Warranty Length Not Related To Drive Life?

Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

Has anyone ever heard anyone claim that the length of a HD's warranty was
simply a marketing and price point decision by the mfg and the warranty
length has nothing to do with expected drive life? Somewhere I think I
remember someone making such a claim and a bunch of trolls tried
unsuccessfully to shoot him down?
34 answers Last reply
More about warranty length related drive life
  1. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    "Ron Reaugh" <ron-reaugh@worldnet.att.net> wrote in message
    news:aNxNc.336547$Gx4.279986@bgtnsc04-news.ops.worldnet.att.net...
    > Has anyone ever heard anyone claim that the length of a HD's warranty was
    > simply a marketing and price point decision by the mfg and the warranty
    > length has nothing to do with expected drive life? Somewhere I think I
    > remember someone making such a claim and a bunch of trolls tried
    > unsuccessfully to shoot him down?

    Well, AFAIK nothing physical changed in ATA drives when manufacturers a
    while ago decided to drop the warranty period from 3 years to 1. And I just
    read that Seagate is now going to *raise* that period to *5* years as an
    inducement to prospective buyers (something I'd certainly take into
    consideration: I deliberately chose a 3-year-warrantied drive last time I
    bought one).

    Even if manufacturers over time might be able to cut corners such that a
    drive would often fail in its second year without too much risk of
    first-year failures, it seems unlikely that the cost savings could make up
    for the resulting bad publicity. So I'd guess that drives should be in the
    bottom of their 'bathtub' curve for several years regardless of what the
    nominal warranty period is: unless they fail (even during that nominal
    period) at a fairly significant rate the savings that the manufacturer can
    realize by shortening it would seem unlikely to be large (though in such a
    cut-throat pricing environment the resulting price difference might
    noticeably affect sales, so if one does it, the rest may have to follow, and
    the same may be true for lengthening the period as Seagate is doing, since
    it would otherwise give them a unique selling point for very little price
    difference).

    Whether similar considerations apply to the terms of service (e.g., duty
    cycle) specified for the drive is less clear: there may be fairly
    noticeable savings in manufacturing a drive for light-desktop rather than
    server-style use, even leaving aside more obscure characteristics such as
    resistance to the need for re-seeking in environments subject to vibration.

    - bill
  2. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    On Tue, 27 Jul 2004 19:11:02 GMT, "Ron Reaugh"
    <ron-reaugh@worldnet.att.net> wrote:

    >Has anyone ever heard anyone claim that the length of a HD's warranty was
    >simply a marketing and price point decision by the mfg and the warranty
    >length has nothing to do with expected drive life? Somewhere I think I
    >remember someone making such a claim and a bunch of trolls tried
    >unsuccessfully to shoot him down?

    What I do understand are that hard drive warranties are often at the
    time they are manufactured and not the date of purchase so if you
    bought a new hard drive today, has a manufactor date of July 20, 2003
    and it breaks down, the manufactor may not take it even though you
    just bought it.

    Most current warranties are for one year and in my experience,
    properly manufactured hard drive should last at least a few years.
    Unfortunately they tended to build them as cheaply as possible so it's
    not unusual for hard drive to die horribly in just months or even
    weeks.

    I still have an old Seagate 20MB drive somewhere that is almost 20
    years old and still works but Maxtor 120GB hard drive that I got new
    last year went south after only 3 months. A 200 GB Western Digital
    hard drive that refused to work properly even though I have 2 more of
    the same models working just fine in the same PC.
    --
    To reply, replace digi.mon with tds.net
  3. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    "Bill Todd" <billtodd@metrocast.net> wrote in message
    news:rvWdnV9fgJxzK5vcRVn-uw@metrocast.net...
    >
    > "Ron Reaugh" <ron-reaugh@worldnet.att.net> wrote in message
    > news:aNxNc.336547$Gx4.279986@bgtnsc04-news.ops.worldnet.att.net...
    > > Has anyone ever heard anyone claim that the length of a HD's warranty
    was
    > > simply a marketing and price point decision by the mfg and the warranty
    > > length has nothing to do with expected drive life? Somewhere I think I
    > > remember someone making such a claim and a bunch of trolls tried
    > > unsuccessfully to shoot him down?

    As usual your pompous jibber below says little.

    The deal is that HD warranties were ALWAYS a marketing and price point
    decision and had little to do with expected HD life. Since the 1 year and 3
    year warranty HDs ALREADY had an expected life of over 5 years. So in a pen
    stroke a company could change its HD warranty length and even retroactively
    without great exposure SINCE the drives were ALREADY going to last for 5
    years anyway as I've always said.

    Seagate simply announced a modest cost change internally and externally
    effectively a modest price decrease, nothing more.

    > Well, AFAIK nothing physical changed in ATA drives when manufacturers a
    > while ago decided to drop the warranty period from 3 years to 1. And I
    just
    > read that Seagate is now going to *raise* that period to *5* years as an
    > inducement to prospective buyers (something I'd certainly take into
    > consideration: I deliberately chose a 3-year-warrantied drive last time I
    > bought one).
    >
    > Even if manufacturers over time might be able to cut corners such that a
    > drive would often fail in its second year without too much risk of
    > first-year failures, it seems unlikely that the cost savings could make up
    > for the resulting bad publicity. So I'd guess that drives should be in
    the
    > bottom of their 'bathtub' curve for several years regardless of what the
    > nominal warranty period is: unless they fail (even during that nominal
    > period) at a fairly significant rate the savings that the manufacturer can
    > realize by shortening it would seem unlikely to be large (though in such a
    > cut-throat pricing environment the resulting price difference might
    > noticeably affect sales, so if one does it, the rest may have to follow,
    and
    > the same may be true for lengthening the period as Seagate is doing, since
    > it would otherwise give them a unique selling point for very little price
    > difference).
    >
    > Whether similar considerations apply to the terms of service (e.g., duty
    > cycle) specified for the drive is less clear: there may be fairly
    > noticeable savings in manufacturing a drive for light-desktop rather than
    > server-style use, even leaving aside more obscure characteristics such as
    > resistance to the need for re-seeking in environments subject to
    vibration.
    >
    > - bill
    >
    >
    >
  4. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    On Tue, 27 Jul 2004 19:11:02 GMT, "Ron Reaugh"
    <ron-reaugh@worldnet.att.net> wrote:

    >Has anyone ever heard anyone claim that the length of a HD's warranty was
    >simply a marketing and price point decision by the mfg and the warranty
    >length has nothing to do with expected drive life? Somewhere I think I
    >remember someone making such a claim and a bunch of trolls tried
    >unsuccessfully to shoot him down?

    I also think this has always been a marketing exercise. Manufacturers
    offset the cost of warranty claims against the increased sales that a
    long warranty generates.

    You also have to consider how many purchasers actually bother to claim
    under warranty when a drive is 2 or 3 years old. By then, larger,
    cheaper, smaller, cooler drives will be available, and most people
    just buy a replacement. The vast majority of drives will last for more
    than 5 years whatever the warranty arrangements.

    Here in the UK, many stores offer extended warranties on consumer
    electronics items for a one off payment. Some offer to refund your
    payment if you don't make a claim. Of course, you have to actually
    *ask* for the refund after 3 or 5 years. Last time I heard, the
    proportion of purchasers that remembered / could be bothered to do
    this was under 10%.

    Best regards, Paul
    --
    Paul Sherwin Consulting http://paulsherwin.co.uk
  5. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Heheh, the wacko troll has stepped up a tree on the psychopathic ladder.
    Now he is Super Wacko.

    "Bill Todd" <billtodd@metrocast.net> wrote in message news:rvWdnV9fgJxzK5vcRVn-uw@metrocast.net...
    >
    > "Ron Reaugh" <ron-reaugh@worldnet.att.net> wrote in message news:aNxNc.336547$Gx4.279986@bgtnsc04-news.ops.worldnet.att.net...
    > > Has anyone ever heard anyone claim that the length of a HD's warranty was
    > > simply a marketing and price point decision by the mfg and the warranty
    > > length has nothing to do with expected drive life?

    > > Somewhere I think I remember someone making such a claim

    And now you know who that someone was.
    When it looks like a trap and sounds like a trap and smells like a trap,
    THEN IT USUALLY IS A TRAP.

    > > and a bunch of trolls tried unsuccessfully to shoot him down?

    And now you are part of that select group.

    >
    > Well, AFAIK nothing physical changed in ATA drives when manufacturers a
    > while ago decided to drop the warranty period from 3 years to 1. And I just
    > read that Seagate is now going to *raise* that period to *5* years as an
    > inducement to prospective buyers (something I'd certainly take into
    > consideration: I deliberately chose a 3-year-warrantied drive last time I
    > bought one).
    >
    > Even if manufacturers over time might be able to cut corners such that a
    > drive would often fail in its second year without too much risk of
    > first-year failures, it seems unlikely that the cost savings could make up
    > for the resulting bad publicity. So I'd guess that drives should be in the
    > bottom of their 'bathtub' curve for several years regardless of what the
    > nominal warranty period is: unless they fail (even during that nominal
    > period) at a fairly significant rate the savings that the manufacturer can
    > realize by shortening it would seem unlikely to be large (though in such a
    > cut-throat pricing environment the resulting price difference might
    > noticeably affect sales, so if one does it, the rest may have to follow, and
    > the same may be true for lengthening the period as Seagate is doing, since
    > it would otherwise give them a unique selling point for very little price
    > difference).
    >
    > Whether similar considerations apply to the terms of service (e.g., duty
    > cycle) specified for the drive is less clear: there may be fairly
    > noticeable savings in manufacturing a drive for light-desktop rather than
    > server-style use, even leaving aside more obscure characteristics such as
    > resistance to the need for re-seeking in environments subject to vibration.
    >
    > - bill
    >
    >
    >
  6. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    "Rod Reaugh" <ron-reaugh@worldnet.att.net> wrote:

    >"Bill Todd" <billtodd@metrocast.net> wrote in message
    >news:rvWdnV9fgJxzK5vcRVn-uw@metrocast.net...
    >>
    >> "RodReaugh" <ron-reaugh@worldnet.att.net> wrote:
    >> >
    >> > Has anyone ever heard anyone claim that the length of a HD's warranty was
    >> > simply a marketing and price point decision by the mfg and the warranty
    >> > length has nothing to do with expected drive life? Somewhere I think I
    >> > remember someone making such a claim and a bunch of trolls tried
    >> > unsuccessfully to shoot him down?
    >
    >As usual your pompous jibber below says little.
    >
    >The deal is that HD warranties were ALWAYS a marketing and price point
    >decision and had little to do with expected HD life. Since the 1 year and 3
    >year warranty HDs ALREADY had an expected life of over 5 years.

    Which is what he said, idiot.

    Classic RonnieRetard - attack someone even when they are in agreement!

    Wacko jibber snipped.
  7. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    chrisv <chrisv@nospam.invalid> wrote in
    news:60dfg01tj2euj0aonr8a9ifdejacrpbt49@4ax.com:

    > Wacko jibber snipped.

    Shouldn't the post be empty, then? There's nothing but wacko ranting coming
    from the ronnie-beast.

    --
    /Jesper Monsted
  8. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Jesper Monsted <newsspam@rootweiler.dk.invalid> wrote:

    >chrisv <chrisv@nospam.invalid> wrote in
    >news:60dfg01tj2euj0aonr8a9ifdejacrpbt49@4ax.com:
    >
    >> Wacko jibber snipped.
    >
    >Shouldn't the post be empty, then? There's nothing but wacko ranting coming
    >from the ronnie-beast.

    True, I didn't snip all of Rod^Hn's jibber. Maybe I should have. 8)
  9. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    On Thu, 29 Jul 2004 01:54:00 +0100, "PM" <none@invalid.address> wrote:

    > Are there problems with the current 15K offerings?

    A frying pan would help as those high speed drives can get really hot.
    Nothing like fresh eggs and bacon while reading emails in the morning.

    Do make sure your PC case have excellent cooling system before
    considering 10K+ drives.
    --
    To reply, replace digi.mon with tds.net
  10. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    "Impmon" <impmon@digi.mon> wrote in message
    news:og2ig01sftlnu4s9gc0l5fvv92gigcnp4l@4ax.com...
    > On Thu, 29 Jul 2004 01:54:00 +0100, "PM" <none@invalid.address> wrote:
    >
    > > Are there problems with the current 15K offerings?
    >
    > A frying pan would help as those high speed drives can get really hot.

    No, users let them get hot. It's the users responsibility to keep the HDs
    cool.

    > Nothing like fresh eggs and bacon while reading emails in the morning.
    >
    > Do make sure your PC case have excellent cooling system before
    > considering 10K+ drives.

    Or use drive coolers.
  11. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    chrisv <chrisv@nospam.invalid> wrote in
    news:ivshg0tnsinblpmfeq3aqgoussb36e50hu@4ax.com:

    > True, I didn't snip all of Rod^Hn's jibber. Maybe I should have. 8)

    Or just snip Ron entirely - let the jihad guys in Iraq have him.


    --
    /Jesper Monsted
  12. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Jesper Monsted <newsspam@rootweiler.dk.invalid> wrote:

    >chrisv <chrisv@nospam.invalid> wrote:
    >>
    >> True, I didn't snip all of Rod^Hn's jibber. Maybe I should have. 8)
    >
    >Or just snip Ron entirely - let the jihad guys in Iraq have him.

    He'd fit right in with those wackos. Pretty soon we'd have terrorist
    attacks on SCSI controller makers, not to mention the heretics at
    storagereview.com... 8)
  13. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    "Ron Reaugh" <ron-reaugh@worldnet.att.net> writes:

    > Has anyone ever heard anyone claim that the length of a HD's warranty was
    > simply a marketing and price point decision by the mfg and the warranty
    > length has nothing to do with expected drive life?

    It cannot be *totally* unrelated. A manufacturer doesn't want a whole
    lot of returns. You can expect that *most* drives will *at least*
    make it through the warranty period.

    A hard drive is free to continue operation well past the warranty
    expiration time. As far as I can tell, the 3 to 1 year standard ATA
    warrenty shorting has had no impact on actual drive lifetimes.

    > Somewhere I think I
    > remember someone making such a claim and a bunch of trolls tried
    > unsuccessfully to shoot him down?
    >
    >

    --
    Johan KULLSTAM
  14. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    chrisv <chrisv@nospam.invalid> wrote in
    news:09ikg0husvp0ts3p0tr9jlkti7f8b3rjda@4ax.com:

    > He'd fit right in with those wackos. Pretty soon we'd have terrorist
    > attacks on SCSI controller makers, not to mention the heretics at
    > storagereview.com... 8)

    Actually, i'd prefer if they'd just implant a large storage system in his
    rectal passage, but seeing him get blown up by a tomahawk missile could be
    fun, too :)

    --
    /Jesper Monsted
  15. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    "Impmon" <impmon@digi.mon> wrote in message
    news:og2ig01sftlnu4s9gc0l5fvv92gigcnp4l@4ax.com...
    > On Thu, 29 Jul 2004 01:54:00 +0100, "PM" <none@invalid.address> wrote:
    >
    > > Are there problems with the current 15K offerings?
    >
    > A frying pan would help as those high speed drives can get really hot.
    > Nothing like fresh eggs and bacon while reading emails in the morning.
    >
    > Do make sure your PC case have excellent cooling system before
    > considering 10K+ drives.
    > --
    > To reply, replace digi.mon with tds.net

    Indeed they can - I have a Lian Li case which, so far, appears upto the job.
    I was looking at getting CD bay drive coolers for any 15k drives I get..

    Thanks for your advice

    PM
  16. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    > Ah, the SCSI drives are Quantum - they've always been good, in my
    > experience :)

    Same here, but now they are a part of Maxtor and, consequently, I wasn't
    sure if you meant they had deteriorated.

    Thank you.

    PM
  17. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    "Johan Kullstam" <kullstj-nn@comcast.net> wrote in message
    news:87zn5hh0xu.fsf@sophia.axel.nom...
    > "Ron Reaugh" <ron-reaugh@worldnet.att.net> writes:
    >
    > > Has anyone ever heard anyone claim that the length of a HD's warranty
    was
    > > simply a marketing and price point decision by the mfg and the warranty
    > > length has nothing to do with expected drive life?
    >
    > It cannot be *totally* unrelated. A manufacturer doesn't want a whole
    > lot of returns. You can expect that *most* drives will *at least*
    > make it through the warranty period.

    Do ya think!

    > A hard drive is free to continue operation well past the warranty
    > expiration time. As far as I can tell, the 3 to 1 year standard ATA
    > warrenty shorting has had no impact on actual drive lifetimes.

    Do ya think!
  18. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    In article <BWyNc.142036$OB3.105819@bgtnsc05-news.ops.worldnet.att.net>,
    Ron Reaugh <ron-reaugh@worldnet.att.net> writes

    >So in a pen
    >stroke a company could change its HD warranty length and even retroactively
    >without great exposure SINCE the drives were ALREADY going to last for 5
    >years anyway as I've always said.

    Yes, just as you "always said" there was nothing wrong with the Deskstar
    75GXPs. We know better now, don't we?

    --
    A. Top posters.
    Q. What's the most annoying thing on Usenet?
  19. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Mike Tomlinson wrote:

    > In article <BWyNc.142036$OB3.105819@bgtnsc05-news.ops.worldnet.att.net>,
    > Ron Reaugh <ron-reaugh@worldnet.att.net> writes
    >
    >>So in a pen
    >>stroke a company could change its HD warranty length and even
    >>retroactively without great exposure SINCE the drives were ALREADY going
    >>to last for 5 years anyway as I've always said.
    >
    > Yes, just as you "always said" there was nothing wrong with the Deskstar
    > 75GXPs. We know better now, don't we?

    What's this "we", White Man? You have a mouse in your pocket?

    --
    --John
    Reply to jclarke at ae tee tee global dot net
    (was jclarke at eye bee em dot net)
  20. Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

    In article <cefchi04r0@news1.newsguy.com>, J. Clarke
    <jclarke@nospam.invalid> writes

    >What's this "we", White Man?

    It's the royal 'we', old chap.

    > You have a mouse in your pocket?

    It's neither a mouse nor a gun. It means I'm pleased to see you.

    --
    A. Top posters.
    Q. What's the most annoying thing on Usenet?
  21. Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

    Mike Tomlinson wrote:

    > In article <cefchi04r0@news1.newsguy.com>, J. Clarke
    > <jclarke@nospam.invalid> writes
    >
    >>What's this "we", White Man?
    >
    > It's the royal 'we', old chap.

    So of what are you the king?

    >> You have a mouse in your pocket?
    >
    > It's neither a mouse nor a gun. It means I'm pleased to see you.

    If you knew why you were being asked if there was a mouse in your pocket you
    would be a good deal less pleased, Round Eyes.
    >

    --
    --John
    Reply to jclarke at ae tee tee global dot net
    (was jclarke at eye bee em dot net)
  22. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Mike Tomlinson <mike@NOSPAM.jasper.org.uk> wrote in
    news:uki49LDpXyCBFwos@jasper.org.uk:

    > Yes, just as you "always said" there was nothing wrong with the
    > Deskstar 75GXPs. We know better now, don't we?

    Oooh, those were almost as nice as the 36LZX 18gig ultrastars. I had 50
    servers delivered, each with one of those drives. I had about 60 exchanges
    on them before my vendor started giving me seagates :)


    --
    /Jesper Monsted
  23. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    In article <Xns9537E7E95F00Fnewsspamrootweilerdk@62.243.74.163>,
    Jesper Monsted <newsspam@rootweiler.dk.invalid> wrote:
    > Mike Tomlinson <mike@NOSPAM.jasper.org.uk> wrote in
    > news:uki49LDpXyCBFwos@jasper.org.uk:
    > > Yes, just as you "always said" there was nothing wrong with the
    > > Deskstar 75GXPs. We know better now, don't we?

    > Oooh, those were almost as nice as the 36LZX 18gig ultrastars. I had 50
    > servers delivered, each with one of those drives. I had about 60 exchanges
    > on them before my vendor started giving me seagates :)

    HP Surestore 2000. Six drives, ten failures, we finally gave up when we
    called HP for a replacement and got routed to some noname 'support'
    organization. Apperently they'd sold the whole drive business, lock, stock,
    and warranty.

    --
    I've seen things you people can't imagine. Chimneysweeps on fire over the roofs
    of London. I've watched kite-strings glitter in the sun at Hyde Park Gate. All
    these things will be lost in time, like chalk-paintings in the rain. `-_-'
    Time for your nap. | Peter da Silva | Har du kramat din varg, idag? 'U`
  24. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Ron Reaugh:
    > Has anyone ever heard anyone claim that the length of a HD's warranty was
    > simply a marketing and price point decision by the mfg and the warranty
    > length has nothing to do with expected drive life? Somewhere I think I
    > remember someone making such a claim and a bunch of trolls tried
    > unsuccessfully to shoot him down?

    Such claim would be wrong...

    As with any product the percentage of failures will
    increase over time; e.g. .5 % during first, 1 % du-
    ring second and 2 % during third year of life.

    Thus increasing the warranty time does indeed cost
    money...

    Manufactures will work hard to get the failure rate
    as much down as possible (less cost, competitive
    advantage).

    --
    Michael Giegerich
  25. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Nonsense. It's the opposite for the first three years: over 1% in first year,
    under 1% in next two.

    "Michael Giegerich" <migieger@web.de> wrote in message
    news:5i2rec.gc22.ln@luva.home...
    >
    > As with any product the percentage of failures will
    > increase over time; e.g. .5 % during first, 1 % du-
    > ring second and 2 % during third year of life.
    >
    > Thus increasing the warranty time does indeed cost
    > money...
    >
  26. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Eric Gisin:
    > Nonsense. It's the opposite for the first three years: over 1% in first year,
    > under 1% in next two.

    Assuming you're right, after how many years the failure
    rate would have come down to 0 % then?

    Believe me, the rate goes up ... (until no drive will
    work anymore, i.e. 100 % failures; just wait - may
    take a few years, but it will go there :-)

    > "Michael Giegerich" <migieger@web.de> wrote in message
    > news:5i2rec.gc22.ln@luva.home...
    > >
    > > As with any product the percentage of failures will
    > > increase over time; e.g. .5 % during first, 1 % du-
    > > ring second and 2 % during third year of life.
    > >
    > > Thus increasing the warranty time does indeed cost
    > > money...

    --
    Michael Giegerich
  27. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Michael Giegerich wrote:

    > Eric Gisin:
    >> Nonsense. It's the opposite for the first three years: over 1% in first
    >> year, under 1% in next two.
    >
    > Assuming you're right, after how many years the failure
    > rate would have come down to 0 % then?
    >
    > Believe me, the rate goes up ... (until no drive will
    > work anymore, i.e. 100 % failures; just wait - may
    > take a few years, but it will go there :-)

    Common phenomenon with electronics of any kind--a high rate of "infant
    mortality" in the first few months of operation, then a relatively low
    failure rate until pieces start dying of old age.

    >> "Michael Giegerich" <migieger@web.de> wrote in message
    >> news:5i2rec.gc22.ln@luva.home...
    >> >
    >> > As with any product the percentage of failures will
    >> > increase over time; e.g. .5 % during first, 1 % du-
    >> > ring second and 2 % during third year of life.
    >> >
    >> > Thus increasing the warranty time does indeed cost
    >> > money...
    >

    --
    --John
    Reply to jclarke at ae tee tee global dot net
    (was jclarke at eye bee em dot net)
  28. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    In article <fucrec.5r22.ln@luva.home>, Michael Giegerich
    <migieger@web.de> writes

    >Believe me, the rate goes up ... (until no drive will
    >work anymore, i.e. 100 % failures; just wait - may
    >take a few years, but it will go there :-)

    It's usually taken for granted that the failure rate is plotted as a
    "bathtub" curve. Here's an attempt at ASCII - view with a monospaced
    font:


    r |
    e |
    l |
    i | ___________________________________________
    a | / \
    b | / \
    i | / \
    l | / \
    i | / \
    t | / \
    y | / \
    +--------------------------------------------------------------
    time
    ^ ^ ^ ^
    1 2 3 4


    The x-axis is time, the y-axis is failure rate.

    What this says is that drive failure is high in their infancy (between
    points 1 and 2). If they see out infancy, they tend to continue working
    (between points 2 and 3) until they reach some point at which they wear
    out, then failure rates increase (between points 3 and 4.)

    Traditionally, the gap between points 2 and 3 has been three years for
    consumer-level (=IDE) drives, and five years for high-end (=SCSI)
    drives, coinciding with the typical warranty period offered on these
    devices. Of course, this is generalising wildly and depends very much
    on how the drive is treated prior to installation, its operating
    conditions, etc. etc.

    --
    ..sigmonster on vacation
  29. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Mike Tomlinson <mike@NOSPAM.jasper.org.uk> wrote in
    news:aSjVg3C08UEBFwdH@jasper.org.uk:

    > the y-axis is failure rate

    ....which means the 'curve' is upside down, but I can guess what you
    intended: reliability = 1/failure rate ???
  30. Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

    "J. Clarke" <jclarke@nospam.invalid> wrote in news:ces2j01ae7
    @news3.newsguy.com:
    > Common phenomenon with electronics of any kind--a high rate of "infant
    > mortality" in the first few months of operation, then a relatively low
    > failure rate until pieces start dying of old age.

    Same thing i've seen in most drives. Once you weed out the badly
    manufactured ones, you'll see little or no failures. After several years of
    non-stop operation, turning them off will make a buttload of them die at
    once, though.

    --
    /Jesper Monsted
  31. Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

    In article <ces2j01ae7@news3.newsguy.com>,
    J. Clarke <jclarke@nospam.invalid> wrote:
    >Michael Giegerich wrote:
    >
    >> Eric Gisin:
    >>> Nonsense. It's the opposite for the first three years: over 1% in first
    >>> year, under 1% in next two.
    >>
    >> Assuming you're right, after how many years the failure
    >> rate would have come down to 0 % then?
    >>
    >> Believe me, the rate goes up ... (until no drive will
    >> work anymore, i.e. 100 % failures; just wait - may
    >> take a few years, but it will go there :-)
    >
    >Common phenomenon with electronics of any kind--a high rate of "infant
    >mortality" in the first few months of operation, then a relatively low
    >failure rate until pieces start dying of old age.

    Nice theory, but, hard drives have a mechanical component, too. Anybody
    got numbers on what dies first/most often?

    >>> "Michael Giegerich" <migieger@web.de> wrote in message
    >>> news:5i2rec.gc22.ln@luva.home...
    >>> >
    >>> > As with any product the percentage of failures will
    >>> > increase over time; e.g. .5 % during first, 1 % du-
    >>> > ring second and 2 % during third year of life.
    >>> >
    >>> > Thus increasing the warranty time does indeed cost
    >>> > money...

    --
    Tom Wilson | Those who do not understand Unix are condemned to
    SAS Institute Inc. | reinvent it, poorly.
    sasctw@unx.sas.com | -- Henry Spencer
    Opinions are mine, not of SAS Institute Inc.
  32. Archived from groups: comp.sys.ibm.pc.hardware.storage (More info?)

    Tom Wilson wrote:

    > In article <ces2j01ae7@news3.newsguy.com>,
    > J. Clarke <jclarke@nospam.invalid> wrote:
    >>Michael Giegerich wrote:
    >>
    >>> Eric Gisin:
    >>>> Nonsense. It's the opposite for the first three years: over 1% in first
    >>>> year, under 1% in next two.
    >>>
    >>> Assuming you're right, after how many years the failure
    >>> rate would have come down to 0 % then?
    >>>
    >>> Believe me, the rate goes up ... (until no drive will
    >>> work anymore, i.e. 100 % failures; just wait - may
    >>> take a few years, but it will go there :-)
    >>
    >>Common phenomenon with electronics of any kind--a high rate of "infant
    >>mortality" in the first few months of operation, then a relatively low
    >>failure rate until pieces start dying of old age.
    >
    > Nice theory, but, hard drives have a mechanical component, too. Anybody
    > got numbers on what dies first/most often?

    It's not "theory", it's practical experience. What difference does it make
    whether the failure is in a mechanical component or a part of a circuit?

    >>>> "Michael Giegerich" <migieger@web.de> wrote in message
    >>>> news:5i2rec.gc22.ln@luva.home...
    >>>> >
    >>>> > As with any product the percentage of failures will
    >>>> > increase over time; e.g. .5 % during first, 1 % du-
    >>>> > ring second and 2 % during third year of life.
    >>>> >
    >>>> > Thus increasing the warranty time does indeed cost
    >>>> > money...
    >

    --
    --John
    Reply to jclarke at ae tee tee global dot net
    (was jclarke at eye bee em dot net)
  33. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    >Not really. Spun up a pair of 380MB ESDI drives last night. Anyone
    >remember them? Seem fine, once the grease warmed up a bit and got
    >spread around a little :)
    >
    >--
    >Paul Repacholi 1 Crescent Rd.,
    >+61 (08) 9257-1001 Kalamunda.

    Still using a pair of 330MB's - 1 FH and 1 HH - in a DOS/WFWG311 system.
    Used once per week only though. Manufacturing date was either 1989 or 1991

    Kev
  34. Archived from groups: comp.sys.ibm.pc.hardware.storage,comp.periphs.scsi,comp.arch.storage (More info?)

    Remember ESDI, blimey, yup.. used to use 150MB ESDI drives many years ago.
    And going back further... 20MB MFM and RLL drives.

    Anyone remember the old AT&T 3B15 150MB drives... you know the ones that
    were 3ft long and had disc brakes in them?

    Mark


    <KevH@yorkie.dabsolLL.co.uk> wrote in message
    news:4121a284$0$37064$ed2619ec@ptn-nntp-reader01.plus.net...
    >
    > >Not really. Spun up a pair of 380MB ESDI drives last night. Anyone
    > >remember them? Seem fine, once the grease warmed up a bit and got
    > >spread around a little :)
    > >
    > Still using a pair of 330MB's - 1 FH and 1 HH - in a DOS/WFWG311 system.
    > Used once per week only though. Manufacturing date was either 1989 or
    1991
    >
Ask a new question

Read More

Storage