On 2021-09-24 7:57 p.m., Aragorn wrote:
> On 24.09.2021 at 23:08, Carlos E. R. scribbled:
>>
>> Me, I don't understand why modern disks don't have multiple actuators.
>> It would reduce the access time and increase i/o.

Added cost is one big reason against it. Apparently it adds power
consumption, too. (Which is an important consideration in a data centre
with thousands of them running.)

> I have recently read about a new hard disk design from one of the usual
> manufacturers that had dual actuators.

The Seagate Exos 2X14 14TB drive has dual actuators. It presents itself
as a pair of 7TB drives. The "2" before the X in the name is indicative
of the dual actuator technology, which Seagate calls "Mach.2". (There's
also a plain "X14" single-actuator drive in their line-up. Don't mix
them up!)

Top and also bottom posting have always both been conventions on Usenet,
and if you encounter a pedant who is up his own fundamentum, then
post both at the top and also at the bottom, as this post, which
should make everybody happy.

On 24/09/2021 13:40, Kerr-Mudd, John wrote:
>>
> Please adopt usenet convention and post your reply text at the bottom, there's a good chap. And some judicious snipping would help too. (Yup I didn't do it last time, mea culpa).
>

Top and also bottom posting have always both been conventions on Usenet,
and if you encounter a pedant who is up his own fundamentum, then
post both at the top and also at the bottom, as this post, which
should make everybody happy.

["Followup-To:" header set to alt.folklore.computers.]
On 2021-09-25, Andreas Kohlbach <ank@spamfence.net> wrote:
>> OK. Learning, havent played with usenet for a while :P
>
> Thanks, perfect. Thank you.
>
> Well, leave an empty line between the quote an your text.

Like this?

--

7-77-777
Evil Sinner!

On 2021-09-25, gareth evans <headstone255@yahoo.com> wrote:
> Top and also bottom posting have always both been conventions on Usenet, and
> if you encounter a pedant who is up his own fundamentum, then post both at
> the top and also at the bottom, as this post, which should make everybody
> happy.

great, but please reformat text, it looks ugly on terminal text readers.

>
> On 24/09/2021 13:40, Kerr-Mudd, John wrote:
>>>
>> Please adopt usenet convention and post your reply text at the bottom,
>> there's a good chap. And some judicious snipping would help too. (Yup I
>> didn't do it last time, mea culpa).
>>
>
> Top and also bottom posting have always both been conventions on Usenet, and
> if you encounter a pedant who is up his own fundamentum, then post both at
> the top and also at the bottom, as this post, which should make everybody
> happy.
>
great, but please reformat text, it looks ugly on terminal text readers.

--

7-77-777
Evil Sinner!

On 25/09/2021 17.35, John-Paul Stewart wrote:
> On 2021-09-24 7:57 p.m., Aragorn wrote:
>> On 24.09.2021 at 23:08, Carlos E. R. scribbled:
>>>
>>> Me, I don't understand why modern disks don't have multiple actuators.
>>> It would reduce the access time and increase i/o.
>
> Added cost is one big reason against it. Apparently it adds power
> consumption, too. (Which is an important consideration in a data centre
> with thousands of them running.)

Sure, speed always need more power.

>
>> I have recently read about a new hard disk design from one of the usual
>> manufacturers that had dual actuators.
>
> The Seagate Exos 2X14 14TB drive has dual actuators. It presents itself
> as a pair of 7TB drives. The "2" before the X in the name is indicative
> of the dual actuator technology, which Seagate calls "Mach.2". (There's
> also a plain "X14" single-actuator drive in their line-up. Don't mix
> them up!)
>

If they are seen as two drives, then that is not what I call a dual
actuator disk.

--
Cheers,
Carlos E.R.

Aragorn <thorongil@telenet.be> wrote:
> On 25.09.2021 at 03:51, Branimir Maksimovic scribbled:
>
>> Corrected.
>
> Not really. Your "signature" contains more than four lines, and your
> message body contains no reference whatsoever as to what you are
> replying to.
>
> Most people on Usenet are subscribed to more than one newsgroup. Your
> style of quoting makes that into a nightmare every time they stumble
> upon your posts.
>
> Usenet netiquette and convention stipulate that you would sbip the
> irrelevant quoted content from your replies and that you write your own
> replies underneath the remaining paragraphs in an interleaved fashion.
>
> Any deviation from that standard makes the Usenet experience only
> harder for everyone else, and needlessly so, because it really doesn't
> cost any effort to do it right.
>

I think I’m just going to have to ignore those posts. I really hope they
don’t contain anything interesting.

--
Pete

Ahem A Rivet's Shot <steveo@eircom.net> wrote:
> On Sat, 25 Sep 2021 02:06:45 -0400
> Andreas Kohlbach <ank@spamfence.net> wrote:
>
>> On Sat, 25 Sep 2021 03:35:28 +0100, Ahem A Rivet's Shot wrote:
>>>
>>> On Fri, 24 Sep 2021 21:02:38 GMT
>>> scott@slp53.sl.home (Scott Lurndal) wrote:
>>>
>>>> Ahem A Rivet's Shot <steveo@eircom.net> writes:
>>>
>>>> Easily resolved if outlook would simply position the cursor at the
>>>> point of the new reply rather than at the start of the message when
>>>> reading new messages.
>>>
>>> You have missed the point - in this context you want the
>>> replies to be at the top and you want a complete copy of the entire
>>> thread. Interleaved replies and snipping in that context is completely
>>> wrong it destroys essential information.
>>
>> This is IMO okay in business environments. But not in the usenet.
>
> Of course. My point was that Outlook is designed for business
> environments not for USENET.
>

Oh, it was *designed*?

--
Pete

On Sat, 25 Sep 2021 12:28:23 -0700, Peter Flass
<peter_flass@yahoo.com> wrote:

>Ahem A Rivet's Shot <steveo@eircom.net> wrote:
>> On Sat, 25 Sep 2021 02:06:45 -0400
>> Andreas Kohlbach <ank@spamfence.net> wrote:
>>
>>> On Sat, 25 Sep 2021 03:35:28 +0100, Ahem A Rivet's Shot wrote:
>>>>
>>>> On Fri, 24 Sep 2021 21:02:38 GMT
>>>> scott@slp53.sl.home (Scott Lurndal) wrote:
>>>>
>>>>> Ahem A Rivet's Shot <steveo@eircom.net> writes:
>>>>
>>>>> Easily resolved if outlook would simply position the cursor at the
>>>>> point of the new reply rather than at the start of the message when
>>>>> reading new messages.
>>>>
>>>> You have missed the point - in this context you want the
>>>> replies to be at the top and you want a complete copy of the entire
>>>> thread. Interleaved replies and snipping in that context is completely
>>>> wrong it destroys essential information.
>>>
>>> This is IMO okay in business environments. But not in the usenet.
>>
>> Of course. My point was that Outlook is designed for business
>> environments not for USENET.
>>
>
>Oh, it was *designed*?

Did somebody add NNTP support to Outlook when I wasn't looking?

On 9/25/2021 12:03 PM, Branimir Maksimovic wrote:
> On 2021-09-25, gareth evans <headstone255@yahoo.com> wrote:
>> Top and also bottom posting have always both been conventions on Usenet, and
>> if you encounter a pedant who is up his own fundamentum, then post both at
>> the top and also at the bottom, as this post, which should make everybody
>> happy.
>
> great, but please reformat text, it looks ugly on terminal text readers.
>
>>
>> On 24/09/2021 13:40, Kerr-Mudd, John wrote:
>>>>
>>> Please adopt usenet convention and post your reply text at the bottom,
>>> there's a good chap. And some judicious snipping would help too. (Yup I
>>> didn't do it last time, mea culpa).
>>>
>>
>> Top and also bottom posting have always both been conventions on Usenet, and
>> if you encounter a pedant who is up his own fundamentum, then post both at
>> the top and also at the bottom, as this post, which should make everybody
>> happy.
>>
> great, but please reformat text, it looks ugly on terminal text readers.
>

Touché

On 2021-09-25, SevenOverSix <hae274c.net> wrote:

> On 9/24/21 1:16 PM, Charlie Gibbs wrote:
>
>> I never saw a drive with multiple actuators, although I
>> wouldn't be surprised if such a thing existed. I did hear
>> about drives that had fixed heads mounted on a few tracks
>> for rapid access.
>
> There were plenty of removable-platter systems with
> multiple heads/actuators. How else would you deal with
> a stack of 4/6/8/10 platters ? They actuators DID
> work independently, it was cool to watch, all those
> little "fingers" probing.

No, there was only one actuator. The head assembly consisted
of multiple heads (10 for the IBM 2311, 20 for the 2314, similar
numbers for later models) mounted on a single frame. All heads
moved across all platters simultaneously. Files were usually
organized in "cylinders"; once you accessed one track in the
cylinder, you could access all other tracks by simply switching
heads without having the actuator move.

--
/~\ Charlie Gibbs | Life is perverse.
\ / <cgibbs@kltpzyxm.invalid> | It can be beautiful -
X I'm really at ac.dekanfrus | but it won't.
/ \ if you read it the right way. | -- Lily Tomlin

On 9/25/21 20:36, Dave Garland wrote:
> On 9/25/2021 12:03 PM, Branimir Maksimovic wrote:
>> On 2021-09-25, gareth evans <headstone255@yahoo.com> wrote:
>>> Top and also bottom posting have always both been conventions on
>>> Usenet, and
>>> if you encounter a pedant who is up his own fundamentum, then post
>>> both at
>>> the top and also at the bottom, as this post, which should make
>>> everybody
>>> happy.
>>
>> great, but please reformat text, it looks ugly on terminal text readers.
>>
>>>
>>> On 24/09/2021 13:40, Kerr-Mudd, John wrote:
>>>>>
>>>> Please adopt usenet convention and post your reply text at the bottom,
>>>> there's a good chap. And some judicious snipping would help too. (Yup I
>>>> didn't do it last time, mea culpa).
>>>>
>>>
>>> Top and also bottom posting have always both been conventions on
>>> Usenet, and
>>> if you encounter a pedant who is up his own fundamentum, then post
>>> both at
>>> the top and also at the bottom, as this post, which should make
>>> everybody
>>> happy.
>>>
>> great, but please reformat text, it looks ugly on terminal text readers.
>>
>
> Touché
>
It even looks ugly on Thunderbird.
But even when we try, it is hard to get really properly formatted text
out of any sort of news-reader/mailer. To begin
to do a decent job the text must be checked from the saved draft
before letting it flutter out to the net.

bliss - Your tag lines (k) were stolen!
(more) There is a puff of smoke!

Quaff a potion of See Invisible.
Smeagol is at it again.
Quaff a potion of Speed.
Pretend you haven't eeen him.
Wait a turn until he tries for these lines.
The smite him with your enchanted dagger...

--
bliss dash SF 4 ever at dslextreme dot com

On 2021-09-26, Bobbie Sellers <bliss@mouse-potato.com> wrote:
> Pretend you haven't eeen him.
> Wait a turn until he tries for these lines.
> The smite him with your enchanted dagger...
>
Great allegory :P

--

7-77-777
evil Sinner!

On Sat, 25 Sep 2021 21:32:28 -0700, Bobbie Sellers wrote:
>
> On 9/25/21 20:36, Dave Garland wrote:
>>>
>> Touché
>>
> It even looks ugly on Thunderbird.
> But even when we try, it is hard to get really properly
> formatted text out of any sort of news-reader/mailer. To
> begin
> to do a decent job the text must be checked from the saved draft
> before letting it flutter out to the net.

You can always set an Expire line in the header otherwise. ;-)

This article should evaporate from the usenet on
27 Sep 2021 23:59:00 -0400 to never be seen again.
--
Andreas

On 26/09/2021 13.36, Andreas Kohlbach wrote:
> On Sat, 25 Sep 2021 21:32:28 -0700, Bobbie Sellers wrote:
>>
>> On 9/25/21 20:36, Dave Garland wrote:
>>>>
>>> Touché
>>>
>> It even looks ugly on Thunderbird.

I haven't noticed any issue in my Thunderbird

>> But even when we try, it is hard to get really properly
>> formatted text out of any sort of news-reader/mailer. To
>> begin
>> to do a decent job the text must be checked from the saved draft
>> before letting it flutter out to the net.
>
> You can always set an Expire line in the header otherwise. ;-)
>
> This article should evaporate from the usenet on
> 27 Sep 2021 23:59:00 -0400 to never be seen again.

We'll see :-)

--
Cheers,
Carlos E.R.

On Sun, 26 Sep 2021 14:50:29 +0200, Carlos E. R. wrote:
>
> On 26/09/2021 13.36, Andreas Kohlbach wrote:
>> On Sat, 25 Sep 2021 21:32:28 -0700, Bobbie Sellers wrote:
>>>
>>> On 9/25/21 20:36, Dave Garland wrote:
>>>>>
>>>> Touché
>>>>
>>> It even looks ugly on Thunderbird.
>
> I haven't noticed any issue in my Thunderbird

Me neither in Gnus.

>>> But even when we try, it is hard to get really properly
>>> formatted text out of any sort of news-reader/mailer. To
>>> begin
>>> to do a decent job the text must be checked from the saved draft
>>> before letting it flutter out to the net.
>>
>> You can always set an Expire line in the header otherwise. ;-)
>>
>> This article should evaporate from the usenet on
>> 27 Sep 2021 23:59:00 -0400 to never be seen again.
>
> We'll see :-)

Or not see (after that date ;-).
--
Andreas

On Sat, 25 Sep 2021 00:07:49 +0000, Branimir Maksimovic wrote:

> UNDER '--', SIGNATURE

IOW, you're a moron. THAT, is NOT what a signature is, nor how you
construct one!!!

--
Testing testing

On 9/26/21 12:25 AM, Charlie Gibbs wrote:
> On 2021-09-25, SevenOverSix <hae274c.net> wrote:
>
>> On 9/24/21 1:16 PM, Charlie Gibbs wrote:
>>
>>> I never saw a drive with multiple actuators, although I
>>> wouldn't be surprised if such a thing existed. I did hear
>>> about drives that had fixed heads mounted on a few tracks
>>> for rapid access.
>>
>> There were plenty of removable-platter systems with
>> multiple heads/actuators. How else would you deal with
>> a stack of 4/6/8/10 platters ? They actuators DID
>> work independently, it was cool to watch, all those
>> little "fingers" probing.
>
> No, there was only one actuator. The head assembly consisted
> of multiple heads (10 for the IBM 2311, 20 for the 2314, similar
> numbers for later models) mounted on a single frame. All heads
> moved across all platters simultaneously. Files were usually
> organized in "cylinders"; once you accessed one track in the
> cylinder, you could access all other tracks by simply switching
> heads without having the actuator move.

I distinctly remember one where the heads moved
independently - because it was So Cool to watch.

But I don't remember who made it or if it had
some special purpose. MIGHT have been on a
Cape K tour when I was still kinda short. My
GUESS, for the timeframe, is that each user
had their own platter and probably one shared
platter for xfers.

On 2021-09-26, yamas <yamas@bloodyhell.no.way> wrote:
> On Sat, 25 Sep 2021 00:07:49 +0000, Branimir Maksimovic wrote:
>
>> UNDER '--', SIGNATURE
>
> IOW, you're a moron. THAT, is NOT what a signature is, nor how you
> construct one!!!
>

Thanks, for compliment, is this now OK?

--

7-77-777
Evil Sinner!

-------- Forwarded Message --------
Path:
buffer1.nntp.dca1.giganews.com!border1.nntp.dca1.giganews.com!nntp.giganews.com!news.misty.com!2.eu.feeder.erje.net!feeder.erje.net!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail
From: John-Paul Stewart <jpstewart@personalprojects.net>
Newsgroups: comp.os.linux.misc
Subject: Re: Obit - Sir Clive Sinclair, Computing Pioneer
Date: Sat, 25 Sep 2021 11:35:22 -0400
Lines: 18
Message-ID: <ir8ttqFos77U1@mid.individual.net>
References: <POCdnc09ErazA9n8nZ2dnUU7-aHNnZ2d@earthlink.com>
<9K33J.43934$md6.4877@fx36.iad> <eli$2109231506@qaz.wtf>
<gW53J.442$2m1.423@fx26.iad> <eli$2109231728@qaz.wtf>
<qpbv1ixkgr.ln2@minas-tirith.valinor>
<ENednfVTVamUzND8nZ2dnUU7-a_NnZ2d@earthlink.com>
<cKn3J.44304$ol1.23517@fx42.iad> <tar12ixb59.ln2@minas-tirith.valinor>
<20210925015711.495244d6@nx-74205>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
X-Trace: individual.net mHHjTDY5zgO4f3eOhnvGmwyHvB0/4qBDXCBPqZbOHsVA6VyVNT
Cancel-Lock: sha1:ICe90ntvUR7Efp3B9fw0pDA6SY4=
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
Thunderbird/78.14.0
In-Reply-To: <20210925015711.495244d6@nx-74205>
Content-Language: en-CA
Bytes: 2046
Xref: number.nntp.giganews.com comp.os.linux.misc:773359

On 2021-09-24 7:57 p.m., Aragorn wrote:
> On 24.09.2021 at 23:08, Carlos E. R. scribbled:
>>
>> Me, I don't understand why modern disks don't have multiple actuators.
>> It would reduce the access time and increase i/o.

Added cost is one big reason against it. Apparently it adds power
consumption, too. (Which is an important consideration in a data centre
with thousands of them running.)

> I have recently read about a new hard disk design from one of the usual
> manufacturers that had dual actuators.

The Seagate Exos 2X14 14TB drive has dual actuators. It presents itself
as a pair of 7TB drives. The "2" before the X in the name is indicative
of the dual actuator technology, which Seagate calls "Mach.2". (There's
also a plain "X14" single-actuator drive in their line-up. Don't mix
them up!)

Yea ... but it's SEAGATE :-)

Haven't bought one of THOSE in a decade, for good reason.

In about five years, SSD tech will come to match magnetics
in capacity/$$$ and re-write endurance. At this point this
discussion will become irrelevant.

On 9/25/21 2:14 PM, Carlos E. R. wrote:
> On 25/09/2021 17.35, John-Paul Stewart wrote:
>> On 2021-09-24 7:57 p.m., Aragorn wrote:
>>> On 24.09.2021 at 23:08, Carlos E. R. scribbled:
>>>>
>>>> Me, I don't understand why modern disks don't have multiple actuators.
>>>> It would reduce the access time and increase i/o.
>>
>> Added cost is one big reason against it. Apparently it adds power
>> consumption, too. (Which is an important consideration in a data centre
>> with thousands of them running.)
>
> Sure, speed always need more power.
>
>>
>>> I have recently read about a new hard disk design from one of the usual
>>> manufacturers that had dual actuators.
>>
>> The Seagate Exos 2X14 14TB drive has dual actuators. It presents itself
>> as a pair of 7TB drives. The "2" before the X in the name is indicative
>> of the dual actuator technology, which Seagate calls "Mach.2". (There's
>> also a plain "X14" single-actuator drive in their line-up. Don't mix
>> them up!)
>>
>
> If they are seen as two drives, then that is not what I call a dual
> actuator disk.

I'd tend to agree - though, with a little finesse, you might
be able to reap the benefits.

Also, SEAGATE. A couple of years back I bought a pre-built
Xeon server. It CAME with two Seagates. Those went directly
into the trash (after magnet extraction) and were replaced
by HGSTs. NOW it's reliable. I like RAID-1 ...

Face it though, in about 5 years SDDs will finally reach
the capacity/$$$ and re-write durability of todays magnetics.
This whole discussion becomes as moot as arguing whether
punch-tape or mag-tape is "better".

On 29/09/2021 02:11, Branimir Maksimovic wrote:
> On 2021-09-26, yamas <yamas@bloodyhell.no.way> wrote:
>> On Sat, 25 Sep 2021 00:07:49 +0000, Branimir Maksimovic wrote:
>>
>>> UNDER '--', SIGNATURE
>>
>> IOW, you're a moron. THAT, is NOT what a signature is, nor how you
>> construct one!!!
>>
>
> Thanks, for compliment, is this now OK?
>
>
the fact that thunderbird deletes it, suggests that it is

--
In a Time of Universal Deceit, Telling the Truth Is a Revolutionary Act.

- George Orwell

On 29/09/2021 05:58, SevenOverSix wrote:
> Face it though, in about 5 years SDDs will finally reach
>   the capacity/$$$ and re-write durability of todays magnetics.
>   This whole discussion becomes as moot as arguing whether
>   punch-tape or mag-tape is "better".

According to my power user friend, this has already occurred.. SSDS now
last longer than hard drives, so the extra cost is offset completely by
the rarer need to replace them.

And, under most wear out scenarios (rather than sudden and total abject
failure) the wear levelling algorithms mean that its like having bad
blocks mapped out until the disk capacity shrinks to noticeable levels
of nastiness.

Spinning rust develops bearing play that means ALL sectors get bad at
the same time. And that may end in a head crash.

I am predicting that provided you take the trouble to monitor the status
on an SSD you will get more life out of it than a conventional disk and
wont lose any data as it nears the beyond the pale level of block loss.

Its like CRT being replaced by LCD. As soon as the price came down there
was no real reason for a CRT any more ever, outside of certain very
niche applications.

--
In a Time of Universal Deceit, Telling the Truth Is a Revolutionary Act.

- George Orwell

The Natural Philosopher <tnp@invalid.invalid> writes:
> On 29/09/2021 05:58, SevenOverSix wrote:
>> Face it though, in about 5 years SDDs will finally reach
>> the capacity/$$$ and re-write durability of todays magnetics.
>> This whole discussion becomes as moot as arguing whether
>> punch-tape or mag-tape is "better".
>
> According to my power user friend, this has already occurred.. SSDS
> now last longer than hard drives, so the extra cost is offset
> completely by the rarer need to replace them.

Dependent on model and usage pattern I would expect, although my
personal experience is consistent: one SSD failure and more HDD failures
than I can count since I started buying SSDs over a decade ago. (And the
SSD was an OCZ...) All my in-use storage is now SSD except for backup
drives.

I’d question whether it’s really relevant at the hobbyist level
though. When you only have a handful of drives the average lifetimes
aren’t as relevant as random variation and bad luck; performance and
cost are likely to be more important.

For a nontrivial organization the reliability considered in bulk starts
to matter more.

> Its like CRT being replaced by LCD. As soon as the price came down
> there was no real reason for a CRT any more ever, outside of certain
> very niche applications.

LCDs started winning on power consumption before unit cost as I
recall. I wouldn’t underestimate the reduction in manual handling risk
and the value of the desk space they free up either l-)

--
https://www.greenend.org.uk/rjk/

On 29/09/2021 13:17, Richard Kettlewell wrote:
> The Natural Philosopher <tnp@invalid.invalid> writes:
>> On 29/09/2021 05:58, SevenOverSix wrote:
>>> Face it though, in about 5 years SDDs will finally reach
>>> the capacity/$$$ and re-write durability of todays magnetics.
>>> This whole discussion becomes as moot as arguing whether
>>> punch-tape or mag-tape is "better".
>>
>> According to my power user friend, this has already occurred.. SSDS
>> now last longer than hard drives, so the extra cost is offset
>> completely by the rarer need to replace them.
>
> Dependent on model and usage pattern I would expect, although my
> personal experience is consistent: one SSD failure and more HDD failures
> than I can count since I started buying SSDs over a decade ago. (And the
> SSD was an OCZ...) All my in-use storage is now SSD except for backup
> drives.
>
> I’d question whether it’s really relevant at the hobbyist level
> though. When you only have a handful of drives the average lifetimes
> aren’t as relevant as random variation and bad luck; performance and
> cost are likely to be more important.
>
Are we just hobbyists? I use computers as part of my daily activity. Its
not a hobby. What I do with them is a hobby, sometimes, but not the
machines themselves. And when I need to check my investments, that's not
a hobby either, but I need a reliable computer to do it.

> For a nontrivial organization the reliability considered in bulk starts
> to matter more.
>
Well I am up to a mere two SSDS and down to a mere 3 large bits of
spinning rust: I still think Id upgrade to all SSDs if I could afford
to. I was sick of fixing computers years ago, Now I just want them to work.

>> Its like CRT being replaced by LCD. As soon as the price came down
>> there was no real reason for a CRT any more ever, outside of certain
>> very niche applications.
>
> LCDs started winning on power consumption before unit cost as I
> recall. I wouldn’t underestimate the reduction in manual handling risk
> and the value of the desk space they free up either l-)
>
Power consumption = cost. Same as LED lights. They are now cost
effective over incandescent over a fairly short lifetime, but not at the
'unit cost' level. I have to confess I don't think at 'unit cost' level.
Downside of running a business ;-)

And the greater reliability of LED lights and SSDs also represents less
maintenance costs.

I've a halogen spot that required a ladder to change it. It's now LED.

At a commercial level the metric is total cost of ownership: Unless you
are totally cash strapped that is sensible at the domestic consumer
level as well.

--
Socialism is the philosophy of failure, the creed of ignorance and the
gospel of envy.

Its inherent virtue is the equal sharing of misery.

Winston Churchill

The Natural Philosopher <tnp@invalid.invalid> writes:
> On 29/09/2021 13:17, Richard Kettlewell wrote:
>> I’d question whether it’s really relevant at the hobbyist level
>> though. When you only have a handful of drives the average lifetimes
>> aren’t as relevant as random variation and bad luck; performance and
>> cost are likely to be more important.
>
> Are we just hobbyists? I use computers as part of my daily
> activity. Its not a hobby. What I do with them is a hobby, sometimes,
> but not the machines themselves. And when I need to check my
> investments, that's not a hobby either, but I need a reliable computer
> to do it.

Matter of definitions ... my point is that for any given user there’s
competing requirements (reliability, cost, performance, physical
space...) and it’s not a given that reliability will trump the rest,
especially given how fuzzy a question it is at small scale.

My NUC ended up with a 4TB SSD because it only has room for one drive
and I wanted to use it for bulk storage as well as compilation etc.
Pricey but fitting everything into a single tiny computer means I get
back office space for something else.

>>> Its like CRT being replaced by LCD. As soon as the price came down
>>> there was no real reason for a CRT any more ever, outside of certain
>>> very niche applications.
>>
>> LCDs started winning on power consumption before unit cost as I
>> recall. I wouldn’t underestimate the reduction in manual handling risk
>> and the value of the desk space they free up either l-)
>
> Power consumption = cost.

Exactly.

--
https://www.greenend.org.uk/rjk/