Loading ...
Sorry, an error occurred while loading the content.

Re: OT: SATA disks & SSDs

Expand Messages
  • David J Taylor
    ... Yes, a less secure plug/socket pair I have yet to see! Super-glue might help hold them in place! [] ... Why does altering the interface speed make
    Message 1 of 7 , May 1 12:14 AM
    • 0 Attachment
      Ulrich G. Kliegis wrote:
      > On 26 Apr 2009 at 15:49, David J Taylor wrote:
      >
      >> Sorry to hear about the
      >> disk - together with fans they are perhaps the hardest working part
      >> of the PC.
      >
      > Although it's a bit off topic here: SATA plugs are still on the top
      > of my list of annoyances.

      Yes, a less secure plug/socket pair I have yet to see! Super-glue might
      help hold them in place! <G>

      []
      > After I changed all my Samsung HDDs (my preference) none of them
      > failed
      > again.
      >
      > And SSDs are here already, although still at LEO prices. Remains the
      > task
      > of plumbing the PC to the house's hot water system :).
      >
      > Cheers,
      > U.

      Why does altering the interface speed make the disk more reliable?
      How may write cycles do SSDs have?

      Cheers,
      David
      --
      SatSignal software - quality software written to your requirements
      Web: www.satsignal.eu
      Email: davidtaylor@...
    • Ulrich G. Kliegis
      ... It seems, but that s only my personal hypothesis, that after some time of non-usage, the disks fall back to their default values and try to re-negotiate. I
      Message 2 of 7 , May 1 4:44 AM
      • 0 Attachment
        On 1 May 2009 at 8:14, David J Taylor wrote:

        >
        > Why does altering the interface speed make the disk more reliable?

        It seems, but that's only my personal hypothesis, that after some time of
        non-usage, the disks fall back to their default values and try to re-negotiate. I
        am not sure if that is true, but it would be an explanation why the problems I
        saw on several computers here, that some disks looked like missing after
        various times, have completely disappeared. Maybe the controllers regards
        the negotiation as done, once it's done, maybe it's ion storms or bad fate, no
        idea.
        The slower speed seems to be more robust, though. Maybe Arne has a
        comment too.

        Cheers,
        U.
      • Ulrich G. Kliegis
        ... The latest generation is supposed to have more cycles than comparable conventional HDDs, but if you take the ratio of cycles/invested capital, HDDs still
        Message 3 of 7 , May 1 4:44 AM
        • 0 Attachment
          On 1 May 2009 at 8:14, David J Taylor wrote:

          > How may write cycles do SSDs have?

          The latest generation is supposed to have more cycles than comparable
          conventional HDDs, but if you take the ratio of cycles/invested capital, HDDs
          still do rule.SSDs are ideal for really mobile computing. But this is getting
          pretty OT here, there are plenty NGs and fora to discuss this.

          Cheers,
          U.
        • a_van_belle
          Hello Ulrich, I have no bad experiences with SATA II (3GB/s) disks. I mainly use samsung and WD in SATA II mode. But you need good cables and support them to
          Message 4 of 7 , May 1 1:10 PM
          • 0 Attachment
            Hello Ulrich,

            I have no bad experiences with SATA II (3GB/s) disks. I mainly use samsung and WD in SATA II mode. But you need good cables and support them to prevent movement in the connector.

            Older SATA cables may not be compatible with SATA II speeds.
            If a cable does not mention SATA II specifically, it may only be suitable for SATA I !

            But most problems occur with older mainboards and new SATA II disks.
            Early SATA I controllers do not support auto-negotiation, causing the disk in SATA II mode while teh controller only supports SATA I.
            Same applies to NCQ mode, older controllers don't know how to handle this.

            As most disks never achieve SATA II speeds, it is safe to jumper a disk down to SATA I mode without loosing performance.
            Only disks with large cache (like 32 MB) or SSD can achieve burst rates that require SATA II.

            Greetings,
            Arne van Belle

            --- In MSG-1@yahoogroups.com, "Ulrich G. Kliegis" <Ulrich.Kliegis@...> wrote:
            >
            >
            >
            > On 1 May 2009 at 8:14, David J Taylor wrote:
            >
            > >
            > > Why does altering the interface speed make the disk more reliable?
            >
            > It seems, but that's only my personal hypothesis, that after some time of
            > non-usage, the disks fall back to their default values and try to re-negotiate. I
            > am not sure if that is true, but it would be an explanation why the problems I
            > saw on several computers here, that some disks looked like missing after
            > various times, have completely disappeared. Maybe the controllers regards
            > the negotiation as done, once it's done, maybe it's ion storms or bad fate, no
            > idea.
            > The slower speed seems to be more robust, though. Maybe Arne has a
            > comment too.
            >
            > Cheers,
            > U.
            >
          • Ulrich G. Kliegis
            ... Arne, your explanation matches with my observations very well. The imprint on the cables just say Serial ATA. Please note, though, that the latest
            Message 5 of 7 , May 2 1:45 AM
            • 0 Attachment
              On 1 May 2009 at 20:10, a_van_belle wrote:

              > As most disks never achieve SATA II speeds, it is safe to jumper a
              > disk down to SATA I mode without loosing performance.

              Arne,
              your explanation matches with my observations very well. The imprint on the
              cables just say Serial ATA.
              Please note, though, that the latest generation of 1 TB-Samsung SATAII-
              HDDs lack the 150 MB/s jumper. You have to use their software utility
              (which is done quickly, simply and is useful, you can also optimize the seek
              speed for maximal speed or minimal noise and do other things). I guess that
              other manufacturers may follow.

              Cheers,
              U. (who is going to order his toroidal mirror these days...)
            • David J Taylor
              ... Ulrich, Arne, Thanks for your observations, and suggestions as to why using SATA I might be a good idea. One other comment which is relevant to the
              Message 6 of 7 , May 2 2:35 AM
              • 0 Attachment
                Ulrich G. Kliegis wrote:
                > On 1 May 2009 at 20:10, a_van_belle wrote:
                >
                >> As most disks never achieve SATA II speeds, it is safe to jumper a
                >> disk down to SATA I mode without loosing performance.
                >
                > Arne,
                > your explanation matches with my observations very well. The imprint
                > on the cables just say Serial ATA.
                > Please note, though, that the latest generation of 1 TB-Samsung
                > SATAII- HDDs lack the 150 MB/s jumper. You have to use their software
                > utility (which is done quickly, simply and is useful, you can also
                > optimize the seek speed for maximal speed or minimal noise and do
                > other things). I guess that other manufacturers may follow.
                >
                > Cheers,
                > U. (who is going to order his toroidal mirror these days...)

                Ulrich, Arne,

                Thanks for your observations, and suggestions as to why using SATA I might
                be a good idea.

                One other comment which is relevant to the data-intensive MSG applications
                is that the newer "EcoGreen" 5400 rpm disks from Samsung do run
                significantly cooler than the earlier non-green 7200 rpm disks, and might
                therefore be expected to be more reliable. Example:

                http://www.novatech.co.uk/novatech/specpage.html?SAM-F21TB

                Look at the drop in disk 1 temperature at the start of week 15 here. That
                was changing my main data disk from a 750GB/7200rpm to a 1TB/5400rpm
                model.

                http://www.satsignal.eu/mrtg/narvik_disk_temp.html

                I had changed because of a number of occurrences of "missing disk" just as
                you had mentioned - perhaps I could have just set the interface speed
                instead!

                Cheers,
                David
                --
                SatSignal software - quality software written to your requirements
                Web: www.satsignal.eu
                Email: davidtaylor@...
              • Ulrich G. Kliegis
                ... At least, here it helped completely - so far (did the change about 6 weeks ago). Cheers, U.
                Message 7 of 7 , May 2 3:47 AM
                • 0 Attachment
                  On 2 May 2009 at 10:35, David J Taylor wrote:

                  >
                  > I had changed because of a number of occurrences of "missing disk"
                  > just as you had mentioned - perhaps I could have just set the
                  > interface speed instead!

                  At least, here it helped completely - so far (did the change about 6 weeks
                  ago).

                  Cheers,
                  U.
                Your message has been successfully submitted and would be delivered to recipients shortly.