Page 1 of 14 123411 ... LastLast
Results 1 to 25 of 348

Thread: Vertex LE vs Crucial C300

  1. #1
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838

    Vertex LE vs Crucial C300

    I've spent too much time generating the charts and iometer results so excuse me for just posting the screenshots, I'll fill in with comments and text as soon as possible.

    All benchmarks are run on my X58A-UD7 rig using the onboard controllers, ICH10R and the Marvell 9128 6G.

    Drives compared
    Vertex LE 100GB FW 1.0
    Vertex LE 100GB FW 1.0 2R0
    Crucial C300 256GB FW 1.0

    Summary of all drives tested will go here in the first post and in the next few posts I'll compare the results.

    I'll be adding the Intel 160GB G2 as well as the X25-E later.

    rr_4KB_512Bvs4KB.PNG

    rr_16KB_512Bvs4KB.png

    rr_64KB_512Bvs4KB.png

    MB/s
    rr_4KB_512Bvs4KB_MBS.png

    Comparing 512B vs 4KB alignment per drive

    rr_4KB_512Bvs4KB_LE.png

    rr_4KB_512Bvs4KB_LE_2R0.png

    rr_4KB_512Bvs4KB_C300.png

    Note: I'll be filling in with more charts and other benchmarks later today.
    Last edited by Anvil; 04-06-2010 at 04:54 PM.
    -
    Hardware:

  2. #2
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    Random write

    Beware that these figures really can't be trusted for the Vertex LE as the iometer test file is easily compressed.
    Using 7Zip the 1GB test file resulted in a 357KB file using normal compression.

    As you probably now by now the SandForce controller uses compression algorithms. In some cases it works great and in some cases not.

    rw_4KB_LE_vs_C300.png

    rw_8KB_LE_vs_C300.png

    rw_16KB_LE_vs_C300.png

    rw_32KB_LE_vs_C300.png

    rw_64KB_LE_vs_C300.png
    Last edited by Anvil; 04-06-2010 at 04:17 PM. Reason: fixed the 8KB random write chart
    -
    Hardware:

  3. #3
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    reserved for more charts
    -
    Hardware:

  4. #4
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    and more charts
    -
    Hardware:

  5. #5
    I am Xtreme
    Join Date
    Nov 2008
    Location
    Warrenton, VA
    Posts
    3,029
    Wow Anvil, much thanks for all the work.
    Looks like 2 vertex LE in R0 does well - looks like 2 c300's might top it?
    I think >60k 4KB random reads 64qd might be better than I can do with the 1231/12xacards!
    I forgot - does w7 automatically give 4k alignment?

  6. #6
    Xtreme Guru
    Join Date
    Aug 2009
    Location
    Wichita, Ks
    Posts
    3,887
    very nice results, the c300 looks impressive. curious that its 512 is almost equal to its 4k, at least if im reading git right.
    "Lurking" Since 1977


    Jesus Saves, God Backs-Up
    *I come to the news section to ban people, not read complaints.*-[XC]Gomeler
    Don't believe Squish, his hardware does control him!

  7. #7
    Banned
    Join Date
    Nov 2005
    Location
    Acreageville, Alberta
    Posts
    1,411
    Great work Anvil! Thanks!!

  8. #8
    Registered User
    Join Date
    Oct 2007
    Location
    CT, USA
    Posts
    29
    Thanks for this...can't wait for the G2!

  9. #9
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    I've updated the second post with Random writes, 4K-64KB.

    @SteveRo,
    If you create a partition using W7 it should be aligned.
    From the looks of it, the C300 should be the winner in an 2R0 array, it's really hard to stay away from the "Buy" button

    @computurd,
    It's a close race up to QD4 but from there on it's game over.
    -
    Hardware:

  10. #10
    Xtreme Addict
    Join Date
    Jun 2006
    Posts
    1,820
    Aha! So the reads are affected quite a bit by the alignment as well
    Any chance you can make a text version of the 4K MB/s chart for my poor eyes that can't tell which blueish color is which?

    Thanks for the tests. I think I'll be gettin' some C300s soon!
    P5E64_Evo/QX9650, 4x X25-E SSD - gimme speed..
    Quote Originally Posted by MR_SmartAss View Post
    Lately there has been a lot of BS(Dave_Graham where are you?)

  11. #11
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    I'll make some adjustments to the charts, I can see it's a bit hard to distinguish between the blue nuances.
    -
    Hardware:

  12. #12
    Xtreme Addict
    Join Date
    Jun 2006
    Posts
    1,820
    So there is quite a difference at higher QD.
    From your chart it looks as if Anand used QD=2 for the random reads test and higher for wrties
    Thanks for the charts, m8.
    P5E64_Evo/QX9650, 4x X25-E SSD - gimme speed..
    Quote Originally Posted by MR_SmartAss View Post
    Lately there has been a lot of BS(Dave_Graham where are you?)

  13. #13
    Xtreme Enthusiast
    Join Date
    Feb 2010
    Posts
    701
    Looks like so far I made the right choice, I wonder how awesome the sf 1500 drives will be.

  14. #14
    I am Xtreme
    Join Date
    Nov 2008
    Location
    Warrenton, VA
    Posts
    3,029
    Anvil, if iometer is skewed due to compression, what is a good test to take compression out of the equation - maybe PCM vantage HDD or full suite test?

  15. #15
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    Steve,

    Here's my observations on the LE regarding sequential writes.

    Files used:
    -iometer test files 10GB
    -Virtual Machine 21GB
    -Images (jpg, nef) 11GB

    sequential_write_test.png

    I added the C300 for comparison. (139GB of VM's)

    The author of AS SSD confirmed that the data used in his Benchmark is hard to compress, and so both the random and sequential read/write throughput reported by AS SSD should be worst case.
    Based on this I'd say the Vertex is a solid performer whichever way you look at it, but performance will vary.
    -
    Hardware:

  16. #16
    Xtreme Enthusiast
    Join Date
    Dec 2009
    Location
    Norway
    Posts
    513
    So virtual machines are compressed to about 80% of original size on SandForce SSDs. Not only will this increase write, but probably also read, and at the same time give more free space for wear leveling and GC.
    Anvil, could you try a couple of copy-tests (to and from a RAM disc) of misc files? Like a couple of folders in program files, the folder Appdata, the Documents folder, etc. A table or diagram would be really interresting.
    It seems i was correct assuming CDM 3.0 and AS SSD uses incompressible (or hardly compressible) data patterns, while ATTO, IOmeter, HDtach, etc uses compressible patterns. And it seems to fit that 140MB/s is the "RAW" speed of the 100GB SandForce drives.


    BTW, i would love some input on this thread:
    Thought experiment: HDD with flash read-cache

  17. #17
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    I'll see what I can do about the copy test.

    For further reading on the "secret sauce" of the SF controller. Link to Anand
    (I'm sure most people have read it but it's good reading)
    -
    Hardware:

  18. #18
    Xtreme Enthusiast
    Join Date
    Dec 2009
    Location
    Norway
    Posts
    513
    They released these details a year ago when they exited stealth mode. Perhaps you remember i made a fuzz at the norwegian forum about SandForce right after that happened? Only too bad it took a year for the products to hit the shelves.
    If they had launched before Intel G2, and at reasonable prices for SF-1200 50/100GB, they could have dominated the market some months.

  19. #19
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    I remember the SF discussions well, I wonder what's next, the SF way caught me by surprise.

    Which RAM drive is recommended for W7 x64?
    (for testing purposes)
    -
    Hardware:

  20. #20
    Xtreme Enthusiast
    Join Date
    Dec 2009
    Location
    Norway
    Posts
    513
    I've used Dataram RAMdisk V3.5.130RC9, worked fine mostly, but i've had it crash 2 times. The first was when i tried to unistall it and the driver wouldn't shut down, but i managed to remove it from a device drivers list. The second time was some error that occured after a couple of months where the program that changed settings for the driver wouldn't open and gave some error, but the RAMdisk worked perfectly anyway. I did a full uninstall and re-install and it has worked fine since.
    http://memory.dataram.com/products-a...wnload-ramdisk
    I see RC10 has been released, so maybe they have fixed those two buggs now. They only occured when i tried to change things after the RAMdisk was in place anyway, and didn't crash the system.

  21. #21
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    It worked just fine, no issues so far.
    I'll be doing some copy test tomorrow.
    -
    Hardware:

  22. #22

  23. #23
    I am Xtreme
    Join Date
    Nov 2008
    Location
    Warrenton, VA
    Posts
    3,029
    c300 appears to be the winner in these tests, how about a pcmark v HDD test?

  24. #24
    Xtreme X.I.P.
    Join Date
    Apr 2008
    Location
    Norway
    Posts
    2,838
    I forgot to run the Vantage test on the C300 (atm it's in my laptop), I'll see what I can do this weekend.

    Vantage score for a single LE was touching 42000.
    -
    Hardware:

  25. #25
    Xtreme Enthusiast Aedubber's Avatar
    Join Date
    Oct 2009
    Location
    TX
    Posts
    961
    Do they have the C300 in less GB? I want to update my SSD i have now but i also dont want to spend 500 or so bucks lol.. I saw this on Newegg's website and would like your thought on it please ?
    Corsair P128 CMFSSD-128GBG2D 2.5" 128GB SATA II MLC Internal Solid State Drive (SSD)
    http://www.newegg.com/Product/Produc...82E16820233087
    HEATWARE
    x58-UD5 ▀▄ i7 950 4.0Ghz ▀▄ ZOTAC GTX 580 ▀▄ Mushkins 1600Mhz

Page 1 of 14 123411 ... LastLast

Tags for this Thread

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •