On Wed, 27 Apr 2016 19:26:34 +0000, David wrote:
> On Wed, 27 Apr 2016 19:39:54 +0100, Java Jive wrote:
>
>> You might be interested to know that recently Seagate have had a
>> particularly poor reputation:
>>
https://www.backblaze.com/blog/what-hard-drive-should-i-buy/
>>
>> On Wed, 27 Apr 2016 19:00:04 +0100, Peter Duncanson
>> <
ma...@peterduncanson.net> wrote:
>>>
>>> I use Seagate drives
>
> Of course, this could be why the drive is so cheap!
This *is* why the drive is so cheap!
>
> I've re-purposed a 500GB Seagate FreeAgent drive for the moment and
> formatted it using the Humax menu.
>
Did you spot the "S" suffix letter in the model number? That means it
was a "Special", basically identical to "S"less models except for a hard
coded 10 or 15 minute power saving spin down (not optional as is standard
in all IDE and SATA drives that inherited this feature from the early IDE
laptop drives -programmable 0-15 minute time out with 0 meaning disabled).
You might suffer exacerbated problems with the "Exact Record" FreeView
feature, like an extra 8 or 10 seconds delay starting a recording due to
the spin up time.
I had a Medion rebranded FreeAgent 500GB drive which kept pissing me off
due to the spin up delay almost every time I needed to access it. I
googled to find a way of turning off this power saving feature but
eventually discovered there was no way to alter it and finally repurposed
it into a PC that had been afflicted with that ironically named Vista OS.
It was a "Match made in Heaven", an OS that just wouldn't stop frigging
around with its HDD and an HDD that needed its pointless spin down
activity stopped dead in its tracks. :-)
I wouldn't have thought a DVR and a Seagate "Special" were a
particularly favourable match in all honesty. :-(
"What was wrong with using Spin Down power saving?" I hear you ask. Good
question and one where the answer gives you a very good reason to boycot
any Seagate product.
In short, Seagate decided, rather misguidedly, to use *Power Saving* as
a method to mitigate against the style over function induced total lack
of ventilation in the FreeAgent external USB enclosures which caused
their already hot running drives to fatally overheat (Click of Death,
anyone?) unless a small desk fan was used to improve the cooling efficacy
of the enclosure by a few measly, but critical, degrees[1] (possibly
reducing the drive temperature by as much as 5 to 10 degrees C) during a
protracted backup or restore operation (a mere 100GB's worth of movie
files, just 20% of the drive's capacity, typically requiring just under
an hour of run time to complete - more than an ample amount of time for
the drive to well and truly cook itself).
Such protracted backup/restore operations totally defeat Seagate's
strategy to avoid fatal temperature limits being exceeded since the spin
down *power saving* will do absolutely nothing under this not untypical
usage case. What's worse, if the run times are reasonably short (say 10
or 15 minutes or so)[2], the drive will experience pretty extreme
temperature cycling which is one of the main drivers of premature failure
in disk drives.
Only a total blithering Idiot could have chosen such a mitigation
strategy to compensate for the piss poor design of the external USB
enclosures used by the FreeAgent model range rather than do the right
thing and have the enclosures completely redesigned to cope with the
unusually higher than normal running temperatures of their primary
product.
When a manufacturer can display such stunning stupidity in their
understanding of their own product's limits (and in another case of
stupidity, claim that not all of their product range is validated for
24/7 NAS/Server usage - once a drive has been spinning for just an hour,
it's about as hot as it can get. An extra 23 hours is unlikely to add
more than another Deg C temperature rise), you do have to question the
reliability of *all* of their product range. The BackBlaze data, afaiac,
is just the 'Icing on The Cake' in the array of evidence against using
Seagate product.
[1] Fortunately, I realised there was a real risk of overheating the
drive whilst my first backup session had been running for ten minutes or
so with an ETA of another hour or so to go. I quickly set up a floor
standing fan (not having a desk fan to hand) to waft a cooling breeze
over the drive enclosure for the remainder of the backup session.
Thereafter, any further protracted backup/restore sessions would see me
setting up the floor standing fan.
When I finally got a chance to check the SMART temperature stats after
fitting it into a Vista afflicted Desktop PC, it had logged a maximum of
60 deg C, right on the upper limit. If I hadn't taken the precaution of
additional cooling, it's likely to have exceeded this limit by at least 5
degrees which, in most cases, is enough to generate bad sectors.
Vista would have kept it spinning non-stop which was a good thing now
that the drive was no longer 'wrapped up in a FreeAgent blanket' and
could 'breath' much more easily in the less confining environment of a
ventilated desktop PC case.
[2] As well as the 'session time', there's also the 10 or 15 minute run
on time of the spin down time out counter to add onto the total. Whilst
this idling time uses slightly less power than normal read/seek activity
and even slightly less power again than during write/seek activity, the
temperature is unlikely to reduce by more than a degree or two (assuming
the drive has reached its limiting temperature by this stage - a 15
minute backup session may see the drive temperature continue to increase
during this spin down idle time out period).
--
Johnny B Good