|
| Notices |
DriverHeaven is currently recruiting for the AWOMO Beta Test / Elite Op Team. AWOMO is a digital download service for games, and we're looking to expand the beta team. If you're interested. Sign up as a member here at DriverHeaven and then head HERE to submit your details. Thanks
For more info on AWOMO visit their site HERE
Welcome to the DriverHeaven.net forums. You are currently viewing our boards as a guest which gives you limited access to view most discussions and access our other features. By joining our free community you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is fast, simple and absolutely free so please, join our community today! If you have any problems with the registration process or your account login, please contact contact us. |
 |
|
Apr 4, 2006, 11:08 PM
|
#1
|
|
DH's Dormant Dragon
Join Date: May 2002
Location: IN Rem-Dormancy
Posts: 24,210
|
Post your HDTach 3 Results!
You know the name of the game, lets see what you've got.
The program is DAMN small and spywareless... so no worries. 
Just got HERE~> http://www.simplisoftware.com/Public...request=HdTach
Click the Download button and download the ity bitty program and run the LONG test (32mb).
Take a snapshot and post it. Use a site like www.imageshack.us for easy quick uploading and posting.
And please, Post JPEGs, bmps are bloated and unnessary.
I'll start.... someone please beat me!
 |
|
|
|
Apr 4, 2006, 11:17 PM
|
#2
|
|
Elite Motherf#$%er
Join Date: Jan 2004
Location: Sacramento, CA
Posts: 2,833
Rep Power: 0
|
Oh man Judas, I totally pwned yours  :
 |
|
|
|
|
Apr 5, 2006, 03:32 AM
|
#3
|
|
DriverHeaven Extreme Member
Join Date: Oct 2003
Location: Gefle, Sweden
Posts: 3,226
|
An old test of my striped Samsung P80's. Will separate them when I switch motherboard as the benefit is not so great after all. Then again the efficiency might be slightly lower because of the aged VIA raid chip on this Socket-A mobo.
Last edited by mkk; Apr 5, 2006 at 03:47 AM.
|
|
|
|
|
Apr 5, 2006, 10:57 AM
|
#4
|
|
DH's Dormant Dragon
Join Date: May 2002
Location: IN Rem-Dormancy
Posts: 24,210
|
hey... nothing wrong with nearly 100MB/s sustained read..... that's roughly twice as fast as my samsung 80gb sata 2 drive here.....
|
|
|
|
|
Apr 5, 2006, 11:56 PM
|
#5
|
|
Roxy Music
Join Date: Jul 2002
Posts: 928
Rep Power: 42
|
The drives.
* Seagate Barracuda 7200.9 ST3300622AS 300GB 7200 RPM SATA 3.0Gb/s x 2 (total 600GB) RAID-0
* Seagate Barracuda 7200.9 ST3160812A 160GB 7200 RPM IDE Ultra ATA100
 |
|
|
|
May 6, 2006, 09:22 PM
|
#6
|
|
Member
Join Date: Mar 2003
Posts: 5,989
|
Judas, do you have ULi SATA RAID results on the RAID0 with 2 Maxtor hard drives?
i'm now running a repair installation for adding RAID drivers on one system, i set up Windows using the same image that i use with other system that uses IDE mode on SATA controller, should be able to run the HDTach in about 20 mins. so i like to see your result of a 2 drives RAID0 setup, and see if i can beat you?
|
|
|
|
|
May 6, 2006, 10:03 PM
|
#7
|
|
DriverHeaven Extreme Member
Join Date: Dec 2002
Location: U.S.A.
Posts: 16,122
Rep Power: 0
|
Heres's mine 2X hitchi deskstar T7K250 250GB SATA2
Quote:
|
Originally Posted by MythicaL
Oh man Judas, I totally pwned yours : |
WTF?? you've somthing running in the background, or?
or useing a strage stripe size? the controler funny?
wierd...... any one explain that one? I always see a basic curve
Raid or no raid....
Quote:
|
Originally Posted by mkk
An old test of my striped Samsung P80's. Will separate them when I switch
motherboard as the benefit is not so great after all. Then again the efficiency might be slightly lower because of the aged VIA raid chip on this Socket-A mobo.
|
I'd suspct the controller your useing is "maybe" lacking.... strage you almost flat out
yet the curve is there in the back ground
what stripe sizes you all useing? Wierd results....
Quote:
|
Originally Posted by PangingJr
Judas, do you have ULi SATA RAID results on the RAID0 with 2 Maxtor hard drives?
i'm now running a repair installation for adding RAID drivers on one system, i set up Windows using the same image that i use with other system that uses IDE mode on SATA controller, should be able to run the HDTach in about 20 mins. so i like to see your result of a 2 drives RAID0 setup, and see if i can beat you?
|
well, hurry up! i'm going to bed soon...
Last edited by The_Neon_Cowboy; May 7, 2006 at 06:10 AM.
|
|
|
|
|
May 6, 2006, 10:23 PM
|
#8
|
|
DriverHeaven Junior Member
Join Date: Dec 2004
Location: Nottingham, UK
Posts: 44
Rep Power: 0
|
Here's mine... 
The drives are 2 Samsung Spinpoint SP2004C drives in R0 through an Intel ICH7R.
 |
|
|
|
|
May 6, 2006, 10:50 PM
|
#9
|
|
Member
Join Date: Mar 2003
Posts: 5,989
|
Quote:
|
Originally Posted by The_Neon_Cowboy
well, hurry up! i'm going to bed soon...
|
lol neon go to bed i don't want to beat you i thought JD still around...
about your comments on how the HDTach's curve...
the curve of sequentail read speed on the HDTach window just came out like that on some systems, just as long as the average read's readouts still decent (not too low), i think the RAID setups are okay...
Last edited by PangingJr; May 6, 2006 at 11:12 PM.
|
|
|
|
|
May 7, 2006, 04:14 AM
|
#10
|
|
Member
Join Date: Mar 2003
Posts: 5,989
|
Seagate ST3808110AS RAID0 (2 hard drives),
2 RAID volumes were created during the firmware RAID set up.
real-time AV scan is on, all Windows and other services (except Remote Registry, System Restore, Automatic Updates, Windows Time and UPnP Device Host) are using default settings.
Two Windows OSes (XP Pro and XP x64 Pro) were installed by using a partition images restoring in mode DOS (Paragon Drive Backup) followed by Windows repair installation to add RAID drivers and then after that the Windows's have been reupdated. and then create a new backup image (Paragon Drive Backup Server edition). some of these partition image will go to my friend who asks me to create them, he also wants me to test the server edition of Paragon Drive Backup program.
anyway, i'll probably check with Seagate to see if they have any suggestion or can help improve the RAID performance, they might be able to help improving those curved shapes abit. |
|
|
|
|
May 8, 2006, 07:00 AM
|
#11
|
|
Member
Join Date: Mar 2003
Posts: 5,989
|
3xSeagate ST3808110AS firmware 3.AAE (P/N xxxx-303) in RAID0, 223.6 GB.
Vol_0 size = 32 GB (2 primary partitions for 2 Windows XP's),
Vol_1 size = 191.6 GB.
> 
HD Tach... for Vol_0...
> 
HD Tach... for Vol_1...
> 
and here is HD Tach... for 1x Seagate ST3808110AS 80 GB (firmware 2AAA P/N xxxx-301) under SATA controller in IDE mode.
> 
|
|
|
|
|
May 8, 2006, 02:31 PM
|
#12
|
|
DH's Dormant Dragon
Join Date: May 2002
Location: IN Rem-Dormancy
Posts: 24,210
|
i think i've got 2 of these maxtors in raid 0 and 3 in raid 0 as well ....somewhere.... benchmarks i mean..
but HDTach i've noticed so far favors lower blocks sizes
However i'm a bit confused on how your setting up the volumes and getting HDTach to see it like that.
If they are indeed raid..... should it combine it into a single large hardrive....
When i load up HDTach, it sees only one available drive.
Course i miss my HighPoint Tech controller, with IDE ATA100 drives, i was getting nearly 1/2 of what i'm getting now... and they were 2mb cache 7200rpm.. generic maxtor (only 2 of them)....
|
|
|
|
|
May 8, 2006, 03:01 PM
|
#13
|
|
Member
Join Date: Mar 2003
Posts: 5,989
|
that's the "Intel Matrix RAID", under ICH7R, RAID0 and RAID1 or RAID5 (Parity) can be combined. suppose to be RAID10 also...
about the RAID subsystem, in some cases... it has to do with hard drives's firmware, not just the controller.
if you want to have the best SATA RAID preformance you might want to consult and work with the hard drives tach supports.
|
|
|
|
|
May 8, 2006, 03:46 PM
|
#14
|
|
Member
Join Date: Mar 2003
Posts: 5,989
|
Judas, have a look here... http://forums.storagereview.net/inde...35&hl=firmware
the link is not working atm, but check it out later. you would see that hard drive's firmware help correcting a 4x 500GB RAID5 performance problem.
anyway, i think i beat your results a little already. you might think i did cheat you a little, since the RAID0 volume size was only 32 GB?
Quote:
|
Originally Posted by Judas
|
|
|
|
|
|
May 8, 2006, 06:26 PM
|
#15
|
|
DriverHeaven Lover
Join Date: Oct 2005
Location: Land of Snow
Posts: 154
Rep Power: 0
|
so what strip size are u guys with raid-0 using, im using 16k and obtaining 105mb average read, seagate 7200.8 x2 raid-0 (too lazy to post screen hehe)
I do believe strip size makes a difference in hdtach
|
|
|
|
|
May 8, 2006, 11:07 PM
|
#16
|
|
Member
Join Date: Mar 2003
Posts: 5,989
|
i've noticed that too. 16K strip size gives best benchmark results,
in many cases, 16K is good for sequential transfer under real-world performance.
the benchmark results can also give indication of real-world performance.
how about when you try a larger strip size like 128K, the benchmark results came out very poor,
so what about the real-world performance?
|
|
|
|
|
May 9, 2006, 12:10 AM
|
#17
|
|
watching 1080i
Join Date: Nov 2002
Location: April 13th 2029
Posts: 19,435
Rep Power: 75
|
good ole ATA 100...
160GB 8mb cache...
EDit- oops here's the long test..
old 40BG 2mb cache...
Still fast enough for me. 
Last edited by BWX; May 9, 2006 at 12:19 AM.
|
|
|
|
|
May 9, 2006, 04:20 AM
|
#18
|
|
Member
Join Date: Mar 2003
Posts: 5,989
|
here's a better HD Tach benchmark results for 3xSeagate ST3808110AS RAID0, 223.6 GB
Vol_0 = 32 GB, strip size = 16 KB
Vol_1 = 191.6 GB, , strip size = 16 KB
i took about 10 minutes to delete/rebuild RAID array using RAID BIOS utility, and restore a Windows partition via mode DOS...
Judas, i like to see the results of your 2x or 3x, can you find them? |
|
|
|
|
May 9, 2006, 09:10 AM
|
#19
|
|
DriverHeaven Lover
Join Date: Oct 2005
Location: Land of Snow
Posts: 154
Rep Power: 0
|
raid 16k makes a difference when i load games and levels. Though i havent tried 64k (which most recommend) to see the differences.
So what strip size are u using judus ?
|
|
|
|
|
May 9, 2006, 11:43 AM
|
#20
|
|
DH's Dormant Dragon
Join Date: May 2002
Location: IN Rem-Dormancy
Posts: 24,210
|
i'm using 64k atm.. it's the largest ULi raid controller supports atm
Running nero Speed test or PcPitstop's HD tests (uncached/cached).. it says i've sustained anywhere from 200-300Mb/s reads..
Still fiddling..
but i know some more recent HD benchmarks that use much much larger files to write/read from, seem really show an accurate result. Benifitting the larger block sizes.
With Raid on my Highpoint tech with 2x 160gb 8mb cache WD hardrives with 2MB block size, transfering videos and recording high def videos was REALLY benificial, 64kb and lower (even 128kb block size) wasn't able to keep up.
HD tach showed good results.. nothing amazing.. but sandras large 1gb file transfering and test should excellent results.
Still fiddling here, but i'm guessing that the intel raid controller and those drives giving hdtach exactly what they need.... and i don't doubt that the raid controller is most likely a fair bit better.
|
|
|
|
|
May 9, 2006, 10:55 PM
|
#21
|
|
DriverHeaven Lover
Join Date: Oct 2005
Location: Land of Snow
Posts: 154
Rep Power: 0
|
i guess the strip size depends on what u do. I'll maybe give 64k size a go some day, but right now raid-0 16k strip size is giving me increased speed in most of the things i use my hd for.
|
|
|
|
|
May 10, 2006, 01:41 PM
|
#22
|
|
DriverHeaven Lover
Join Date: Jan 2005
Posts: 119
Rep Power: 0
|
2x WD2500YD 250Gig 16M cache western digital caviar re RAID 0 32k strip
I think they should go faster, probably just the raid controler is lousy.
 |
|
|
|
|
May 10, 2006, 01:46 PM
|
#23
|
|
DH's Dormant Dragon
Join Date: May 2002
Location: IN Rem-Dormancy
Posts: 24,210
|
yeah i think the uli isn't bad.. but it's not top preforming....
I'd still like to see Abit or other company get ahold of HighPoint Technologies chipsets... they are the fastest i've ever seen.
|
|
|
|
|
May 10, 2006, 04:17 PM
|
#24
|
|
Member
Join Date: Mar 2003
Posts: 5,989
|
8K/16K stripe sizes work best with benchmarking tools, but it might be misleading.
the benchmark results could be just the hard drive's and controller's raw performance capabilites,
but a poor indication of real-world system performance.
128K or 64K strip sizes for RAID0 is usually recommended or set as default on the onboard controllers,
but anyway, the best strip size for each system is really based on and depends on the panned drive usage.
best benchmark results, but poor real-world system performance.
[/img]http://img95.imageshack.us/img95/4907/4xr0poorperformance4kc.png[/img]
| |
