Ruzzen fruzzen mglle frggle news server...
Andrew Maizels
Uwe Sattelkow wrote:
> I need general information about CPUs, about architectures, etc. Anything
> relating to the topic "What makes a CPU fast?", nothing on specific CPUs.
> Could anyone tell me, where I can get information from? (webpages, ...)
The critical thing is to use fresh CPUs. It's well known that CPUs slow
down over time; for example, my SGI O2 was quite zippy when I bought it
back in 1997, today it's rather slow, and I expect within another year
or two it will be almost unuseable.
This has given rise to a tremendous scam that has netted major computer
vendors billions of dollars over the years. This is how it works:
New CPUs are fresh and clean, with wide-open pathways that allow
electrons to zoom freely from pin to pin. As time passes, the
occasional electron will get stuck in a tight corner, or overshoot an
output buffer and hit the insulator and shatter. These particles -
electrons and fragments of electrons such as deutrinos and kleptons -
are known in the business as "cruft".
The "cruft" gradually builds up and clogs the once-wide paths, so that
the problem starts to accelerate. Soon the once-fast system is getting
old and slow, and at this point the vendor steps in and offers an
"upgrade".
In fact, this "upgrade" is nothing more than a fresh, clean CPU. Even
worse, the vendor will then take your old CPU, clean out the "cruft"
with compressed air (they used to use freon, but this has since been
banned), and - get this - re-sell the refurbished CPU as an "upgrade" to
another customer!
What's more, vendors have over time been *deliberately* making the
pathways on chips ever-narrower! The broad 3-micron boulevardes of the
old days have been replaced by twisty 0.18 micron alleys, and if they're
not stopped, we'll soon be trapped within goat tracks just 0.1 micron
wide, forcing the electrons to move sideways!
I encourage everyone to write to their local media outlet or political
representative to protest this outrageous behaviour!
Andrew.
--
There's only one game in town.
You can't win.
You can't break even.
You can't quit the game. -- The four laws of thermodynamics.
Andrew Maizels
> Well, let's see if that upstream server is working today...
Looks like it is! Good wombat, have a biscuit!
Andrew Maizels
>
> Andrew Maizels wrote:
>
> > Well, let's see if that upstream server is working today...
>
> Looks like it is! Good wombat, have a biscuit!
And the tired little wombat finally arrives back at his burrow, having
made a round trip of 40,000km on nothing more than a biscuit and a cup
of orange-flavoured sports drink.
Juan Pérez
--
Are you hypo-(er)- thermically challenged?
Are you often delusional?
Andrew Maizels <and...@one.net.au> wrote in message
news:3A96F833...@one.net.au...
Juan Pérez
Andrew Maizels <and...@one.net.au> wrote:
> Well, let's see if that upstream server is working today...
>
> Uwe Sattelkow wrote:
>
> > I need general information about CPUs, about architectures,
etc. Anything
> > relating to the topic "What makes a CPU fast?",
* nitrogen *
* Canned Farts work better. *
Andrew Maizels
>
> Andrew? Will you be alright?
I'll probably survive.
> Are you hypo-(er)- thermically challenged?
Hyper. Too damned hot at the moment.
> Are you often delusional?
Not nearly often enough!
People actually *read* alt.test.wombat these days?
A.
PPHMII
Spanked out upon the buttocks of barbie in morse code:
/"Juan Pérez" wrote:
/>
/> Andrew? Will you be alright?
/
/I'll probably survive.'
Oh well, things could be worse then
/
/> Are you hypo-(er)- thermically challenged?
/
/Hyper. Too damned hot at the moment.
I'll meet you half way, I'm fucking freezing up here.
/
/> Are you often delusional?
/
/Not nearly often enough!
Get a more relyable dealer.
/
/People actually *read* alt.test.wombat these days?
No, You are deluded.
/
/A.