Jump to content

What do CrystalDiskMark results actually mean? (Samsung 860 EVO - Sata III & Samsung 850 EVO - USB 3.0)

I've decided to run the CrystalDiskMark benchmark on my SSD's (and external HDD for the fun of it) - but what do these results actually mean? 

 

What does each category actually mean? 

 

Sequential Read/Write means how fast the drive can read/write 1 large file, correct? But what does "Q32T1" mean?

 

Whats does 4KiB - Q8T8, Q32T1, Q1T1 mean?

 

These following results are on my 500gb Samsung 860 EVO - and are comparatively below par, presumably because it is my boot drive running my OS.

CDM_860evo.PNG.72e3c51784ab40dfb0c5198417750ce2.PNG

 



-----------------------------------------------------------------------
CrystalDiskMark 6.0.2 x64 (C) 2007-2018 hiyohiyo
                          Crystal Dew World : https://crystalmark.info/
-----------------------------------------------------------------------
* MB/s = 1,000,000 bytes/s [SATA/600 = 600,000,000 bytes/s]
* KB = 1000 bytes, KiB = 1024 bytes

 

   Sequential Read (Q= 32,T= 1) :   471.231 MB/s
  Sequential Write (Q= 32,T= 1) :   324.382 MB/s
  Random Read 4KiB (Q=  8,T= 8 ) :   393.900 MB/s [  96167.0 IOPS]
 Random Write 4KiB (Q=  8,T= 8 ) :   321.975 MB/s [  78607.2 IOPS]
  Random Read 4KiB (Q= 32,T= 1) :   384.970 MB/s [  93986.8 IOPS]
 Random Write 4KiB (Q= 32,T= 1) :   326.315 MB/s [  79666.7 IOPS]
  Random Read 4KiB (Q=  1,T= 1) :    33.779 MB/s [   8246.8 IOPS]
 Random Write 4KiB (Q=  1,T= 1) :   134.580 MB/s [  32856.4 IOPS]

 

  Test : 1024 MiB [C: 26.4% (122.8/465.2 GiB)] (x5)  [Interval=5 sec]
  Date : 2018/12/15 0:48:36
    OS : Windows 10  [10.0 Build 17134] (x64)

 

Also, I found a 250gb Samsung 850 EVO at work, which I formatted and I have hooked up via a USB3.0 Case Adapter. Results are presumably below par as the USB3.0 adapter is bottle necking - was too lazy to unplug and open my case to hook up to SATA - besides, wanted to get actual USB 3.0 results that I could expect - as the setup is being used specifically for my Xbox One to get some better performance out of certain games.

CDM_850evo-usb3.0.PNG.98da181a4d49604c6cb2443d46bd51ea.PNG

 

Finally - just for the hell of it, I ran the test on my 1TB Western Digital Passport, external HDD - and WOW! I had no idea how slow HDD's were compared to SSD's!!!! (Also presumably slower than a HDD connected thru SATA)

CDM_WesternDigitalPassport-usb3.0.PNG.43a4ae9f3c89aa371705e3347e7ed6e9.PNG

 

 

So - I am curious as to what these different categories actually mean? I've tried to look on youtube and online - only to find basic information.

 

Also - I am running the tests "stock" in the default parameters:

 

Test Count: 5

Test Size: 1GB

Interval Time: 5 seconds

 

Should those be changed at all?

 

 

THANK YOU FOR YOUR HELP!!!!

 

 

 

CPU: Intel i5-9600k | MoBo: Gigabyte Aorus Elite z390 | RAM: 16gb (4 x 4gb) Crucial Ballistix Sport LT DDR4-2400

GPU: Gigabyte Aorus Xtreme 1080ti | Storage: 500gb Samsung 860 vNand SSD x2 & 1tb WD Caviar Blue HDD

Chassis: NZXT h700i White w/ RGB LED | Cooling: Corsair H100i Pro RGB AIO & 6x Corsair AF120 fans White LED

Screens: 2x 27" Acer HA270 Ultra Slim LED | Peripherals: MSI Interceptor RGB DS4200 Key & D200 Mouse

 

Link to comment
Share on other sites

Link to post
Share on other sites

31 minutes ago, hazeyez said:

So - I am curious as to what these different categories actually mean? I've tried to look on youtube and online - only to find basic information.

Sequentical = access data in a row, so for example sector 1,2,3,4,5,6 in that order.

 

Random = access data randomly, so 2,5,3,6,1,4 for example.

 

Queue depth is how many requests the drive has at one time. Normally on a desktop Its normally 1-4, heavy io uses on servers can queue depth upto around 16, the queue depth 32 test is pretty unrealistic.

 

If a drive has more items in a queue its faster as there is more it can work on at once. Think of it like going to the store to get one item vs getting 10 items at a store. 

 

Threads is how many processes are accessing the drive at once, more processes, higher queue depth, more speed.

 

34 minutes ago, hazeyez said:

Test Count: 5

Test Size: 1GB

Interval Time: 5 seconds

 

Should those be changed at all?

For normal testing this is fine.

 

Some drives has a cache, so a larger test will make the cache less effective and will be slower. The more the counts, the more confident you are in a result(these are averages)

 

Interval allos the drive to rest where it normally will empty the cache between tests.

 

 

 

If you want to go more indepth, crystal disk mark is just a gui for window's disk speed and you can set those value to what ever you want.

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×