Overclock.net › Forums › Industry News › Hardware News › [techwarelabs]Modifying the Seagate 1.5TB Hard Drive: Unleash the Hidden Performance
New Posts  All Forums:Forum Nav:

[techwarelabs]Modifying the Seagate 1.5TB Hard Drive: Unleash the Hidden Performance - Page 4

post #31 of 48
7/17/09 at 10:36am
Quote:
Originally Posted by denrocks2 View Post
You guys saying you can do better?

Paying twice as much isn't worth a measly 3ms.
What you fail to realize is the millions of files that get affected by a 3ms seek time.


So, yes, it is worth paying for.
Advanced All-In-One HTPC+Server+Gamer from AMD+ATI
(13 items)
  		  
CPUMotherboardGraphicsRAM
AMD Phenom 9950 Black Edition @ 3.0Ghz Gigabyte mATX 785G? HD3200 IGP *Passively cooled* 2x1GB WINTEC Crap, 2x2GB Patriot | DDR2 800 1.9v 
Hard DriveOSMonitorKeyboard
Western Digital AAKS Series 500GB Windows 7 Enterprise x64 Custom Cooked for SPEED Samsung 58" B650 1080p Plasma Logitech DiNovo Mini 
PowerCaseMouse
Silverstone 350Watt Silent PSU Cooler Master 590 Logitech DiNovo Mini 
View all
  hide details  
Reply
Advanced All-In-One HTPC+Server+Gamer from AMD+ATI
(13 items)
  		  
CPUMotherboardGraphicsRAM
AMD Phenom 9950 Black Edition @ 3.0Ghz Gigabyte mATX 785G? HD3200 IGP *Passively cooled* 2x1GB WINTEC Crap, 2x2GB Patriot | DDR2 800 1.9v 
Hard DriveOSMonitorKeyboard
Western Digital AAKS Series 500GB Windows 7 Enterprise x64 Custom Cooked for SPEED Samsung 58" B650 1080p Plasma Logitech DiNovo Mini 
PowerCaseMouse
Silverstone 350Watt Silent PSU Cooler Master 590 Logitech DiNovo Mini 
View all
  hide details  
Reply
post #32 of 48
7/17/09 at 10:57am
Quote:
Originally Posted by XFreeRollerX View Post
What you fail to realize is the millions of files that get affected by a 3ms seek time.


So, yes, it is worth paying for.
It may be worth paying for, but it's not ever worth paying twice as much for.

Unless you like wasting money.
First Intel Build ($250ish total cost)
(13 items)
  		  
CPUMotherboardGraphicsRAM
E2160 @ 3ghz Biostar TP35D2-A7 9600 GSO 96sp 4gb (4x1GB) OCZ Platinum 
Hard DriveOptical DriveOSMonitor
160gb Seagate 2 DVD Burners XP Home 17" CRT 
PowerCase
400 Watt Dynex Antec 900 
View all
  hide details  
Reply
First Intel Build ($250ish total cost)
(13 items)
  		  
CPUMotherboardGraphicsRAM
E2160 @ 3ghz Biostar TP35D2-A7 9600 GSO 96sp 4gb (4x1GB) OCZ Platinum 
Hard DriveOptical DriveOSMonitor
160gb Seagate 2 DVD Burners XP Home 17" CRT 
PowerCase
400 Watt Dynex Antec 900 
View all
  hide details  
Reply
post #33 of 48
7/17/09 at 11:00am
Quote:
Originally Posted by ErBall View Post
Not to hijack but I thought the PERC cards were supposed to yield massively better raid results. I have 3 comparable dries raid-0'd. (640gb blacks) running on ICH10r. Yet I only short stroked to 1TB.



what gives?
A couple of things.... the 7200.12 and Blacks aren't the same drive. The Blacks may just perform sequential read at the default block size slightly better than the 7200.12. Note that my random access is faster by 1.5ms.

As I just alluded.... this is only one benchmark using a specific test size. HD benchmarking is a bit more complicated as performance can greatly differ at different block and I/O counts. I need to run http://www.iozone.org ....

Here's a sample result of the iozone....



Also, RAID0 does not benefit that much from a RAID controller. I got this PERC 6/i for the future..... ICH10R can only do around 650MB/s while the PERC 6/i can do at least 1GB/s.




Quote:
Originally Posted by trueg50 View Post
That 2.2 gb/s is the write back (IE running purely off the drives 16mb cache).
The write-back cache is the system memory. It is the memory that the RAID controller is using to cache. The drive's individual memory are what each drive uses to cache.


Quote:
Originally Posted by denrocks2 View Post
It may be worth paying for, but it's not ever worth paying twice as much for.

Unless you like wasting money.
Worth is subjective.
Edited by DuckieHo - 7/17/09 at 11:04am
Once again...
(13 items)
  		  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
View all
  hide details  
Reply
Once again...
(13 items)
  		  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
View all
  hide details  
Reply
post #34 of 48
7/17/09 at 11:32am
Quote:
Originally Posted by DuckieHo View Post
True, but if you know you won't ever need the space and just want the performance....
Yes, if you'll never use the space, then partial stoking can greatly reduce the degfragmentation you need to maintain performance.

There may be no downside to short stroking for you, but there is next to no upside for those willing to perform a bit of regular maintenance.

Quote:
Originally Posted by XFreeRollerX View Post
What you fail to realize is the millions of files that get affected by a 3ms seek time.


So, yes, it is worth paying for.
There are no files beyond the first 25% of my first volume. Seeking anywhere there is actually data on my drive takes quite a bit less time than the random seeks to utterly empty areas that these benchmarks perform.

If I were to short stroke further, I would see zero difference in real world performance.

I defrag before and after every major software install, and once a week on top of that.

Quote:
Originally Posted by DuckieHo View Post
That's 3ms per random seek..... How many random seeks does your hard drive perform every few seconds?
Since all my files are lined up on the outer tracks of my drives, they do zero random seeks to the inner tracks, except when I defrag.
Primary
(15 items)
  		  
CPUMotherboardGraphicsRAM
i7 970 3012B113 @ 4.2GHz, 1.36v, HT On Gigabyte GA-X58A-UD5 rev 2.0 Radeon 6950 (unlocked) @ 950/1450, 1.205v, OCP off 6x4GiB Samsung MV-3V4G3 @ 1680, 8-8-8-24-T2, 1.5v 
Hard DriveOptical DriveCoolingOS
Kingston V+100, Samsung F3, 4x Hitachi 7K1000.C LG 22X in Vantec eSATA enclosure Prolimatech Genesis Windows Server 2008 R2 SP1 
MonitorKeyboardPowerCase
Hanns-G HZ281 27.5" 1920x1200 Filco Majestouch Tenkeyless (Cherry MX brown) Antec True Power New 750 Cooler Master HAF 932 
MouseMouse PadAudio
Zowie EC1 X-Trac Ripper XXL ASUS Xonar DS 
View all
  hide details  
Reply
Primary
(15 items)
  		  
CPUMotherboardGraphicsRAM
i7 970 3012B113 @ 4.2GHz, 1.36v, HT On Gigabyte GA-X58A-UD5 rev 2.0 Radeon 6950 (unlocked) @ 950/1450, 1.205v, OCP off 6x4GiB Samsung MV-3V4G3 @ 1680, 8-8-8-24-T2, 1.5v 
Hard DriveOptical DriveCoolingOS
Kingston V+100, Samsung F3, 4x Hitachi 7K1000.C LG 22X in Vantec eSATA enclosure Prolimatech Genesis Windows Server 2008 R2 SP1 
MonitorKeyboardPowerCase
Hanns-G HZ281 27.5" 1920x1200 Filco Majestouch Tenkeyless (Cherry MX brown) Antec True Power New 750 Cooler Master HAF 932 
MouseMouse PadAudio
Zowie EC1 X-Trac Ripper XXL ASUS Xonar DS 
View all
  hide details  
Reply
post #35 of 48
7/17/09 at 11:38am
Is this not the same thing (in the end at least) as short-stroking the drive? It appears some agree with me and others do not. This just seems to be a more round-about way of doing that.
BladeRunner v2.2
(13 items)
  		  
CPUMotherboardGraphicsRAM
Intel Core i7 920 D0 @ 4.202 (200x21) w/o HT ASUS P6T VisionTek HD 5870 3x2GiB OCZ XMP Ready DDR3 1600 @ 1678 9-9-9-27-1T 
Hard DriveOptical DriveOSMonitor
2x120GB Vertex 3 MaxIOPS in RAID-0, 750GB WD AAKS LG WH08LS20 (BD-RW), Samsung SH-S203N (DVD-RW) Windows 7 Home Premium 64-bit BenQ V2400W 24" Monitor 
KeyboardPowerCaseMouse
Logitech G15 Ultra X3 1000W Antec 1200 Logitech G5 
Mouse Pad
Generic 
View all
  hide details  
Reply
BladeRunner v2.2
(13 items)
  		  
CPUMotherboardGraphicsRAM
Intel Core i7 920 D0 @ 4.202 (200x21) w/o HT ASUS P6T VisionTek HD 5870 3x2GiB OCZ XMP Ready DDR3 1600 @ 1678 9-9-9-27-1T 
Hard DriveOptical DriveOSMonitor
2x120GB Vertex 3 MaxIOPS in RAID-0, 750GB WD AAKS LG WH08LS20 (BD-RW), Samsung SH-S203N (DVD-RW) Windows 7 Home Premium 64-bit BenQ V2400W 24" Monitor 
KeyboardPowerCaseMouse
Logitech G15 Ultra X3 1000W Antec 1200 Logitech G5 
Mouse Pad
Generic 
View all
  hide details  
Reply
post #36 of 48
7/17/09 at 11:45am
Quote:
Originally Posted by stargate125645 View Post
Is this not the same thing (in the end at least) as short-stroking the drive? It appears some agree with me and others do not. This just seems to be a more round-about way of doing that.
It is short stroking, by the lowest level means cnsumers have access to.

- Setting drive capacity via manufacturers utilities. (this article)

- RAID volumes less than total drive size.

- First partition less than total drive size.

- Just keeping your data well defragged.

They all keep the read write heads away from the slowest areas of the drive, and all offer similar performance benefits.
Primary
(15 items)
  		  
CPUMotherboardGraphicsRAM
i7 970 3012B113 @ 4.2GHz, 1.36v, HT On Gigabyte GA-X58A-UD5 rev 2.0 Radeon 6950 (unlocked) @ 950/1450, 1.205v, OCP off 6x4GiB Samsung MV-3V4G3 @ 1680, 8-8-8-24-T2, 1.5v 
Hard DriveOptical DriveCoolingOS
Kingston V+100, Samsung F3, 4x Hitachi 7K1000.C LG 22X in Vantec eSATA enclosure Prolimatech Genesis Windows Server 2008 R2 SP1 
MonitorKeyboardPowerCase
Hanns-G HZ281 27.5" 1920x1200 Filco Majestouch Tenkeyless (Cherry MX brown) Antec True Power New 750 Cooler Master HAF 932 
MouseMouse PadAudio
Zowie EC1 X-Trac Ripper XXL ASUS Xonar DS 
View all
  hide details  
Reply
Primary
(15 items)
  		  
CPUMotherboardGraphicsRAM
i7 970 3012B113 @ 4.2GHz, 1.36v, HT On Gigabyte GA-X58A-UD5 rev 2.0 Radeon 6950 (unlocked) @ 950/1450, 1.205v, OCP off 6x4GiB Samsung MV-3V4G3 @ 1680, 8-8-8-24-T2, 1.5v 
Hard DriveOptical DriveCoolingOS
Kingston V+100, Samsung F3, 4x Hitachi 7K1000.C LG 22X in Vantec eSATA enclosure Prolimatech Genesis Windows Server 2008 R2 SP1 
MonitorKeyboardPowerCase
Hanns-G HZ281 27.5" 1920x1200 Filco Majestouch Tenkeyless (Cherry MX brown) Antec True Power New 750 Cooler Master HAF 932 
MouseMouse PadAudio
Zowie EC1 X-Trac Ripper XXL ASUS Xonar DS 
View all
  hide details  
Reply
post #37 of 48
7/17/09 at 12:25pm
Of course you can always short stroke a Velociraptor too. I tried that and got my access times down to about 5.5ms.
Sandy Bridge is here
(13 items)
  		  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K @ 4.6GHz Gigabyte GA-P67X-UD3-B3 Diamond 7970 @ 1055/1500 16GB Corsair Vengeance 
Hard DriveOSMonitorKeyboard
128GB Crucial M4 + 640GB WD Black Windows 7 Ultimate x64 24" Samsung SyncMaster 1080p LED Microsoft Digital Media Pro 
PowerCaseMouse
Corsair 1000w Antec P-182 Logitec MX518 
View all
  hide details  
Reply
Sandy Bridge is here
(13 items)
  		  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K @ 4.6GHz Gigabyte GA-P67X-UD3-B3 Diamond 7970 @ 1055/1500 16GB Corsair Vengeance 
Hard DriveOSMonitorKeyboard
128GB Crucial M4 + 640GB WD Black Windows 7 Ultimate x64 24" Samsung SyncMaster 1080p LED Microsoft Digital Media Pro 
PowerCaseMouse
Corsair 1000w Antec P-182 Logitec MX518 
View all
  hide details  
Reply
post #38 of 48
7/17/09 at 12:54pm
This article is interesting even if it only shows that the more sophisticated controllers and higher capacity platters have a lot of room for tweaking and optimizing. Even if it is only on synthetic benchmarks. I still stand by the idea that if anybody wants true performance they should get a good SSD.
No Compromises
(13 items)
  		  
CPUMotherboardGraphicsRAM
AMD A8 3850 ASUS F1A75-M PRO Micro ATX HD 6550 2x4gb G skill rip jaws DDR3 
Hard DriveOptical DriveOSMonitor
2x120gb intel 320 SSDs SATA DVD Burner Windows 7 x64 SP1 2 x 23" TN 1920x1080 
KeyboardPowerCaseMouse
Dell wireless USB 270 watt Aptek TFX APEX DM-387 Black Steel Media Center Logitech Optical 
Mouse Pad
Leather Asus Pad 
View all
  hide details  
Reply
No Compromises
(13 items)
  		  
CPUMotherboardGraphicsRAM
AMD A8 3850 ASUS F1A75-M PRO Micro ATX HD 6550 2x4gb G skill rip jaws DDR3 
Hard DriveOptical DriveOSMonitor
2x120gb intel 320 SSDs SATA DVD Burner Windows 7 x64 SP1 2 x 23" TN 1920x1080 
KeyboardPowerCaseMouse
Dell wireless USB 270 watt Aptek TFX APEX DM-387 Black Steel Media Center Logitech Optical 
Mouse Pad
Leather Asus Pad 
View all
  hide details  
Reply
post #39 of 48
7/17/09 at 1:44pm
Thread Starter 
Thanks guys for the different opinions , and technical info. Me personally never used a raid setup , and therefore I don't have any knowledge about it . Could some one post a link with some good tech info about raid and its benifits?
Gladiator
(13 items)
  		  
CPUMotherboardGraphicsRAM
PhenomII X4 B50 BE GA-MA790GP-UD4H GV-R577SO 4gb OCZ Reaper_2gb OCZ Flex XLC DDR2 @ 1150 
Hard DriveOSPowerCase
WD_640_Black_ WD 320_Blue Win 7 _ 7600 _X64 BFG_E_550W_PSU CM Gladiator 600 
View all
  hide details  
Reply
Gladiator
(13 items)
  		  
CPUMotherboardGraphicsRAM
PhenomII X4 B50 BE GA-MA790GP-UD4H GV-R577SO 4gb OCZ Reaper_2gb OCZ Flex XLC DDR2 @ 1150 
Hard DriveOSPowerCase
WD_640_Black_ WD 320_Blue Win 7 _ 7600 _X64 BFG_E_550W_PSU CM Gladiator 600 
View all
  hide details  
Reply
post #40 of 48
7/17/09 at 1:48pm
Thread Starter 
Quote:
Originally Posted by OpTicaL View Post
Interesting.

Drop the parenthesis and use open/close bracket in your title.
Thanks
Gladiator
(13 items)
  		  
CPUMotherboardGraphicsRAM
PhenomII X4 B50 BE GA-MA790GP-UD4H GV-R577SO 4gb OCZ Reaper_2gb OCZ Flex XLC DDR2 @ 1150 
Hard DriveOSPowerCase
WD_640_Black_ WD 320_Blue Win 7 _ 7600 _X64 BFG_E_550W_PSU CM Gladiator 600 
View all
  hide details  
Reply
Gladiator
(13 items)
  		  
CPUMotherboardGraphicsRAM
PhenomII X4 B50 BE GA-MA790GP-UD4H GV-R577SO 4gb OCZ Reaper_2gb OCZ Flex XLC DDR2 @ 1150 
Hard DriveOSPowerCase
WD_640_Black_ WD 320_Blue Win 7 _ 7600 _X64 BFG_E_550W_PSU CM Gladiator 600 
View all
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [techwarelabs]Modifying the Seagate 1.5TB Hard Drive: Unleash the Hidden Performance