In comp.os.linux.advocacy, High Plains Thumper
on Wed, 09 Aug 2006 11:00:56 +0900
> Roy Schestowitz wrote:
>> Dell laptop blew up the truck
>> ,----[ Quote ]
>> | There seems to be no doubt in Forqueran's mind that his Dell
>> | Inspiron 1300/B130 set his truck ablaze.
>> | "I want people to know what happened to me," he said. "These laptops
>> | are dangerous and people need to know."
>> This points to some dodgy Web site. Take this with a grain of salt.
> What is not stated is whether the laptop was connected to the truck's
> electrical system, to charge the laptop battery. Also what is not stated,
> is whether the laptop was powered up or not.
> With radio controlled electric aeroplanes, there have been documented
> instances of LiPo batteries exploding and bursting into fire whilst
> charging. Also there have been instances of batteries bursting into fire
> during use in flight.
> It is best to be present whilst LiPo batteries are charging. If something
> goes wrong, it can be dealt with right away. Since batteries are nothing
> more than corrosion cells, they are most active when charging or
An interesting and correct way of putting it. There's one
character who used to claim that caps can outperform
batteries -- he didn't last all that long. :-)
Batteries don't do well against gasoline engines,
either -- hybrids are taking advantage of a quirk in
the reciprocating piston engine's efficiency curve.
The methane-fueled batteries look interesting for such
applications as laptops, and I'll admit to wondering how
well continuously variable transmissions will do against
the current hybrid crop. I compute that we should see 100
mpg vehicles -- grossly underpowered compared to today's
cars, but about the same size and speed otherwise --
but nothing bigger, simply because of air drag. If we
can get into smaller, bullet-shaped single-seater units,
we might get higher mpg. Motorcycles have their own
problems, though, and then there's the pasta equivalency
issue. The Tour de France is a little extreme, perhaps:
but presumably the bicycle commuter would still have to
pay for the equivalent of fuel, in this case pasta fuel.
I'll admit to wondering how much desktop computing power
a kilowatt can maximally buy (most PCs are about 300W or
so now, but rather variable since it depends on whether
they're sitting there doing nothing, chewing on heavy
computation tasks such as SETI, Distributed.net's RC5-72
and OGRp2 projects, or maybe a BBC-sponsored climate
analysis project, or swapping like crazy because it's an
overloaded server), and how best to measure said power.
It's worth noting that 0.1 micron = 100 nm is getting
rather close to atomic size -- a 100 nm strip of aluminum
metal would be about 700 atoms wide, if not less, though
the exact number depends on a variety of factors.
IBM has an interesting structure, though scaling up might
be a problem:
Their diagram should ideally be fuzzier, since 6nm = about
42 atoms. Compared to today's transistors it's a little
complicated -- and power density is going to be interesting,
to say the least.
We'll see how far Moore's law can go. :-) With luck Linux
will help it get there; while NT has solved the "busyloop"
problem of Win95 (which had Linux running parts up to 30%
cooler at one point!), it appears to me that Linux will
stress transistors more evenly, with lower power. Then
again, we'll just have to see.
Windows Vista. Because it's time to refresh your hardware. Trust us.