Email Robert |
Daynotes
Journal
Week of 16 August
1999
Sunday, 22 August 1999 08:10
A (mostly) daily
journal of the trials, tribulations, and random observations of Robert
Bruce Thompson, a writer of computer books. |
TTG Home
Robert Home
Daynotes
Home
Special
Reports
Current Topics
|
Monday,
16 August 1999
[Last
Week] [Monday] [Tuesday]
[Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next
Week]
The Register reports
this morning that Opera is gaining share in the browser market. According
to their latest figures, IE has 43.7%, Navigator 35.4%, and Opera 9.46%.
Opera claims that last figure has now climbed to 10.2%. That may be true,
although I doubt it. I just checked my web stats for last week. On my site
last week, IE had 62.97%, Navigator 32.09%, and Opera only 1.03%. The
remaining 3+% was accounted for by a bunch of different browsers,
including several running on unusual operating systems. My guess is that
my figures are more reliable than those that The Register published, which
they admit come from a site visited by people who are "mostly
developers, Web site designers, journos and browser nuts". In fact,
my figures are probably high, too. I'd guess that my readers are much more
likely to use Opera than is the general population. Overall, I'd be
surprised if Opera has more than a 0.05% share of the browser market.
The Register also published an article
saying that the Pentium II is now officially dead, something I'd declared
several weeks ago. So those of us still running Pentium II systems are now
officially using legacy processors, I guess...
* * * * *
This from Bo Leuf [bo@leuf.com]:
"What's depressing is that I know some women who are into PCs
in a big way. Their work areas invariably resemble Barbara's."
If you lend me your Olympus, I'd post some
comparative pictures <g>. I am blessed with a wife who is not
compulsively neat, but she does sometimes mention that she regrets not
having the dining-room area free instead of an office space there.
I fight a sometimes win, more often lose
battle against clutter, and my wife and I share in the battle against
paper clutter -- stacks grow on any flat surface, because invariably you
have something in hand and don't know where to put it, so you put it on
the nearest surface "for later". Next time you look that way
again, there is a stack of totally unrelated papers inches deep. Never
fails. Then you stand with the stack in hand, wondering where to put
that to go through it, and the phone rings, so the stack goes on top of
another stack...
There is something to learn from that.
Nature's way of archiving? Perhaps just a cosmic law for sedimentation
processes?
/ Bo
--
"Bo Leuf" <bo@leuf.com>
Leuf fc3 Consultancy
http://www.leuf.com/
I like it. The sedimentary theory of office organization. Perhaps
millions of years from now they'll be mining paperstone from our former
office locations. Entropy: it's not just a good idea. It's the law.
* * * * *
This from Tim Werth [twerth@kcnet.com]:
I can wholeheartedly agree w/everyone's
stories about wife's cleaning up offices. Both from my own experience
and also from observing my parents over the years. My question is more
generic about the female species however. Why is it that most women that
I've known insist on keeping the house clean but at the very same time
trash their cars??? I can think of several exceptions to this but as a
whole it tends to hold up. Everything has to be in its place in the
house but in a car women just throw stuff over their shoulder. I know,
why ask why. ;-)
Hmm. You may be right statistically, but I must confess that I'm
no more prone to cleaning up my car than my office. As a matter of fact,
that's one of the things that I miss most about my Jeep CJ. I used to be
able to clean out the inside by removing the drain plugs from the floor
boards and hosing it down. I can't do that with my Trooper. Well, I could,
but the carpet would get all soggy and hard to light. Barbara, on the
other hand, has a neat car as well as a neat office. Her Trooper usually
looks like it just came off the showroom floor. Mine, on the other hand,
is lucky if it gets washed every couple of years. Of course, I keep it in
the garage most of the time, and put an average of about 25 miles a month
on the odometer (literally).
* * * * *
This from Robert Rudzki [rasterho@pacbell.net]:
i like your office just as is, along the
equipment closet, i myself ran out of space in our shared 13 x 13 foot
spare bedroom/office/computer lab/armory, etc. so i set up a caseless pc
on the coffee table in the living room! It is just amazing how easy it
is to troubleshoot a pc when you don't have a cheap case [$19.95 at
Fry's!] to get in the way and cut your fingers and hands to shreds!
i am sure the first woman living with the
first caveman in a cave had a strong interest in sweeping the place out
regularly from bone and hide scraps, they attract insect pests and
scavenger animals, then she found some pretty flowers down by the creek
when she was getting fresh water to bath in, so she brought them home
and before you know it, Martha Stewart evolved from that first woman
living with the first big lug of a caveman... and it was only a blink of
the eye in GMT [Geologic Martha Time...] =8^-)
What a good idea. A PC in the living room. Actually, I've brought
up from time to time my idea of putting 100BaseT jacks in our den so that
Barbara and I could each have a PC near where we sit in the evening. I
mentioned it again last week, and for the first time Barbara didn't object
to the idea. She still doesn't think much of my idea of putting jacks in
the bathrooms, though.
* * * * *
This from Jan Swijsen [qjsw@oce.nl]:
>Hmm. Lucky she didn't shoot.
Lucky someone in the car didn't pull a gun
either.
On the cleanup of your office.
I think you should clean up just a little.
For example if you move out all the (empty?) boxes you could get just
that much more equipment in. <g>.
I did notice a loose box on the ground
somewhere on the photo of Barbara's office. Couldn't you use this as an
argument, pointing out it is just a gradation of cleaned-up-ness ?
<g>.
I don't have a regular office at home, I
have all my equipment in my bedroom resulting in a mixed chaos of
clothes and books and computer bits distributed on all free surfaces.
(The definition of a free surface here is 'any surface that stuff
doesn't fall off from'). I am moving the computer stuff over to the
attic but I am afraid the chaos will just follow me there. I do, just
like your Barbara, put the extraneous bits back into the box and close
it. Just because I don't have free space to drop it and closing the box
provides a flat usable surface. When I look at my brothers workspace (he
is not in computers but in DIY and woodworking) I see the same kind of
chaos. His partner produces even more chaos. She was teaching arts and
she gets the whole house overflowing with bits and pieces of art and
school work and 'potentially useful' items. She does however complain
about the unordered layout of his tool shed :) . It is not the computers
that cause this and it clearly is not only a guy thing.
It is, as you said, the task focused way one
lives.
BTW If all our offices look 'a bit
disorganized' is it a wonder that (MS) Office is a bit (crashingly)
disorganized ?
Entropy wins. All times, every time.
Svenson.
Actually, you're right. I should clean up a little, and I
probably will. I know it's time to do that when I start finding that I
can't immediately put my hands on something I want. The other day, I was
scratching around looking for some of the components I need to build my
new system. I knew that there were two Pentium III/550 CPUs around here
somewhere, but I couldn't find them. Barbara said, "how can you
possibly lose two Pentium III/550s?" Good point. As far as the loose
box on the floor of Barbara's office, would that it were true. In fact,
that's a very neatly organized box of CDs.
|
TTG Home
Robert Home
Daynotes
Home
Special
Reports
Current Topics
|
Tuesday,
17 August 1999
[Last
Week] [Monday] [Tuesday]
[Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next
Week]
A more-or-less Y2K issue hits this Saturday at 8:00 p.m. If you have an
older Global Positioning System (GPS) receiver, it may go berserk or stop
functioning then. See this page
for more details.
* * * * *
This from Bo Leuf [bo@leuf.com]:
In passing, apropos your figures...
"IE had 62.97%, Navigator 32.09%, and Opera only 1.03%."
Note that Opera by default reports itself
"as being Netscape" (actually Mozilla) for server
compatibility reasons -- not because it is Netscape compatible (it
isn't) but simply in order not to invoke IE-specific functionality/pages
(which by now seems often to be default server/site behavior). The full
HTTP_USER_AGENT string for my system is actually: "Mozilla/4.0
(Windows NT 4.0;US) Opera 3.60 [en]", but a lot of analytic code
will stop at the Mozilla part and class this as Netscape 4. I think
earlier versions did not even report the Opera version, only the Mozilla
part.
/ Bo
--
"Bo Leuf" <bo@leuf.com>
Leuf fc3 Consultancy
http://www.leuf.com/
Well, yes, but nearly all versions of any mainstream browser
identify themselves as Mozilla for just the reason you mention. I use
Analog to analyze my logfiles, and it does indeed differentiate between
the different browsers.
* * * * *
This from Robert Rudzki [rasterho@pacbell.net]:
Well, the old coffee table PC just could not
see any of the 3 nic's, so I tried 2 more old tired mobos I had sitting
in the junk box they had various problems that I barely remembered so I
wasted several more hours of my life, but did watch the VIPER movie with
the main spousal unit Taylor while I played with the hardware. Do they
have teenage boys write and direct TV action movies these days?, it was
really juvenile and technically impossible in at least 10 different
areas.
All the old cases and mobos went into the
trash today, Wednesday is pickup day... I saved the 70 Meg of EDO and
FPM of generic memory in case I need to waste more hours of my life at a
later date. =8^-)
I read Dr. Pournelle's Byte column today and
his remark about the 2 far-end connectors on the new SCSI internal
cables being different piqued my interest, do I load and lock straight
tracer then alternating AP and HE into the water-cooled twin fifties, or
do I do a little research before I shoot him out of the sky?
Heh, good thing I did the research
or I would have had egg on my face...
Discretion is the better part of valor, to
coin a phrase.
I see that the Wall Street Journal today had
an article about predictive dialers now hanging up when they hear a real
person but recording their message on your answering machine, even to
sounding like real people talking with the 'ah's' and 'em's'. They use
special algore-rythms to detect live speech and hang up, yet record if
the program detects a synthetic voice.
Janet Reno told us today that gun buyers
need to take written tests to see if they know shooting kids at a church
run school is illegal and a reporter asked if that would have prevented
the recent Buford R. Furrow from shooting all those kids and she refused
to answer the question.
The famous Brady Bill would not have stopped
Hinckley for a New York Second, since he had purchased the gun 4 months
before shooting Ron Reagan and all his mental health records were
confidential since his Mom and Dad were rich and had him committed to
private clinics all the many times he had gone into the deep end before.
He had no criminal record on file anywhere.
Robert Rudzki
rasterho@pacbell.net
gewebe@pacbell.net
http://home.pacbell.net/rasterho
I'm still not sure what Pournelle is talking about on the SCSI
cables. I've never seen one like he described. I've seen self-terminating
cables, but those don't have two connectors on the end. They have one
connector and a lumpy terminator at the very end of the cable. In theory,
terminating twice shouldn't hurt anything. That is, if the final device is
terminated, having the second terminator at the end of the cable should
have no effect. The terminated device itself should stop the SCSI signals
dead, and they should never get to the final termination.
|
TTG Home
Robert Home
Daynotes
Home
Special
Reports
Current Topics
|
Wednesday,
18 August 1999
[Last
Week] [Monday] [Tuesday]
[Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next
Week]
Updates are going to be short this week. I'm working hard. We did get
one minor problem with Barbara's new system fixed yesterday. I was working
away yesterday morning when I heard an anguished cry from her office. As
it turned out, Barbara can't stand the keyboard we installed on theodore.
Thoth used an AT system board, and theodore of course has an
ATX. Barbara wanted her old keyboard back. I knew that I had an
AT-keyboard-to-PS/2 adapter around here somewhere, but I couldn't put my
hands on it, as usual. Barbara was running some errands, anyway, so she
stopped and picked one up. She now has her old keyboard back, and is
happy.
I see that Tom's Hardware has run an article on Athlon motherboards.
Tom complains that it's hard to find Athlon motherboards and impossible to
find Athlon processors. According to Tom, there's some great conspiracy
led by Intel to prevent motherboard manufacturers from shipping Athlon
motherboards. I think it's much more likely that most motherboard
manufacturers have noticed AMD's past failures to ship product in a timely
manner and have simply decided to wait and see before they devote
resources to manufacturing boards for which processors are scarce.
* * * * *
This from Robert Rudzki [rasterho@pacbell.net]:
I saw that Pournelle's home page is getting
cleaner and better looking, now if only he would deep-six those left
column blue icons in favor of text labels.
Jerry AKA 'Wade Curtis' =8+] is plugging the
Science Fiction convention in Anaheim, he provided a link:
I went there to see that he is a guest speaker, it's only 40+ miles from
my house maybe I should go to hear him in person...
Heh, after I heckle him that the only sci-fi
writer alive worse than Niven is Pournelle they'll probably haul me off
or the adoring crowd will mob me and I will die the death of 10,000
paper cuts from bad science fiction hardbacks. I have read exactly one
of Niven's books back in 1972, NEUTRON STAR it was and it was terrible.
That has to be one of the ten worst sites I
have seen in my 6 years of Web surfing, but maybe my vision is going...
I think every kid should read all of RAH at
least once, in order but when you get to be old and grey you think about
the $3.4 billion cost of the Cassini space-junk sling-shotting back
around the Earth to spend the next 7 years flying to Saturn. Then I
begin to wonder about the cost-benefit of these interplanetary probes.
263 million Americans paid $128 each for the privilege on a one-year
basis. What do we get in return?
I am all for the unmanned/unwomaned
commercial and military use of space. GPS is great, 500 channels of TV,
The Red Chinese now being able to track our boomer nuke subs hundreds of
feet underwater on deep patrol quiet. But to save money, use unmanned
heavy lift boosters like the Titan IV, scrap the Shuttle and all the
interplanetary probes today.
I saw in the paper a column by our 'Dave
Barry' clone only not as funny, about a local kid who went to college
back in Nashville, had a local ISP account there and then moved back to
Riverside, CA and his mom's house to study for his bar exam. [just what
we need in California, yet another lawyer!] He was staying logged in his
dial-up connection doing on-line tutorials for 12 hours per day, using
his Nashville, TN local ISP number long distance! The total phone bill
was over $2700! I only hope he never takes any telecommunication law
cases...
The Board of LA County Supes has decided to
ban all gun sales on County land including the Pomona County Fair Gun
Show at 4 times per year one of the biggest gun shows on the Left Coast.
They sent undercover BATF and local fuzz with $4000 to buy the dreaded
illegal assault style weapons and ran out of money real fast since there
were so 'many' of them at the show... Since you can't get a decent
Colt Sporter or HK for much less than $1000 out here, $4000 does not go
very far.
And the guy selling Class III weapons did
not have them at the show, you had to go to his gun shop back room to
see and buy these guns. But you would have trouble seeing that from the
TV news unless you were watching carefully.
Robert Rudzki
rasterho@pacbell.net
gewebe@pacbell.net
http://home.pacbell.net/rasterho
Well, as far as the left menu bar icons, I agree with you, but de
re gustibus non est disputandum. As far as your opinion that Pournelle and
Niven are the worst living SF authors, I think you're in a very small
minority. They are generally acknowledged to be two of the best. Mr.
Heinlein himself said that Mote was the best SF novel he'd ever read,
including his own. I've read all of their joint works, all of Pournelle's
individual works, and many of Niven's individual works. I think Pournelle
is by far the better author of the two.
|
TTG Home
Robert Home
Daynotes
Home
Special
Reports
Current Topics
|
Thursday,
19 August 1999
[Last
Week] [Monday] [Tuesday]
[Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next
Week]
The Seagate ST150176LW Barracuda 50 drive I'd been waiting for showed
up yesterday. That's 50 GB of U2W disk. Fifty thousand megabytes. That's
incredible. And it all fits in a 3.5" drive with a 1" form
factor. I plan to install this drive in an audio/video server that will
sit next to our television, VCR, and audio gear. That raises all sorts of
questions. I'm already checking into wireless keyboards (Barbara says she
won't put up with a keyboard cable stretching from our sofa to the
equipment). What about computer video output? I suspect there's no way
that my JVC television has the bandwidth to display even at 640X480, so
how do I display computer video? Has anyone done anything like this
before?
* * * * *
I thought it was about time to do a quick update on my experiences with
the Olympus D-400 Zoom. On balance, this camera is just about perfect. The
resolution is more than adequate for what I use it for, but there are a
few things I would change:
- Flash - the built-in flash works well, but is somewhat
anemic. I would like to see provision made for using an external
flash. This could be something as simple as a hot shoe that could be
used with a "smart" flash unit, or as complex as full
integration with high-end flash units.
- USB support - the FlashPath adapter works well, but it would
be nice if the D-400 Zoom had a USB port and allowed one simply to
connect a USB cable and download images automatically.
- Lens - the current 3X zoom lens is pretty good, but I'd like
to see something with a bit broader range, say 6X to 8X, from somewhat
wider than the current widest to quite a bit longer than the current
longest. Say something on the order of 24mm to 200mm equivalent.
But those are all nit-picking. In fact, the D-400 Zoom is a wonderful
digital camera as it is, and I recommend it highly. I'd intended to
include a summary of my findings on battery life here, but the camera is
still running on its first charge. I've taken about 125 shots total so
far, with moderate use of the flash and light use of the LCD screen. Most
of those shots were at SHQ, although perhaps 25 early ones were done at
SQ, a dozen or so at HQ, and a couple with no compression.
* * * * *
This from Dave Farquhar [farquhar@lcms.org]:
I think there's another explanation for the
relative scarcity of Athlon motherboards and CPUs right now. Since VIA,
ALi, and SiS don't have Athlon chipsets ready yet, AMD's supplying both
the CPUs and the chipsets. Big companies like IBM and Compaq are
interested in the Athlon and have announced systems based on the new
chip. Now, if you're AMD and have a limited supply of CPUs and chipsets,
are you going to sell them to IBM and Compaq, or are you going to sell
them to brokers? If I have enough chips to build a half-million systems
and the likes of Compaq and IBM want all of them, I'm gonna give the
chips to them because Compaq and IBM help my reputation, while
second-tier manufacturers can hurt it. How many people believe AMD chips
aren't compatible with Intel chips because second-tier manufacturers
have slapped AMD chips onto cheap motherboards with incompatible
chipsets and el cheapo video cards and not loaded the proper drivers
(assuming they exist)...? I've seen a large number of these cheap
systems that had compatibility problems that I eventually traced down to
drivers not being loaded, or the wrong drivers being loaded. That's not
AMD's fault, but AMD shoulders much of the blame since their chip is the
biggest-name component in the system.
AMD had the same initial shortage problem
with each iteration of the K5 and K6 series as well, though it was
compounded by manufacturing woes in both cases.
The Intel conspiracy theory is kind of hard
to buy. Is Intel paranoid enough to do such a thing? Maybe. Is there any
need for them to do so? No.
Dave Farquhar
Microcomputer Analyst, Lutheran
Church-Missouri Synod
farquhar@lcms.org
Views expressed in this document are my own
and, unless stated otherwise, in no way represent the opinion of my
employer.
Your explanation strikes me as very possible. I take anything
that Dr. Pabst says about Intel versus AMD with a grain of salt. Tom's
Hardware has always shown a strong pro-AMD bias, and this latest strikes
me as more of the same. As far as the problems with K6-* systems, I'm sure
the factors you point out have something to do with it. But I think a much
more important factor is the demonstrable inferiority of the Socket 7
platform (and non-Intel chipsets) to GTL+, Slot 1, Socket 370, and the
Intel BX/ZX chipsets. Socket 7 was simply never intended to run at the
speeds it has now been pushed to.
I also found an interesting article on The Register this morning.
Apparently, an independent benchmarking lab is claiming that AMD
"cooked" the benchmarks to show the K7 was faster than the
Pentium III. I figure it'll all come out in the wash anyway.
* * * * *
This from Robert Rudzki [rasterho@pacbell.net]:
In matters of taste there is no dispute, I
like your Latin.
As a child, I used to go to the Main Library
in Chicago with my parents, the checkout limit was ten books, so using
my mom's card was hard if she and I wanted to get more than ten books
collectively we had to negotiate.
So one day the checkout librarian said there
is a children's room and he can get his own library card if you will
sign the form and make good any books lost or damaged. So at the age of
7 I had my very own library card at a very large and important library.
I was interested in science and gadgets and discovered RAH and Asimov
early on, read all their teen books then started on the adult books. I
normally browsed the adult sections, occasionally my mom would not let
me check out a book for unspecified reasons [when you're older you'll
thank me for this...] and the checkout librarian would raise an eyebrow
at some of my choices but never said anything, to me anyway.
Once we arrived in California, I promptly
got a library card at my local branch and got ten books at a time, 5 on
each side of the dropped handle bars of my 10-speed bike on off to home
and a good read. I read Analog when it was small and when it was big and
then small again, I found Campbell's essays made me think and widen my
horizons.
Basically I read the entire collection of
sci-fi they had and all the back issues of Analog, tons of other books
as well, so by the time I reached 17 I had read more sci-fi than most
people I have met.
In fact I burned out on it, and when
Pournelle and Niven finally came into my sights I wondered what all the
noise was about.
But then Howard Stern and Dr. Laura are
highly regarded by a lot of people, as were Roman circus arenas and
European public executions were widely attended, so much so they had to
move them into prison yards the crowds got so unruly and boisterous.
I remember the first time I read THE MOON IS
A HARSH MISTRESS I thought RAH had gone completely around the bend, and
his later works continued the decline IMO. His last book and it is a
sign I cannot remember the title, was just awful and I wish I had never
read it.
Anthony Burgess, Elmore Leonard and John D.
Macdonald are my favorite authors now, Len Deighton is a close '2nd'.
Taylor likes Leonard, Ross Macdonald and Dick Francis and Anna McCaffrey
with all those improbable flying dragons on the covers... I cannot get
her to read Burgess she says he uses too many big words.
The 2 worst non-science fiction authors IMO
are Arthur Haley and Alistair MacClean the latter's books are even worse
than the movies made from them which are really bad...Ludlum can get
pretty silly and not tie up loose ends by the end of the book, Robbins
is great for really sick and kinky babes and hot sex.
PS: I found it odd that Pournelle in his 3
[!] trips to see the Corpse Plant at the Huntington told us although he
had 'research' privileges there he had not visited in nearly 20 years.
We went there at least once every couple of months on Saturdays until
the crowds became impossible, and the large families with undisciplined
children running, climbing on and 'decorating' all the flat surfaces and
statuary. The Huntington finally instituted a reservation system and
when the lots are full, they lock the gates. By this time we had moved
to Riverside and both of us were driving a lot of miles on the freeway
to work and the last thing we wanted to do on the weekend was get on the
freeway one more time...
I did stop there during the week for lunch
if I was at the Pasadena store for tech support work on the POS system
it was really quiet and cool on the lunch pavilion patio.
Robert Rudzki
rasterho@pacbell.net
gewebe@pacbell.net
http://home.pacbell.net/rasterho
Well, again, favorite authors and books are obviously a matter of
personal preference. I haven't read as much science fiction as you have, I
guess, but I'd certainly put Pournelle and Niven right up there with
Heinlein among my favorite authors. Heck, for that matter, I kind of like
Lois McMaster Bujold's Barrayar series, although I guess it doesn't
qualify as "serious" SF. As far as _The Moon is a Harsh
Mistress_, I'd have to say that you're in a minority opinion again. It did
win a Hugo, and many, including me, consider it to be one of Heinlein's
best. Nor do I share your opinion that he showed any decline in his later
books, although many thought that _The Number of the Beast_ was not among
his better works. I always liked it. I've never read Haley, but I agree
that Alistair McLean is not a particularly good author.
* * * * *
This from Mark Grieshaber [mvgrie@shute.monsanto.com]:
Hello Robert:
I found a page of your online journal when I
was searching for info regarding DEC RX-50 floppies. You mentioned that
you still had some used ones lying about since you never throw anything
away. I've just picked up a DECmate computer (older cousin to the DEC
Rainbow), and am trying to get it up and running for nostalgia. Both the
DECmate and Rainbow used RX50 floppies, but I am having a hard time
finding any these days, as they are of course, completely obsolete.
Any chance you would be willing to part with
yours? Used is fine with me, as I would be reformatting them to start
with. I'm also interested in the old DEC hardware, as well... I am
curious as to what equipment you used the RX50 floppies on?
I'd be happy to pay for packing and shipping
(I'm in St. Louis, MO). UPS will come to your home for a pickup, if that
would be more convenient for you than a trip to the Post Office.
I also note you were planning on using the
Tyvek sleeves from floppies to protect cds. If interested, I have a
modest number of the flexible cd sleeves (transparent front, soft back)
that I could send your way.
Also, plenty of jewel boxes, but you
indicated you didn't like them so much (still, a stock of a dozen spare
goes a long way towards covering breakage...).
Mark Grieshaber
mvgrie@shute.monsanto.com
Hmmm. I think they're still around, but I can't swear to it. I do
let Barbara clean things out periodically, and that may be something that
has been pitched, although I don't think so. Once I can clear a path back
there, I'll check and let you know. If you don't hear from me for a while,
you'll know there was an avalanche and I'm trapped beneath a pile of
obsolete computer stuff. As for machines, I used RX-50's on all three of
DEC's "kind-of-PC" machines--the DECmate II, the Rainbow, and
the Pro 350/380.
|
TTG Home
Robert Home
Daynotes
Home
Special
Reports
Current Topics
|
Friday,
20 August 1999
[Last
Week] [Monday] [Tuesday]
[Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next
Week]
Interesting stuff this morning. The Register broke a story
about the Intel utility newspeed.exe, which allows overclocking a
processor from the command line. The bad news is that it works only for
processors that are not locked, those marked "Intel
Confidential." I have some of those, but I've never been a fan of
over-clocking so I probably won't both to try the program, except perhaps
to verify that it works. If you somehow have gotten your hands on an Intel
Confidential CPU and want to try this program, you can download newspeed.exe
from HardOCP here.
* * * * *
My morning newspaper tells me that the US Navy is raising the alarm
about Y2K problems. The results of their study say that widespread outages
of electrical power, water, and natural gas are "possible" or
"likely" in many cities. The government says that the Navy
conclusions are flawed because they assumed that any system that had not
been verified would fail. That's perhaps not an unreasonable assumption.
At any rate, I plan to lay in a cord of firewood. We have both wood and
natural gas fireplaces, so we'll be able to stay warm no matter what
happens. And a cord of wood is not a bad thing to have around anyway.
* * * * *
This from bdenman [bdenman@FTC-I.NET]:
Good morning. I have been out of pocket for
awhile (was in Texas for 10 days) but have tried to catch up on your
column including discussions regarding AMD. I am not sure about Tom
being pro AMD. And if someone cooked the Athlon's benchmarks...it won't
remain a secret and certainly will hurt.
When it comes to AMD vice Intel, I
personally think it comes down to capability, performance, support and
value. AMD and the Celeron are both viable options. But I do almost
agree now that the Celeron chip is probably the better value now for the
low to middle system. The price factor has been minimized and there may
be still be a few minor compatibility issues with the K-6... actually
more of a 3rd party driver support problem related to the cpu's 3DNow
instructions and/or chipsets from VIA, ALI, etc.
Doing some research over the past couple of
months, I saw various comments regarding driver isssues and chipsets for
AMD. This was of interest as I was in the market for a new graphics
card. My old STB Lightspeed 128 (with 2.25MB video ram) was working fine
with my K6-2/400 but I wanted to run 1024/768 at true color (24 bit) so
needed more ram. Anyway, I kept reading that some video drivers worked
poorly with various chipsets. Last week I finally bought an ATI Xpert
128 on sale (based in part on JerryP's experience). I did find their
standard drivers work fine for me on my system so who knows. I did try
their Beta drivers for the AMD though were not stable enough for
me..lockups occurred.
Now I also would like a new PCI sound card.
I leaned heavily towards Creative's SB Live Value until I read various
comments in the SB Live newsgroup. A number of people there were seeking
help as there are apparently compatibility problems with the VIA
chipset. So I have put that purchase on hold until that is resolved. I
will continue to make do with my basic ISA AOpen sound card which still
works fine.
So, bottom line...going with Intel probably
is better. Drivers for AMD seemingly maybe be late or not optimized
(makes sense that most companies code for Intel cpus and chipsets
first). This is not to say that good combinations are not available.
Just not the universal compatibility one gets with Intel no matter what
the cost.
Having said all that, bottom line is next
time I will probably go with Intel. Not so much for the
support/compatibility issues (which I think are minor) but for the
reason that I believe that AMD is doomed and won't be around much
longer. I do not think they will succeed even with the K-7/Athlon. Intel
has too much of a presence/brand preference/market bias on the high and
profitable end and AMD cannot generate enough income working the low to
middle end. So; I expect AMD will call it quits sometime next year.
I also would not be suprised to see Intel
then slowly raise entry level prices as new CPUs are introduced. As
older slower chips go away, new and faster ones will naturally be more
expensive but there will not be a corresponding drop in existing chip
prices. Oh well.
Later
Bruce
bdenman@ftc-i.net
http://web.infoave.net/~bdenman
I have no religious issues about processors. The simple fact is
that Socket 7 is an inadequate platform for processors running at modern
speeds. Anand just ran a review of a Tyan Socket 7 motherboard, announcing
more or less that he'd finally found a stable Socket 7 motherboard, now
that it was just about too late. I agree with Anand about that, although
the EPoX Socket 7 motherboards are the best of the bunch in my opinion. I
wouldn't use a Tyan on a bet. The Athlon changes the equation a lot,
because Slot A is competitive with Slot 1/Socket 370 and GTL+ in terms of
stability and suitability for running a fast processor. The question is
whether AMD can execute, which they haven't done well in the past. The
secondary questions are whether motherboard manufacturers will deliver
high quality Athlon motherboards in quantity, and whether Via and other
chipset manufacturers can deliver stable, full-featured chipsets. That
last is a major question in my opinion.
* * * * *
This from Chuck Waggoner [waggoner at gis dot net]:
I've only had passing contact with the ATI
All-in-Wonder video card--my use of it on a client's computer was to
create text files of the 'caption' content of certain TV programs for a
project I was working on,--but it has connectors for standard NTSC video
output, if your TV will accept that.
Can't attest to the quality, but I've also
come across some people who use it as a means to get computer video
output into large television monitors for presentation purposes, and
they seem satisfied with it.
Thanks. I'll check it out.
* * * * *
This from Dave Farquhar [farquhar@lcms.org]:
I saw that Register article too. I'm
wondering how many people are surprised--it's extremely common practice
to cook your benchmarks to make you look good. Intel does it, AMD does
it, Cyrix and IDT did it... Apple and Motorola certainly do it. I never
buy anything based on manufacturers' benchmarks. Watch the hardware
sites, figure out what it is you're going to do with the computer, then
see how the chips in question perform in the roles you need. Then figure
out what you're willing to spend, and buy the chip that meets your
budget that does the best job, being sure to leave enough money in your
budget for adequate memory, a good video card and a good hard drive (all
of which are more important than the CPU in most cases anyway).
I can cook a benchmark to make whatever chip
you want look as bad as you want. Want the Pentium III to look bad?
Well, let's see how it runs with my ISA Paradise VGA video card, circa
1991... Hmm, that's too good. I've got an ISA IDE controller and an old
Seagate 120-megabyte IDE drive hanging around here somewhere... Now
where'd I put that 8-meg PC100 DIMM? Seems to me that Robert Collins did
a piece on benchmark cooking over at www.x86.org
a couple of years back that's still relevant.
The most important thing is that the
performance of these chips is very close, and now that Intel's being
attacked at the high end, we can get 500-MHz CPUs for less than $300. It
wouldn't matter if the K7 were slightly slower than a P3 running at the
same speed, because consumer perception is that megahertz is all that
matters, and the clock speeds are comparable. Once K7 motherboards are
manufactured in quantity and the price comes down to a reasonable level
(who cares if a K7 costs $100 less than a comparable P3 when the K7
motherboard costs $100 more than a comparable P3 board?), things will be
better for all of us, regardless of which camp we belong to. Intel's
sweating, and what's bad for Intel right now is good for consumers.
Dave Farquhar
Microcomputer Analyst, Lutheran
Church-Missouri Synod
farquhar@lcms.org
Views expressed in this document are my own
and, unless stated otherwise, in no way represent the opinion of my
employer.
Yes, I agree with you about benchmarks, which is one of the
reason I don't pay a whole lot of attention to them. That and the fact
that speed is usually the least important issue. Chances are that no one
sitting in front of a given machine will know (or care) whether it has a
Pentium III/600 or an Athlon/600 in it. If you can't tell the difference,
what difference does it make? Far more important are issues like
stability, availability of bug-free drivers, etc. etc. Stuff that most
people pay scant attention to.
* * * * *
And another from Dave Farquhar [farquhar@lcms.org]:
Here's another article
on the benchmarks that goes into a bit more detail than The Register's
story:
What AMD did is shady, but little different
from Intel's standard practice of optimizing its code with MMX or SSE
instructions, then comparing it to unoptimized code for whatever they
see as competition (be it AMD/Cyrix/IDT/Rise CPUs or older Intel CPUs).
A truly fair benchmark would be optimized for each CPU in question,
using whatever special features are appropriate for that CPU, and that
would give you an idea of not only PIII vs. K7 performance but also SSE
vs. 3DNow! performance, but the real-world usefulness of even that
benchmark would be questionable because there are some applications
and/or device drivers that are optimized for one or the other but not
both.
Mark Twain would probably say there are
lies, damned lies, and then there are benchmarks.
I don't really think that doing a benchmark optimized for, say,
SSE is cheating. The benchmarks that Intel supplies allow one to compare
an application running on an SSE processor with and without SSE
optimizations to that app. I think that's a valid method, especially given
Intel's dominance and the likelihood that most software vendors will
support SSE for apps where it makes sense to do so. What outrages me is
the vendors (some video card vendors are notorious for this) that optimize
their drivers for a specific benchmark, such as the PC Magazine WinBench
benchmark. That would be questionable if all they were doing was actually
optimizing the drivers for the benchmark, but what they do is spoof the
benchmark, which is simply cheating.
* * * * *
This from Werth, Timothy P [timothy.werth@eds.com]:
Thought you would both find this
interesting.
Indeed. For those who do not view the article, Mr. Moody is
commenting on the fact that computers are supposed to have increased
productivity dramatically, but no such increase shows up in the overall
productivity statistics. He uses some rather tortured reasoning to blame
it on Microsoft Windows. I think it's attributable to something much
simpler. Computers have indeed had a huge impact on productivity, as
anyone who uses one knows. But there's been another equally ubiquitous
factor counting the productivity gains due to computers. The government
via ever-increasing laws and regulations has been killing productivity. If
PCs did not exist, this would be evident as a huge and ongoing drop in
overall productivity year by year. The fact that productivity has remained
more or less constant over the last decade or two simply means that the
huge productivity increases made possible by ubiquitous PCs has been
offset by the huge and growing productivity decreases caused by government
regulation.
* * * * *
This from a regular reader who asked to remain anonymous:
Don't attribute it to me, but I thought you
might be interested to know what Microsoft's up to-- this trick also
works for Minesweeper and Solitaire, but not for 3D Pinball.
From:
<mailto:ryancoody@mail.utexas.edu>Ryan Coody
Newsgroups: ms.beta.win2000.general,ms.beta.win2000.tips_n_tricks
Sent: Saturday, August 14, 1999 3:14 PM
Subject: The "boss key" is back!
Anyone remember the "boss key" from the old DOS games? A key
that you could
hit that would either immediatly exit the game, blank the screen, or
bring
up a fake screen that looked like you were working.
In the Win2k version of FreeCell you can hit 'ESC' to minimize it and
replace the title with "budget.xls"
387729
Ryan Coody
Thanks. I'll have to remember that in case I catch myself playing
Solitaire when I walk into my office.
|
TTG Home
Robert Home
Daynotes
Home
Special
Reports
Current Topics
|
Saturday,
21 August 1999
[Last
Week] [Monday] [Tuesday]
[Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next
Week]
The morning paper tells me that a new study says that adult soccer
players (football, for you non-US readers) are likely to suffer mental
impairments, presumably from striking the ball repeatedly with their
heads.
* * * * *
The Washington Post ran a truly frightening article.
Apparently, the Justice Department is annoyed that encryption technology
is making it possible for us ordinary people to protect our documents on
our PCs. They want to allow agents to break into our homes to disable
encryption on our PCs preparatory to a later visit to actually seize the
information.
* * * * *
Bo Leuf is in the midst of making some changes to his leuf.com domain.
If you have problems accessing it, note the following message from Bo:
I am moving the LeufCom hosting. The DNS update
should be seamless, but in case it is not, the "raw" subweb
URL to the new location is http://www.leuf.net/leufcom/
and the pages will all be there some time during week 34.
* * * * *
One or two of the following letters engage in ad hominem attacks on
Robert Rudzki. Ordinarily, I'd not have published those letters, or would
have removed the references, but I figured that by engaging in ad hominem
attacks himself, Mr. Rudzki had at least tacitly approved others doing the
same. And I figure he's a big boy.
Incidentally, thanks to those of you who have started sending me mail
in rich-text or HTML form. It makes it much easier to get them posted here
because I don't have to worry about line ends and paragraph breaks.
* * * * *
This from cc [cc@carnagepro.com]:
Boy are you having a good time. I have cut
Dr Pournelle's bookmark out of my humour section, as I don't have a
"right wing idiot" section he's gone. The Cold Warrior bit did
it for me.
I see you have been corresponding with
Robert Rudzki. An even bigger fool than Jerry. His writing is even more
self-indulgent than JP's but I have too high a standard in my humour
section for him.
To get to the point here. I am amused by
your reaction to Mr Moody's piece. He is no fan of MS and so blames the
lack of increase on that, you decide it is big government and reveal
your predudice. Am I the only one with a sharp axe?
Lets get real here. Computers are for
networking and the reason we network em' is to play Quake.
That's why those of us who do it for a
living are amused by all this nonsense about production and
productivity;).
CC
--
Upgrade to Linux...the penguins are hungry!
Chris Carson aka "GreyDeth"
250-248-0142
http://carnagepro.com
Well, I don't think I'm showing any prejudice when I blame
increasing government regulation for a decrease in productivity. Simply
ask anyone who runs a business how much time and effort is required to
comply with government regulations now versus ten years ago and twenty
years ago. The impact of government intrusiveness on productivity is not
in question, at least to anyone who has to deal with it.
* * * * *
This from Matt Beland [mbeland@itool.com]:
No snide comments from Robert Rudzki today?
Gee, that's too bad... :)
I don't have anything useful to say today,
other than to note that Tom Syroid is right about "virtual
friendships," at least for so-called "geeks"; I spend an
average of 14 hours a day online between work and home, and my best
friends (aside, of course, from the lovely and talented Mrs. Me) are
people I've never met, rarely "spoken" to, and sometimes know
only by screen name or alias. It's odd, completely new and different,
and yet somehow very old. After all, after reading RAH and
Niven/Pournelle's books so many times, some of my best friends from
childhood had to be Whitbread and Staley, Renner and Renner's Motie,
Juan Rico, Matt and Tex from Space Cadet, and a host of others. Everyone
wanted to know Mama Maureen - some better than others :) - and of
course, there was Tom and Huck from Mark Twain, Peter Pan, the Swiss
Family Robinson, and a host of others. We laughed with them, cried with
them, and in our minds, at least, fought dangers and celebrated
victories with them. Well, we all laugh with you and the others at the
antics of the Border Collies, or the stories of Roberta bombing the
Great Hall with trash bags of hardware, or Danielle forgetting the
lessons of her parents, and we are perplexed with you when the latest
software/hardware problem crops up. The difference is that you answer
back in more than our imaginations, and we don't have to wait for the
next book release to hear from you.
And so, friend, we thank you and appreciate
everything you do for us.
Matt Beland
Thanks. We had many of the same friends growing up.
* * * * *
This from a reader who requests anonymity:
Why do you continue to print Rudzki's
ravings? The man is imbalanced and all you're doing is encouraging him.
I don't know what Dr. Pournelle ever did to him but he seems to have
some kind of vendetta against him.
Well, I print letters that I consider interesting. I don't
necessarily agree with everything (or anything) that any particular reader
sends me--I certainly don't agree with much of what Mr. Rudzki has to
say--but I like to keep things stirred up. And I will say that at least
Mr. Rudzki is willing to put his name to the stuff he writes. I almost
didn't print your letter for that reason. Requesting anonymity so as not
to upset your employer or something similar is one thing. Requesting
anonymity for a personal attack is quite another. But, on balance, I
decided to go ahead and publish your letter. I don't think I'll publish
any similar ones, however.
* * * * *
This from another reader who requests anonymity:
I think if you decide to put this up, I'd
like to remain anonymous, thanks :-)
"I think it's attributable to something much simpler.
Computers have indeed had a huge impact on productivity, as anyone who
uses one knows. But there's been another equally ubiquitous factor
counting the productivity gains due to computers. The government via
ever-increasing laws and regulations has been killing
productivity."
Government regulation may take some of the
heat, but there's another factor at work that I think both you and Moody
have overlooked.
That is that computers enabled a lot of
things to be done that couldn't be done in the old days because of the
sheer cost in time and labor. A given document might have been better if
it contained a diagram -- but producing the diagram might take two or
three man-days' fussing with pencil, paper, T-square and templates. No
diagram. Now there's Visio, and the diagram you didn't do before is not
only doable but required. The document it took you a man-day to produce
as plain typed text now is much fancier, festooned with six different
fonts and bullets and numbers and embedded links to a table of contents
and of course an embedded diagram -- and it takes a man-day to produce.
Likewise writing a memo meant drafting the
thing in longhand and then getting it typed, or maybe even typing it
yourself. If you wanted to send it to a distribution list you had to
have it copied, or make copies, and send it through the corporate snail
mail system. That was a lot of work, enough so that it wasn't often done
for trivia (unless you had a secretary you could stick with it). Now
anybody struck by some absolutely essential bit of trivia can pound it
out in Word and dump it into the in-boxes of everybody in the department
or the division or the company with a few mouse clicks. The sort of
corporate bumf that used to go up on bulletin boards now floods e-mail
systems from sea to shining sea.
The bean counters and paper pushers were
particularly enchanted with the new possibilities that desktop
automation offered; all sorts of things that were simply impossible in
the old days could now be done, and if they had anything to say about it
(and they always do) they would be, too. Everybody could now write a
weekly status report, and a weekly time sheet accounting for every
minute of his or her time. Their managers could take their subordinates'
status reports and write status reports for _their_ managers, and so on
right up the chain. Every process could be made lengthier and more
complicated, with more documentation required at every step -- the most
trivial project could now have its very own six-page task list and of
course a constantly updated Gantt chart! What's more, the ease and speed
with which this information could be shared meant that every step of the
process could be subjected to review and comment! Gloryosky! More
meetings and conference calls and paper-passing! And when it comes to
CYA, we're in heaven -- our butts aren't only covered, they're
positively _fortified_ in paper!
This is not to say that government
regulation has had no impact, but I think you're underestimating the
ability of corporate America to take up the slack of the productivity
gains computers offered with the busy work that the same computers made
possible. in my own case, and I hardly think I'm untypical, anywhere
from a half day to a full day's potentially productive time, each and
every week, is sunk into this kind of tomfoolery. And that isn't even
counting the added complexity and extra steps now required to get actual
work done. As far as I can see very little of this comes out of
government regulation -- companies do it to themselves, and have some
kind of suicidal impulse to do more and more of it all the time.
Actually, I hadn't overlooked that factor, although I didn't
explicitly mention it. But even considering the amount of useless work
that computers encourage, it's clear to anyone who remembers using a
typewriter, slide rule, or adding machine that computers have dramatically
cut the amount of effort needed to achieve a desired result.
Government has damaged productivity in two primary ways. First,
of course, is the effect of burdensome and useless regulations. Those are
damaging in themselves, and more so because they require that personnel
and other resources be devoted to ensuring compliance. Forsyth County, NC
government, for example, now has a full-time staff member who spends most
of his time visiting all county sites and cutting the heads from two-prong
extension cords. I am not making this up. Second is the fact that
government regulations make it increasingly harder to cut deadwood from
the employee roster, as Dilbert makes plain. That's always been a problem
in government agencies, and has always been a problem to a limited extent
in private business, but it's becoming much more of a problem everywhere.
* * * * *
This from Robert Rudzki [rasterho@pacbell.net]:
It's been hot here a lot, no surprise for
Southern California in August. The wife Taylor was sitting reading in
her favorite chair and sort of listening to "The Rockford
Files" on UHF Canale 56 [she likes Jimbo Garner] when she wrinkled
her nose and said something has died outside. Most of our windows are
open when we're home, we run the A/C only on the really hot days.
This was my cue to go out and look in the
side germanium flower bed that edges the northern side of the house. I
found a baby possum dead and real stinky that I had seen coming through
the cat door in the kitchen a couple of days ago, so I got a shovel and
covered him with 5 spade-fulls of soil. We have several vicious stray
cats who wander around here, ours just stare in amazement at young
possums who come to eat and drink water from the cat dishes.
I have a friend who cut his Winchester Model
75 target rifle barrel down to 18", 11 degree reverse taper crown
on the muzzle. He has target blocks and a 12x Unertl Ultra on it, with
.22 CBee Longs, 29 grains @ 700 fps it sounds as if you dropped a heavy
book flat on a hardwood floor, more than one stray has died instantly of
'natural causes AKA lead poisoning' in front of it. I may have to borrow
it.
I am glad I am not part of the emergency
rescue crews in Turkey it has been hot there as well.
I went to our local community college again
to add a class, ART-483 Web Graphic Design. Naturally the entire college
computer system was down again, they did a huge cut-over to a new
admissions system and Y2K upgrade in July and are still debugging the
inevitable problems. Monday is the next time they promise all will be
well again, back I go with all my certificates and transcripts...
I hit The Doctor's page today. Taylor, I
shouted you have to come see this Pournelle has a bottle bobbing around
for his new email icon! She came over and started giggling, I like she
said. So I get no respect even in my own house from my own wife. Oh
well, life is hard then you die.
PS: I hate the bottle even more than the
blimp.
Robert Rudzki
rasterho@pacbell.net
http://home.pacbell.net/rasterho
"If the 1st Amendment applies to the all the States, why
doesn't the 2nd...?"
I must admit that I don't understand why you bother to read Dr.
Pournelle's page if it annoys you so much. Why not just say no? It seems
to me that you'd save yourself a lot of aggravation.
|
TTG Home
Robert Home
Daynotes
Home
Special
Reports
Current Topics
|
Sunday,
22 August 1999
[Last
Week] [Monday] [Tuesday]
[Wednesday] [Thursday]
[Friday] [Saturday]
[Sunday] [Next
Week]
Okay, I spent quite a bit of time yesterday reading and responding to
mail from people who typically said something like, "It's your page,
but I'd really appreciate it if you'd stop posting Rudzki's Rants."
As a matter of fact, the sentiment was unanimous, which is very unusual
for anything around here. Although I agree with very little of what Mr.
Rudzki has to say, which I think I've made clear, I've continued to post
his messages because (a) he took the time to write them, (b) some of what
he has to say is interesting in a different kind of way, and (c) I don't
believe in censoring what people say or think. But this is my page, and my
readers are telling me they've had enough. So, from this day forward, I
will no longer post any messages from Mr. Rudzki. Those who would like to
read what Mr. Rudzki has to say can read his own web page.
* * * * *
Intel slashes prices tomorrow. The Pentium III/600 remains at $700 in
quantity 1,000, but the Pentium III/550 has dropped to $490, and the
Pentium III/500 to $255. Most impressive, the Pentium III/450 has dropped
to $187, which puts it in the Celeron class in terms of bang-for-the-buck.
In another way of looking at it, the 600 costs $1.17/MHz; the 550 $0.89;
the 500 $0.51; and the Pentium III/450 only $0.42. That means you can put
together a dual Pentium III/450 system for less money than a single
Pentium III/550 system, even considering the additional cost of the
dual-CPU motherboard. I know which one I'd rather have. A $175 EPoX KP6-BS
and a couple of $187 Pentium III/450 CPUs forms the foundation of an
inexpensive system that will blow the doors off any of the $2,500 systems
that Dell or Gateway sells.
* * * * *
This from Chuck Waggoner [waggoner at gis dot net]:
I'll tell you how the computer has changed
the way I work. For every project, I used to have 2 full-time assistants
and part of a secretary. Nowadays, for every project, I've got one
assistant 3 days before taping, and maybe part-time during editing.
The work load hasn't changed; I'm just doing
a LOT more of it by myself.
I don't know where these low productivity
figures come from. When fewer people are turning out the same amount of
work in the same period of time, I would think that meant an increase in
productivity.
And I haven't met anybody in any line of
work that didn't think computers were being used to get them to crank
out more work in the same period of time, not to ease their load and
reduce the work week, as was the promise.
Exactly. And for nearly any other information handling endeavor,
computers have been of equal benefit. And as for why you're getting more
work done and overall productivity isn't increasing, it's because those
who would otherwise be doing useful work are in fact doing
government-mandated busy-work instead, whether on or off the government
payroll. That and people who do little or no useful work but continue to
draw a paycheck.
|
|