The carbon cost of MS Windows?
Richard Torrens
very clear to me that MicroSoft have cunningly built in obsolescence. All
of the MS computers I have experienced have a tendency for the software to
rust and seize up as they age, forcing the user to buy a newer faster lot
of hardware. It's very clear that MS are deliberately driving hardware
evolution - which means high carbon cost!
What really bugged me about Windoze, apart from the really boggy
performance, even on fast hardware, is the fact that they keep
reconfiguring themselves - without asking the user.
My daughter bought a Vista machine. New, shiny, fast hardware. But it's as
slow as paint drying. And I cannot get it networking.
RISC OS has never done that. If I change the configuration, it stays
changed.
Apple computers do not do that either! If I reconfigure an Apple, it stays
reconfigured. I can lock the config to stop any further changes.
RISC OS hardware may be old and slow. But in most things, it's quicker to
do things that a far faster PC running Windoze. A 3 year old Apple is much
more responsive than either XP or Vista! Similar to RISC OS.
If you update the software on either Apple or RISC OS, the performance
improves.
So an Apple computer may cost twice as much as a Windoze machine. But it
lasts 3, 4 5 times as long.
So the carbon cost of a Windoze machine is relatively huge. And the cost
of owning Windoze is higher than Apple or RISC OS.
--
Richard Torrens - 0000...@Torrens.org.uk
This time limited email address and must not be added to any mailing list: A charge will be invoiced for handling any unsolicited mailing list emails received.
www.4qd.co.uk www.4qdtec.com www.torrens.org.uk
Rob Kendrick
Richard Torrens <News+...@Torrens.org.uk> wrote:
> So the carbon cost of a Windoze machine is relatively huge. And the
> cost of owning Windoze is higher than Apple or RISC OS.
A pretty meaningless blanket statement :) If what ever you use your
computer for involves encoding of audio or video, the modern PC
hardware used on Windows and Apple boxes will finish long before the
RISC OS box, and may actually end up being more power efficient because
of it.
(To name one obvious example of many.)
B.
Jonathan Schneider
> If what ever you use your computer for involves encoding of audio or video
but hardly representative of most machines do most of the time !
Jon
Aly
news:4fca23e14e...@Torrens.org.uk...
> Running a network of RISC OS, XP, Vista and now an Apple computer, it is
You're quite right of course. Windoze is a beast to be tamed more than
anything else. Norton Ghost is the key, or just rebuild the things
periodically.
When I *used* to be in a support environment I often encountered IT Managers
(on paper) who were so obsessed with getting to the bottom of problems.
9/10 no one has the foggiest clue what Windoze is doing in the background.
It's often easiest all round just to backup the data (which should be backed
up already or on the network!), format the hard disk, and start again.
Windoze is utterly dire. You can have a floor of IDENTICAL hardware and
builds, yet there will always be a couple of machines that start
misbehaving. You haven't got a clue why, so you just start again.
Frankly I'm glad that I'm out of it.
PaulB
A good description of why 2nd line support sucks. And you moved on to do
what?
Paul
Aly
news:rE0mk.134655$8_2....@newsfe08.ams2...
>
> A good description of why 2nd line support sucks. And you moved on to do
> what?
>
> Paul
>
I would honestly LOVE to tell you, but this Internet thing that I helped
create has me running for cover. ;-) We grew up with these insipid things
remember. :-)
Richard Porter
write:
I think Richard's point was that you have to keep upgrading Windoze
PCs more frequently or they just get choked by their own software, so
they create more waste, pollution and emissions from manufacturing.
Having tried with audio, I wouldn't use RISC OS for encoding audio of
video if I had an alternative. There are some applications that are
just so cpu-intensive that you have to have a fast processor, but they
aren't typical PC applications.
--
Richard Porter
rich@ / www. richardporter.me.uk
"You can't have Windows without pains."
Cwatters
alone. It tends to get left on as it takes an age to start up. Today I
fitted a 13 plug to a table lamp while waiting for it to boot. :-( My
RiscOS machine never took that long.
Jonathan Schneider
One possible solution is to make it standy/hibernate possibly
automatically. As long as the drivers are ok with it you might get a
much quicker powerup. Another good thing about auto-hibernate is that
you don't lose state even if power fails (when hibernated). I am
surprised more coorporate machines aren't set up like this.
Jon
druck
> Running a network of RISC OS, XP, Vista and now an Apple computer, it is
> very clear to me that MicroSoft have cunningly built in obsolescence. All
> of the MS computers I have experienced have a tendency for the software to
> rust and seize up as they age, forcing the user to buy a newer faster lot
> of hardware.
Thats why Microsoft and Intel are making billions, and Acorn/RISC OS
died years ago, even if most people still running RISC OS 3.7 on their
Risc PCs haven't quite realised yet.
Dont cross post this to misc, follow ups to csa.advocacy only.
---druck
--
The ARM Club Free Software - http://www.armclub.org.uk/free/
The 32bit Conversions Page - http://www.quantumsoft.co.uk/druck/
John M Ward
druck <ne...@druck.freeuk.com> wrote:
> On 5 Aug 2008 Richard Torrens <News+...@Torrens.org.uk> wrote:
> > Running a network of RISC OS, XP, Vista and now an Apple computer,
> > it is very clear to me that MicroSoft have cunningly built in
> > obsolescence. All of the MS computers I have experienced have a
> > tendency for the software to rust and seize up as they age, forcing
> > the user to buy a newer faster lot of hardware.
> Thats why Microsoft and Intel are making billions, and Acorn/RISC OS
> died years ago, even if most people still running RISC OS 3.7 on
> their Risc PCs haven't quite realised yet.
I do my bit to support our market by buying every significant new
machine when I can. The present line-up here includes an Iyonix pc, an
A9home and 3 RiscPCs, two of which are running RISC OS 4.
All are networked peer-to-peer (Access+) along with the portable
(running VA-RPC) and the one and only Wintel desktop machine. The
A9home consumes less than three Watts of power when the display is
switched off, which it is most of the time.
These computers all have their uses, though a couple of them are
switched on only occasionally (e.g. the bedroom machine is really used
only for taking snapshots off the TV, via its RiscTV podule linked to
the cable TV set-top box). It's all very elegant and consumes
relatively little power in normal use.
--
John Ward in Medway, Kent - using RISC OS since 1987
Now using an Iyonix, an A9home, 2 RiscPCs and Virtual-RPC!
Acorn/RISC OS web page: www.john-ward.org.uk/personal/john/computers
David Pitt
> In article <23ad45ca...@druck.freeuk.net>,
> druck <ne...@druck.freeuk.com> wrote:
> > On 5 Aug 2008 Richard Torrens <News+...@Torrens.org.uk> wrote:
>
> > > Running a network of RISC OS, XP, Vista and now an Apple computer, it
> > > is very clear to me that MicroSoft have cunningly built in
> > > obsolescence. All of the MS computers I have experienced have a
> > > tendency for the software to rust and seize up as they age, forcing
> > > the user to buy a newer faster lot of hardware.
>
> > Thats why Microsoft and Intel are making billions, and Acorn/RISC OS
> > died years ago, even if most people still running RISC OS 3.7 on their
> > Risc PCs haven't quite realised yet.
>
> I do my bit to support our market by buying every significant new machine
> when I can. The present line-up here includes an Iyonix pc, an A9home and
> 3 RiscPCs, two of which are running RISC OS 4.
I have done (oops, past tense) my bit, I have a similar setup, somehow it
seems as if that was not enough.
As for my carbon costs it might be the case that the LAN components, which
are left on all day, do more damage that the individual computers which can
sleep or hibernate.
My first thought on seeing "The carbon cost of MS Windows" was does that
mean when I set fire to it, something of a temptation from time to time.
--
David Pitt
Roland Perry
21:12:37 on Tue, 5 Aug 2008, Cwatters
<colin.wat...@TurnersOakNOSPAM.plus.com> remarked:
>I worked out this week that my PC costs 75p a day to run in electricity
>alone.
Crikey! 75p is 7.5Kw hours, or 312 watts for 24 hours. I've just checked
my large tower-format desktop PC and it's only taking 110w plus another
35w for the LCD monitor.
--
Roland Perry
PaulB
news:Deidnc6y-u5SMwXV...@bt.com...
You are a flirt and ICMFP. You're hardly going to get stalked are you?
Paul
John M Ward
Interesting! I come from an electronics background, so have long
realised that even though PC-type computers tend to come with a minimum
350 Watt power supply -- and often higher -- this includes a decent
degree of safety margin. It also allows for internal expansions up to
the full capacity of the machine (drives, PCI cards) and even the
relatively small current that can be drawn from the USB ports.
This is why, for example, my "A9home" desktop computer comes with a
(physically tiny) 20 Watt power supply, even though it takes less than
three Watts itself and runs more-or-less silently and cold.
Aly
>
> You are a flirt and ICMFP. You're hardly going to get stalked are you?
>
> Paul
Five Pounds!!??
Oh no no no, that's not what I mean at all. I wasn't suggesting THAT at
all.
From my perspective I view this t'Internet thing as 'not real' :-) In some
respects I do actually wonder if it's at least partly responsible for the
social mess we're all living in now. For the first time you have all sorts
of different people, able to readily find others like themselves. Be it
they a facination with knives, fraud, or whatever. That I find incredibly
worrying.
Fast Rewind 20-years, and the last thing any of us would have done would
have been to put a large bulk of our personal information into a single
computer, let alone an interweb.
It's not computers that I don't trust. Nor is it the core academics. It's
when these things became accessible to the general populous sundry an all.
I grew up with these things, and loved the complexity of these things. Yet
I feel today that they are quite unhealthy indeed.
"Our computers made a mistake," say the large corporates in response to your
missing order.
Computers don't make mistakes, people do, and then they try to cover for
their own stupidity (which blowing their trumpet throughout) by blaming it
on something that the majority do not understand themselves.
If computers 'made mistakes' of the kind that so many companies blamed them
for, then frankly nothing would work, they wouldn't even switch on.
Computers don't selectively lose people's contact information, humans do
that. So it's either the computer or it isn't.
10am. Better do some work!! :-)
Roland Perry
2008, John M Ward <jo...@acornusers.org> remarked:
>> >I worked out this week that my PC costs 75p a day to run in
>> >electricity alone.
>>
>> Crikey! 75p is 7.5Kw hours, or 312 watts for 24 hours. I've just
>> checked my large tower-format desktop PC and it's only taking 110w
>> plus another 35w for the LCD monitor.
>
>Interesting! I come from an electronics background, so have long
>realised that even though PC-type computers tend to come with a minimum
>350 Watt power supply -- and often higher -- this includes a decent
>degree of safety margin. It also allows for internal expansions up to
>the full capacity of the machine (drives, PCI cards) and even the
>relatively small current that can be drawn from the USB ports.
My PC has two extra HDDs and a fairly upmarket video card, but nothing
drawing power from USB ports. A lot of the power margin is probably for
starting up HDDs (older ones, anyway).
--
Roland Perry
John M Ward
S'funny: my one-slice RiscPCs have 70 Watt PSUs, and the 2-slice has a
103 Watt unit. The 2-slice has now been running trouble-free for just
about 13 years (I bought it in August 1995), it has three podules, two
(modern) hard drives and is generally left on 24/7 (it monitors the
'phone and fax lines, and logs all calls) -- not at all bad for what is
after all a 1994 design...
Would a third hard drive and an "upmarket video card" require hundreds
of Watts extra power, multiplying my machine's power requirements rather
than merely adding a little? Admittedly, if yours also has a pizza
oven, kitchen sink and ten 'slices', then it would undoubtedly require
more power than normal:
www.john-ward.org.uk/personal/john/computers/html/rocket.html
I well recall seeing that beast in action, and its baby brother (with a
mere nine slices!) at shows. It all worked via an infra-red remote
control.
If anyone thinks that such a project is completely beyond the reach of
ordinary users, be aware that my two-slice became a four-slice for a
couple of years (when it had a couple of SCSI hard drives and Jaz drive
inside), and one can still buy spare slices and locking bolts for up to
seven slices from CJE Micro's.
Duncan Wood
>> nothing drawing power from USB ports. A lot of the power margin is
>> probably for starting up HDDs (older ones, anyway).
>
> S'funny: my one-slice RiscPCs have 70 Watt PSUs, and the 2-slice has a
> 103 Watt unit. The 2-slice has now been running trouble-free for just
> about 13 years (I bought it in August 1995), it has three podules, two
> (modern) hard drives and is generally left on 24/7 (it monitors the
> 'phone and fax lines, and logs all calls) -- not at all bad for what is
> after all a 1994 design...
>
> Would a third hard drive and an "upmarket video card" require hundreds
> of Watts extra power, multiplying my machine's power requirements rather
> than merely adding a little? Admittedly, if yours also has a pizza
> oven, kitchen sink and ten 'slices', then it would undoubtedly require
> more power than normal:
>
Potentially
http://www.xbitlabs.com/articles/video/display/power-noise_3.html#sect1
Stuart
> realised that even though PC-type computers tend to come with a minimum
> 350 Watt power supply -- and often higher -- this includes a decent
> degree of safety margin. It also allows for internal expansions up to
> the full capacity of the machine (drives, PCI cards) and even the
> relatively small current that can be drawn from the USB ports.
And allowance for peak currents that may be drawn on occasions.
The load on the PSU is far from constant so you choose a PSU that will
handle all of that and ensure longer life by choosing a PSU of higher
rating than the minimum needed, so that it is not over-stressed.
The actual KWh will be much lower than that suggested by multiplying the
PSU maximum rating by time.
--
Stuart Winsor
For Barn dances and folk evenings in the Coventry and Warwickshire area
See: http://www.barndance.org.uk
Brian Morrison
Indeed, my UPS tells me that the three PCs, ADSL modem, router and 2
wireless APs that it supplies take about 120W between them. Not
including monitors, but they're frequently in power save mode.
--
Brian
Aly
news:4fca8913afinval...@invalid-domain.co.uk...
>>
>> 10am. Better do some work!! :-)
>
>
> Paul
>
> --
> Life, the Universe, RISC OS Help and Everything - www.vigay.com
> Share and discuss ideas or chat about the above - www.vigay.com/forum
> Quality Internet Services, Broadband & Hosting -
> www.orpheusinternet.co.uk
Blaming it on the weather now. It's muggy again. :-(
10am to 2pm. 4-hours work, I think I need a rest. :-)
Aly
news:4fca8db...@acornusers.org...
>
> Would a third hard drive and an "upmarket video card" require hundreds
> of Watts extra power, multiplying my machine's power requirements rather
Yes.
The power requirements of some of those cards is ridiculous. They draw so
much power that they take their feed straight off of the PSU rather than the
slot.
Running an MSI Nvidea 8600GT PCIE16 here, one of the last to still take it's
power from the slot. And that soaks up about 70-watts maximum I think it
is.
Give em all DOS 5.0 I say and Word/WordPerfect. Hell, Minix and a copy of
VI.
The Natural Philosopher
Aly
news:12180347...@proxy02.news.clara.net...
>>
>> Give em all DOS 5.0 I say and Word/WordPerfect. Hell, Minix and a copy
>> of VI.
> To be honest, that's probably more than most people want or need.
Oh the joys of the 80's and 90's. A Dec VAX was the first major O/S I was
exposed to. Dialing in at 1200/300 to add the day's figures. It was a
little evening job I had when I was about 12!!
Those were the days, when merely using a computer was seen as an
achievement. Now you have to develop databases containing 100-million
records which cross reference every patient in the U.K. No prizes for
guessing why they don't work.
People have been watching far too much Star Trek and far too many Hollywood
movies.
Martin Saxon
> ... Dialing in at 1200/300 ...
1200/75 surely? or 300/300. Who would have thought you could get (or
need) a faster rate than that :-)
--
Martin
Music at www.martinsaxon.co.uk
magwitch
> "The Natural Philosopher" <a@b.c> wrote in message
> news:12180347...@proxy02.news.clara.net...
>>> Give em all DOS 5.0 I say and Word/WordPerfect. Hell, Minix and a copy
>>> of VI.
>> To be honest, that's probably more than most people want or need.
>
>
> People have been watching far too much Star Trek and far too many Hollywood
> movies.
>
>
Why they end up patients in the first place — too much sitting and
watching or sitting and computing.
...which reminds me ;-)
Aly
news:874p5y6...@news2.kororaa.com...
>
> The DEC VAX-11 was the machine, not the OS; I presume you mean VAX/VMS.
>
Can't quite remember actually. I know that the *thing* that I used to do
the data entry was a DEC VT100 terminal? It was just a dumb terminal. And
then it dialed in to a UNIX style CLI. All very basic stuff.
Aly
news:77ednd3HuOmESgTV...@pipex.net...
> Aly wrote:
>> ... Dialing in at 1200/300 ...
>
> 1200/75 surely? or 300/300. Who would have thought you could get (or need)
> a faster rate than that :-)
>
Not sure. Might be. The 300-baud figure sounds more plausible though. But
then again, the 1200 also rings some bells too. I know that it didn't have
automatic negotiation. There was also an acoustic coupler that you pushed
the phone into.
Now and then you'd get rubbish so would drop back the speed and ring in
again.
Then there was the TELEX......... Punched tape!!! They used to receive
messages from the Queen on that! :-/ Or at least it seemed like it. If you
think of the James Bond films where the submarines would get their
instructions on a teletype, then that's it!!
:-)
Aly
"Aly" <sf333ddf@sfsss'=-3498hfn2nmdf0xdh3222a]].c> wrote in message
YES!! It was definitely a DEC VT100. Just searched for it on Google
Images.
PaulB
Nothing wrong with telex. Kept me employed on ships for years during the
80s. All good stuff. SITOR/ARQ works very well over HF and telex is the
ideal format.
Paul
Tim Ward
news:_ZGdnUsHnodCcQTV...@bt.com...
>
> But then again, the 1200 also rings some bells too. I know that it didn't
> have automatic negotiation. There was also an acoustic coupler that you
> pushed the phone into.
Very sophisticated, that was, having your own portable terminal and, with
the aid of the acoustic coupler, dialling into the Post Officer's EPSS
(experimental packet switching system). Around 1977 (but I used it to access
a mainframe, not a VAX). Thermal bog rolls for output.
> Now and then you'd get rubbish so would drop back the speed and ring in
> again.
Now that's *much* more sophisticated than any such terminal I used - either
it worked or (more often) it dropped the line.
> Then there was the TELEX......... Punched tape!!! They used to receive
> messages from the Queen on that! :-/ Or at least it seemed like it. If
> you think of the James Bond films where the submarines would get their
> instructions on a teletype, then that's it!!
Like I said, the bog roll / acoustic coupler 300 baud terminal was *very*
sophisticated compared to the 110 baud ASR 33. The point about the paper
tape was that if the mainframe crashed during your attempt to type in four
lines of input you could just re-run the paper tape an hour or two later
when the operators had succeeded in rebooting it.
--
Tim Ward - posting as an individual unless otherwise clear
Brett Ward Limited - www.brettward.co.uk
Cambridge Accommodation Notice Board - www.brettward.co.uk/canb
Cambridge City Councillor
The Natural Philosopher
I worked on some telex software once. 50 baud is telex.
To test it, I grabbed a telex directory, and randomly dialled all parts
of the world apologising for a test message. sort of telex spam.
Never got a reliable connection to Zaire, but did manage to contact my
old company in Johannesburg. He even replied!
> :-)
>
>
The Natural Philosopher
emulator uses by default.
I remember when companies first discovered that a PC on Ethernet
running telnet was faster than a 38.4k baud connected wyse 50.
R.C. Payne
Though for submarines running submerged, ELF is still the best way of
getting a message to them, and the data rates for that make TELEX look
positively fast. It would not surprise me for such communications still
to use paper tape.
Robin
John M Ward
Paul Vigay <invalid-em...@invalid-domain.co.uk> wrote:
> In a dim and distant universe <4fca8db...@acornusers.org>,
> John M Ward <jo...@acornusers.org> enlightened us thusly:
> [Snippety snip]
> > If anyone thinks that such a project is completely beyond the reach
> > of ordinary users, be aware that my two-slice became a four-slice
> > for a couple of years (when it had a couple of SCSI hard drives and
> > Jaz drive inside), and one can still buy spare slices and locking
> > bolts for up to seven slices from CJE Micro's.
> I had a seven slice Risc PC until not so long ago.
Crumbs! I hope you took photos of it for posterity...
> In the end I dismantled it for space and converted it back into four
> separate Risc PCs. :-)
Surely that would take /more/ space (as it would necessitate four base
units -- plus the same number of slices -- rather than one), and
certainly a far greater overall footprint?
John M Ward
Aly <sf333ddf@sfsss'=-3498hfn2nmdf0xdh3222a]].c> wrote:
> "John M Ward" <jo...@acornusers.org> wrote in message
> news:4fca8db...@acornusers.org...
> >
> > Would a third hard drive and an "upmarket video card" require
> > hundreds of Watts extra power, multiplying my machine's power
> > requirements rather
> Yes.
Crazy! Doesn't technology advance any more? My more modern RISC OS
machines run virtually cold, as the memory chips etc are less wasteful
of power than they were back in 1994 (which was to do with why I
mentioned that date in respect of my two-slice RiscPC).
> The power requirements of some of those cards is ridiculous. They
> draw so much power that they take their feed straight off of the PSU
> rather than the slot.
Although there is some merit in having greater resources (processing
power, memory et al) on graphics cards, it does seem to have become an
over-the-top race to outdo the competition, so they are just getting
more of everything, rather than better. This is a common error --
confuding "more" with "better"...
> Running an MSI Nvidea 8600GT PCIE16 here, one of the last to still
> take it's power from the slot. And that soaks up about 70-watts
> maximum I think it is.
One of the Radeon cards apparently takes a total of well over 100 Watts,
which really is very silly, bearing in mind what it actually is -- just
a graphics card.
> Give em all DOS 5.0 I say and Word/WordPerfect. Hell, Minix and a
> copy of VI.
Why go backward? I've always preferred to go forward, but sensibly and
intelligently, not the "more, more!" greedy-eyed but unintelligent
thinking that seems to be the target of today's marketplace.
John M Ward
Duncan Wood <nntp...@dmx512.co.uk> wrote:
[quoting John Ward thus:]
> > Would a third hard drive and an "upmarket video card" require
> > hundreds of Watts extra power, multiplying my machine's power
> > requirements rather than merely adding a little? Admittedly, if
> > yours also has a pizza oven, kitchen sink and ten 'slices', then it
> > would undoubtedly require more power than normal:
>
> Potentially
> http://www.xbitlabs.com/articles/video/display/power-noise_3.html#sect1
Amazing! I suppose it's like American cars: they seem to like 'em big
and inefficient. Ego first, brains second, it all strongly suggests.
It's like having all that memory (and a huge pagefile) -- the system
spends so much time and effort trying to mange (cope with?) it all, that
it frequently chuggles away to itself for several seconds before even
doing something as simple as putting up a requested menu...
I am so accustomed to everything happening at the precise moment I
request it, that it is still something of a "culture shock" to switch to
the Wintel machine and try to work on that. Sometimes it's okay, but
much of the time it has to think about its own internal workings to such
an extent that the user becomes very much a second-class player. I
imagine it must be like taking a lorry to do the shopping...
All the power and all the space and other resources in the world won't
change that behaviour (and never have, during the past two decades of
experience -- it just grows worse). It is inelegant, and shows a lack
of understanding of how to implement a proper system -- which is why I
have so often referred to my RISC OS machines as "my proper computers".
The Wintel ones aren't, and never can be.
Duncan Wood
wrote:
> In article <3qqdnakNZYS_JATV...@bt.com>,
> Aly <sf333ddf@sfsss'=-3498hfn2nmdf0xdh3222a]].c> wrote:
>> "John M Ward" <jo...@acornusers.org> wrote in message
>> news:4fca8db...@acornusers.org...
>> >
>> > Would a third hard drive and an "upmarket video card" require
>> > hundreds of Watts extra power, multiplying my machine's power
>> > requirements rather
>
>> Yes.
>
> Crazy! Doesn't technology advance any more? My more modern RISC OS
> machines run virtually cold, as the memory chips etc are less wasteful
> of power than they were back in 1994 (which was to do with why I
> mentioned that date in respect of my two-slice RiscPC).
>
>> The power requirements of some of those cards is ridiculous. They
>> draw so much power that they take their feed straight off of the PSU
>> rather than the slot.
>
> Although there is some merit in having greater resources (processing
> power, memory et al) on graphics cards, it does seem to have become an
> over-the-top race to outdo the competition, so they are just getting
> more of everything, rather than better. This is a common error --
> confuding "more" with "better"...
>
I can't help but suspect your machine would however take some weeks to
render a basic Wysiwyg lighting render.
John M Ward
Duncan Wood <nntp...@dmx512.co.uk> wrote:
> On Thu, 07 Aug 2008 10:46:07 +0100, John M Ward
> <jo...@acornusers.org> wrote:
[some snipped]
> > Although there is some merit in having greater resources
> > (processing power, memory et al) on graphics cards, it does seem to
> > have become an over-the-top race to outdo the competition, so they
> > are just getting more of everything, rather than better. This is a
> > common error -- confuding "more" with "better"...
>
> I can't help but suspect your machine would however take some weeks
> to render a basic Wysiwyg lighting render.
Fortunately, that is not the kind of thing I do here. If it were, I'd
have a specialist machine or add-on for that purpose. The rest of the
time, I'd work on the Iyonix or a RiscPC, which for the tasks that
occupy computers on Planet Earth for at least 98% of the time, would no
doubt knock the spots off any Wintel machine that has ever been built.
As Scotty would have said: "The right tool for the right job, laddie!"
(The Undiscovered Country).
Rob Kendrick
John M Ward <jo...@acornusers.org> wrote:
> Fortunately, that is not the kind of thing I do here. If it were, I'd
> have a specialist machine or add-on for that purpose. The rest of the
> time, I'd work on the Iyonix or a RiscPC, which for the tasks that
> occupy computers on Planet Earth for at least 98% of the time, would
> no doubt knock the spots off any Wintel machine that has ever been
> built.
Jingoism if ever I heard it :) Of course, should you want to web
browse, view Flash videos on YouTube etc, encode your music collection
to MP3s, store said MP3s to your iPod or whatever, it's pretty
sub-optimal.
Plus, all that money you save on electricity is most likely spent again
because of the high price of such equipment, and the software to run on
it.
Why can't you be content to say you use it because you prefer it?
B.
John M Ward
Rob Kendrick <nn...@rjek.com> wrote:
> On Thu, 07 Aug 2008 10:16:07 GMT
> John M Ward <jo...@acornusers.org> wrote:
> > Fortunately, that is not the kind of thing I do here. If it were,
> > I'd have a specialist machine or add-on for that purpose. The rest
> > of the time, I'd work on the Iyonix or a RiscPC, which for the
> > tasks that occupy computers on Planet Earth for at least 98% of the
> > time, would no doubt knock the spots off any Wintel machine that
> > has ever been built.
> Jingoism if ever I heard it :) Of course, should you want to web
> browse, view Flash videos on YouTube etc, encode your music collection
> to MP3s, store said MP3s to your iPod or whatever, it's pretty
> sub-optimal.
I don't do any of that apart from web browsing, most of which works well
on NetSurf. Remember: we're not all here as people of leisure, with
time to indulge in purely recreational pursuits. Most of what I do is
serious stuff, so I need a serious tool.
> Plus, all that money you save on electricity is most likely spent
> again because of the high price of such equipment, and the software
> to run on it.
It was never a question of money.
> Why can't you be content to say you use it because you prefer it?
<Vir> "That too!" </Vir> (Babylon 5, series 1, Parliament of Dreams)
Rob Kendrick
John M Ward <jo...@acornusers.org> wrote:
> I don't do any of that apart from web browsing, most of which works
> well on NetSurf. Remember: we're not all here as people of leisure,
> with time to indulge in purely recreational pursuits. Most of what I
> do is serious stuff, so I need a serious tool.
Sorry, I was confused by you saying that your machine could knock the
socks off from Wintel machines for what 98% of Planet Earth's computers
do.
B.
John M Ward
Correct! I work with the real people of this world. They go to work to
earn money. They enter data, they crunch numbers, they produce
accounts, they respond to emails (unfortinately, most work with
Microsoft trash, hence the moronic "top posting" style that now
predominates -- thicko writing, as anyone with any clue would of course
realise) and myriad other real world day-to-day tasks.
I have here a Husky Hunter -- the best in-the-field data collection
device ever created. British, because we tend to be far better at
understand needs and delivering appropriate and well thought out
responses to those needs. It is virtually indestructible, simple in
use, and many years old. There is nothing better that has ever been
produced for this purpose, because people cannot (or will not) see
beyond the ends of their noses.
Most humans are essentially one-dimensional in their thinking, which is
why this world still doesn't have teleports, for example, and filths up
the planet with primitive centuries-old technology that is rubbish.
They never could understand what is worth keeping, and what is better
scrapped, and they nearly always get it the wrong way round.
Back to today's technology at its uses, though...
For leisure people have games _consoles_! Wii, PlayStation or similar.
The right tool for the job. Yes, many have iPods (an Apple technology,
by the way -- not Microsoft) but again these are just leisure toys --
something for Toys'R'Us perhaps, not for serious purposes. People have
to work, first and foremost, to be able to provide for the ancillaries.
It's a matter of priorities -- but one has to be adult enough to
understand that, of course. Put down the foundation first, then build
on that. You cannot do that if you are more interested in iPods and
suchlike than real productivity.
I was the DTI's IT Support Team (a team of one!) for their entire London
and south-east regional structure for more than six years. I saw both
there, and at the businesses with whom we worked, the real world -- not
an isolated corner of it. I saw and learned how a nation can be built
on a solid foundation if it has its principles right, regardless of (in
spite of?) the specific technology. The one think we had right from the
outset was a proper attitude. I lent them one of my RISC OS computers
for a few months, and they were keen to pursue this; but the Treasury
issued an edict that made it impossible.
They'd been bought.
Learn that lesson, all you juniors in Life who are reading this. You
get junk because vested interests issue orders to endure that you have
no choice. I know: I was there -- you weren't. The above is just one
example of many I could quote...
Rob Kendrick
John M Ward <jo...@acornusers.org> wrote:
> > Sorry, I was confused by you saying that your machine could knock
> > the socks off from Wintel machines for what 98% of Planet Earth's
> > computers do.
>
> Correct! I work with the real people of this world. They go to work
> to earn money. They enter data, they crunch numbers, they produce
> accounts, they respond to emails (unfortinately, most work with
> Microsoft trash, hence the moronic "top posting" style that now
> predominates -- thicko writing, as anyone with any clue would of
> course realise) and myriad other real world day-to-day tasks.
Perhaps I'm unusual, but most people I know use computers mostly to
either web browse and use email (for which PCs, Macs, and whatnot are
either better or just as good), use office suites (of which the stuff
available on PCs, Macs, and whatnot are either better or just as good,
and have better compatibility with others), use for computationally
expensive processes (such as compilation/software development, CAD,
media) (for which PCs, Macs and whatnot are vastly superior), or use
speciality software that's just not available elsewhere.
I think blaming Microsoft for people's writing styles is another
example of factless and baseless jingoism, though.
B.
Jules
> One of the Radeon cards apparently takes a total of well over 100 Watts,
> which really is very silly, bearing in mind what it actually is -- just
> a graphics card.
Although back in the 70s and 80s, this is how serious graphics-capable
machines did it - half the machine's logic was dedicated to graphics
processing, containing as much horsepower as the main CPU (with power
consumption to match).
We're currently (as usual with computing!) reinventing the wheel... in
the PC world it's just unfortunate that we're saddled with the
compatibility issues, as otherwise I bet someone could come up with a 'PC'
that was low-power *and* capable of out-performing the current generation
of machines.
cheers
Jules
Duncan Wood
You could market it as a games console...
Jules
>> Although back in the 70s and 80s, this is how serious graphics-capable
>> machines did it - half the machine's logic was dedicated to graphics
>> processing, containing as much horsepower as the main CPU (with power
>> consumption to match).
>
> You could market it as a games console...
Hmm, maybe Sony would have some luck if they marketed a keyboard, mouse
and Linux distribution for the PS3...
Alan Braggins
>On Thu, 07 Aug 2008 09:46:07 +0000, John M Ward wrote:
>> One of the Radeon cards apparently takes a total of well over 100 Watts,
>> which really is very silly, bearing in mind what it actually is -- just
>> a graphics card.
>
>Although back in the 70s and 80s, this is how serious graphics-capable
>machines did it - half the machine's logic was dedicated to graphics
>processing, containing as much horsepower as the main CPU (with power
>consumption to match).
And when the first Apple LaserWriter was introduced, it had a more powerful
CPU than Macintoshes of the time (in order to run PostScript).
Tony Houghton
John M Ward <jo...@acornusers.org> wrote:
> In article <op.ufiglxv5haghkf@lucy>,
> Duncan Wood <nntp...@dmx512.co.uk> wrote:
> > On Thu, 07 Aug 2008 10:46:07 +0100, John M Ward
> > <jo...@acornusers.org> wrote:
>
> [some snipped]
>
> > > Although there is some merit in having greater resources
> > > (processing power, memory et al) on graphics cards, it does seem
> > > to have become an over-the-top race to outdo the competition, so
> > > they are just getting more of everything, rather than better.
> > > This is a common error -- confuding "more" with "better"...
They are better though. The more powerful cards can render more complex
scenes with more realism, do it quicker etc etc. And decode video much
more efficiently than a CPU.
> > I can't help but suspect your machine would however take some weeks
> > to render a basic Wysiwyg lighting render.
>
> Fortunately, that is not the kind of thing I do here. If it were, I'd
> have a specialist machine or add-on for that purpose.
Or a PC with a graphics card. People who don't want advanced 3D
rendering, for games or otherwise, don't have to have one of these
flashy graphics cards just because it's a PC; they can use a more basic
model with on-board graphics.
> The rest of the
> time, I'd work on the Iyonix or a RiscPC, which for the tasks that
> occupy computers on Planet Earth for at least 98% of the time, would
> no doubt knock the spots off any Wintel machine that has ever been
> built.
You don't seem to have much idea of what people use their PCs for and
how much things have changed since RISC OS machines had similar amounts
of power to the competition. If you don't think computers should be used
for leisure/entertainment you're very much in a minority.
> As Scotty would have said: "The right tool for the right job, laddie!"
> (The Undiscovered Country).
And not being a good tool for programmers is a big problem for RISC OS.
A lot more could be done even with the limited hardware it runs on if
programmers were better catered for instead of driven away.
--
TH * http://www.realh.co.uk
Jules
> And when the first Apple LaserWriter was introduced, it had a more powerful
> CPU than Macintoshes of the time (in order to run PostScript).
Indeed. Funny how things have gone backwards in that regard, and
'home' printers now are just cheap pieces of junk which rely on the host
machine to do all the work.
cheers
J.
Tony Houghton
In the old days the processing power needed to render a printout would
have tied up the PC (or whatever) and been inconvenient, but it's fairly
insignificant for today's CPUs, so why not take advantage of them to
make printers cheaper?
The Natural Philosopher
Well getting rid of bloatware operating systems and eye candy would be
a start.
So thats windows, and x-windows ..GONE.
Now lets throw out Postcript. Another total waste of computing power..
And how about tossing object oriented programming as well.
We are back down to not needing more than a few megs of RAM..
> cheers
>
> Jules
>
>
Al Grant
> Running a network of RISC OS, XP, Vista and now an Apple computer, it is
> very clear to me that MicroSoft have cunningly built in obsolescence. All
> of the MS computers I have experienced have a tendency for the software to
> rust and seize up as they age
I've been running Windows 2000 on the same machine
for eight years and it's fine. It copes with all the new
hardware I plug in.
The eight-year-old version of Red Hat on the same machine
looks ancient by comparison. It doesn't do itself any favours
by being unable to drive the screen in as high a resolution as
Windows can - which makes it look a generation older.
>, forcing the user to buy a newer faster lot
> of hardware. It's very clear that MS are deliberately driving hardware
> evolution - which means high carbon cost!
Evolution doesn't necessarily mean more power consumption.
E.g. moving graphics off-chip to graphics accelerators can
reduce the need for the CPU to burn cycles doing repetitive
tasks. Sophisticated power management systems can reduce
overall power consumption. An OS which doesn't keep up with
these is environmentally unfriendly.
If RISC OS was a credible operating system there would be
people right now porting it to the latest low-power ARM cores
with vector processing unit and graphics accelerators.
Yet the core support roadmap seems to stop somewhere in
the last century. So what if your Archimedes or whatever is
as responsive as today's. I got sub-second response out of
my ZX80. The question is how much power you have on tap
to do things in the background or to make compute-intensive
tasks (e.g. media processing) go fast.
Tim Ward
> The Natural Philosopher <a@b.c> writes:
>
>> And how about tossing object oriented programming as well.
>
> crossed my desk recently indicated that 30% of our customers are using
> C++. And this is deeply embedded DSP stuff. Mind you, 2% were using
> BASIC, and 1% claimed to be using XML, so perhaps my customers are
> simply insane.
There's nothing wrong with using C++ for embedded *if* you know what you're
doing, ie you know the difference between one piece of source code that
emits 0 instructions and a similar looking piece that takes millions of
cycles to execute.
There's nothing wrong with XML (well, there is, but not in embedded
performance terms) as a specification language (eg for MIBs or protocols
where the embedded stuff has to match other implementations on other
platforms written in other languages), *provided* you generate something
sensible from it (tables, C++ code, whatever) during the build process to
use at run time, and don't try to parse it on the fly.
(I'm not going to defend BASIC for embedded. Other than VB for lashing up
quick hack management systems to run on the PC.)
John M Ward
Stuart <Spa...@argonet.co.uk> wrote:
> In article <4fca855...@acornusers.org>,> > Interesting! I come from an electronics background, so have long
> > realised that even though PC-type computers tend to come with a
> > minimum 350 Watt power supply -- and often higher -- this includes
> > a decent degree of safety margin. It also allows for internal
> > expansions up to the full capacity of the machine (drives, PCI
> > cards) and even the relatively small current that can be drawn from
> > the USB ports.
> And allowance for peak currents that may be drawn on occasions.
Yup: that too.
> The load on the PSU is far from constant so you choose a PSU that
> will handle all of that and ensure longer life by choosing a PSU of
> higher rating than the minimum needed, so that it is not
> over-stressed.
> The actual KWh will be much lower than that suggested by multiplying
> the PSU maximum rating by time.
All that is true enough, but it is a case of "sauce for the goose". The
same principle will apply to all machines, not just one type. My
A9home's 20 Watt PSU is still going strong, although I have had it for
only (a little under) three years so far. I shall be surprised if it
should fail in less than at least ten years, probably twice that.
Roland Perry
2008, August West <aug...@kororaa.com> remarked:
>> And how about tossing object oriented programming as well.
>
>crossed my desk recently indicated that 30% of our customers are using
>C++. And this is deeply embedded DSP stuff. Mind you, 2% were using
>BASIC, and 1% claimed to be using XML, so perhaps my customers are
>simply insane.
There's nothing wrong with BASIC! Plenty of early business software for
PCs was written in BASIC, and if you wanted it to run faster: compile.
One of my BASIC programs I hand compiled into 8086 assembler - you'd be
surprised how few machine code instructions most BASIC statements
translate into (having debugged the logic using an interpreted version).
--
Roland Perry
Tim Ward
news:BxGh52GP...@perry.co.uk...
>
> There's nothing wrong with BASIC! Plenty of early business software for
> PCs was written in BASIC, and if you wanted it to run faster: compile. One
> of my BASIC programs I hand compiled into 8086 assembler - you'd be
> surprised how few machine code instructions most BASIC statements
> translate into (having debugged the logic using an interpreted version).
Having written BASIC interpreters and compilers I'd not be surprised if many
BASIC statements compiled into a single instruction ... being a call into
the runtime system, which takes as long as it takes, thus the compiled
version runs at pretty well the same speed as the interpreted version.
Tim Ward
> "Tim Ward" <t...@brettward.co.uk> writes:
>
>> "August West" <aug...@kororaa.com> wrote in message
>> news:87proj3...@news2.kororaa.com...
>>> The Natural Philosopher <a@b.c> writes:
>>>
>>>> And how about tossing object oriented programming as well.
>>>
>>> It's everywhere, you know, like a creeping weed. A customer survey that
>>> crossed my desk recently indicated that 30% of our customers are using
>>> C++. And this is deeply embedded DSP stuff. Mind you, 2% were using
>>> BASIC, and 1% claimed to be using XML, so perhaps my customers are
>>> simply insane.
>>
>> There's nothing wrong with using C++ for embedded *if* you know what
>> you're
>> doing, ie you know the difference between one piece of source code that
>> emits 0 instructions and a similar looking piece that takes millions of
>> cycles to execute.
>
> the code my customers' engineers (and I use word losely) write. Believe
> me, they don't know the difference.
That's their fault not yours. (Provided that you've documented any clever
tricks your optimiser gets up to that wouldn't be obvious to an experienced
C++ engineer - you wouldn't want to go to all the trouble of writing a
clever optimisation and then find it wasn't used because the programmers
instinctively avoided that construct.)
John M Ward
Jules <jules.rich...@remove.this.gmail.com> wrote:
> On Thu, 07 Aug 2008 09:46:07 +0000, John M Ward wrote:
> > One of the Radeon cards apparently takes a total of well over 100
> > Watts, which really is very silly, bearing in mind what it actually
> > is -- just a graphics card.
> Although back in the 70s and 80s, this is how serious graphics-capable
> machines did it - half the machine's logic was dedicated to graphics
> processing, containing as much horsepower as the main CPU (with power
> consumption to match).
I well recall the Sun, Apollo and Silicon Graphics workstations; and of
course even the dear old Amiga had four specialist chips for sound and
graphics which might explain why it came second only to the Archimedes
in PCW's wall poster of benchmark results back in late 1987/early 1988.
I still have mine (poster, not Amiga!) somewhere in the loft...
> We're currently (as usual with computing!) reinventing the wheel...
> in the PC world it's just unfortunate that we're saddled with the
> compatibility issues, as otherwise I bet someone could come up with a
> 'PC' that was low-power *and* capable of out-performing the current
> generation of machines.
It is getting to the time when a completely new design will be the only
sensible way forward. Perhaps true 3-D (holographic) displays will
become viable and desirable, for example, or it could be something else
entirely -- but something will trigger a new era in the field, I feel
certain, and quite soon now.
It would be great if British inventiveness and ingenuity were to not
only create this next generation device, but were also to capitalise on
it properly this time. It should certainly be based around a more
modern processor core architecture, so no Intel traditional types should
be considered.
Rob Kendrick
John M Ward <jo...@acornusers.org> wrote:
> It would be great if British inventiveness and ingenuity were to not
> only create this next generation device, but were also to capitalise
> on it properly this time. It should certainly be based around a more
> modern processor core architecture, so no Intel traditional types
> should be considered.
What, like Itanic? :)
B.
The Natural Philosopher
> "August West" <aug...@kororaa.com> wrote in message
> news:87proj3...@news2.kororaa.com...
>> The Natural Philosopher <a@b.c> writes:
>>
>>> And how about tossing object oriented programming as well.
>> It's everywhere, you know, like a creeping weed. A customer survey that
>> crossed my desk recently indicated that 30% of our customers are using
>> C++. And this is deeply embedded DSP stuff. Mind you, 2% were using
>> BASIC, and 1% claimed to be using XML, so perhaps my customers are
>> simply insane.
>
> There's nothing wrong with using C++ for embedded *if* you know what you're
> doing, ie you know the difference between one piece of source code that
> emits 0 instructions and a similar looking piece that takes millions of
> cycles to execute.
The advantage of assembler, is that a casual glance rather than several
years of study, will elicit that ;-)
>
> There's nothing wrong with XML (well, there is, but not in embedded
> performance terms) as a specification language (eg for MIBs or protocols
> where the embedded stuff has to match other implementations on other
> platforms written in other languages), *provided* you generate something
> sensible from it (tables, C++ code, whatever) during the build process to
> use at run time, and don't try to parse it on the fly.
>
> (I'm not going to defend BASIC for embedded. Other than VB for lashing up
> quick hack management systems to run on the PC.)
>
BASIC works reaosnably well if all yiou want to do are imple Thibgs, and
you compile it.
Now who's for some FORTH?
The Natural Philosopher
arena.
What is needed is a stripped down sort of NAC.just runs browser, mail, a
worm processor and a spreadlegs.
Roland Perry
>>>crossed my desk recently indicated that 30% of our customers are using
>>>C++. And this is deeply embedded DSP stuff. Mind you, 2% were using
>>>BASIC, and 1% claimed to be using XML, so perhaps my customers are
>>>simply insane.
>>
>> There's nothing wrong with BASIC!
>
If someone has written a BASIC for it, then there must be a useful
market of some kind.
--
Roland Perry
Roland Perry
Aug 2008, Tim Ward <t...@brettward.co.uk> remarked:
>> There's nothing wrong with BASIC! Plenty of early business software for
>> PCs was written in BASIC, and if you wanted it to run faster: compile. One
>> of my BASIC programs I hand compiled into 8086 assembler - you'd be
>> surprised how few machine code instructions most BASIC statements
>> translate into (having debugged the logic using an interpreted version).
>
>Having written BASIC interpreters and compilers I'd not be surprised if many
>BASIC statements compiled into a single instruction ... being a call into
>the runtime system, which takes as long as it takes, thus the compiled
>version runs at pretty well the same speed as the interpreted version.
The program of mine that I referred to has no runtime system, it's pure
assembler. The only thing it calls is the low level MSDOS operating
system to open and close files etc.
As you say, many of the BASIC statements compile to one assembler
instruction: "IF x>80 then goto 100".
ps I agree that some assemblers do produce nothing more than a set of
calls to a runtime library.
--
Roland Perry
John M Ward
The Natural Philosopher <a@b.c> wrote:
> BASIC works reaosnably well if all yiou want to do are imple Thibgs,
> and you compile it.
I have to admit to having tried some imple Thibgs over the years, but I
have never compiled my code :-)
Al Grant
> Having written BASIC interpreters and compilers I'd not be surprised if many
> BASIC statements compiled into a single instruction ... being a call into
> the runtime system, which takes as long as it takes, thus the compiled
> version runs at pretty well the same speed as the interpreted version.
That's less a function of BASIC than of 8-bit register-poor
targets where simple things like integer addition are done
with calls to runtime support routines. When you start being
able to keep your working set in registers then the difference
between interpreted and compiled becomes more significant.
For complicated floating-point expressions involving long-latency
operations you need to have multiple instructions in flight for
maximum performance. No reason you can't do that for BASIC,
but you do need a proper compiler.
For similar reasons, FORTH is a hopeless way to program
these architectures - at least if you want to exploit more than
a tiny fraction of the performance available to you.
Tim Ward
news:b380be03-dc3e-4e96...@34g2000hsh.googlegroups.com...
> On 8 Aug, 10:55, "Tim Ward" <t...@brettward.co.uk> wrote:
>> Having written BASIC interpreters and compilers I'd not be surprised if
>> many
>> BASIC statements compiled into a single instruction ... being a call into
>> the runtime system, which takes as long as it takes, thus the compiled
>> version runs at pretty well the same speed as the interpreted version.
>
> That's less a function of BASIC than of 8-bit register-poor
> targets where simple things like integer addition are done
> with calls to runtime support routines.
Yes, it was that long ago ...
Rupert Moss-Eccardt
> Aly wrote:
>> "Aly" <sf333ddf@sfsss'=-3498hfn2nmdf0xdh3222a]].c> wrote in message
>> news:4ZCdnSr7Nqp...@bt.com...>>> news:874p5y6...@news2.kororaa.com...
>>>> The DEC VAX-11 was the machine, not the OS; I presume you mean VAX/VMS.
>>>>
>>> Can't quite remember actually. I know that the *thing* that I used
>>> to do the data entry was a DEC VT100 terminal? It was just a dumb
>>> terminal. And then it dialed in to a UNIX style CLI. All very basic
>>> stuff.
>>>
>>
>> YES!! It was definitely a DEC VT100. Just searched for it on Google
>> Images.
>>
>>
> Ah: the terminal that set the standard that every 'glass teletype'
> emulator uses by default.
Except that most VT100 emulators got it wrong. For example many of them
had more than four function keys, supported printing and the screen
handling was mostly done to match a VT102 or, more often, a VT220.
Alan Braggins
>
>worm processor and a spreadlegs.
I'm sure builtin support for porn is needed for widespread adoption,
but do we really need malware support too? Or is a worm processor
a limited firewall? :-)
Al Grant
> "Al Grant" <algr...@myrealbox.com> wrote in message
> > That's less a function of BASIC than of 8-bit register-poor
> > targets where simple things like integer addition are done
> > with calls to runtime support routines.
>
> Yes, it was that long ago ...
Nothing to do with time. I wrote my first program on a
36-bit DEC-20. The 2Gb pen disk I bought several
decades later has an 8-bit 8051 in it. (But think how
much floor space we'd have saved if we'd had it then...)
The Natural Philosopher
The Natural Philosopher
that's because by and large todays chips are designed to run C...
Jules
I've always thought that right track might be to make the assumption that
any one computer is usually connected to several others. Base the OS
around that, and throw it into non-volatile storage within a low-powered,
fanless, small machine - which in a lot of cases won't even need any
local "user" storage.
Unforutunately various companies have tried that and failed over the years
because the interconnection wasn't there outside of specialist
environments (Apollo's Domain was wonderful, and X-terminals came close
to ideal desktop systems - albeit whilst needing a big server on the far
end)
We might be at that point where it could really work now, though - enough
home machines are connected via broadband to the 'net, and of course the
business environment has been all set for years.
My main worry is that Microsoft will end up controlling the software side
again, and the likes of Intel will push for conventional PCs on the
hardware side - resulting in God-awful proprietary software running on
hardware that was nowhere near as efficient as it could be. But the *idea*
is nice, I think...
cheers
Jules
The Natural Philosopher
solid stare disked browser and email, bundle in star office and run
linux under it all..
Jules
> almost any company with some hardware skills could make a stripped down
> solid stare disked browser and email, bundle in star office and run
> linux under it all..
Yep, they could. But could they market it and gain acceptance, whilst
fending off the big corporations from stealing the concepts and
reinventing them for their own purposes?
Probably not - I bet we'd see cut-down versions of conventional PCs
running MS OSes within a year, things would advance hardly at all, and the
public would eventually decide they may as well just fork out for a
normal PC and MS OS.
cheers
J.
The Natural Philosopher
> On Fri, 08 Aug 2008 16:59:04 +0100, The Natural Philosopher wrote:
>> almost any company with some hardware skills could make a stripped down
>> solid stare disked browser and email, bundle in star office and run
>> linux under it all..
>
> Yep, they could. But could they market it and gain acceptance, whilst
> fending off the big corporations from stealing the concepts and
> reinventing them for their own purposes?
>
That might not matter *if it was cheap*.
Lets say you and a bunch of chums team up with Fukyumikro-soft and put a
single board palmtop together based on a someaht old low power cg
intel compatible chipset..strip linux down to the bare necessities, port
Firefox, star office and Thunderbird.. and the whole shebang goes out
with an ethernet and wireless connector for £150 ad a small but decent
screen and keyboard..and USB ports.
> Probably not - I bet we'd see cut-down versions of conventional PCs
> running MS OSes within a year, things would advance hardly at all, and the
> public would eventually decide they may as well just fork out for a
> normal PC and MS OS.
>
If they wanted teh stiff that came with it. The Macintosh shows you
there is a niche market for PCS that don't do very much, but do it
fairly well.
> cheers
>
> J.
>
John M Ward
Jules <jules.rich...@remove.this.gmail.com> wrote:
> On Fri, 08 Aug 2008 16:59:04 +0100, The Natural Philosopher wrote:
> > almost any company with some hardware skills could make a stripped
> > down solid stare disked browser and email, bundle in star office
> > and run linux under it all..
Ah, the "thin client" concept, as per Oracle's initiative a handful of
years ago, with the n|c hardware produced for them by Acorn...
> Yep, they could. But could they market it and gain acceptance, whilst
> fending off the big corporations from stealing the concepts and
> reinventing them for their own purposes?
No. Not even Oracle seems to have achieved that.
> Probably not - I bet we'd see cut-down versions of conventional PCs
> running MS OSes within a year, things would advance hardly at all,
> and the public would eventually decide they may as well just fork out
> for a normal PC and MS OS.
That's the danger. It is effectively a monopoly, so -- as always when
there is little or no genuine competition -- there is very little
progress, just bolting-on this month's new gizmo and patching the most
embarrassing security holes. I have always found the Microsoft world
boring -- there is simply nothing exciting about it.
Andrew Mobbs
>
>single board palmtop together based on a someaht old low power cg
>intel compatible chipset..strip linux down to the bare necessities, port
>Firefox, star office and Thunderbird.. and the whole shebang goes out
I reckon they could do all that and get a WiFi connection in there
too. They might even decide to call it the Eee PC 700
--
Andrew Mobbs - http://www.chiark.greenend.org.uk/~andrewm/
John M Ward
Rob Kendrick <nn...@rjek.com> wrote:
> On Thu, 07 Aug 2008 11:56:07 GMT
> John M Ward <jo...@acornusers.org> wrote:
> > > Sorry, I was confused by you saying that your machine could knock
> > > the socks off from Wintel machines for what 98% of Planet Earth's
> > > computers do.
> >
> > Correct! I work with the real people of this world. They go to
> > work to earn money. They enter data, they crunch numbers, they
> > produce accounts, they respond to emails (unfortinately, most work
> > with Microsoft trash, hence the moronic "top posting" style that
> > now predominates -- thicko writing, as anyone with any clue would
> > of course realise) and myriad other real world day-to-day tasks.
>
> Perhaps I'm unusual, but most people I know use computers mostly to
> either web browse and use email (for which PCs, Macs, and whatnot are
> either better or just as good), use office suites (of which the stuff
> available on PCs, Macs, and whatnot are either better or just as good,
> and have better compatibility with others), use for computationally
> expensive processes (such as compilation/software development, CAD,
> media) (for which PCs, Macs and whatnot are vastly superior), or use
> speciality software that's just not available elsewhere.
Over the years, including right up to just weeks ago, I have supported
real-world users by the hundred, in Government, at home, running small
businesses -- all sorts. They all have the non-choice of millions, so
yes they email using Outlook Distress, which all knowledgeable
commentators have condemned as very poor (and that's putting it kindly).
Even the alternatives, which I have checked out, are messy and clumsy,
so no -- it is not a good platform for email.
No wonder users I encounter all the time have so much difficulty
managing their messages and their systems in general. Very few can
handle even the simplest things (e.g. mailing lists, killfiles,
attachments) and I have to take them through it step by step. I am
still frequently sent very poorly prepared messages in any one (or more
than one) of a number of ways. It is rare (though not unknown) for a
RISC OS user to send emails that are a mess.
This is nothing to do with systems coping more broadly, it is just plain
wrong, sloppy and bad practice. I have extracted material from emails
sent to others that their systems were unable to read, by taking a
complete copy and processing it here. No professional would accept such
low standards in their own sphere of work, so I'm not interested in
anyone trying to find excuses. It's a case of GIGO.
> I think blaming Microsoft for people's writing styles is another
> example of factless and baseless jingoism, though.
Top-posting, broken/ignored sig separators, HTML copies -- all
originated with Microsoft's well-documented attempts to corrupt
standards for their own benefit. If you want to put your head in the
sand that is our choice. Just don't expect anyone ever to take you
seriously again.
John M Ward
Andrew Mobbs <and...@chiark.greenend.org.uk> wrote:
> The Natural Philosopher <a@b.c> wrote:
> >
> >Lets say you and a bunch of chums team up with Fukyumikro-soft and
> >put a single board palmtop together based on a someaht old low
> >power cg intel compatible chipset..
Must it still be Intel-based? That was probably a mistake that Larry
Ellison made with the n|c, after developing it (along with a range of
companies producing their own variants) on the more modern and far more
numerous ARM.
Especially bearing in mind the vast embedded market (already anticipated
back then) of which ARM processors are the dominant type -- including in
mobile 'phones -- that would have been the 21st-century route to have
taken, rather than stagnating. Traditional Intel-type CPUs are just
/so/ last Millennium...
> > strip linux down to the bare necessities, port Firefox, star office
> >and Thunderbird.. and the whole shebang goes out with an ethernet
> >and wireless connector for £150 ad a small but decent screen and
> >keyboard..and USB ports.
With suitable (relatively modest) investment, all that could be done.
We already have a port of Firefox, though it will of course need
continuous updating.
> I reckon they could do all that and get a WiFi connection in there
> too. They might even decide to call it the Eee PC 700
Well, I already have a Risc PC 700...
Tony Houghton
John M Ward <jo...@acornusers.org> wrote:
> In article <0yB*o7...@news.chiark.greenend.org.uk>,
> Andrew Mobbs <and...@chiark.greenend.org.uk> wrote:
> > The Natural Philosopher <a@b.c> wrote:
> > >
> > >Lets say you and a bunch of chums team up with Fukyumikro-soft and
> > >put a single board palmtop together based on a someaht old low
> > >power cg intel compatible chipset..
>
> Must it still be Intel-based? That was probably a mistake that Larry
> Ellison made with the n|c, after developing it (along with a range of
> companies producing their own variants) on the more modern and far
> more numerous ARM.
>
> Especially bearing in mind the vast embedded market (already
> anticipated back then) of which ARM processors are the dominant type
> -- including in mobile 'phones -- that would have been the
> 21st-century route to have taken, rather than stagnating.
> Traditional Intel-type CPUs are just /so/ last Millennium...
Unfortunately, credible web browsing platforms are almost forced to use
x86 if they want a decent Flash plugin, and Intel can apparently offer
better efficiency/price than VIA or AMD.
The technology used to implement that crappy old Intel instruction set
is considerably more advanced than ARM and still being developed faster.
ARM isn't even keeping up with the demands of mobile phones and is being
relegated to a minor role feeding data to a GPU on the same chip.
Meanwhile Intel and AMD aren't just putting GPUs and CPUs on the same
chip:- they're working on chips where the GPU is an extension of the
CPU's instruction set (a bit like MMX or SSE).
--
TH * http://www.realh.co.uk
Jules
> Jules wrote:
>> On Fri, 08 Aug 2008 16:59:04 +0100, The Natural Philosopher wrote:
>>> almost any company with some hardware skills could make a stripped down
>>> solid stare disked browser and email, bundle in star office and run
>>> linux under it all..
>>
>> Yep, they could. But could they market it and gain acceptance, whilst
>> fending off the big corporations from stealing the concepts and
>> reinventing them for their own purposes?
>>
>
> That might not matter *if it was cheap*.
>
> Lets say you and a bunch of chums team up with Fukyumikro-soft and put a
> single board palmtop together based on a someaht old low power cg
> intel compatible chipset..strip linux down to the bare necessities, port
> Firefox, star office and Thunderbird.. and the whole shebang goes out
> with an ethernet and wireless connector for £150 ad a small but decent
> screen and keyboard..and USB ports.
The problem I see is that it doesn't appeal to gamers - but at the same
time it doesn't appeal to business types, who largely mistrust Linux, Open
Office, and anything else you throw at them which doesn't either have
Microsoft on the label or (at a stretch) someone like Sun. When it
comes to computing, business types seem to hate trying something new, no
matter how good the benefits might look on paper.
>> Probably not - I bet we'd see cut-down versions of conventional PCs
>> running MS OSes within a year, things would advance hardly at all, and
>> the public would eventually decide they may as well just fork out for a
>> normal PC and MS OS.
>>
> If they wanted teh stiff that came with it. The Macintosh shows you
> there is a niche market for PCS that don't do very much, but do it
> fairly well.
True, and I think if there's going to be a radical change it'll come from
Apple if anywhere. But they seem to be going the route of increasing
bloat and of using the same power-hungry CPUs as everyone else, coupled
with marketing over-priced stuff that doesn't really do anything new or
innovative. I fear they're going to spend the next decade or two trying to
be the same greedy company as Microsoft turned out to be.
cheers
J.
Jules
> In article <pan.2008.08.08....@remove.this.gmail.com>,
> Jules <jules.rich...@remove.this.gmail.com> wrote:
>> On Fri, 08 Aug 2008 16:59:04 +0100, The Natural Philosopher wrote:
>> > almost any company with some hardware skills could make a stripped
>> > down solid stare disked browser and email, bundle in star office
>> > and run linux under it all..
>
> Ah, the "thin client" concept, as per Oracle's initiative a handful of
> years ago, with the n|c hardware produced for them by Acorn...
Yep, and before them there were x-terminals, and before that there were
character-based terminals... the concept's nothing new. The problem I
think is that we went too down the fat client road, that one OS software
company achieved market dominance, and that silicon's become so cheap that
few people care.
Sure, people will gripe and moan about power consumption and component
lifetime and cruddy software, but unless a company with both the product
*and the marketing budget to make people aware of it* comes along, we're
stuck with that setup for the foreseeable future.
cheers
Jules
Martin Bazley
spouted the following:
> I think blaming Microsoft for people's writing styles is another
> example of factless and baseless jingoism, though.
>
You cannot, however, deny that Outlook Express is responsible for a
large part of the top-posting and HTML problems we have today. I'm not
saying that you don't have a choice - I don't have enough experience
with the program to comment - but unfortunately, most people stick with
the defaults because they're too thick to do otherwise.
--
__<^>__ === RISC OS is a work of art. Some people adore it, ===
/ _ _ \ === others can't see the point of it, and it's really ===
( ( |_| ) ) === expensive. ===
\_> <_/ ======================= Martin Bazley ===================
Rob Kendrick
You keep thinking that if you want. Most are more than able to write
dreadful emails, regardless of where their mail client puts the caret.
B.
John M Ward
Rob Kendrick <nn...@rjek.com> wrote:
> On Fri, 08 Aug 2008 17:13:15 GMT
> John M Ward <jo...@acornusers.org> wrote:
> > > I think blaming Microsoft for people's writing styles is another
> > > example of factless and baseless jingoism, though.
> >
> > Top-posting, broken/ignored sig separators, HTML copies -- all
> > originated with Microsoft's well-documented attempts to corrupt
> > standards for their own benefit. If you want to put your head in
> > the sand that is our choice. Just don't expect anyone ever to take
> > you seriously again.
>
> You keep thinking that if you want.
I shall. I have archives here of thousands of emails from all comers,
and it would be simple enough (apart from the sheer size of them all,
and somehow hiding the content and email addresses) to demonstrate the
veracity of my claim. I've just flicked through some of them as a
reminder. Yup: I was right.
druck
[56 lines of quoting removed]
[3 lines of reply]
Please learn to snip.
---druck
--
The ARM Club Free Software - http://www.armclub.org.uk/free/
The 32bit Conversions Page - http://www.quantumsoft.co.uk/druck/
druck
> The Natural Philosopher <a@b.c> wrote:
>>Lets say you and a bunch of chums team up with Fukyumikro-soft and put a
>>single board palmtop together based on a someaht old low power cg
>>intel compatible chipset..strip linux down to the bare necessities, port
>>Firefox, star office and Thunderbird.. and the whole shebang goes out
>>screen and keyboard..and USB ports.
> I reckon they could do all that and get a WiFi connection in there
> too. They might even decide to call it the Eee PC 700
You are assuming this discussion is take place in 2008, most poeple
here wont accept a world post 1997.
Jules
> On 8 Aug 2008 The Natural Philosopher <a@b.c> wrote:
>
> [56 lines of quoting removed]
>
> [3 lines of reply]
>
> Please learn to snip.
Gotta use those CPU cycles for something...
Theo Markettos
> You are assuming this discussion is take place in 2008, most poeple
> here wont accept a world post 1997.
Where by 'here' you mean comp.sys.acorn.advocacy. It's most amusing,
reading this thread, on watching the interaction between the
comp.sys.acorn.* regulars and the cam.* regulars.
So, csa.* people, what about browser plugins?
So, cam.* people, what about guided buses?
Oh, so we're already talking about both of those...
:-)
Theo
The Natural Philosopher
A few years of recession and the token will pass to china.
You can bet they will be far happier with thinner clients and all the
data under state control ;-)
> cheers
>
> J.
>
>