Overclock.net › Forums › Components › Hard Drives & Storage › SSD › Do's and don'ts when benchmarking SSD's
New Posts  All Forums:Forum Nav:

Do's and don'ts when benchmarking SSD's

post #1 of 3
8/1/12 at 11:01am
Thread Starter 
Very important in bold

Do not use AS_SSD (writes between 3-5 GiB) on default settings.

Do not use CDM (writes between 15-19 GiB) on default settings.

Do not use AS_SSD and CDM together in the same Logged On session.

Do not run FF pattern on ATTO. Default settings are ok.

Run IOmeter with 4KiB 'sector' alignment. Do not run it to fill the drive.

Do not run Random patterns straight after Sequential or vice versa.

Best to run individual tests with small file sizes and not full suite default.

Do not use 'Full' Benchmarks more than once in the same Logged On session.

Do not use any Benchmark (compressible data) more than once a week.

Do not use any Benchmark (un-compressible data) more than once a month.
Sandy Babe
(17 items)
  		  
CPUMotherboardGraphicsRAM
Intel i7 2600k MSI P67A GD55 WC PowerColor 7970 16gb Corsair Vengeance 1866mhz 
Hard DriveCoolingCoolingCooling
OCZ Vertex 4 Alphacool UT 60 240mm EK Supremacy Alphacool NexXxus 7970 block 
CoolingOSMonitorKeyboard
Phobya DC12-260 Windows 7 64bit 25" HH251D Logitech G15 
PowerCaseMouseMouse Pad
Xigmatek 600w 80+ CM Storm Scout Razer Death Adder CM Storm Speed RX Mousepad 
View all
  hide details  
Reply
Sandy Babe
(17 items)
  		  
CPUMotherboardGraphicsRAM
Intel i7 2600k MSI P67A GD55 WC PowerColor 7970 16gb Corsair Vengeance 1866mhz 
Hard DriveCoolingCoolingCooling
OCZ Vertex 4 Alphacool UT 60 240mm EK Supremacy Alphacool NexXxus 7970 block 
CoolingOSMonitorKeyboard
Phobya DC12-260 Windows 7 64bit 25" HH251D Logitech G15 
PowerCaseMouseMouse Pad
Xigmatek 600w 80+ CM Storm Scout Razer Death Adder CM Storm Speed RX Mousepad 
View all
  hide details  
Reply
post #2 of 3
8/1/12 at 11:46am
A full run of AS SSD writes 5GiB to the SSD. 1GiB for the sequential write and 2GiB each for the random 4K and 4K QD64.

CDM, basically, multiply the number of test runs by the length of each test run and that's how much data it'll write to the SSD for each write benchmark segment.

Seriously, you can run AS SSD once a day and you still won't kill your SSD within the warranty period.
Lucifiel
(13 items)
 
Metatron
(13 items)
  		 
CPUMotherboardGraphicsRAM
Core i7-860 Gigabyte GA-H55N-USB3 EVGA GTX 460 1GB GDDR5 Kingston 2x4GB DDR3 1333 
Hard DriveOSPowerCase
G.SKILL Phoenix Pro 120GB Windows 7 Ultimate x64 Silverstone ST45SF 450W Silverstone Sugo SG05B 
View all
CPUMotherboardGraphicsRAM
Core i7-860 Biostar TH55B HD MSI GT 240 1GB GDDR5 Kingston 2x2GB DDR3 1333 
Hard DriveOSMonitorPower
Intel X25-M 120GB Windows 7 Ultimate x86 Samsung 22" Antec EA-380D 
Case
Rosewill R101-P-BK 
View all
  hide details  
Reply
Lucifiel
(13 items)
 
Metatron
(13 items)
  		 
CPUMotherboardGraphicsRAM
Core i7-860 Gigabyte GA-H55N-USB3 EVGA GTX 460 1GB GDDR5 Kingston 2x4GB DDR3 1333 
Hard DriveOSPowerCase
G.SKILL Phoenix Pro 120GB Windows 7 Ultimate x64 Silverstone ST45SF 450W Silverstone Sugo SG05B 
View all
CPUMotherboardGraphicsRAM
Core i7-860 Biostar TH55B HD MSI GT 240 1GB GDDR5 Kingston 2x2GB DDR3 1333 
Hard DriveOSMonitorPower
Intel X25-M 120GB Windows 7 Ultimate x86 Samsung 22" Antec EA-380D 
Case
Rosewill R101-P-BK 
View all
  hide details  
Reply
post #3 of 3
8/2/12 at 12:06am
LOL, maybe for a SF drive but practically any marvelle drive can be hammered all day and will recover in very short time so you can do it all again. biggrin.gif
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: SSD
Overclock.net › Forums › Components › Hard Drives & Storage › SSD › Do's and don'ts when benchmarking SSD's