Sparc vs i386 architecture

[Ted Mittelstaedt]

>-----Original Message-----
>From: owner-freebsd-questions at freebsd.org
>[mailto:owner-freebsd-questions at freebsd.org]On Behalf Of Danial Thom
>Sent: Tuesday, January 10, 2006 5:28 AM
>To: freebsd-questions at freebsd.org
>Subject: RE: Sparc vs i386 architecture
>
>
>
>
>--- Ted Mittelstaedt <tedm at toybox.placo.com>
>wrote:
>
>> 
>> 
>> >-----Original Message-----
>> >From: owner-freebsd-questions at freebsd.org
>> >[mailto:owner-freebsd-questions at freebsd.org]On
>> Behalf Of Gerard Seibert
>> >Sent: Sunday, January 08, 2006 7:02 AM
>> >To: freebsd-questions at freebsd.org
>> >Subject: Re: Sparc vs i386 architecture
>> >
>> >
>> >Danial Thom <danial_thom at yahoo.com>
>> >
>> >> Frankly, people who spend $9000. worth of
>> time
>> >> dicking around with some old piece of junk
>> to
>> >> avoid buying a $400. computer crack me up.
>> :)
>> >
>> >Yes, it is amassing. I have a friend who has
>> spent thousands of dollars
>> >keeping his old car running. He could have
>> purchased a new one with a
>> >new warranty, etc. and have saved all that
>> money, but he refused. For
>> >some individuals, the challenge is the real
>> thrill that they crave.
>> >
>> 
>> That is an interesting, if very inaccurate,
>> analogy, and as a car guy
>> that does my own wrenching, let me tell you
>> why.
>> 
>> Computer gear every year gets cheaper and
>> faster and better.  Cars
>> by contrast, have not improved much over the
>> last 20 years - unless you
>> count larger cupholders as an improvement - and
>> espically they haven't
>> changed at all over the last 10.  Ever since
>> EFI and airbags became
>> standard on vehicles there just haven't been
>> any compelling or significant
>> improvements.  In fact for many models, the
>> engine designs themselves
>> are the same as 20 years ago.  For example the
>> 2.4 Turbo used in 
>> the 2006 Chrysler PT Cruiser went into
>> production in 1994, the 4 speed
>> computer-controlled transmission used in that
>> car went into production
>> in 1989.  Many parts for that transmission in
>> fact are the same - how
>> many computers do you know that you can use a
>> 17 year old part in?
>
>you need to get out more, Ted, if you think cars
>haven't improved much in 20 years. I know you
>can't afford a BMW or Mercedes, but you can test
>drive them for free if you put on a nice suit.
>Drive a '95 BMW and a 2005 and you'll see there
>are huge differences in power, handling, brakes
>and a whole lot of stuff you can't see, not to
>mention safety. Anti-skid brakes were a joke when
>they first came out; now I can stop on a dime.
>Traction control actually works now.
>

There have been improvements in suspension - however
suspension is largely a preference thing.  You 
for example probably would think the suspension
tuning I prefer is way too harsh, I by contrast would
probably think the suspension in a new BMW or
Mercedes that you prefer feels like a speedboat
wallowing in the river.  As for brakes, a lot of the
so-called "improvement" in stopping power in brakes
is achieved by larger rotors - and to get these 
without making the wheel bigger and heavier you
have to make the tire low profile - which means a
more expensive tire, and a lot more frequent incidents
of bent rims if the roads in your area have a lot of
potholes (which is common on the east coast during the
winter as the road salt and such destroys the asphalt.

And safety, yes there's more airbags, and yes the
bodies of cars are safer but once again, thats
a tradeoff too.  To make the car bodies safer the
frame is strategicly weakened in areas to increase
the "compressibility" of the body so that in a massive
crash, the car folds up around you.  The downside is
that in low-speed 10-15Mph collisions where you would
survive them in a less compressible body anyway, now the
car has to be totaled out because the body simply folds
up if it's barely tapped.  Once more your getting
a small increase in survivability by making
everyone else have a lot more expensive-to-fix car,
which drives up insurance rates.  That might be
agreeable until you look at the percentage of the
major collisions and discover most of them were caused
by drunks, who possibly the society as a whole would
be better off if the drunk had died in the accident
instead of being saved by the air bags.  (since
quite often the non-drunk people in these kinds of
collisions are killed while the drunks survive,
due to their bodies being more relaxed)

Most of what your seeing as improvements are merely
changes in the tradeoffs in automotive designs that
have always existed.  In the olden days, people
cared more about lower lifetime maintainence costs,
so manufacturers were more apt to choose a simpler
and more bullet proof design, today by contrast
people view cars as disposable if they go wrong,
so manufacturers are more apt to choose the complex
and less-bullet proof design if it creates some
minor feature that they think will help them sell
a car.

The only real new things that have come down the
pike as it were is production hybrid powertrains,
but the only thing that made those a reality is massive
government subsidies.  20 years ago if we had the same
government subsidies in place for hybrids, we would
have had them then.

>But computers are not comparable to cars. It
>might make sense to keep an old car going as a
>new car depreciates much more than your time is
>worth in the first year, and they aren't cheap.
>

No, they definitely aren't!  And another sad fact
is that as they get more computerized and more
complex, the typical car problems that are appearing
now more and more are beyond the capabilities of
the average mechanic to fix.  Just imagine for
example what the Toyota Prius owners are going to
be doing once their roach coaches are all out from
under the Toyota warranty.

Ted