Jump to content

davismccarn

Member
  • Posts

    18
  • Joined

  • Last visited

  • Donations

    0.00 USD 
  • Country

    United States

Posts posted by davismccarn

  1. Careful! Hard disk drive quality is a moving target. IBM sold its storage division to Hitachi (2002ish?) because of the infamous "deathstar" drives; but, Hitachi has since reengineered their drives and current models are very good. Seagate has been my vendor of choice; but, they currently have a defective series which came from a plant in Thailand and some are still on retailer's shelves. Western Digital has accounted for more than 50% of the data recoveries I have seen in the past 7 years, followed by Maxtor at about 20%.

    Right now, I would choose Hitachi, Seagate, Samsung, or Fujitsu with Hitachi edging out Seagate because of their handling of the defective drives.

    P.S. I have been doing data recovery since 1979 (on cassettes and floppies!) and have seen well over 20,000 failed drives.

  2. We don't know; but, if he uses cable internet, most cable modems assign an ip address that is not one of the standard, private ones. (Time-Warner uses 71.xxx.xxx.xxx and Earthlink 24.xxx.xxx.xxx) Setting him up to use a crossover cable would clobber his ability to connect to the internet.

    And yes; you would need several switches to get 255 PC's on one router; but, it works just fine.

    Excepting some Netgear routers which don't acta as DNS forwarders, every other router I know of works great if you just plug in the PC's and use automatic everything with the advantage being they remain as flexible as possible. If the router fails in a thunderstorm, just plug in another one; if its a laptop, they still work great on the road, and so on....

    I use static ip's for servers, printers, and gateway devices; leaving everyone else to be automatic.

  3. OK, basic CPU 101....

    The bigger the cache, the less often you have to come out of it to access slower things

    AND

    The FSB (Front Side Bus) speed affects the communication with the rest of the system.

    The E8400 has an FSB of 1333Mhz (almost double the 800 of the E5400) and the 6MB cache is as big as it gets.....

    If you install it yourself (easy, as you found out), it's $167 at ZipZoomFly (they've been around forever) http://www.zipzoomfly.com/jsp/ProductDetai...rodlist=froogle

  4. OK; you're right, he doesn't say he wants to share the internet; but, we're going to have him set static ip addresses so the crossover works and then undo that every time he wants to use the internet or buy another network card so he can share the connection on the desktop when a router is plug it in and it works?

    Now; technically, his scanner will need to support network scanning (which is ify unless it is an expensive model).

  5. Most cable internet companies will require you to turn the modem off for two minutes+ before it will connect to a new device. Did you do that? And/or, did somebody clone your mac address to the router aeons ago to get it working? If they did, reset the router to defaults, turn the modem off for two minutes, and see if it isn't a lot better.

  6. C'mon guys!

    What you need is a router which will then let you have up to 255 PC's, sharing an internet connection and folders or printers.

    The key to getting started is to make sure the systems are in the same Workgroup. Go to the Computer name tab in System Properties (right-click My Computer, left-click properties) and, if necessary, change the workgroup name to be the same.

    http://www.practicallynetworked.com has a lot of help.

  7. The beginning of this is informational....

    Both Vista and XP SP3 have flawed disk drivers. Vista will completely hang on a CRC error which exacerbates a failing HDD and XP, service pack 3, gives "device io errors" if it has fallen back to PIO mode. I do not know how to get these problems corrected; but, add a hard disk drive with other bugs and the mix is disastrous....

    Details:

    I have two identical data recovery/forensics setups with ASRock 865 chipset motherboards and only one difference; one is XP SP2, the other SP3. I generally use WinHex to sector copy failing drives and have a Promise 133 card so all drives are on their own channel. On the SP2 box, I can copy with no errors excepting the read errors on the source drive. On the SP3 box, I get "device io errors" and timeout errors on the known good drive I am copying to unless I reset the controllers to eliminate the entries leading to PIO fallback mode.

    Recently, on a Vista system, what was happening was it would simply stop. You could click on anything; but, the only response would be the spinning arrow (forever!). It was a 7200.11 drive, so I cloned it to another then updated the firmware. Because the drive only reported 11 bad sectors during the clone and no reallocated sectors in SMART, I forced rewrites of the "bad sectors" which fixed the CRC errors, tested it extensively, and put it back in the clients system. That was 6 weeks ago and it has been flawless, ever since.

    If you want more details, please ask; but there be bugs in MS land which make a failing drive much worse.

    Now, a question.......

    Somehow; between the firmware issue and the buggy drivers, these drives can write thousands of "bad sectors" without complaint until power is cycled and then the stuff hits the fan..........

    I have a 7200.11, 1TB drive, for example, which reads the last 70% perfectly and the first 15% with trivial problems; but, the range from 15-30% is a nightmare! It will read 10,000 to 150,000 sectors with no problems and then hit a stretch of 10-200 sectors which may take over 12 hours to give up on. Because there is no pattern to the errors that matches any possible physical layout, I have to believe the problems are created by the bugs in both the firmware and the OS. (BTW, I have been doing data recovery since 1979)

    Gradius; I have a terminal setup; but, little experience with the Seagate commands. What commands will disable retries and/or tell the drive to ignore CRC errors altogether? (I suspect there is good data in many of the "bad sectors" and it is only the CRC which is wrong)

    And, does anyone else have any suggestions or experience with the same problem?

×
×
  • Create New...