Results 1 to 5 of 5

Thread: SeriTek/2SE2-E is G5 and PM-bootable, MacPro is supported.

  1. #1
    Join Date
    Feb 2003
    Location
    ex-Cupertino, CA, USA. Now HU+SRB (Europe).
    Posts
    163

    Default SeriTek/2SE2-E is G5 and PM-bootable, MacPro is supported.

    Rick - just in the case you missed the news...

    http://www.amug.org/amug-web/html/am.../firmtek/2se2e

    This is the fastest booting SATA card. And the only one which boots on PCI-e machines. (It does not boot on Mac Pro - yet).

    It supports Port Multiplier - and it is the only SATA card which is capable of booting from drive(s) attached to Port Multiplier.

    Unlike similar cards sld with "free" Mac drivers it is actually now a RAID-5 card on the PC.

    Who has the card with firmware 5.1.5 - 5.1.7 please don't be shy for a free upgrade.

  2. #2
    Join Date
    Aug 2001
    Location
    Grangeville, ID USA
    Posts
    9,122

    Default

    We have been testing the Seritek card in a MacPro 2.66 Ghz machine. We have a CalDIgit 4 Port card in slot 4 at 8 Lanes and the Seritek in Slot 2 with 2 lanes.

    I am having some trouble getting acceptable performance from the Seritek card with port multiplier equipped in striped RAID. The card seems to be limited to 120-125 MB/sec no matter how many drives are connected to a single port multiplier board.

    A single lane card is limited to a little over 200 MB/sec by the PCIe bus, so that shouldn't be the issue. I have tested direct connect drives, Hitachi, WD drives and compared exactly the same setup on the CalDIgit card. With 4 drives I get over 200 MB/sec read and write on the Caldigit, with the Seritek I get 115 to 120 MB/sec. I have tried every configuration I can think of.

    The Seritek is stable, fast as the CalDIgit when the drives are mounted JBD. But in a RAID0 everything is low. I wonder if we have something in DIsk Utility that doesn't get along with?

    Testing continues, the card is certainly usable, just not for Port Multiplier based speed....... yet. Hopefully it is some little detail that can be fixed or a problem in my testing.

    Rick
    molṑn labe'
    "I am a mortal enemy to arbitrary government and unlimited power. I am naturally very jealous for the rights and liberties of my country, and the least encroachment of those invaluable privileges is apt to make my blood boil."
--Ben Franklin

  3. #3
    Join Date
    Feb 2003
    Location
    ex-Cupertino, CA, USA. Now HU+SRB (Europe).
    Posts
    163

    Default

    Rick,

    The SeriTek/2SE2-E is using the entry-level SiliconImage 3132 SATA processor. What you see is the top bandwidth this SATA chip is capable. Please compare the SeriTek/2SE2-E with other cards having the same processor. Caldigit card you testing is using Silicon Image 3124 PCI-X processor with Intel 8x lane PCI-X - to - PCI-e bridge and costs 3 times more.

    You should compare Apples with Apples, and not Geo Metro with Ferrari Testarossa. Be fair to us and please compare this one: http://www.caldigit.com/FASTA-2e.asp with SeriTek/2SE2-E.

    BTW: the 3132 and 3124 driver model is identical. Same driver set will be much faster on 3124. But the SATA processor + bridge are pretty expensive.

  4. #4
    Join Date
    Aug 2001
    Location
    Grangeville, ID USA
    Posts
    9,122

    Default

    I wasn't testing for comparison or to bash anything. We test all our products so we can post speeds and expected performance. I didn't know if the speed issue was chip based ceiling or firmware issue and possibly fixable, so I asked.
    molṑn labe'
    "I am a mortal enemy to arbitrary government and unlimited power. I am naturally very jealous for the rights and liberties of my country, and the least encroachment of those invaluable privileges is apt to make my blood boil."
--Ben Franklin

  5. #5
    Join Date
    Feb 2003
    Location
    ex-Cupertino, CA, USA. Now HU+SRB (Europe).
    Posts
    163

    Default

    Yes, if you look at 3132 drivers made by SImage they won't fly faster with PM. It's unfortunately a chip issue. The 3124 is great - as long as you use it in PCI-X. Try it in PCI, and ouch. One of the reasons we were so reluctant to give up the Intel GD31244 because it's one of the fastest if not the fastest chips on PCI-based machines, even without the PM support.

    For Mac Pro or with the bridge on PCIe-based machines (see caldigit) 3124 is fine.

    We are not so sure, the very expensive development caldigit made "mating" the 3124 with a PCIe bridge will return the money. But we can always let people upgrade the driver caldigit uses from SiliconImage for some money and get the bootability, for instance + S.M.A.R.T. support + better handling of hot-plug (in the case of RAID-10). There are some extra stuff and more potential in UltraTek/SeriTek driver family compared with other. Question, do we want? And if yes, how to do to ensure, the user will pay? How about liability? If these questions are solved - I won't mind.

Posting Permissions

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