Ken Lucke wrote:

I have surely appreciated the crazy explosion in flash memory capacity,
but hard drives are not keeping pace. I was just looking for a drive
today and was disappointed that hard drive prices haven't fallen more
since I bought my last drive a few years ago.

For instance:
<http://www.pcworld.com/article/id,77543-page,1/article.html>

"Less than a year ago, a $300, 80GB desktop drive was considered huge;
today you can find a 160GB drive for the same price."

That was in January of 02'! According to this, between January of '01
and January of '02 capacity doubled to 160 GB while the price remained
constant. If that trend had continued, you could now buy a 5 terabyte
drive for $300. Instead it's $300 for 750GB.

Consulting pricewatch from 4 years ago, I came up with the following
annual growth rates:

Flash: 138%
Hard drive: 32%
RAM: 19%

For comparison, Moore stated his "Law" at 100% per year in 1965,
and at 41% (doubling every other year) in 1975. It is often quoted at
doubling every 18 months, which would be 59% per year.

Yeah, I'm spoiled. But compared to last century, this one isn't doing
so hot. (Except for Flash).





For the interested, here's the data I used:

Look at pricewatch on the internet archive from 4 years ago:
<http://web.archive.org/web/20030128022327/http://www.pricewatch.com/>
A 120GB drive was $117, and the maximum available size was 250GB.

Now, 4 years later, that same $120 will buy you a 400 GB drive and the
max available size is 750GB.
<http://www.pricewatch.com/hard_drives/>
So in 4 years, price is basically constant while capacity has gone up by
a little more than a factor of 3.

Meanwhile, using the same sources, 4 years ago a 256 MB usb flash drive
was $75. Today for $79 you get an 8 GB flash. That's a factor of 32!

In 2002, 1GB of PC133 RAM would set you back $144.
In 2007, for the same price you can get 1 GB ddr2-1066 1gb or 2GB
DDR2-400. The cheapest 1GB module now is PC100 for $60.
So in 4 years, RAM has only doubled in capacity for the same price.


<http://en.wikipedia.org/wiki/Moore's_law>

timeOday <timeOday-UNSPAM@theknack.net> wrote:
That isnt as true with the best after rebate prices.

timeOday wrote:

There has been a lot of consolidation in the hard drive industry,
resulting in an oligopoly. Without competitive pressures, prices
will not decline at the same pace.


--
The e-mail address in our reply-to line is reversed in an attempt to
minimize spam. Our true address is of the form che...@prodigy.net.

CJT <abujlehc@prodigy.net> wrote:
Have fun explaining how come you dont get the same
effect with flash ram which has the same problem.

CJT wrote:
Kinda gotta consider where you shop as well.

Right now, a retail-boxed internal 300GB/16MB Maxtor SATA drive (exact model
unspecified) would cost me CDN$236(!!!) at Staples... meanwhile my regular
retailer sells a 500GB/16MB SATA-II packed only in an antistatic bag, for
CDN$169 (they don't even list a 300, but a 250 is $85-$95). That's a lotta
extra cash for a cardboard box and "installation instructions".

Also, check your warranties. If Maxtor is anything like Intel you get really
poor warranties with OEM (plastic bag only) items. If you buy a retail CPU
and a retail motherboard Intel gives you the warranty through them for 3
years on each. If you go OEM they don't warrant anything and leave it up to
the place you bought it. For example an OEM CPU from TigerDirect has a 1
year warranty and it is through them.

So make sure that the extra you pay isn't for warranty. Though honestly the
price difference still doesn't warrant that. If a drive or CPU or
Motherboard is going to fail it would surely do it within a year. Just make
sure that you do get a warranty and if it is handled by the place you buy
from that you can count on them being in business for at least the length of
warranty.

TGC Ltd.


"Matt Ion" <soundy106@gmail.com> wrote in message
news:uNIGh.1221431$5R2.461484@pd7urf3no...

On Sat, 03 Mar 2007 22:44:15 -0700, timeOday
<timeOday-UNSPAM@theknack.net> wrote:

Moore didn't state a "law."
He made an observation.
And the observation was concerning chips, and had nothing to do with
hard drives.
Trying to compare the density of transistors on chips and aerial
density of hardd rives is comparing apples and potatos.

--
John Edwards was warned by Democrats
in Nevada Thursday that his support
for a bill to ban betting on college
sports will cost him the state. You
have to love Nevada. It's the only
place where a personal injury lawyer
has the moral high ground.

On Mon, 05 Mar 2007 14:24:44 -0700, Bill Funk <BigBill@there.com> wrote:

In comp.sys.ibm.pc.hardware.storage Bill Funk <BigBill@there.com> wrote:
Indeed. The ''law'' was made out of his observations by the press, I
beliveve.

More like apples and fish, really.

Arno

In comp.sys.ibm.pc.hardware.storage AZ Nomad <aznomad.2@premoveobthisox.com> wrote:

To some degree for capacities. Not at all for interface speeds.
And they are not his observations for HDDs, since he did not
observe HDDs at all.

Arno

On 5 Mar 2007 22:15:44 GMT, Arno Wagner <me@privacy.net> wrote:



Know of many consumer hard drives that did 80MB/s in 1995? In 1997, a $500
10Krpm scsi-3 drive was only good for 3-4MB/s. In 1985, consumer PC hard
drivess were good for maybe 500KB/s. Right now I'm getting 80MB/s on a $140
500GB drive. Sure sounds like a geometric progression to me.

Pretty much, drive speeds have gone up in step with drive capacities.
You get greater capacity either with more platters or a higher data density.
Both translate to greater throughput given the same rotational speed.

I was getting 3.5 MB/s STR out of 500 MB 3600 RPM (?) IDE Quantum ($350) in
1994. This was on a IDE card connected to VESA bus.

"AZ Nomad" <aznomad.2@PremoveOBthisOX.COM> wrote in message
news:slrneupa7o.rup.aznomad.2@ip70-176-155-130.ph.ph.cox.net...

AZ Nomad wrote:

Uh, adding platters does nothing for STR. More throughput, but all of
the data about speed in this thread has been about STR.

{Long long ago, in a galazy near near to us, more platters meant higher
STRs for a special class of expensive HDs that did parallel reads from
multiple heads. But everything made today reads (and writes) one head
at a time. If you want RAID0, the economics of HDs make it cheaper to
deploy multiple HDs than multiple heads on the same HD, at least for now.}

HD STR has never maintained a 60% CAGR over any period of a decade or so
for mainstream HDs AFAIK.
--
Cheers, Bob

On Mon, 05 Mar 2007 21:39:35 GMT, AZ Nomad
<aznomad.2@PremoveOBthisOX.COM> wrote:

Lead balls dropped in air acellerate at a certain rate.
Do you, therefore, expect feathers dropped in air to acellerate at the
same rate?
But you *observed* the lead balls.
Moore's observation had/has nothing to do with hard drives.

--
Hillary Clinton went to a civil
rights ceremony in Selma Sunday
and spoke to a local black Baptist
Church congregation. She mentioned
her husband at the top of her
speech. Whenever you face a tough
crowd you have to open with your
best joke.

In comp.sys.ibm.pc.hardware.storage Bill Funk <BigBill@there.com> wrote:
Good analogy.

Arno

On Tue, 06 Mar 2007 09:07:18 -0700, Bill Funk <BigBill@there.com> wrote:

On Tue, 06 Mar 2007 21:46:16 GMT, AZ Nomad
<aznomad.2@PremoveOBthisOX.COM> wrote:

Not at all.
Moore's observation was about transistor counts on chips.
That's what he observed.
He made no observation about hard drives.
Chips=lead ball.
Hard drive=feathers.
Get it?

--
Hillary Clinton went to a civil
rights ceremony in Selma Sunday
and spoke to a local black Baptist
Church congregation. She mentioned
her husband at the top of her
speech. Whenever you face a tough
crowd you have to open with your
best joke.

timeOday wrote:
<edited, for brevity>

Hello, timeOday:

Hard disks have been around far longer, than flash memory. Hence, it's
unrealistic to expect the same kinds of rapid price/performance gains,
from such mature technology.


Cordially,
John Turco <jtur@concentric.net>

On 06 Mar 2007 22:26:03 EST, John Turco <jtur@concentric.net> wrote:

However using USB or firewire greately limits the speed. Now you can
install an external SATA 3 that will run as fast as the internal
drives.

I paid $10 USD less than that each for 4 500 Gig SATA 3 drives.
$139 X 4 = $556 for 2 Terabytes.

Flash memory hasn't come close to being able to match hard drives for
capacity. Yes, I can purchase a 1 Gig CF for about $20 to $30
compared to the $70 at discount I paid for them when first available.
Stores were selling them for $129 at the time.

But at any rate, when you compare the changes in HDs from the many
thousands of dollars for a 10 Meg Winchester drive of the 80's to the
1 Terabyte USB external drives for $400 give or take about $50 those
HDs certainly have kept pace with drive capacity for a reasonable
price about trippeling in the last year.

You've been going to the wrong stores. A few years ago HDs were
expensive and running in the 120 Gig range. I have one left that is
that small. I have over 6 terabytes...make that 7 between 5
computers.
Those 5 500 Gig SATA 3 drives with 16 meg caches cost about 1.5 times
the price of the 120 Gig drive I mentioned. You can now purchase 80 to
120 Gig drives for around $50 pt $60 with a bit of shopping. Actually
Best Buy had 250s for $89 a while back.

I just looked at New Egg and they have a 250 Gig SATA 3 WD (8 meg
cache)for $70, a 320 Seagate SATA 3 for $90 and a 500 Gig SATA 3 WD
for $139. The last two have 16 Meg caches.

BTW they have a 1 Terabyte external with 32 Meg Cache set up for both
USB2 and Firewire and fan cooled for $379 after rebate. ($9 shipping)

What we are seeing is still a rapid increase in capacity and prices
*still* coming down. One Terabyte external (with fan) for about the
same price as a 400 Gig just 6 months ago.

(N833R, S# CD-2 Worlds oldest Debonair)
www.rogerhalstead.com

John Turco wrote:

Consider, also, that HD capacity increases are subject to big leaps
when a new technology is introduced, followed by a plateau until the
next leap. Flash-memory capacity, on the other hand, inexorably
increases according to Moore's Law.