10:42
- I played around with some work stuff over the weekend, but I
pretty much took the weekend off. That was the first in a very long
time. I'm getting caught up with mail this morning.
Autumn has finally arrived in Winston-Salem, although most of the trees
are still completely green. The temperature this morning was about
44°F (7°C). The highs this week are to be just under 60°F
(15°C) and the lows around 40°F (4°C). We haven't turned on
our furnace yet. The indoor temperatures are hovering around 65°F
(18°C). But we may yet have frost on the pumpkin.
And, this morning for the first time this season, I'm drinking hot tea
instead of Coke.
When I finish a book, I generally take a short break before beginning
work on the next project. I haven't done that this time, because I'm
trying to increase my output. For 2006, my goal is to have two new
titles and one revised title in the bookstores, and one more title
nearing or at completion. I'd like to maintain that 3 to 3.5 books/year
pace for the next ten years or so, so that Barbara and I can build up
our retirement nestegg. When we retire, I'll cut back to only one or
two books a year, and I may consider writing a novel or three.
08:50
- This is very strange. Every morning, I back up our working
data to a DVD+RW disc. I use 4X discs, and the backup generally runs
pretty close to 4X. After the write completes, the K3b software opens
the tray and then closes it again to reload the disc to do a verify
pass.
A few days ago, I cleaned things up after finishing the last book, so
our working data directories are down to about 2 GB. Ordinarily, that 2
GB would take 6 minutes or so to write at 4X. When I started the backup
this morning, though, the disc ejected after only a
couple minutes. I figured the disc was bad, but when I looked at
the K3b status screen, I saw that K3b had somehow apparently written
the disc at 17.53X rather than the normal 4X. Hmmm.
Not expecting much, I let the verify pass run. At the end of it all,
K3b said all the files matched. The only thing I can figure out is that
K3b overwrote only changed files and then did a compare against all of
the files, changed and unchanged. I'm not sure how it could that,
though.
Wednesday, 26 October
2005
08:50
- I spent most of yesterday completing a marketing survey for
the new book. O'Reilly's marketing folks need information about who the
book will appeal to, how to reach those prospective buyers, lists of
bullet points, and so on. It sounds simple, but it's one of the hardest
things I have to do.
Another one bites the dust. Other than my notebook, which runs
Windows because we use it in the field to run Windows-only astronomy
software, the only Windows box in the house had been ripper. No more. I finally got
disgusted with Windows XP and installed Xandros 3.0 Deluxe to replace
it. It took a bit of tweaking to get the software I needed to run under
Crossover Office, but I eventually managed to get it working.
Interestingly, it's not only more stable under Xandros, it's actually
faster as well. I do wish the guy who wrote the application and
distributes it freely would release the source code and put it under an
open-source license, though. We'd soon have a native Linux version.
I finally gave up on the Aria system, which was stable under Linux, but
crashed frequently under Windows XP. So I swapped that out for the
original ripper, an AMD
Sempron system that had been--you guessed it--stable under Linux but
crashed frequently under Windows XP. I blew away Windows on that system
and reinstalled Xandros 3.0 Deluxe. After some tweaking, I got the
Windows ripping application running just fine under Xandros with
Crossover Office.
The striking thing is how much faster a Prescott-core Pentium 4/3.2
processor is at compressing and encoding than a Sempron 2800+. I mean,
close to a factor of two. The Pentium 4/3.2 typically took 40 to 50
minutes to compress/encode, while the Sempron routinely takes 90 to 100
minutes.
Which brings up an interesting juxtaposition of articles I read
yesterday. The first, from ExtremeTech, is The
Right Desktop Processor: CPU Price/Performance. In it, the
Prescott2M-core Pentium 4 630 does extremely well in price/performance
against the AMD competition. The second, from AnandTech, is Intel's 65nm
Processors: Overclocking Preview. In it, AnandTech concludes, "Unfortunately,
we will have to wait for Intel's next-generation
processors for a true competitor to AMD's low power Athlon 64s."
The difference, of course, is that AnandTech focuses on gaming, for
which AMD processors are unquestionably superior. But, as quick glance
at the benchmarks in the ExtremeTech article shows, AMD doesn't have
the playing field all to itself. I only wish they'd extended their
comparison to include the Celeron D and Sempron processors. The
bang-for-the-buck ratios would have been much more interesting.
Thursday,
27 October
2005
10:45
- The
Empire Strikes Back. As I said before, this one is going to get
nasty. Office format lock-in is one of the major pillars that the
Microsoft monopoly is founded upon. If that goes away, Microsoft's
empire begins to crumble. So you can bet that Microsoft is bringing
every gun to bear on this, and every Massachusetts politician and
bureaucrat that money can buy. Microsoft will do its best to crush
Quinn and the other ODF advocates by fair means or foul. Expect the
Microsoft FUD factory to be working three shifts from now on. They
can't afford to lose this one.
11:58
- I realize must get at least one more production system built.
I need a Windows XP system on my desk to run the astronomy software
that I'll be using for one of the books I'm currently working on, and I
suppose I can kill two birds by building one of the project systems for
the next hardware book. I'd like that to be the "budget" system that I
plan to build around one of the new Intel motherboards that uses
the ATi chipset with integrated graphics, but even though that
motherboard is shown on the Intel web site, it isn't shipping yet and
may not be for several weeks. So I suppose I'll build an Athlon 64
system instead.
We were hoping to go out observing tonight, but the Clear Sky Clock
isn't looking promising. Tomorrow night is the public Mars exhibition
sponsored by Forsyth Astronomical Society and SciWorks, our local
nature-science center. There are likely to be hundreds of people there,
and FAS is pleading for all club members to come out and help. Barbara
and I won't attend, though, mainly because we think it's a dumb
idea.
At its largest, Mars is a tiny object that requires excellent seeing
and high magnification. Even given that, most non-astronomers will be
underwhelmed by the view. Our atmosphere is very turbulent right
now--the jet stream is right over us--and is likely to be so tomorrow
night as well. That means it's useless to use high magnifications,
because the atmospheric instability makes the image so blurry as to be
useless. And, to top it all off, there's a huge dust storm on Mars
right now, which pretty much obscures all of its surface features. That
means, at best, all the public visitors will see is a tiny, blurry,
salmon-colored dot. The morning newspaper, of course, mentioned that
Mars reaches closest approach this Saturday evening, and included a
Hubble Space Telescope shot of it. Nothing like raising unrealistic
expectations among the public.
Our friend Paul Jones had an excellent suggestion. Instead of doing a
public Mars observation, we should be doing public observations of
Jupiter and Saturn once they're well placed for an evening session.
Jupiter and Saturn are both impressive objects, even for
non-astronomers. If we want to get people interested in astronomy, it
makes a lot more sense to show them something interesting than
something as singularly unimpressive as Mars.
Saturday,
29 October
2005
00:00
-
00:00
-
Copyright
© 1998,
1999, 2000, 2001, 2002, 2003, 2004, 2005 by Robert Bruce Thompson. All
Rights Reserved.
