The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.
← Previous page

    When are the new Intel SSDs ariving?

    Discussion in 'Hardware Components and Aftermarket Upgrades' started by kwantz, Feb 25, 2011.

  1. chimpanzee

    chimpanzee Notebook Virtuoso

    Reputations:
    683
    Messages:
    2,561
    Likes Received:
    0
    Trophy Points:
    55
    I am wondering, couldn't the same be achieve via standard NTFS compression ? The net result is still 'less write' to the SSD/NAND.
     
  2. gvonnatur

    gvonnatur Notebook Enthusiast

    Reputations:
    0
    Messages:
    31
    Likes Received:
    0
    Trophy Points:
    15
    How do you think a WD SSD would hold up against the new Intel? I use Raptors in my PC, and they're awesome.
     
  3. Judicator

    Judicator Judged and found wanting.

    Reputations:
    1,098
    Messages:
    2,594
    Likes Received:
    19
    Trophy Points:
    56
    In theory, certainly, although doing it that way is going to result in extra CPU load instead. Also, given the way NTFS reserves the full space of the uncompressed file, it still may result in greater writes to the NAND (or confusion on the part of the SSD controller). Sandforce also uses proprietary compression algorithms, which are probably better than the LZ77 based compression algorithm of NTFS.

    The currently released WD SSDs, the SiliconEdge Blues, are decidedly average in performance. They're nowhere near the performance of even the G2 Intels, let alone the new 510 or the upcoming G3.
     
  4. chimpanzee

    chimpanzee Notebook Virtuoso

    Reputations:
    683
    Messages:
    2,561
    Likes Received:
    0
    Trophy Points:
    55
    CPU cycle is cheap and under utilized in most machines and going forward(Quad being the standard) the trend will continue. Sure different algorithm and approach would result in different compression ratio but that difference IMO would not be that large as one requirement of real time compression/decompression is that it has to be very fast which limits what you can do.

    In fact, I briefly went through SF's patent claim and they instead of pattern their algo, they just patent the method. So even if they are using proprietary algo, they didn't patent it which would make me wonder, how different it is.

    I am very sure their approach is more efficient(after all, dedicating a controller for dedicated task) but the difference may not be that big.
     
  5. Judicator

    Judicator Judged and found wanting.

    Reputations:
    1,098
    Messages:
    2,594
    Likes Received:
    19
    Trophy Points:
    56
    Well, the LZ77 algorithm came out in 1977, so it's kinda old. :p We won't be able to tell exactly what the difference is, though, unless someone can find some real numbers. Hm. That could be an interesting test. I wonder if it would be possible to get a review site to compare, say, a Vertex 2 using a regular file system, and an Intel X-25M using NTFS compression. That does bring up another possible point; can you compress your OS files? I don't know enough about NTFS compression to have any idea.
     
  6. Pirx

    Pirx Notebook Virtuoso

    Reputations:
    3,001
    Messages:
    3,005
    Likes Received:
    416
    Trophy Points:
    151
    You can, in principle, but it's not worth it: Those binaries are already stored quite densely, and do not give you very significant compression ratios.
     
  7. chimpanzee

    chimpanzee Notebook Virtuoso

    Reputations:
    683
    Messages:
    2,561
    Likes Received:
    0
    Trophy Points:
    55
    Playing the devil's advocate, those are also the files that would only be read from which in the case of NTFS compression yield the largest benefit as NTFS compression seems to suffer from write performance(according to some sources I read but didn't bother to test). And decompression is very light on CPU.

    BTW, my Windows directory has a compression ratio of about 85%, Program Files(x86) about 75% and Program files 80%.
     
← Previous page