HD Tach Scores - Page 2

jbloggs · 12-08-2006, 06:50 PM

HDTach Scores:

2 x 74GB WD Raptors (8MB cache) in RAID0

Sequential Read Peak = 143Mb/s
Burst Speed = 228MB/s
Average Read = 130.1MB/s
CPU Utilization = 5%
Random Access = 7.7ms

Seagate Barracuda 320GB SATAII

Sequential Read Peak = 80Mb/s
Burst Speed = 233MB/s
Average Read = 66.4MB/s
CPU Utilization = 2%
Random Access = 13.5ms

System:

Asus P5WDG2 WS Pro
Intel Core 2 Duo E6600
2GB (2 x 1GB) Geil DDR2 6400C4
MSI Nvidia Geforce 7900GTO
2 x 74GB WD Raptors in RAID0
Seagate Barracuda 320GB SATAII

**Kalim** · 12-27-2006, 03:00 AM

Here's one that performed better than expected when carried out on my main play system (4 yr old):

Long Bench 32MB Zone

Maxtor 82GB 7200RPM 8MB Buffer UDMA 6 (3 yr old, 62GB full)

AMD Duron 1.3GHz
300W Generic PSU
1.5GB DDR400

Sequential Read Speed Peak = 61MB/s
Burst Speed = 100.8MB/s
Random Access = 14.9ms
CPU Utilization = 8%
Av. Read = 49.3MB/s

Repeated with my professional system

Kentsfield 2.66GHz Quad core
OCZ Ballistix Tracer DDR2800 4GB
Seasonic 650W PSU

Sequential Read Speed Peak = 72MB/s
Burst Speed = 121.5MB/s
Random Access = 11.0ms
CPU Utilization = 1%
Av. Read = 57.9MB/s

Some difference there. The rest of the drives here with me are ITB and 1.5TB external through FireWire800 and SCSI (which aren't compatible with the older spec AMD), so I'll leave them for now.

**koala** · 12-27-2006, 01:15 PM

Athlon 64 X2 4200+
Asus M2V mobo
2GB RAM (DDR2 800mhz)
250GB Maxtor SATA2

Quick 8MB test
Sequential Read Peak = 80 Mb/s
Burst Speed = 133 MB/s
Average Read = 64 MB/s
CPU Utilization = 2%
Random Access = 14.4ms

**Kalim** · 01-01-2007, 01:53 AM

It won't count but I have some better RAM back in now from another system that was half way around the world and these are the new tests I ran this morn

Quick 8MB Zone

System: In left side info.

Seagate Cheetah 73.4GB 8MB Cache Ultra320 SCSI 15,000 RPM
Average Read = 89.2MB/s
CPU Utilization = 0%
Random Access = 5.1ms

Seagate Cheetah 146GB 8MB Cache Ultra320 SCSI 15,000 RPM
Average Read = 78.7MB/s
CPU Utilization = 1%
Random Access = 5.5ms

74GB WD Raptor 10,000 RPM 8MB Cache in RAID 0
Sequential Read Peak = 145MB/s
Burst Speed = 231.4MB/s
Average Read = 130.1MB/s
CPU Utilization = 0%
Random Access = 7.0ms

150GB WD Raptor 10,000 RPM SATA 16MB Cache
Sequential Read Peak = 110MB/s
Burst Speed = 138MB/s
Average read = 67.1 MB/s
CPU Utilization = 0%
Random Access = 10.3ms

Maxtor IDE 82GB 7200RPM 8MB Cache
Sequential Read Peak = 69MB/s
Burst Speed = 133MB/s
Average Read = 56.5MB/s
CPU Utilization = 0%
Random Access = 11.0ms

Thanks

**Kalim** · 01-06-2007, 08:37 AM

2.67GHz quad core overlocked to 3.5GHz

Quick 8MB Zone

150GB WD Raptor 10,000 RPM SATA 16MB Cache
Areca ARC-1210 - PCIe x8 RAID Controller

Sequential Read Peak = 289MB/s
Burst Speed = 683MB/s
Average read = 160.3MB/s
CPU Utilization = 1%
Random Access = 8.0ms

!!! I've done this three times now to confirm. This is my professional system setup usually running off the PCIe X8 slot. The difference is double.

**smz** · 03-24-2007, 05:51 AM

FYI: the current link to download HD Tach is coming up dead.

I was going to contribute some of my drives to this thread but want to make sure you are using the same version as myself.

Quote:

Originally Posted by blackduck30

This thread was started to help give people some indication of speed and read/ write time of different drives ( be it raid or non raid ) in what i hope to be various systems.
This thread will hopefully allow people to post there scores in HD Tach only to keep things uniform. It will also allow you to post your system specs for others to compare scores with systems. THIS IS NOT A CONTEST

Please be aware thats scores can vary in two systems with the same specs and hard drives, this is meant as a reference point and can help you decide and gauge the difference between say a raid 0 and a non raid and may help you decide if it is worth your outlay and time or if indeed your system is way out of tune.

If you do not have HD Tach you can download a free version here

http://www.simplisoftware.com/Public...request=HdTach

Here are my scores for my Intel Raid 0 system drive running the quick 8MB zone test

Sequential Read Peak = 115Mb/s
Burst Speed = 270MB/s
Average Read = 96.9 MB/s
CPU Utilization = 6%
Random Access = 15ms

My 200G Maxtor storage drive ( Non raid ) running the quick 8MB zone test

Sequential Read Peak = 75Mb/s
Burst Speed = 210.7 Mb/s
Average Read = 57.3 Mb/s
CPU Utilization = 3%
Random Access = 16.3ms

System,
Asus P5WD2 Premium
Intel Pentium D 940 3.2 mhz
2G corsair value select DDR2
Gainward,Nvidia 7900GT
2 X maxtor 200G Sata 2, intel Raid 0 ( system drives )
2 X maxtor 200G Sata 2 storage drives
Silverstone Zeus 650W Power Supply

**blackduck30** · 03-25-2007, 02:44 AM

Quote:

Simpli Software will be back soon. We are transferring to a new web server and didn’t quite get the migration right.

We apologize for the inconvenience.

thanks for letting us know about the dead link, I presume they will be back up and running soon

I added an alternative link from MajorGeeks

**blackduck30** · 03-25-2007, 03:16 AM

OK here are the results from my new build, same drives more or less but added a 250G Samsung SATA 2 and an External 80G drive to keep a clone of my O/S on

nvidia raid0 2X200G Maxtor SATA2's

Sequencial read 115mb/s
Burst Speed 267.3MB/s
Average Read 96.0MB/s
CPU Utilization 4%
Random Access 15.0ms

Samsung 250G SATA2 Storage drive

Sequencial read 75mb/s
Burst Speed 173.3MB/s
Average Read 62.2MB/s
CPU Utilization 3%
Random Access 14.9ms

Seagate Barracuda SATA 1 Storage drive external via usb2

Sequencial read 35mb/s
Burst Speed 34.1MB/s
Average Read 34.3MB/s
CPU Utilization 11%
Random Access 12.9ms

Maxtor 200G SATA2 Storage drive

Sequencial read 75mb/s
Burst Speed 220MB/s
Average Read 57.5MB/s
CPU Utilization 2%
Random Access 15.8ms

ASUS P5N32-E SLI Plus
Intel E6400 @ 2.4
Gainward 7900Gt
Audigy 4 sound
2G corsair DDR2 533
Swiftech watercooling
Silverstone Zeus 650W PSU
TT armor Case

sarcan · 04-16-2007, 06:48 AM

Seagate ST3500630NS3 SATA/300
Controller Sil 3112
Computer Dell Precision 530 Dual Xeon 2.4GHz 2GB Rambus

Sequential Read max 82 MB/s (read from graph)
Burst 102.7 MB/s
Random access 13.6 ms
CPU 2%
Average read 65.4 MB/s

This drive should, according to several reviews, be the fastest thing my money can buy. But it is MUCH slower than my old 32 GB SCSI ultra160, for example going into hibernate on the old drive takes about 3 seconds while on this new one over 12 seconds!
I'm aware that this is just a place to exchange test results but please allow me to ask if anyone has any suggestion what the bottleneck might be and / or how to solve it.

**mattlock** · 04-16-2007, 07:41 AM

Quick Bench

Western Digital WD1600AAJS 160GB single platter

Sequencial read= 85 mb/s
Burst Speed= 201.2 MB/s
Random Access= 13.2 ms
CPU Utilization= 1%
Average Read= 68.2 MB/s

Seagate Barracuda 160GB ATA100

Sequencial read= 59.5 mb/s
Burst Speed= 68MB/s
Random Access= 14.9 ms
CPU Utilization= 4%
Average Read= 47.2 MB/s

**linderman** · 04-16-2007, 07:47 AM

@ Scarcan

Sata II drives (300mbs) are not yet faster in real world use than their 150mbs brother. The 150 & 300mb/s speeds are theoretical speeds.
You drive is performing to speeds that would be expected!

SCSI still has advantages over SATA drives just not in the price structure. Sata has however, closed the gap between IDE and SCSI to a point where justifing the cost of SCSI is very difficult.

sarcan · 04-16-2007, 08:33 AM

My problem are not as much the HD Tach results, but more the real life lack of speed feeling. It can't be normal that I have to wait 4x as long before the hibernate blue bar fills up? Starting every day progs like Paintshop pro, firefox, Acrobat are all equally slow. My old Seagate Cheetah hit the market around the turn of the millenium, 7 year old technology, not even SCSI U320. My brand new Barracuda 500GB SATA/300 should be faster according to several reviews.

I'm not the type of person to get bothered by a little bit less performance than reviews promise. But the presence of a bottleneck is in my case so self evident that it's really worrying.

**linderman** · 04-16-2007, 09:06 AM

have you tried disabling power management features so see if that speeds the drives feel up?

sounds as if the drive might be running in the wrong mode ?

go to your device manager

right click on IDE controllers & sata controllers

go to properties

then advanced

what are the drive speeds set at ? what are the current transfer modes set at ???

sarcan · 04-16-2007, 01:49 PM

Found the problem. There appears to be a tiny incompatibility issue between Sil3112 controllers and Seagate disks, known as the Mod15write problem. So all I need is another controller.
Thanks Google,
thanks forum members for suggestions

**linderman** · 04-16-2007, 05:57 PM

Thanks for sharing

easy solution for that problem too

sarcan · 04-17-2007, 12:31 AM

actually not easy to find at all. Not a word on Seagate website, not a word on siliconimage.com. When I dived deep into my registry I found
HKLM\SYSTEM\CurrentControlSet\Services\si3112\ProblemDevices
A lot of string values, all Seagate drive model numbers, all with data Mod15Write.
After searching Google for Mod15Write I found a lot of forum sites, mainly Linux developers, discussing the mod15write problem in such programmers lingo that it still took me a lot of time before I got the picture.
see for example http://home-tj.org/wiki/index.php/Sil_m15w which is in relatively clear simple lingo.
So to let this forum be the first to have the whole picture for non-linux users in plain English (with a Dutch accent)
Lots of Seagate disks are not 100% compatible with si3112 controllers. The results can be system hangups and bad performance. One workaround is adding your drive type to the list in the above registry key. This prevents system hangups but at a cost: even worse performance.
The other workaround is either another controller or another drive.
The brands using si3112 chips AFAIK are ATI, Dell, Gigabyte, IBM, Logitech, MSI, NEC, Sitecom, Sony, System Talks, Sweex and even Intel.
Not happy and very curious how these companies have been able to hide this fact for so long.

**linderman** · 04-17-2007, 08:56 AM

I have heard of such problems with NCQ (native command queing)and some controllers, but to my knowledge I have seen no similar problems with western digital drives.

I personally would email Seagate and maybe they could exchange drives with a model not effected by this incompatability ?

12-27-2006, 03:00 AM	#22 (permalink)
Kalim Roaming To Help Join Date: Nov 2006 Posts: 5,484 OS: Many	Here's one that performed better than expected when carried out on my main play system (4 yr old): Long Bench 32MB Zone Maxtor 82GB 7200RPM 8MB Buffer UDMA 6 (3 yr old, 62GB full) AMD Duron 1.3GHz 300W Generic PSU 1.5GB DDR400 Sequential Read Speed Peak = 61MB/s Burst Speed = 100.8MB/s Random Access = 14.9ms CPU Utilization = 8% Av. Read = 49.3MB/s Repeated with my professional system Kentsfield 2.66GHz Quad core OCZ Ballistix Tracer DDR2800 4GB Seasonic 650W PSU Sequential Read Speed Peak = 72MB/s Burst Speed = 121.5MB/s Random Access = 11.0ms CPU Utilization = 1% Av. Read = 57.9MB/s Some difference there. The rest of the drives here with me are ITB and 1.5TB external through FireWire800 and SCSI (which aren't compatible with the older spec AMD), so I'll leave them for now. Last edited by Kalim : 12-27-2006 at 03:02 AM.

12-27-2006, 01:15 PM	#23 (permalink)
koala Moderator Hardware Team Join Date: Mar 2005 Location: UK Posts: 7,844 OS: XP/UBUNTU My System	Athlon 64 X2 4200+ Asus M2V mobo 2GB RAM (DDR2 800mhz) 250GB Maxtor SATA2 Quick 8MB test Sequential Read Peak = 80 Mb/s Burst Speed = 133 MB/s Average Read = 64 MB/s CPU Utilization = 2% Random Access = 14.4ms __________________ New members: Subscribe to your thread (Thread Tools) to receive an instant email notification when you get a reply. TSF Folding@Home Team 85015 - details here

03-25-2007, 03:16 AM	#28 (permalink)
blackduck30 Moderator Join Date: Sep 2004 Location: Wollongong/Australia Posts: 4,227 OS: XP pro SP3/Vista Ultimate My System	Re: HD Tach Scores OK here are the results from my new build, same drives more or less but added a 250G Samsung SATA 2 and an External 80G drive to keep a clone of my O/S on nvidia raid0 2X200G Maxtor SATA2's Sequencial read 115mb/s Burst Speed 267.3MB/s Average Read 96.0MB/s CPU Utilization 4% Random Access 15.0ms Samsung 250G SATA2 Storage drive Sequencial read 75mb/s Burst Speed 173.3MB/s Average Read 62.2MB/s CPU Utilization 3% Random Access 14.9ms Seagate Barracuda SATA 1 Storage drive external via usb2 Sequencial read 35mb/s Burst Speed 34.1MB/s Average Read 34.3MB/s CPU Utilization 11% Random Access 12.9ms Maxtor 200G SATA2 Storage drive Sequencial read 75mb/s Burst Speed 220MB/s Average Read 57.5MB/s CPU Utilization 2% Random Access 15.8ms ASUS P5N32-E SLI Plus Intel E6400 @ 2.4 Gainward 7900Gt Audigy 4 sound 2G corsair DDR2 533 Swiftech watercooling Silverstone Zeus 650W PSU TT armor Case __________________ Blackduck30 Time is like money and milk, It's always running out Any Donations Help Keep TSF Free For All Last edited by blackduck30 : 03-25-2007 at 03:20 AM.

04-16-2007, 06:48 AM	#29 (permalink)
sarcan Registered User Join Date: Apr 2007 Posts: 5 OS: xp sp2	Bottleneck obvious? HD Tach Scores Seagate ST3500630NS3 SATA/300 Controller Sil 3112 Computer Dell Precision 530 Dual Xeon 2.4GHz 2GB Rambus Sequential Read max 82 MB/s (read from graph) Burst 102.7 MB/s Random access 13.6 ms CPU 2% Average read 65.4 MB/s This drive should, according to several reviews, be the fastest thing my money can buy. But it is MUCH slower than my old 32 GB SCSI ultra160, for example going into hibernate on the old drive takes about 3 seconds while on this new one over 12 seconds! I'm aware that this is just a place to exchange test results but please allow me to ask if anyone has any suggestion what the bottleneck might be and / or how to solve it.

04-16-2007, 07:41 AM	#30 (permalink)
mattlock moderator hardware team Join Date: Dec 2005 Location: Ohio Posts: 3,703 OS: XP SP2, Ubuntu Gutsy Gibbon My System	Re: HD Tach Scores Quick Bench Western Digital WD1600AAJS 160GB single platter Sequencial read= 85 mb/s Burst Speed= 201.2 MB/s Random Access= 13.2 ms CPU Utilization= 1% Average Read= 68.2 MB/s Seagate Barracuda 160GB ATA100 Sequencial read= 59.5 mb/s Burst Speed= 68MB/s Random Access= 14.9 ms CPU Utilization= 4% Average Read= 47.2 MB/s __________________ Matt Everest-Thermal Paste Instruction-Power Supply Info-Sensorview PSU Calculator-Speedfan-Bench Test Help TSF Fight Disease. It's Free! Last edited by mattlock : 04-16-2007 at 08:02 AM.

12-08-2006, 06:50 PM	#21 (permalink)
jbloggs Registered User Join Date: Oct 2004 Posts: 24 OS: Windows XP Pro	HDTach Scores: 2 x 74GB WD Raptors (8MB cache) in RAID0 Sequential Read Peak = 143Mb/s Burst Speed = 228MB/s Average Read = 130.1MB/s CPU Utilization = 5% Random Access = 7.7ms Seagate Barracuda 320GB SATAII Sequential Read Peak = 80Mb/s Burst Speed = 233MB/s Average Read = 66.4MB/s CPU Utilization = 2% Random Access = 13.5ms System: Asus P5WDG2 WS Pro Intel Core 2 Duo E6600 2GB (2 x 1GB) Geil DDR2 6400C4 MSI Nvidia Geforce 7900GTO 2 x 74GB WD Raptors in RAID0 Seagate Barracuda 320GB SATAII

01-01-2007, 01:53 AM	#24 (permalink)
Kalim Roaming To Help Join Date: Nov 2006 Posts: 5,484 OS: Many	It won't count but I have some better RAM back in now from another system that was half way around the world and these are the new tests I ran this morn Quick 8MB Zone System: In left side info. Seagate Cheetah 73.4GB 8MB Cache Ultra320 SCSI 15,000 RPM Average Read = 89.2MB/s CPU Utilization = 0% Random Access = 5.1ms Seagate Cheetah 146GB 8MB Cache Ultra320 SCSI 15,000 RPM Average Read = 78.7MB/s CPU Utilization = 1% Random Access = 5.5ms 74GB WD Raptor 10,000 RPM 8MB Cache in RAID 0 Sequential Read Peak = 145MB/s Burst Speed = 231.4MB/s Average Read = 130.1MB/s CPU Utilization = 0% Random Access = 7.0ms 150GB WD Raptor 10,000 RPM SATA 16MB Cache Sequential Read Peak = 110MB/s Burst Speed = 138MB/s Average read = 67.1 MB/s CPU Utilization = 0% Random Access = 10.3ms Maxtor IDE 82GB 7200RPM 8MB Cache Sequential Read Peak = 69MB/s Burst Speed = 133MB/s Average Read = 56.5MB/s CPU Utilization = 0% Random Access = 11.0ms Thanks

01-06-2007, 08:37 AM	#25 (permalink)
Kalim Roaming To Help Join Date: Nov 2006 Posts: 5,484 OS: Many	2.67GHz quad core overlocked to 3.5GHz Quick 8MB Zone 150GB WD Raptor 10,000 RPM SATA 16MB Cache Areca ARC-1210 - PCIe x8 RAID Controller Sequential Read Peak = 289MB/s Burst Speed = 683MB/s Average read = 160.3MB/s CPU Utilization = 1% Random Access = 8.0ms !!! I've done this three times now to confirm. This is my professional system setup usually running off the PCIe X8 slot. The difference is double.

04-16-2007, 07:47 AM	#31 (permalink)
linderman Moderator, Hardware Team Join Date: May 2005 Posts: 11,953 OS: win 2000 pro / Win XP pro My System	Re: HD Tach Scores @ Scarcan Sata II drives (300mbs) are not yet faster in real world use than their 150mbs brother. The 150 & 300mb/s speeds are theoretical speeds. You drive is performing to speeds that would be expected! SCSI still has advantages over SATA drives just not in the price structure. Sata has however, closed the gap between IDE and SCSI to a point where justifing the cost of SCSI is very difficult. __________________ IF YOUR HARD DRIVE DIED RIGHT NOW; WOULD YOU CRY ??? ASK US FOR HELP DEVELOPING A DATA BACK-UP METHOD

04-16-2007, 08:33 AM	#32 (permalink)
sarcan Registered User Join Date: Apr 2007 Posts: 5 OS: xp sp2	Re: HD Tach Scores My problem are not as much the HD Tach results, but more the real life lack of speed feeling. It can't be normal that I have to wait 4x as long before the hibernate blue bar fills up? Starting every day progs like Paintshop pro, firefox, Acrobat are all equally slow. My old Seagate Cheetah hit the market around the turn of the millenium, 7 year old technology, not even SCSI U320. My brand new Barracuda 500GB SATA/300 should be faster according to several reviews. I'm not the type of person to get bothered by a little bit less performance than reviews promise. But the presence of a bottleneck is in my case so self evident that it's really worrying.

04-16-2007, 09:06 AM	#33 (permalink)
linderman Moderator, Hardware Team Join Date: May 2005 Posts: 11,953 OS: win 2000 pro / Win XP pro My System	Re: HD Tach Scores have you tried disabling power management features so see if that speeds the drives feel up? sounds as if the drive might be running in the wrong mode ? go to your device manager right click on IDE controllers & sata controllers go to properties then advanced what are the drive speeds set at ? what are the current transfer modes set at ??? __________________ IF YOUR HARD DRIVE DIED RIGHT NOW; WOULD YOU CRY ??? ASK US FOR HELP DEVELOPING A DATA BACK-UP METHOD

04-16-2007, 01:49 PM	#34 (permalink)
sarcan Registered User Join Date: Apr 2007 Posts: 5 OS: xp sp2	Re: HD Tach Scores Found the problem. There appears to be a tiny incompatibility issue between Sil3112 controllers and Seagate disks, known as the Mod15write problem. So all I need is another controller. Thanks Google, thanks forum members for suggestions

04-16-2007, 05:57 PM	#35 (permalink)
linderman Moderator, Hardware Team Join Date: May 2005 Posts: 11,953 OS: win 2000 pro / Win XP pro My System	Re: HD Tach Scores Thanks for sharing easy solution for that problem too __________________ IF YOUR HARD DRIVE DIED RIGHT NOW; WOULD YOU CRY ??? ASK US FOR HELP DEVELOPING A DATA BACK-UP METHOD

04-17-2007, 12:31 AM	#36 (permalink)
sarcan Registered User Join Date: Apr 2007 Posts: 5 OS: xp sp2	Re: HD Tach Scores actually not easy to find at all. Not a word on Seagate website, not a word on siliconimage.com. When I dived deep into my registry I found HKLM\SYSTEM\CurrentControlSet\Services\si3112\ProblemDevices A lot of string values, all Seagate drive model numbers, all with data Mod15Write. After searching Google for Mod15Write I found a lot of forum sites, mainly Linux developers, discussing the mod15write problem in such programmers lingo that it still took me a lot of time before I got the picture. see for example http://home-tj.org/wiki/index.php/Sil_m15w which is in relatively clear simple lingo. So to let this forum be the first to have the whole picture for non-linux users in plain English (with a Dutch accent) Lots of Seagate disks are not 100% compatible with si3112 controllers. The results can be system hangups and bad performance. One workaround is adding your drive type to the list in the above registry key. This prevents system hangups but at a cost: even worse performance. The other workaround is either another controller or another drive. The brands using si3112 chips AFAIK are ATI, Dell, Gigabyte, IBM, Logitech, MSI, NEC, Sitecom, Sony, System Talks, Sweex and even Intel. Not happy and very curious how these companies have been able to hide this fact for so long.