On 7/4/2010 12:29 AM, C.H.Ling wrote:
> If you representing the typical consumer than the world of quality has ended,
> everyone just find the cheapest product to buy and rely on warranty to cover
> his purchase and think it is the right way.
I only wrote about a specific product, Hard Drives. I don't use the same
principle for everything. It depends on the
nature of the product. My late brother, who ended up as a free lance designer
of portable computers, was both MUCH more
knowledgeable about digital electronics than I and better at math and
statistics.
He first gave me this idea when he did some analysis of auto frequency of
repair data that Consumer Reports magazine
collects and publishes. He proved to his own satisfaction, and he was picky
about such things, that choosing a car based
on that data did nothing significant to increase one's chance of getting a car
that would require fewer repairs than
another brand/model.
The problem occurs, and this is true, only when frequency of failure/repair and
frequency of purchase are quite low . If
I were buying HDs for a server farm, my approach would be different.
> Fortunately most people still buy according to the reputation of the
> manufacturer if they want good
> quality product. Of couse it doesn't count many buy according to fashion,
> look and new features....
Recently, I decided to buy a new toaster oven, to replace one over 30 years
old. I read reviews, both by "pros" on web
review sites and by individual users. I spent a lot of time on it, and am both
delighted with the performance so far and
pretty confident it will last a long time.
That's a product where the basic technology changes very slowly and most model
changes are essentially marketing driven
and mostly or completely cosmetic. For example, I bought exactly the same
product, except for a top/side cover that's
painted, instead of stainless steel, from a discount chain, rather than a
higher end retailer.
> You are TOTALLY wrong if you think the manufacturer have not idea on the life
> time of a product until there are many thousands in service.
I must respectfully disagree. I'm sure you are right for most manufacturers,
most of the time. Nevertheless, Toyota,
which had he best record and reputation for reliability for many years, has
recently had two major recalls for
significant problems, and looks like it will have a third shortly, for stalling
engines in Lexuses. I wouldn't be
surprised if it costs them $Billions.
I'm pretty sure they thought their parts and systems were safe and reliable
both in design and after prototype testing.
Unfortunately, it wasn't true. The world simply isn't entirely controllable.
> A reasonable good manufacturer knows very well each component they used, such
> as the life time of the capacitor on their product, not only the well known
> short life electrolyte cap. even the high quality film cap will die just in a
> few years in the power circuit, it die with cap. value decrease and in the
> worse cases even short circuit.
>
> Failure cames in three ways: [snip excellent descriptions]
I don't know what kind of lighting manufacturer you worked for. Most lighting
manufacture is based on very old, well
known technology, as far as I know. If I buy a new ballast for the fixture in
my basement, I'll bet it is very little
different, if at all, from the 20+ year old one it replaces. The principles you
list above apply very well there.
In other fields, new designs must be made and tried all the time, perhaps
several times a year. HDs recently went
through a big change, to vertical magnetic domain recording techniques, to get
up to and over 1TB in 3.5" drives. some
of the earlier ones had more troubles than the older designs. Now they seem to
be pretty stable.
I don't know how carefully you read my original post. I suggested buying one
step below the latest designs - for exactly
those reasons. To quote Chuck on the same subject:
On 7/4/2010 4:57 AM, Chuck Norcutt wrote:
> I do have one rule that I follow that I think makes for a somewhat better
> chance of getting a reliable drive... don't get the highest capacity drive
> available. Although some capacity increases are done
> with rather conventional methods (I think the recent increase from 1.5TB to
> 2.0TB was done using 5 platters rather than 4) sometimes it's done with very
> new technology and under competitive pressure. Working OK in the lab and
> working OK off the end of a high volume manufacturing line perhaps with
> totally new and untested components are different things. I prefer to give
> the technology a bit of time to settle.
Consider a drive design that switches from 4 to 5 platters in the same form
factor. That will means many components that
have to be redesigned, spindle, head placement arms, heads themselves, perhaps
platters, and undoubtedly other
mechanical parts. In a competitive market, it is impossible to throughly test
these new designs long enough and in
enough environments to be sure of the level of reliability that the last
generation of the technology has by then achieved.
One has to do one's best in design, manufacturing design and testing and get to
market before others have eaten into
your market share. It's a delicate, risky game. Sometimes, like Toyota, you
lose.
As Chuck and I suggest, when I last bought HDs, i went with 1 TB when 1.5 TB
ones had been on the market for a few
months, and apparently were working.
> In real life every company could do differently and concentrated on different
> areas. Cost reduction pressure is high and changing to a smaller supplier
> with less overhead/cost is a common way, this always cause problem. So if you
> just looking for the cheapest product you will end up with more
> trouble.
As I said above, I think that depends on the product and the market.
Moose
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|