Cloning so a SATA drive, boting form a SATA drive?

Discussion in 'Computer Information' started by R. Giggs., Nov 7, 2012.

  1. R. Giggs.

    R. Giggs. Guest

    OK so I cloned my drive to a SATA drive , then I remoed ny original
    IDE boot drive and tried to boot with just the SATA connected.

    Problem was it did n't work, aster the POST screen it just went to
    a black screen.

    I read something about drivers, maybe the computer can't boot
    from a SATA drive because it has no drivers for one as it expected
    and IDE one?
    R. Giggs., Nov 7, 2012
    1. Advertisements

  2. R. Giggs.

    VanguardLH Guest

    After removing your IDE hard disk (which your BIOS could find), and
    during the reboot, did you go into your BIOS to reconfigure to boot from
    the *SATA* hard disk? Look at the boot drive load order option in your
    BIOS. You'll have to move SATA to the top (although it might work if
    it's included as any type in the list of boot devices).
    VanguardLH, Nov 7, 2012
    1. Advertisements

  3. R. Giggs.

    R. Giggs. Guest


    First I just removed the IDE drive, I expect it to just find the SATA drive
    boot from tat, but I just got a black screen.

    Then I powered down and went to the bios boot menu, (where there was just
    SATA drive listed (IDE drive was disconnected) and tried to boot like that
    the same black screen and no boot.

    Then I reconnected the IDE drive and went into the bios to a boot order menu
    the cd drive was one group then another group was the hard drives, it was
    to boot from the IDE drive but I selected that and changed it to the SATA
    drive. Then I reboted and it booted up OK.

    However I think it still booted from the IDE drive, not sure why I think
    this but I do.
    FOr starters the desktop I have is on the IDE drive, that's why I think it
    booted from the
    IDE drive.

    So just to make sure I will remove the IDE drive (disconnect it) and try
    and boot.

    I expect it to fail to boot. We shall see I am going to try it again.

    The whole point of doing this is to be abl to boot from the SATA drive

    Slighly odd (to me) thing is that the IDE drive has letter I: for one
    partition and F: for the
    next partition, but the drive was always listed as I: when it was just a
    blank drive.
    I expect it's a conventionfor the first SATA drive to be I:

    Will try a reboot now on justthe SATA drive, don't expect to succeed!!!
    R. Giggs., Nov 7, 2012
  4. R. Giggs.

    er3er3 Guest

    Hi there!!! This is me!! ie R.Giggs, I am using a new account I
    What happened was I could not boot on the SATA drive (as I expected)
    but I too the oppertunity to boot up on an only Ubuntu CD I had made
    ago, I forgot the loging to my original account (all the info is on
    windows) so I created this new account so I could post.

    I am quite pleased with this because it means if I screw up my hard
    drive I can get on the net via the ubuntu CD, whihc is quite nice, it
    means I have a second way to access the internet, which give me a bit
    of a fall back in case
    of drive failure!! Very nice!!
    er3er3, Nov 7, 2012
  5. R. Giggs.

    er3er3 Guest

    OK going back to window now lol.
    er3er3, Nov 7, 2012
  6. R. Giggs.

    Paul Guest

    For its first boot, a cloned drive should be by itself.
    I think you're following that rule.

    Because of that, I personally see no benefit to connecting
    both original and cloned drive, at the same time, and attempting
    to boot the cloned drive. This causes the very problem, that
    the "cloned drive should boot by itself the first time"
    rule is supposed to prevent.

    When the drive is cloned, you should be verifying, with
    Disk Management, that the two drives have the same
    content. The Disk Management should be showing the same
    partitions, something like "Healthy (Active)" for
    the cloned disk OS partition (the one with the boot
    flag set). If the new drive was completely empty for example,
    you'd want some portion of Sector 0 to get copied over, to
    support booting.

    In the BIOS, it's important that the SATA and IDE drive settings,
    are selected according to the drivers already loaded in Windows.
    When you cloned the drive, there should have been a driver at
    that time, to access the SATA drive. So both an IDE and SATA driver
    must have been there. The only way to get around such a situation,
    is if you cloned with an OS other than the one that is being
    copied to both disks.

    Now, say the SATA drive, freshly cloned again, was connected
    to the motherboard, and the IDE is disconnected. Now, you're
    saying "I can't boot - got black screen, flashing cursor".
    At this point, I'd boot my WinXP installer CD (if the OS was
    WinXP), and use the "recovery console". That looks like a
    MSDOS prompt when you get there. When you do that, the installer
    CD loads its set of drivers, and at some point, it prompts you
    with "type a number to select the partition to log into". With
    just the one disk connected, there is only one choice here,
    and that is to select the OS on Disk 1. The installer CD will
    ask for the administrator password (as that's an excellent way
    of verifying you're attacking the correct partition). I have
    two OSes on my computer, and I see two choices in the menu,
    and the password on the two OS administrator accounts is different.

    Once there, doing "fixmbr" would repair the 440 bytes of
    boot code for the cloned disk. That doesn't touch the four
    primary partition slots in sector 0.

    Sector 0
    446 bytes boot code (on Windows version, searches for boot flag)
    64 bytes = (4) 16 byte primary partition table entries
    2 bytes = 0xAA55 signature bytes, meaning "somebody put a valid
    MBR in sector 0". If this field is zeroed out, any OS
    looking at it, would treat the other info with extreme

    When the disk was cloned, the partition table has to be set up.
    That means the 0xAA55 type signature is there. The 4x16 area
    has to be defined. But the 446 byte area, I don't know if
    there's a guarantee that gets copied. If the copied partition
    was "active", then it makes sense that the 446 bytes should be
    copied as well.

    In any case, if the new drive, plus a CD drive are connected,
    you boot from the CD drive using the Windows installer disc,
    log into the recovery console, and do a "fixmbr", that should
    remove all doubt.

    You can also use "diskpart" in the recovery console, if you
    need to list the various partitions.

    And the fact you could "log into" the cloned C:. proves there
    is enough OS-like info there, to convince the recovery console
    that you're actually repairing an OS partition.

    Now, the next command of interest, is fixboot. This is less
    likely to be needed in your case. When you format a partition,
    the partition boot sector is overwritten. When you install an
    OS, the partition boot sector is loaded. So when you installed
    Windows on C: of the original drive, a special sector near
    the front of the partition gets defined.

    The cloning software, upon realizing it's an OS partition,
    can copy the file system and the partition boot sector.

    Now, I use "non-cloning" software sometimes. I format C:
    (erasing the partition boot sector), then I use Robocopy
    to copy files from an archive, back onto C:. This looks
    for all the world, like a C: partition. But the problem is,
    Robocopy has no interest in partition boot sectors, so now
    I "don't got one". To fix that, I boot my WinXP installer
    CD and do "fixboot C:" or whatever drive letter the recovery
    console happens to think it is. As long as you correctly
    identify the partition, the partition boot sector will be
    defined for you.

    What other things prevent booting ? On WinXP, it would be
    boot.ini. Using another OS, you can edit boot.ini if you want.

    Let's say the original disk has C: as the third partition.
    When the cloning software clones it, it's supposed to put
    C: as the third partition of the new disk. Now, say something
    already occupies that slot. The clone software may still
    complete its job, by copying the third partition on the
    original disk, to the second partition on the new disk.

    When the new disk tries to boot, the boot.ini has the wrong
    ARC path inside it. If you hand-edit the boot.ini, you can
    change the original reference to the third partition, and
    make it equal to the new second partition location.

    [boot loader]
    [operating systems]
    multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Microsoft Windows XP Professional" /noexecute=optin /fastdetect /numproc=2

    My partition structure - four primary partitions are defined.

    First slot = WinXP <--- this is partition(1)
    Second slot = data partition
    Third slot = data partition
    Fourth slot = data partition

    Now, if my boot.ini says "partition(2)" for some reason,
    I would edit it with a text editor and change it to partition(1)
    to match the partition table definitions. One way to view these
    entries, is with PTEDIT32 for Windows. This removes the ambiguity
    about spatial position, versus slot number (the four partition
    table entries, do not have to be in spatial order - the second
    partition in, can be loaded into the third slot for example - this
    tool reports the slot number for each partition).

    While "bootcfg" exists in the recovery console, to recompute boot.ini,
    I prefer to fix it manually, rather than have this command trash
    something I don't have a backup copy of.

    The list of recovery console commands is here. I had trouble
    getting to this page on the first try.

    Anyway, those are some ideas about fiddling with repair work
    to get that pig booted. I think you have a driver present for
    the disk, because you were able to clone the disk, and that
    driver can later be used for booting the OS on the clone.
    There are ways to clone a disk, without using Windows, and
    if you were to do that, we'd have no proof the necessary
    drivers was absolute and for sure, present and ready to go.

    Paul, Nov 7, 2012
  7. R. Giggs.

    R. Giggs. Guest

    HI there!! Thanks, I am back on my old account and my PC running windows.

    Yes I did initially try the cloned drive by itself.
    Well I when I put both on I never really expected the SATA drive to boot
    but I need both on to have a look at both.

    I didn't do that, mind you I never did that when I cloned previously
    to that now faulty drive.

    I am not sure I folow this, I have both drives now, I have drives
    c: healthy (system) ntfs 60 gig
    f: healthy (active) ntff 400 gig
    (recovery) d: (healthy) fat32 6 gig
    (recovery) i: (healthy) fat32 6 gig

    basically one physical drive is c: and d: (80GB) and f: and i: (500GB)
    Well if's clone with the same system basically
    Maybe it is, I think it was just like an underscore ie _
    I never pressed return or anything like that.

    I have never used my winXP installer CD before so I am unfamillilar with
    Hmmmm... quite a lot of info to take in there, I am not at all sure what I
    am supposed
    to be doing and why.

    I am not sure what I shououdl be doing here, I am trying to think.

    1) I have some sort of windows XP disk
    2) I have a couple of 'recovery' CD's knocking about.

    I am unsure if those disks will still work in the CD drive, it's a bit hit
    and miss.
    I have just bought some new DVD's and none of them work, however I do have
    some old DVD's which do work, ie

    So I am not too sure what the best plan is now.

    I guess I can try with just the sata drive in and boot on the recovery
    and see if it can do anything?

    Anyhow I think I will try rebooting with just the sata and a recovery CD and
    what I get, at least I will find if that works or not, and I can see what I
    can do or see from there.
    R. Giggs., Nov 7, 2012
  8. R. Giggs.

    Paul Guest

    OK, here's the deal.

    Step 1: Copy old disk to new disk.
    Operation is not a success, unless new disk now boots by itself,
    without the old disk present.

    Step 2: If you know that's going to happen (you've already demonstrated a
    lack of boot, with the clone by itself), you need a safe environment
    that won't upset the clone. Booting with old and new disks connected,
    and selecting the new disk, doesn't count as a success. Booting with
    old and new connected, and using old to boot, is roughly back to
    step one. But personally, I want step 1 to be "clean", and for the
    thing to boot with the other disk absent.

    Step 3: The WinXP Recovery Console, is a boot system you can use, while only
    the clone is connected. You have your new drive connected, as well as
    an optical drive you can use for the WinXP disc. All the recent
    Windows OSes have a flavor of this. And to some extent, you can even
    use the other discs to do stuff. I can work on WinXP from a Windows 7
    recovery console from the Windows 7 DVD. But there are a few, select
    commands, that may be missing that way.

    So the reason I'm suggesting that method, is booting the old disk and working
    from there, we don't really know whether the clone has been affected by anything
    you're doing or not. Whereas, running the recovery console, that's not
    going to screw the clone up.

    And my long procedure, that's example of forensic tests or repair procedures
    a person could try, in an effort to try to figure out why the cloning
    operation is failing. You don't have to do any of that, if you know
    there is something wrong with your recipe. If you know why it's broken,
    then go find another cloning method.

    When I've used Partition Magic in the past, to copy an OS from one
    disk to another, it changed the partition slot in the MBR, and that
    screwed up the boot.ini. So that's an example of something I'd
    now be looking for, based on discovering the bug in the past. If
    the tool you used, has a "reputation", or reviews or a forum
    where you can learn stuff about it, it may already be documented
    somewhere as to why it might not work.

    Paul, Nov 7, 2012
  9. R. Giggs.

    R. Giggs. Guest

    Well, OK I have tried a few things noe of which worked unsurprisingly
    I did learn a bit.

    One thing I learned is that the windows disk I had knocking around was
    a windows98 disk!! Obviously that was not much good except I could perhaps
    installed windows98, but I restisted the temptataion fairly easilly.
    That 98 disk was from my old system, I have never even used it. The reason
    why I have
    never used it is because it is a HP system which has it's own recovery
    system on it
    in a seperate partition, so I am not at all familiar with using original
    Indeed I have no windowsXP disk, but I do have a recovery disc for XP
    which you are advised to create.

    Anyhow I tried usingthe recovery disc without any success. IT gets a bit
    confusing using it,
    it gives you 3 options, 1) normal windows, 2 recovery console 3 ubuntu
    (I had a ubuntu partion (wubi) on the machine at the time).

    Anyhow it comes up with can't frin system32/hal.dll

    I remember having that before, however I am not sure wwhetehr it is
    lookingon the recovery CD
    for it or on the hard drive, at first I thought it was lookingon the hard
    drive, but now I think it
    may be lookingon the CD (or DVD).

    There is a hal.dll on the SATA drive I just checked.

    I think it can't find the file on the recvoery CD. I just tried to explore
    recovery CD and it locked up and the drive has disappeared off the menu so
    that disc is
    pretty useless.

    hence I am pretty keen to egt it out of the drive, I will have to reboot to
    do that as the drive
    will not open.

    I will do that now before I forget and is causes problems.
    R. Giggs., Nov 7, 2012
  10. R. Giggs.

    R. Giggs. Guest

    Well I have rebooted and got the disk out. I am not 100% sure what happened
    with the
    recovery disc, I think it had problems reading itself as opposed to the hard

    I have one idea that it can't read the SATA drive. The system seems to like
    to call the
    SATA drive i: for some reason, at least when itis connect as a slave. I
    can't see what it
    is called when connected as a master of course because it will not boot up.

    I also have DOS drive so I think I will try and boot up on that if it still

    I am just wondering if there was anything else I did last time when I cloned
    the drive,
    however I suspect the problem may be related to it being a SATA drive.
    R. Giggs., Nov 7, 2012
  11. R. Giggs.

    R. Giggs. Guest

    Well it's seems my recovery DVD does not work properly, if I could burn
    I could create anther one but unfortunately I have no blank DVD's the drive
    will read or write to.
    I will have to buy another brand of DVD's and see it the drive wil read
    I really do not like optical drives they have so many problems not
    recognising disks,
    I mean I have just bought a pack of 25 and it does not read any of them!!!
    ALthough I did burn (linux image) to one of them by ignoring the fact it it
    could not read it.
    I can see it has been burned to by the marks on the disc.
    I recalled the burning software also verified the drive yet the operating
    does not recognise anything in the drive. ????
    Maybe it is usuing a different driver, but that is just typical of the
    problem you get with
    optical drives.
    R. Giggs., Nov 7, 2012
  12. R. Giggs.

    R. Giggs. Guest

    And just to prove my earlier point, "that is just typical of the
    problem you get with optical drives.", I just put the DVD in I had burned
    earlier and now it reads it despite sayingit was blank earlier!!!

    So maybe I will be able to use the new dvds after all?

    Well I tried a couple again and it does not recognise them, but I but the
    one I burned in again and it still recognises it, so may burning the other
    dead DVD wil work, I will have to try and ceate a recovery DVD and see.
    R. Giggs., Nov 7, 2012
  13. R. Giggs.

    VanguardLH Guest

    If Giggs had not first installed the SATA driver into his working OS
    running on the IDE drive and then cloned it to the SATA drive, could he
    boot using *just* the SATA drive and use the Windows install disc to do
    a Repair install to get the F6 prompt to insert a floppy to install the
    SATA driver?

    I also have to wonder about the boot.ini file. The entries for an IDE
    hard disk are different than for a SATA hard disk. The SATA hard disk
    looks like a SCSI device so its device and partition definition for an
    entry in the boot.ini would be different.

    In the boot.ini file on the system partition, the OP's old IDE drive
    would be defined using the "multi" definition to locate the physical
    disk and partition of where was the Windows boot partition. For SATA
    drives, the "scsi" definition is used. What's confusing is that I've
    read some users of SATA still have the multi() definition to find that
    device. Apparently it appears dependent on how the mobo implemented
    SATA. If it is a separate controller chip or card, it's treated as a
    SCSI-like device. If the mobo's chipset incorporates SATA support then
    it might be treated in the BIOS similarly to how it treats an IDE device
    with the result that multi() gets used for SATA. Giggs never mention
    WHAT brand and model of mobo he has; otherwise, we could go look at its
    specs, pics, and manual to see how SATA was implement there and what
    BIOS settings apply to it.

    Note to Griggs: Microsoft names the partitions in backwards order to
    what users think is intuitive. The "system" partition is where are the
    files to find the boot sector and loader while the "boot" partition is
    where is the rest of the OS. You boot using the system partition which
    then loads the rest of the OS from the boot partition. Yeah, go figure.
    In most setups, the system and boot partitions are the same (users
    typically install Windows into one partition) but they can be different.
    The boot.ini is in Windows' system partition.

    From that same MS article, the ntbootdd.sys driver needs to be used when
    the scsi() notation is used in boot.ini. During installation, and if
    scsi() is used then the installer includes the ntbootdd.sys driver. On
    my IDE-based install, that file is not in my Windows system/boot
    partition. It appears that file shows up if you *install* onto a SCSI
    device (whether actual SCSI or SATA which gets treated as SCSI). I'm
    not sure how you are going to get the ntbootdd.sys driver into an
    instance of Windows installed on and running from an IDE hard disk.
    Maybe you have to specify the SATA disk as the boot device but use the
    Windows install disc to perform a Repair install (aka in-place upgrade)
    so it sees the device is SCSI-like and includes the ntbootdd.sys driver.

    Hmm, does the Repair install (by booting using the Windows XP install CD
    and selecting Repair [twice, I believe]) issue the F6 prompt to install
    a driver? The following article indicates the F6 prompt will appear
    which is when Giggs could install the SATA driver (provided his computer
    has a floppy drive since, I believe, that's the only place that the
    installer will look at later for the driver when it prompts for it).

    I forget who made it but remember I once had a utility that would read
    the boot.ini from a partition, let you edit it locally, and then replace
    the modified copy back in the NTFS partition. If you changed the
    physical layout (disk number, partition number) by moving around hard
    disks, this let you modify the boot.ini file to get the entry to match
    there for where the hard disk was now.

    On my 6-year old computer with its old BIOS (for which there are no
    updates since Abit bit the dust), SATA support is iffy for the boot
    device. I ran into too many errors, especially in backups to other
    device types, and an occasional hang during the initial setup to detect
    drives (before Windows even starts to load) that I gave up and went back
    to using an IDE drive. However, it is unlikely that I bothered to clone
    and IDE drive onto a SATA drive and instead very likely had only the
    SATA drive powered up (with the IDE drives either powered down or their
    IDE controller disabled in BIOS so they couldn't be found) and did a
    fresh install. I prefer a fresh install rather than carrying along any
    pollution from an old installation. Alas, I don't remember what was in
    my boot.ini file regarding the entry for the SATA drive and partition
    (i.e., I don't know if it use the multi() or scsi() notation). Giggs
    might be trying to use cloning to save himself time for migrating from
    IDE to SATA but if he spends more than 3 evenings on troubleshooting
    that migration then the time would've been better spent doing a fresh
    install on the SATA drive (and with it being the only detectable drive),
    fresh installs of his apps (which means he'll probably exclude those he
    found weren't really that important), and restore his data from backups;
    however, if he doesn't have a floppy drive to use after hitting F6
    during the install to get the SATA driver installed, other workarounds
    will be needed.
    VanguardLH, Nov 7, 2012
  14. R. Giggs.

    Paul Guest

    If you burn DVDs with "multisession", they need to be closed.

    What you need at the moment, is something to analyse the discs you've
    already written. I suspect there isn't that much wrong with them.

    "It has the ability to access "deleted" data
    on multisession optical discs"

    That suggests it may be able to analyse the discs you've already
    made, and suggest what's wrong with them or what step is missing.


    When you get a new optical drive, it may need a firmware update.
    So checking the firmware situation for the processor inside
    the optical drive, is your first step.

    If no reviewers comment on the need or availability of better firmware,
    then the next step is running test media on the drive, and later scanning
    it for errors. This is an example of an error scanning tool. Several
    of the Nero utilities are free.

    Discs are protected with multi-dimensional Reed Solomon error
    correcting codes. When you see errors as in that scan, most of
    those will be correctable. If the graph gets to thousands or tens
    of thousands on the Y axis (and at that point the drive is not
    making any forward progress reading and could be "stalled"),
    then you know the errors are now uncorrectable. But in any case,
    you need to do some research on those charts, and what's a good chart
    and what's a bad chart. There are a couple forums (CDfreaks) where they review
    optical drives and present error scans for media burned with them.

    So testing the burn quality of the optical drive, is the "dialing in" phase.
    You might test "three-packs" for example. If, say, some Verbatim
    seems to work, you might buy a small spindle and give it a whirl.
    So some amount of preparation (about a week of work), is needed
    to make an optical drive "trustworthy". Some drives are crap,
    and they never get very far. Scanning for errors, after a burn,
    is one way to determine whether the drive will ever make
    good media for you. The last two drives I bought were
    very good, so with a bit of care to select a drive with
    good reviews, you can be happy with it.

    I don't think your problems are of that sort. Probably just
    a session needs closing... or something. I'm not an optical
    disc expert. Maybe IsoBuster can give some hints about it.

    Nero Discspeed is free for download, as well as being provided
    with certain Nero suites. The author (Eric Deppe), used to have
    his own web site for distribution of that software, but now
    the acquisition process is more torturous. You have to find
    a site you can trust, to download it. Depending on how bloated
    it has become, you might have to go back to an older version.
    The second link here, should pop up a save dialog.

    ( )

    This version would be a year older.

    Paul, Nov 7, 2012
  15. R. Giggs.

    Paul Guest

    Yes, you could hammer it with a repair install, but where's the fun
    in that ? My one experience with a "repair install", was that while
    it is claimed to preserve "100% of user settings", in fact it doesn't.
    The result was still annoying, and required fixing things that had
    been fixed before. And with the little details about uninstalling
    advanced versions of Internet Explorer first, can be a little

    I would sooner invest the time in simple minded forensics,
    than do all the research needed to guarantee a successful
    repair install.

    As for the SATA drive and the boot.ini ARC, there's probably a couple
    outcomes there. SATA can be virtually identical to IDE,
    with the right hardware and setting (clone from IDE to SATA
    and keep the ARC). Or in the case of some of the early add-on chips
    for SATA, it does appear as pseudo-SCSI. It made writing the
    first drivers easier.

    I don't do anything special for my SATA here. I just avoid
    AHCI/RAID and stick with Compatible or Enhanced. And on my
    first motherboards with SATA, I probably used IDE drives
    for the most part. As SATA was relatively new at the time.
    I waited a long time, before I got my first SATA drive.
    Perhaps three or four SATA motherboards went by, before my
    first SATA drive showed up.

    Paul, Nov 7, 2012
  16. R. Giggs.

    R. Giggs. Guest

    Thanks, what seems to be the case is that a lot of blank discs seem to be
    'invisible' untill something is written to them.

    I do have some older discs while will somehow disable the DVD and
    make it disappear off the list of drives somehow.

    I used "Reflect" to create a recovery CD, there were tow options, a LINUX
    one ot a windows one, The windows one requires a large down load so
    I chose the LINUX one, I was a bit surprised it was only 12mb in size
    thought. I will create a windows one later.

    Few things I tried if I recall correctly. I could run a SMART test from
    the BIOS on the SATA drive alone, so the BIOS can see the drive OK for sure.

    I also booted in an old MSDOS CD, howver I could see no other drives
    bar the one I was one, that might be a perculairty of DOS though.

    I have just had thought though, I never checked the pins on the
    drive ie for master slave or cable select. Howver it seems it does not
    matter about that for SATA drves.

    Another thing is I was allowed to only create one recovery disc on
    my PC from HP and that does not see to work. Indeed I just out that
    recover disc into the drive and now the drive has locked itself and
    off the drive list.

    I find stuff like that pretty unacceptable really, which is why I had pretty
    stopped using optical drives, it is just way too much trouble!!
    I mean there is basically some sort of design faulty when things like that

    I may try recloning the drive but that should not really make any
    R. Giggs., Nov 7, 2012
  17. R. Giggs.

    Paul Guest

    *Every* DVD I've burned here, exists on a hard drive as an ISO9660.

    If the DVD media goes bad, I can burn another.

    When making "recovery media" from a new computer, you immediately
    "image" all the freshly burned discs. Then, store those ISO9660
    files on a hard drive or two (i.e. also on your external
    backup disk drive).

    I used to have the problem you describe, with Memorex CDRW discs.
    Burn today, gone tomorrow. So it does happen. But I haven't had that
    problem recently, with the small amount of DVDs I burn (about a 6 inch


    SATA doesn't have Master, Slave, Cable_Select (~ 10 pin block).
    SATA has Force150 and Spread Spectrum, and jumpers for those
    are only necessary in specific situations (VIA chips and Mac computers).

    The BIOS shows "Master" and "Slave" designations for SATA cabling,
    because of the emulation being done there. Some aspects of the BIOS
    are still in "IDE land" when it comes to the program logic. So they
    just arbitrarily label one port a Master and another a Slave. The
    designation that way, is not a function of some jumper on the drive.


    MSDOS supports FAT file systems. Just about any version
    should support something like FAT12 or FAT16. (At least, any
    version you're likely to have in hand. Not something you
    borrowed from the Smithsonian.)

    The MSDOS floppy you make in Win98 ("sys a:"), that one
    may support FAT32.

    MSDOS doesn't support NTFS. There are things like NTFS4DOS
    that you can add to an MSDOS floppy, but they're not exactly
    convenient. LFS (long filename) is not supported, only 8.3
    shorthand for the NTFS file listings (useless for practical

    If you need to format a huge disk (2TB) to FAT32, there
    is the ridgecrop formatter. It does things, that the Microsoft
    one won't. While this page is covered in advertising (to pay for
    web site bandwidth), the actual files are here. (not essential)

    I only mention that, in case some day you happen to be running from
    MSDOS, and want an extremely large partition that MSDOS can
    read and write.


    It is possible to create a "bare, minimal" Linux OS, that fits in
    a relatively small space. The Reflect one (12MB), isn't the smallest
    possible. I think possibly DBAN had a smaller image, which seemed
    quite capable (had lots of storage drivers). When you get
    a 700MB Ubuntu disc, most of that is utilities and fancy GUI
    stuff. For fixed function boot discs, you can make the core
    quite tiny.

    I don't know what the attraction would be, of downloading the
    huge WAIK kit from Microsoft, and making some kind of WinPE
    boot CD. When the Linux one is ready to go.

    Paul, Nov 7, 2012
  18. R. Giggs.

    R. Giggs. Guest

    Well that sounds kind of interesting, however I am not really familiar with
    a lot
    of this stuff. FOr one I am not too sure what you mean about the SATA
    By the way I don't think I have a windows install disc, so I don't think I
    can do that.
    The computer came without a windows disc, instead it came with a recovery
    on the drive. You were given the chance of also make a recovery DVD, but you
    only allowed to make one of thos and the one I made seems to be faulty.

    The thing about the boot.ini sounds significant if it different on a SATA
    drive, I mean
    that sounds like you could not do a clone and expect it to work, and it
    certainly does
    not work lol. I will have to look into that, for a start I am not sure what
    a boot.ini
    is or does anyway, so I need to google that.

    OK I see wha it is it is just this file

    [boot loader]
    [operating systems]
    multi(0)disk(0)rdisk(0)partition(2)\WINDOWS="Microsoft Windows XP Home
    Edition" /fastdetect /NoExecute=OptOut /usepmtimer
    C:\CMDCONS\BOOTSECT.DAT="Microsoft Windows Recovery Console" /cmdcons

    OK sorry about that, but I will mention it now!!! (if I can remember/find

    My system

    HP Pavilion Desktop PC

    Mobo specificiation:-

    There is a very INTERESTING STATEMENT in there:-
    Please note that users cannot install OS, either WinME or Win98,

    in their SATA hard drives. Under these two OSs, SATA can only be

    used as an ordinary storage device.

    OK that does not mention XP so perhaps not as interesting as I thought!!!

    OK I will post this now and comment on the rest later so you can see the


    Not sure I follow this, I have two partition usually, the system partition
    the system is booted from and aslo a revovery partition (because I don't
    have a windowsXP
    recovery disc or whatever it is called), however I do note the partitiions
    seem to
    be named backwards ie the last partion is given the lower letter.

    R. Giggs., Nov 7, 2012
  19. R. Giggs.

    R. Giggs. Guest

    I guess I should have done that with the recovery DVD.
    I am pretty sure I loaded it recently to look at and it seemed 'all there'.
    If I an load it again I will make a fresh copy.
    I havee never had call to use the recovery CD in the past so I was not too
    bothered about it. I was not aware I was only allowed to makeone atthe time
    Also the drive worked well back then, pretty much perfect.

    So is there a master and slave SATA connector on the board?
    Mine are labled SATA 1 and SATA 2, not sure which I am using off hand.
    A quick looks says SATA 2, maybe I can try SATA 1 and see if that helps.
    It says DOS 6.22 on the disc. I wrote that on it at some point.
    I booted on the LInux disc, the 12mb one, but it asked me for the
    ISO image, which was on the disconnected drive - lol.
    ANyway I have burned that to a DVD now, although it was
    a lot smaller than the 1.6GB of data I downloaded, it was about 126MB IIRC
    which confused me.
    To be honest I do not know what reflect thing was doing after it downloaded,
    it was
    doing all sort of stuff which I had no clue about, I think there was
    about windows PE or something, I dunno. Bit of a mystery.
    AH yes that's it WinPE, no idea what that was all about.

    The Linux one was not ready to go anyway wher as far as I can see, at least
    not without the ISO image, I have that now, good job you remined me.
    It's on a DVD (I hope). I have also copied it onto the SATA drive if that
    will help.
    I could see the SATA drive when I booted from the Linux DVD.

    I do get the feeling none of this will work though because I don't think it
    sees the drive so I don't think it matters what I put on the drive.

    It is like when I tried to boot from the faulty drive in a way, same
    black/grey screen
    with an underscore at the topleft whci just sits there.

    Anyhow I will try a few things,

    1) switch to sata 1

    2) Boot on the linux dvd and give it the ISO it asks for and see what

    (not expecting much - lol).
    R. Giggs., Nov 8, 2012
  20. R. Giggs.

    Paul Guest

    R. Giggs. wrote:

    That's interesting.

    Your Southbridge (the part that connects to disks) is SB400.
    Without a heatsink on it, you might see "IXP400" printed on top.

    I cannot find a block diagram. (Not the official kind AMD
    has provided for other chips.)

    And I think I can see a potential reason (conspiracy theory).
    The chip may have made use of third-party IP blocks (and,
    more than one of them, as the USB implementation in that
    chip is inferior).

    The SB400 seems to support two IDE connectors (four drives)
    and four SATA connectors (four more drives). Not all motherboards
    solder four SATA connectors on, to access those interfaces. One
    laptop, didn't have any SATA connectors at all. So some designs
    don't use the "max ports" available.

    In my searches, I ran into a reference to the usage of a
    "SATA_SIL" Linux driver with the SB400. Now, normally, SIL would
    mean SIL3112 or Silicon Image SIL3112 chip. The fact that
    Linux driver works with SB400, implies ATI bought an intellectual
    property block straight from Silicon Image, and put it in the
    SB400. Since there are four SATA ports, there could be two
    SIL3112 blocks present.

    The SIL3112 is SATA I (150MB/sec). I cannot find any references
    to it not working with SATA II drives (a problem the VIA chips had).
    But the BIOS code shipped with SIL3112 equipped boards, there was
    a bug in there, where drives larger than 500GB or so, would cause
    the BIOS to hang during boot. I don't think the cursor would even
    flash, because the code is in a tight loop in the SIL module. For
    the affected motherboards, if you connect a SATA drive larger than
    500GB (like a 750GB or 1TB) drive, the BIOS screen freezes before
    the OS can boot.

    Silicon Image eventually issued a bug fix, but for motherboards,
    it would be the responsibility of the motherboard maker to
    incorporate a new code module into a release BIOS. On Asus
    A7N8X motherboards, it took forever, before a newer module
    of that sort was included. I think at least one person, was
    hacking in their own code update, to speed up the process.

    Now, if the SB400 uses

    1) Silicon Image IP blocks in hardware.
    2) It would be natural to use Silicon Image BIOS rom code for
    Extended INT 0x13 boot code for the BIOS. This is the
    code with the "large drive bug".
    3) The "chipset driver package", would include Silicon Image
    RAID management code. For systems with enough of the
    SATA ports soldered to the motherboard.

    If I look in , I
    see VEN/DEV for the SATA ports of SB400.

    1002 Advanced Micro Devices [AMD] nee ATI
    4379 IXP SB400 Serial ATA Controller

    And using a driver file here, I see the following
    in a SIL3112 package (file SI3112.inf). The first
    entry is for vanilla SIL3112 add-on chip. (I got
    SI3112.inf from file
    104,771 bytes.)

    PCI\VEN_1095&DEV_3112.DeviceDesc="Silicon Image SiI 3112 SATALink Controller"

    But in the same file I can find this. Which implies
    the hardware inside the SB400 SATA, is very very similar
    to the SIL3112. It also means, if you're aware of bugs
    in SIL3112 stuff, they could also affect SB400.

    PCI\VEN_1002&DEV_4379.DeviceDesc="ATI 4379 Serial ATA Controller"

    From the same file...

    ; List of controller subsystem IDs to be supported by the SI3112.mpd driver on
    ; Windows 98/ME systems.

    which tells me that you might be able to get Win98 working
    as well. (I've only got one SIL3112 here, and don't ask
    me to dig out that motherboard, which is retired :)
    I don't plan on any Win98 installs this week.)


    In any case, this has nothing to do with the cloning problems.

    If you were able to copy the old drive to the new drive, then
    a driver had to be present for that. If the cloning program
    reboots the computer to do the copy, then the SATA driver
    *could* be missing (since a "foreign" OS could be doing the
    clone copy). If the clone operation is done "hot",
    using Volume Shadow Service (VSS - macrium reflect does this),
    then you might very well be effectively proving the SATA driver
    is present.

    With the old disk booted, right now, go to Device Manager, and
    have a look for

    "ATI 4379 Serial ATA Controller"

    Maybe the string won't be exactly that, but you need some proof
    that a driver has been installed for the SATA port. And
    Device Manager is one way to look for proof. Do properties,
    check for driver files. One of the driver files might be Si3112.sys.

    One way to get to Device Manager, is Start : Run : devmgmt.msc

    Paul, Nov 8, 2012
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.