
Attachments:
Create Multiple Channels duration.vi ‏17 KB

Hi SteveP,
The DAQmx task contains data structures representing channels, which in turn contain other data structures representing channel properties and such. Most of the execution time of DAQmx Create Channel is spent allocating and initializing these data structures. They are separate from the task's input buffer (where data is stored until DAQmx Read is called). The input buffer is allocated when you reserve the task (which is normally done by DAQmx Start Task), and it is indeed allocated as a single large array.
DAQmx support for repeated channels focused on the SCXI use case, where hundreds of milliseconds of channel creation time seemed to be an acceptable tradeoff for measuring thousands of thermocouples. During development of NI-DAQmx 7.0, we used 3072 channels as our large system benchmark for channel creation, since that's the maximum number of channels supported by an SCXI task: 8 chassis x 12 modules x 32 channels (ignoring CJC channels). Repeated channels also make your use case possible, but this wasn't the original goal, otherwise we would have tried to make it easier to use for this purpose.
When replying to your previous thread on this topic (DAQmx AI Task with many samples CH0, then many CH2, then many CH3), I assumed that you were doing this because you were having problems when you took the straightforward approach to averaging, but Ben is right to point out that there are ways to deal with ghosting, cross talk, etc. What is the source impedance of the signals that you are measuring? How Do I Eliminate Ghosting From My Measurements? lists some options for dealing with high source impedance.
Failing that, is it possible for you to reduce the frequency at which your program creates and destroys tasks? If you can extend the task lifetime by reusing the same task more than once, you can avoid the channel creation overhead, or at least amortize it over the lifetime of the task. Note that you don't have to clear one task before you create a second one; you only have to make sure that two tasks don't have the same hardware reserved at the same time--if you do, you will get a reservation error. In most cases, stopping the first task before starting the second is sufficient to avoid this. It is possible to create multiple tasks up front, then switch between them throughout the rest of your program.
Brad
Brad Keryan
NI R&D

Need some techical information about Satellite P100-10P

Hello,
I own a Satellite P100-10P laptop.
My question is:
How much gigaflops can this laptop produce (not in normal usage, but technically)?

Hi
First of all you will not find here any Toshiba technicians but simply, common Toshiba notebooks users. This is a user-to-user forum!
Regarding your question;
Why you dont use a benchmark application to measure floating-point performance.
Maybe a LINPACK 64bit benchmark application could helps. I write could helps because I never used and tested this program. ;)
http://linux.softpedia.com/get/System/Benchmarks/High-Performance-Linpack-1396.shtml
But this program was written for Linux OS. It looks like you will be not able to run it on Win OS.

BIG PSD and slow Progress Bar!

To open a 200meg PSD file in CS3 takes about 20 secs, I get this progress bar come up saying Reading Photoshop Format... is this normal?
I have a fast P4 processor, 2 gigs of Ram and large separate hardrive I'm using as the scrtch disk.
I am curious becuase my friend has a less efficient system running CS3 and he gets no progress bar or delay when opening this same image or any large image for that matter?
Thanks

On Tue, 29 Apr 2008 00:25:36 -0700, [email protected] wrote:
> To open a 200meg PSD file in CS3 takes about 20 secs, I get this progress bar come up saying Reading Photoshop Format... is this normal?
> I have a fast P4 processor, 2 gigs of Ram and large separate hardrive I'm using as the scrtch disk.
> I am curious becuase my friend has a less efficient system running CS3 and he gets no progress bar or delay when opening this same image or any large image for that matter?
> Thanks
The times you report show that you may be getting only 10MB / second, which
is not fast by today's standards.
The first thing I would check is your disk throughput, using a program such
as DiskTT:
http://www.softpedia.com/get/System/Benchmarks/Disk-Throughput-Tester.shtml
When using DiskTT, I generally save time with a buffer size of 50 megs, and
use the write times only when I want to make a quick comparison. The
larger default buffer size will give a more accurate reading, particularly
for read times.
Test your internal drive, as well as the external one. If your internal
drive is faster, you may see better performance if you use it as your
scratch drive and save your images on the the external drive.
BTW, I'm not questioning the other person's individual experience, but IMHO
it's very unlikely that this has anything to do with your default printer
setting.
Mike Russell - http://www.curvemeister.com

Update requires a valid UpdateCommand when passed DataRow collection with modified rows.

It is my
understanding that when using a TableAdapter and selecting the proper
options, Insert, Delete and Update statements are generated. The
Update statement should work just using the following:
myTableAdapter.Update(changes);
Where changes is the DataSet containing the modified data.
This logic works great for one of the tables in the DB but not for another.
Learning this new stuff is Confusing at best - Frustrating at worst.
Help appreciated,
Michael

ISiM to organize COMAD 2009 - the 15th International Conference - Call for papers
Dear All,
The 15th COMAD is organized by the International School of Information Management (ISiM), University of Mysore during December 9-11, 2009 at Infosys Technologies campus, Mysore. Similar to previous years, COMAD   2009’s scope  will include all   areas in the data management space including database management systems, Web and Information Retrieval and Data Mining. For details: http://www.isim.ac.in/comad2009
For close to two decades COMAD – the International Conference on Management of Data, modeled along the lines of ACM SIGMOD, has been a premier international database conference hosted in India. The first COMAD was held in 1989, and it has been held on a nearly annual basis since then. COMAD always had significant international participation with about 30% of the papers from outside India including Europe, USA, and East/South East Asia.
Call for Papers
Original research submissions are invited not only in traditional database areas but also in Data Quality, Web, Information Retrieval and Data Mining.
We invite submission of original research   contributions as well as proposals   for demonstrations, tutorials,   industrial presentations, and panels.
Areas of interest include but are not limited to:
Data Management Systems:
* Benchmarking and performance evaluation
* Data exchange and integration
* Database monitoring and tuning
* Data privacy and security
* Data quality, cleaning and lineage
* Data warehousing
* Managing uncertain, imprecise and inconsistent information
* Multilingual data management
* Novel Data Types
* Parallel and distributed databases
* Peer-to-peer data management
* Personalized information systems
* Storage and transaction management
Web and Information Retrieval:
* Categorization, Clustering, and Filtering
* Document Representation and Content Analysis
* Information Extraction and Summarization
* IR Theory, Platform, Evaluation
* Question Answering and Cross-Language IR
* Web and IR
* Social Network Analysis
Data Mining
* Novel data mining algorithms and foundations
* Innovative applications of data mining
* Data mining and KDD systems and frameworks
* Mining data streams and sensor data
* Mining multi-media, graph, spatio-temporal and semi-structured data
* Security, privacy, and adversarial data mining
* High performance and parallel/distributed data mining
* Mining tera-/peta-scale data
* Visual data mining and data visualization
To ensure wide visibility of material published at the conference, we plan
to make arrangements with ACM SIGMOD for including the proceedings of the conference in the SIGMOD on-line and CD-ROM archives. Two awards- for Best Paper and Best Student Paper, will be presented at the conference.
Important Dates:
·         Research and Student Papers-July 31, 2009
·         Industrial Papers, Demo, Panel and Tutorial Proposals-August 28, 2009
·         Notification to authors-September 25th, 2009
·         Camera-ready copy due-October 30, 2009
·         Early-Bird Registration Deadline-December 1, 2009
·         Conference-December 9-11, 2009
Conference Organization:
General Chair
·         Shalini Urs, ISiM, University of Mysore, India
Organizational Chair
·         Srinath Srinivasa, IIIT-Bangalore, India
Program Chairs
·         Sanjay Chawla, University of Sydney , Australia
·         Kamal Karlapalem, IIIT-Hyderabad, India.
Contact Details:
Dr. Shalini Urs
COMAD 2009 – General Chair
Executive Director and Professor
International School of Information Management
University of Mysore, Manasagangotri
Mysore – 570 006
Tel: +91-821-2514699; +91-821-2411417
Fax: +91-821-2519209,
Email: [email protected]
Regards

Promise RAID vs. Via RAID

When I was setting up my system I wasn't really sure which RAID controller I should use in order to get the best performance...I opted to go with the Via controller as it is an integrated part of the KT800 chipset whereas the Promise RAID controller is an optional add-in chip (or such is my understanding), so I figured using the Via controller would give the best performance...but I'm curious as to how the Promise controller performs. Has anyone tried setting an array up both ways and testing to see which controller performed better?
I'm running two 36 GB WD Raptor SATA drives in RAID-0 off the Via controller, and I get a score of 61 MB/sec in Sandra 2004's File System benchmark...if anyone is running a similar array off the Promise controller and could post their benchmark results, I'd appreciate it.

I've 'tested' both controllers.
Before the K8T, I was using a KT4 ultra with a Promise 376 and two raptors RAID0 (around 60-70Mb/s with HDTach)
Then I switched to the Promise 378 of the K8T (I keep promise to keep data). A little bit better than with the 376 of the KT4V but not very clear. I recently put the two raptors on the VIA. A little bit better but again, not very clear. It seems however that with the VIA the output is more 'unstable' .. I mean: with Promise, HD Tach gives me a straight line from the begining of the disk array to the end; with VIA, it's less straight.
I think I will crash my RAID 0 to test with one raptor on the VIA and one on the Promise. OSsystem for a disk, software and apps for the other one. I'm not sure that it's a goo idea

Stuck at startup in FCPX

When i last worked at my project I accidentally quit the program (not a force quit). I think it happened during a background transcoding, but I'm not sure. All I know is that the shutdown of FCPX tok long time.
Now when I try to open it, it stops at the loading screen with the message -Updating "Discus" event. All goes well until it gets to loading the timeline i was working on when I made the accidental quit. Then it just hoggs all processor power and locks at -Updating "Discus" event.
Have i lost all my data? Or can i "just" erase the chapter that it gets stuck on?
Best regards
Joakim

Mark - I had the same problem when I upgraded my system to a 6600GT recently. I had a Radeon 9600 and the BIOS set on TURBO, got good scores, stable system. When I installed the 6600GT, it locked up when it tried to find my disk drives. Only clearing the BIOS would let me continue. I now have all the settings back the way I want them - except TURBO. The system benchmarks significantly slower in memory throughput now - that is, the Memory Bandwidth Benchmark (SiSoft Sandra 2005 SP1). My 3dMark2005 went through the roof so it's a decent trade-off, but I want my PC to be all it can be. How do I get my TURBO setting back?!?
Madoc Owain
http://www.madocowain.com
http://www.playmaille.com

IPhoto Performance After Leopard

Has anyone noticed any reduced speed (editing, comparing, etc.) in iPhoto '08 after installing Leopard? I'm interested in getting the new OS, but am always wary of the slow downs that can occur with an upgrade. iPhoto is one of the more intensive app's I use, so I thought that would be a good check. Web review seem to suggest that Leopard is quite efficient, but I thought I'd check for the real user reactions.

I had a contrary experience, which I attribute to running on an older G4 chip. Any graphics-intensive work was noticeably slower in Leopard, particularly iPhoto and Quicktime. System benchmarks bore this out; my machine is 17% faster running Tiger than Leopard.
That said, I think that this is entirely due optimizations made for the Intel chips in Leopard, and that it does seem to be faster for everyone on more recent Macs.

KT4 Ultra "RAID?" controller

The S-ATA and ATA133 Promise controller on KT4 Ultra
1)Is it a RAID or an ATA (and S-ATA) controller?
2)Does it have RAID 0 performance level with ATA133 HDDs, same as that of the KT3 Ultra 2 RAID Controller, or it is just a RAID for S-ATA HDDs?

Good, that's what I wanted to hear actually. While I never admired the RAID controllers I got comfused when I say the FILE SYSTEM BENCHMARK in Sandra 2003 saying that RAID level 0 is a lot faster than NON-RAID, givving a bandwitch of about 45Mbyte/sec to 8MB cache 80GB HDD in NTFS WINXP, compared to quite less without the use of RAID. In my NTFS WINXP with NO RAID it counted 30MByte/sec on the 80GB 8MB cache WD HDD. The disk transfer rate that I did count in pure format was counted at 45Mbyte/sec

MSI Super Raid (Raid 0)

Hi I've got two sandisk-x110-128gb-mlc-sata-iii-sd6sf1m ssd disk in my MSI GE70 Apache Pro I have this disks in Raid 0= MSI Super Raid but i have only read about 650 and write about 500 I ask in sandisk forum and they told me to upgrade firmware for this ssd I tried sandisk software for upgrade but software don't recognize ssd..also they told me these ssd is OEM manufactured..and MSI provide update of firmware.. How can I update firmware on these ssd? For better performance :-)

Quote
Hi I've got two sandisk-x110-128gb-mlc-sata-iii-sd6sf1m ssd disk in my MSI GE70 Apache Pro I have this disks in Raid 0= MSI Super Raid but i have only read about 650 and write about 500
Test it with ATTO: http://www.softpedia.com/get/System/Benchmarks/ATTO-Disk-Benchmark.shtml
and post bench results

Help! Slow SATA Performance (865PE Neo2-FIS2R + Raptor)

As stated in my subject.. the performance of my WD SATA Raptor seems very under par, especially when compared to my Maxtor ATA133 HD.
The Raptor is connected to the Intel SATA connector. (During boot-up, it's listed as UDMA-5???)
I have 2 DVD drives hooked up to IDE1, and a CD drive hooked up to IDE2
My Maxtor is hooked up to the Promise ATA133 (during boot-up, it's listed as UDMA-6!!).
I'm running Bios 1.5, with the Bios set to:
Native, S-ATA Only, Keep P-ATA Enabled (Yes, Both), and SATA as Raid (No).
Promise Controller is set to As SATA (I found it odd that only settings available are Disabled, SATA, or RAID. Since I'm not running a Raid but I AM using the controller for the Maxtor, I set it to SATA)
I have tried to install the Intel Accelerator but it says uncompatible hardware (I'm assuming because I'm not technically running a RAID setup on the chipset... so it just uses the standard XP Professional driver). I have also installed the latest Intel INF.
Benchmarking my system, I'm using the SiSoftware File System benchmark. I don't understand how the Raptor could do so poorly, especially the buffered Read/Write??
Here are the results ;(
WD SATA Raptor
Overall Score 25646 kB/s
Buffered Read 97 MB/s
Seq. Read 36 MB/s
Random Read 10 MB/s
Buffered Write 72 MB/s
Seq. Write 33 MB/s
Random Write 12 MB/s
Access Time 5ms
Maxtor ATA133
Overall Score 27589 kB/s
Buffered Read 872 MB/s
Seq. Read 42 MB/s
Random Read 8 MB/s
Buffered Write 867 MB/s
Seq. Write 25 MB/s
Random Write 12 MB/s
Access Time 6ms

Quote
Originally posted by jaeger66
You can't use Sanda to bench hard drives. It doesn't give anything close to relevant numbers. And why are you using both the Promise and Intel controllers with only 2 drives?
What software should I use to bench HD? Even if Sanda shouldn't be used... why would the Maxtor score higher?
The only reason why I'm using the Promise controller is because (from what I can tell from the MSI manual) only the Promise IDE connector supports ATA133, while the Intel IDE connectors only support ATA100.
Intel IDE 1 (Master, Slave) = DVD RW, DVD
Intel IDE 2 (Master) = CD-RW
Intel SATA (Boot/OS) = WD Raptor
Promise IDE 3 (Master/Data files) = Maxtor

Upgraded to BIOS V1.9 and lost 15% HD performance

Hi.
I upgraded my 865PE NEO2-S BIOS from V1.5 to V1.9 for enhancements in hyperthreading code & improved stability and using Sandra Sisoft have found:
small (2-6%) improvement in CPU benchmark (ALU & iSSE2 marks especially)
0.5% decrease in memory bandwidth
major 15%! decrease in C:\ hard drive file system benchmark
(from 31270 legacy/ 31438 native SATA mode to 26346 legacy/ 26399 native SATA)
This was after I had reset the new BIOS settings to match my original settings.
While the small changes in CPU & memory benchmarks may be within the margins of error, the large drop in HDD performance is very clearly real.
In the BIOS CMOS the disk is configured as AUTO. There is no setting to check if 32-bit/ block transfer is enabled. In WindowsXP device manager the IDE controller properties are:
primary channel device 0 - currently using DMA mode 2 (my CD-RW I think)
secondary channel device 0 - currently using DMA mode 5 (my SATA HD)
Does anyone know why my HDD performance has decreased with the V1.9 BIOS... has anybody else had this. Is this a function of the 'increased stability'? Are there any changes I can make to improve my HDD performance back to the level I was getting with V1.5.. apart from going backwards to V1.5?
thanks for your help
system specs below
Nothing overclocked. Performance mode turbo.
Sanata
I have tried using both legacy or native IDE modes (my HDD is a Seagate SATA 80Mb 7200rpm 8Mb cache)

For what it's worth (perhaps not too much!), I've seen significantly different results consistently reported by the Sandra File System benchmark for identical Seagate 160GB SATA drives. #1 was about 25% full and Sandra consistently reported around 31000. #2 was completely empty and Sandra consistently reported around 36000. Now, with both drives in a RAID 1 array, Sandra is reporting around 29000 for the combination. All of this was measured under BIOS 1.8.
I suspect some detailed analysis might show that the Sandra file system benchmark is sensitive to the amount and placement of data recorded on the drive.

System benchmark

Similar Messages

Maybe you are looking for