jp.jpg (13389 bytes)

CHAOS MANOR MAIL

Mail 195 march 4 - 10, 2002 

 

HOME

VIEW

MAIL

Columns

BOOK Reviews

read book now

emailblimp.gif (23130 bytes)mailto:jerryp@jerrypournelle.com

CLICK ON THE BLIMP TO SEND MAIL TO ME

 

LAST WEEK                          Current Mail                           NEXT WEEK

  The current page will always have the name currentmail.html and may be bookmarked. For previous weeks, go to the MAIL HOME PAGE.

FOR THE CURRENT VIEW PAGE CLICK HERE

If you are not paying for this place, click here...

Highlights this week:

IF YOU SEND MAIL it may be published; if you want it private SAY SO AT THE TOP of the mail. I try to respect confidences, but there is only me, and this is Chaos Manor. If you want a mail address other than the one from which you sent the mail to appear, PUT THAT AT THE END OF THE LETTER as a signature. In general, put the name you want at the end of the letter: if you put no address there none will be posted, but I do want some kind of name, or explicitly to say (name withheld).

Note that if you don't put a name in the bottom of the letter I have to get one from the header. This takes time I don't have, and may end up with a name and address you didn't want on the letter. Do us both a favor: sign your letters to me with the name and address (or no address) as you want them posted.

I try to answer mail, but mostly I can't get to all of it. I read it all, although not always the instant it comes in. I do have books to write too...  I am reminded of H. P. Lovecraft who slowly starved to death while answering fan mail. 

Day-by-day...
Monday -- Tuesday -- Wednesday -- Thursday -- Friday -- Saturday -- Sunday
 
atomz search

Search: type in string and press return.

 

or the freefind search

 
   Search this site or the web        powered by FreeFind
 
  Site search Web search

read book now

Boiler Plate:

If you want to PAY FOR THIS PLACE I keep the latest information HERE.  MY THANKS to all of you who sent money.  Some of you went to a lot of trouble to send money from overseas. Thank you! There are also some new payment methods. I am preparing a special (electronic) mailing to all those who paid: there will be a couple of these. I have thought about a subscriber section of the page. LET ME KNOW your thoughts.
.

If you subscribed:

atom.gif (1053 bytes) CLICK HERE for a Special Request.

If you didn't and haven't, why not?

If this seems a lot about paying think of it as the Subscription Drive Nag. You'll see more.

Search: type in string and press return.

 

line6.gif (917 bytes)

read book now
This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

read book now

TOP

Monday  March 4, 2002

USB Flash

Regarding the responses to this subject:

1) The 4MBit/sec WRite, 7.2Mbit/sec Read speed is directly off of the packaging and the Transcend (www.transcendusa.com ) web site. I have no way to verify this speed, but I suppose there are benchmarks somewhere that would do this. But the transfer speed is not that important to me.

2) I wouldn't (and didn't) recommend the device for data backup, but as an easy way to transfer files between computers ("sneakernet") as when one would work on files off-site, or to use for portable presentations (a PowerPoint presentation or Autocad drawing can easily fit on the device).

3) My use (and recommendation) for the devices was for 'sneakernet', as a way to work on data files at home (for instance, during web page development). Good backups are a must, of course; for that I use the CD Writer to store important data at home (at work, we have a server dedicated to tape backups).

Thanks...

Rick Hellewell

Indeed. And these are wonderful gadgets. Niven and I use them now on trips: we can transfer a novel between machines without setting up a Zip drive. I still use ZIP where the drives are already set up. And of course by using the gadget to transfer we DO get a useful backup independent of both laptops.

The following concerns Everquest...

Hail!

Seems that many of us in EQ land are having trouble since the Shadows of Luclin expansion came out. Random Lockups are going to kill EQ if they don't address the hardware issues that are demonizing their game. Rather than fighting with my hardware, I may have to give up my 2+ year investment in EQ and go play something else. I had Win98, a Voodoo 5 with a Tyan 400 (1854) motherboard with a VIA chipset and a P3800EB with 512 MB SDRAM when the SoL came out. No random lockups there, just extreemy choppy and unplayable in large groups. So, I read some reviews and got an AMD 1800+ with a Soyo Dragon + and a Geforce Ti 200 (oh, and 512 MB 2700 OCZ DDR [256x2]). Having trolled every tidbit and tweak Google could find for me, nothing works. Yesterday, I went and got a ATI 8500 but have yet to put in a session on EQ. I swear, if this doesn't work, it will be a painful farewell to my friends on EQ. Seems that the people that make the game have hardware requirments that are met (along with the right O/S) but seem to not care that the players are having this trouble (most are, arguably, with newer equipment). The video card companies say "our cards cant share IRQ's" but don't have lists of reccomended motherboards they can go on and motherboard companies don't give a rats rear about the videocard people... ad infinitum. If you still play EQ, what hardware do you use?

Thanks!

Rusty Coleman Baton Rouge, LA

This will be addressed in the March column. 

I have had none of those problems with an Intel system based on an Intel D815EEAL2 board and a 933 MhZ Pentium 3; the video board is a random (MSI actually, $79 from New Egg) GeForce 2. I have also had no problems with Intel 2 GHz systems and GeForce boards. 

I have had some problems, none as severe as yours, with an AMD Athlon 1.2 and a GeForce 2, and I can only conclude that some of the well known bridge problems in most motherboards built for Athlon chips (not the chips themselves) have been responsible. This comes from considerable experience of systems side by side sharing an internet connection and having identical video boards: the Athlon has more Everquest problems than the others. It is a sample of one, of course, but I also know about some of the northbridge problems with the motherboards built for AMD systems.

I am experimenting with some other combinations including an AMD Dual built from the AMD chipset mentioned in today's BYTE , and we will see what comes of that. 

I do have link problems you probably don't have since I don't have DSL and the satellite won't connect to Everquest so I get linkdeath more often than I like, but in general I don't have the problems you describe, and the Luclin 3d images are very nice...

I think the following needs no comment:

Dear Dr. Pournelle,

Would you like to hear my flight story? About a fortnight ago I went out to the airport; I was some two hours early for the departure time, so I had to wait in line at check-in for about 15 minutes. Before that, bags through the X-ray, self through that doorway thing, followed by a pat down by a security guard. After check-in, through to airside, browse in the duty-free, then, after a meal, to the gate, walk down the tunnel, board the aircraft. Which took off about five minutes early. After a long flight, dismebark, through Immigration, pick up bag that went in the hold, through Customs, who did not want me to open either piece of baggage, taxi into the city and about my business. Return trip was almost exactly the same, except that, arriving three hours before the flight, I did not have to stand in line at Check-in at all. That plane also took off a few minutes early. Coming back, there was a stopover for about ninety minutes in another city, where I was allowed off to go and get a cup of coffee and browse the shops. Straight back on to the plane, and on to my destination.

Two other things. One, discounting the women and children, quite a large proportion of the passengers were what is usually described as sallow or olive-skinned, and I wouldn't be surprised if quite a large number of them were called Mohammed. Two, the airports in question were Bahrain and Hong Kong, and the stopover city was Dubai. Do you suppose that the US authorities are guilty of over-reaction?

The quote, I believe, is Rousseau. "Man is born free yet is everywhere in chains." Well, not in all countries.

Tony E.

[and see below]


Hi Jerry, The wheels of bureaucracy grind on. From our local paper:

"When it came to disposing of his waste oil, Brian Creegan did everything right, and he did it to the letter of the law.

He knows he could have easily dumped the oil from his small business illegally, but that would have been wrong.

And he's about to pay for doing the right thing.

Creegan knows a bill is coming his way soon, courtesy of the U.S. government. It will be a big bill for a small business owner -- at least $11,855.20, probably more -- levied for the environmental sins of someone else..."

< http://search.newschoice.com/ArchiveDisplay.asp
?story=d:\index\newsarchives\lo\midsen\fpg\20020303
\1390266_tfront.txt&source=www%2Esentinelandenterprise
%2Ecom&puid=&paper=Sentinel+%26+Enterprise+Online
  >

(I hope this long URL works!)

As if we needed more proof that "the Law is an ass"!

Cheers, Rod Schaffter

It is likely we will see much more of this sort of thing. 


Concerning the current column:

Jerry,

I have an issue with this statement.

I have a minor ethical dilemma here: I believe in competition, and I certainly want Intel to have at least one major competitor. At the same time, I just can't recommend any single-processor AMD system. It's not the AMD Athlon chips, which are so far as I can tell quite stable and solid; it's the support chip sets, which aren't.

Early chipsets for the Athlon had stability problems. That has not been the case for years. This affected the very first AMD 750 and VIA KX133 based motherboards. Since Socket A came about the stability problems have gone away. I have no idea what you are basing this statement on. I have tested chipsets from VIA, Ali, SiS, Nvidia, and soon ATI. They do not have stability problems. The stability problems out there stem from user error in most cases. You have to install drivers for all of these chipsets, including AMD’s. Stating that the chipsets are not stable is really doing a great disservice to AMD and the 3rd party chipset vendors. I build the systems for our Governor’s Campaign, and the box he uses at the capitol and the governor’s mansion. I use VIA chipsets, and guess what, there are no problems. Also, whoever that tall guy was at the Compaq booth at Comdex was a jerk. I would expect more from Byte.

 

Thanks,

Chris Tom

www.amdzone.com www.socketa.com 

Statements like "I would expect more" from partisans are common but not helpful. I'm glad you have no problems with VIA chipset systems. Unfortunately too many have. Bill Godbout used to say "if an error rate is high enough to measure, it's too high." The fact is that the northbridge problems of systems with VIA chipsets are measurable. The error rate isn't terribly high and won't show up at all with many combinations of software and peripherals, but there are errors, they are repeatable, and I for one don't want to take the chances. Most of this is going in the column.

Clearly one should install the proper drivers for systems. I don't recall ever giving contrary advice.

I don't know who the tall guy at the COMDEX booth was. Byte didn't have a large presence at COMDEX since we were not giving the awards. I was there, but apparently this wasn't me.

You have given me no reasons to change my views.

Hiya! "...I have a minor ethical dilemma here: I believe in competition, and I certainly want Intel to have at least one major competitor. At the same time, I just can't recommend any single-processor AMD system. It's not the AMD Athlon chips, which are so far as I can tell quite stable and solid; it's the support chip sets, which aren't..."

Pray tell, where did you come up with this fallacy? And, tell me, how many professionally built AMD powered systems have you tested?

In the summer of 2000, my partner and I decided to switch all our uniprocessor production to Athlons, partly because of the average lower cost than Intel silicon, but also because of superior performance. We did experience some issues with ALI chipsets and have not used them since. However, we have never had a problem with the original AMD 750 Irongate chipset on Gigabyte boards (we have two UNIX servers powered by that combination) nor have we had any trouble with the AMD 760, Via KT133 chipsets and their successors. To date we have built some forty uniprocessor servers and over 300 PCs with Athlon power. None has ever exhibited any stability that could be attributed to a chipset issue. In fact, one of our clients who purchased 16 Athlon powered PCs and an Athlon powered server last year placed another order for PCs and specified that they be Athlons. To me, that's a ringing endorsement.

Objectivity is a very expensive commodity in the computer press. The price often is too high for many journalists and they tend to make claims based upon some kind of hidden agendum. If you are going to make such a generalization as you did in your article, could you please offer some supporting data to back up your allegations? Otherwise, you can't protest too loudly when you are flamed for being in bed with Intel. You aren't a shill for them, are you?

Bill Brier, the Prez BCS Technology Limited

Bill Brier [steggy@bcstechnology.com]

No. 

Hello Jerry, I can't believe that you would say such a thing in your article. I have a minor ethical dilemma here: I believe in competition, and I certainly want Intel to have at least one major competitor. At the same time, I just can't recommend any single-processor AMD system. It's not the AMD Athlon chips, which are so far as I can tell quite stable and solid; it's the support chip sets, which aren't.

I was a charter subscriber to Byte and still have issue # 1. I've read your articles since you've been writing them for Byte. Frankly, I think you ought to be more knowledgeable about this stuff. I can understand being biased, I'm very biased about AMD but to tell an out and out lie about AMD chipsets is wrong. Do you even have any experience with Athlon systems? If you've had a bad experience at least say so. I've had bad experiences with both Intel and AMD systems. I had a Socket7 System that was unstable. Mostly because of the ATI video card. But those were issues with ATI and VIA chipsets.

I've had three athlon systems and have not had a single problem with any of them. My company sells Athlon / Duron systems and we have far fewer returns on them than we do on Intel systems. Please look at the data before making such claims in the future.

Your long time fan,

Michael Roberts.

 

========================================== Michael Roberts Director Product Development 1st CTI LLC Nashville Research & Engineering Labs.  mikeroberts@1cti.com http://www.1cti.com ==========================================

I have had half a dozen AMD systems over the years. I even built one to go to the Persian Gulf on the USS Tripoli. They have all "worked" and most haven't appeared to have more problems than anything else: the noise levels from software and operating systems often overwhelm the contributions of hardware to failures.

Clearly most people haven't had problems. I still have no reason to revise what I said. 

And 

How do you not get flamed when you say supporting chipsets for athlons arent stable? Which chipsets are you talking about? Thats an extremely vague remark to make without giving one chipset to reference. Your basically dropping a hint that someone shouldnt get an Athlon. Im a system builder and I have yet to have a single customer have any problem with any system that has been built for them. I use brand from VIA, SIS and others and still have not have any problems since the early ASUS KZM bios issues. Thanks Todd

I fear I do not know what to say to that. Or to the following:

 

You wierd fu*k*r! [Ed note: original not bowdlerized] What year do you think it is? Athlon chipsets are the backbone of the enthusiast market. We're not exactly tolerant of unstable products, and we certainly arent tolerant of spurious bastards like yourself.

Somebody's getting spam!

Weyland Corporation [aleksei1282@hotmail.com]

-------------------------------------------------------------------------------- Join the world’s largest e-mail service with MSN Hotmail. Click Here

But I find it interesting. Presumably this is testimony to the integrity of the Weyland Corporation whatever that is.

It was followed by:

 

Greetings Wierd fu*k*r! [Ed note original was not bowdlerized]

We wanted to let you know that Worst Nightmare created a SPAM.com electronic post card for you. Click this link to see it, or copy and paste this link into your Web browser's address line:

http://www.spam.com/12/GAL5.16_21.cfm?jlw051258=8771 

If you'd like to send your own SPAM electronic post card, use the "send your own SPAM post card" link at the bottom of the e-card collection page.

Enjoy!

SPAM.com

P.S. To us, your privacy is a priority. Here's a direct link to our online Privacy Policy statement if you'd like to review it.

http://www.spam.com/hp/hp_lg.htm 

I believe that California has some laws on this.

 

 

As to

> At the same time, I just can't recommend any single-processor AMD system. It's not the AMD Athlon chips, which are so > far as I can tell quite stable and solid; it's the support chip sets, which aren't. >

I think you will find that the SIS 735 Socket A chipset is completely stable and offers very high performance at low cost. I have an ECS K7S5A MB with a Duron 750 and it has been the most stable MB I have used to date.

Cheers

Duncan Munro

Thanks for the data. I'll try to look into it. SIS has not always had the reputation of being the highest quality chipsets on the block, but I am told they are a good match to the newest Athlons.

As I have said often, at the lower end there is little price competition between Intel and AMD. They all work pretty well, too, and the driver situation is better every week. 

At the very highest end the applications and drivers get very important and there are a lot of instabilities in the software side of things.  Until recently there wasn't any SMP competition to begin with, and systems with VIA chipsets had small but measurable problems. 

The bottom line is that for standard systems doing standard things go with what you like and your prejudices toward or against Intel or AMD will likely not matter. As you move toward more complicated things with more intensive tasks, the motherboard bridge connections and timings become far more important. I would myself stay with as conservative a hardware design as I could find if I were doing things that caused my system problems due to the sheer speed of operations.

At the speeds the high end systems operate in now, memory quality, power quality, everything is at the edge: things have to go just right. It's amazing that they do. If you are comfortable with AMD at that level, by all means stay with what you like.

But Intel has more test engineers than AMD has engineers. Think on that when you think about being out at the bleeding edge.

I wish AMD well, and for standard systems I doubt anyone would notice the difference between an AMD box and and Intel box: as I have said many times both here and in the column. And with the new dual processor chipset boards it looks good for genuine competition between AMD and Intel. But I still have no reason to change what I said.

If I were to change my views it would be because of this kind of thing:

Hi!

I just read your column, and I was wondering what kind of basis you have for saying that the motherboard chipsets for Athlons are unstable?

I'm the editor of the Temple Of Technology website, www.templeoftech.com, and I've tested quite a few AMD motherboards (both for the site, personal systems and systems I've built for friends) in the last two years. I've found all the current Athlon motherboards to be very stable and fast performers. Even the $60 ECS K7AMA is very stable.

For my personal system, I'm using the EPOX 8K7A+ (AMD 761 based), and it is rock solid. I run a lot of heavy duty applications, and it NEVER crashes..

regards,

Vidar Langberget

But once again we are dealing with low probability problems and small numbers of cases. Most people won't experience problems anyway.

And those who want to build their own system can use whatever the like: they don't need my approval or permission. As to me, at higher end systems intended for complex operations, I will stay with Intel, a position I have had for some time: but I found the new chip sets interesting enough to point out that the dual processor motherboards look not to have difficulties. The result is that I have been put on spam lists by AMD enthusiasts. Really makes me fond of the company and writing about it.

And one final comment by Joel Rosenberg and we're done with this since the time I might spend with letters is going to have to be invested in fighting the spam that the AMD supporters have arranged for me.

Your argument about the bleeding edge/AMD thing makes sense to me, and I think one of the lessons is to stay away from the edge unless you've got a good reason to do otherwise.

("I've got the time, money, and the desire to play," is IMHO, a perfectly fine reason; so, obviously, is "I'm doing forty-five minute compiles three times a day, and it looks like I can cut that time in half by upgrading.")

Even when it works right -- which, most of the time, for most people, it seems to -- that Extra Edge Tax can be high. Not just for the latest equipment, either -- there was a time when the better performance of SCSI drives was so much better than their IDE equivalents that there were some interesting cost/benefit tradeoffs, most of which have vanished. (I'm told that if you stripe a RAID-0 partition across four 7200 RPM IDE disks, with a sensible controller/cable setup and you'll be hard put to find a two-disk SCSI RAID setup it won't leave in the dust, unless you've got a immense amount of money to throw at the latest and widest and fastest.)

My experience, generally pretty far from the bleeding edge (my 850mhz Athlon was, I believe, two steps down from the fastest chip AMD chip in the store when I bought it), but when I did my latest upgrade (added a CD burner and a couple hard drives -- my root directory is now on a RAID-5 array, which means that I've got to lose two drives to lose any data) there were some . . . interesting moments during the in-store testing, until we upgraded the power supply to a more robust one, after which things worked like a charm, and are still working just fine.

Then again, I'm not in the business of exploring the bleeding edge, nor do I do a lot of heavy-CPU graphics manipulation very often.

If I were to replace the motherboard, I'd probably still go for something a couple of steps back from the fastest available, although a two-cpu motherboard would be awfully tempting, given reports I've gotten on the Linux SMP kernel, and I've still got the notion of playing with a Mosix array on my "gee, that'd be fun, and it wouldn't be too horribly expensive" list.

Still, the old rule applies: the best is the enemy of good enough, and for what I do, what I've got is more than good enough.

-- ------------------------------------- There's a widow in sleepy Chester Who weeps for her only son; There's a grave on the Pabeng River, A grave that the Burmans shun, And there's Subadar Prag Tewarri Who tells how the work was done. -------------------------------------

I don't find one word to argue with there. And of course there are millions of AMD systems working quite well: I have built and reported on about 10 of them in the course of this column.

And let's wrap this up with a comment that is worth paying attention to:

Jerry:

I read your column every week, in fact, it is one of the highlights of my week. I don't beleive, however, that you have been exposed to the current range of chipsets and boards available for AMD Athlons -- if you had, you couldn't possibly throw out such a blanket recommendation.

It is true that the early VIA chipsets had some compatibility issues and flaky BIOS's. No one in their right mind will argue that. The Athlon also generates a fairly high thermal load and has a high power demand. If you try and force on a cooler designed for a PIII, you will shatter your core and void your warranty. These issues have caused early adopters and the ham fisted some long nights of frustration.

The latest VIA chipsets are quite stable, as are the latest from Nvidia. I've seen several of these in the last few months. The trick is, as it has always been, to buy a quality mainboard, power supply, video card and case and assemble with due care. A couple of months ago, I built a new PC with an ECS mainboard with a SIS chipset and a Duron CPU for myself. It's been flawless -- board $99, CPU $90 in Canadian dollars.

I'm pretty much an agnostic when it comes to buying computer gear -- there are Intel zealots, AMD zealots, Linux zealots, etc. etc.

I try to use the best piece of technology for the task at hand. I have Pentiums and Athlons and K6's and even a Centaur kicking around. I use Windows and Linux and different flavors of BSD. I still have an active account on one of my company's Alpha servers running VMS.

My opinion, right now, I wouldn't buy a Celeron or PIII; the P4 to be worth the money looking forward has to be the new core (2 or 2.2 GHz). The best low end buy is Duron/Athlon. There is no best buy at the high end, as Intel is due to ramp speed soon, and performance will unlikely ramp with current chipsets, and AMD is due to ramp to Hammer soon.

Rick Spencer

I have several times said I look forward to Nvidia and AMD working together. And my view is more or less yours, one uses the tools that do the job. 

And of course I can't see everything. I don't have the 30 editors and 9 technicians BYTE used to have in Peterborough. There's me and my advisory group, and I try to give the best advice I can: and my wrapup is that just as few butter adds tell us that it tastes just like margarine, there are few Intel adds that say their chips are AMD-compatible. Absent pretty good reason -- and cost is of course a pretty good reason -- I advise people to stay with the standards, and don't overclock the system; the potential gains outweigh the losses. And of course there will be people who try everything and overclock it all and some of them send me reports I trust and pay attention to, and it's a lot of fun. 

For most purposes AMD systems work fine. My problems have been with screwy operations, USB peripherals, and odd software like, for example, Everquest. None of them have been so severe that I have retired the Athlon systems. But if I really want Intel compatibility there's a simple way to be sure I have it: and Intel still has more test engineers than AMD has engineers.

The interesting thing is I am now on a hundred porno spam lists because I wrote to say that despite problems in the past I am recommending the new AMD systems.  No good deed goes unpunished.

And you are correct: the statement I made was intended to be taken in the context of many columns which have reported that I use Athlon systems and have had few problems with them. I am aware that most people don't have problems and I have said so often enough, or thought I had.

This came off stronger than intended. So now I have to set my rules to reject the porn offers. It has certainly made me love AMD users more. And yes, I am aware that this is the work of a very few people. The AMD press relations people are horrified. Me, I have to reset my rules fairly often anyway so it's not that big a deal. But it has been enlightening.

 


 

On another topic:

<i> [I think the following needs no comment:]

Dear Dr. Pournelle,

Would you like to hear my flight story?.....the airports in question were Bahrain and Hong Kong, and the stopover city was Dubai. Do you suppose that the US authorities are guilty of over-reaction?....

Tony E.</i>

Well, actually it does need a comment. While I am just as disgusted with the illogical and downright stupid approach that our airport authorities have been taking regarding security, it is certainly not fair to compare us to airports in nations that, as far as I am aware, are a considerable distance farther down the list of radical Islamic targets than the United States--if they are on the list at all. No skyscrapers have fallen in Hong Kong, nor do I expect to hear about planes being flown into buildings in Bahrain and the United Arab Emirates.

Tom Brosz

Clearly.

Roland says:

Let it Be.

http://zdnet.com.com/2110-1106-850296.html 

I was for a while an enthusiast for the Be operating system, which was a good development environment; but it never caught on, and seems to be going away. RIP

 

U should have ur PHD REVOKED!!!

Obviously, you didn't get your PHD in a computer-related field 'cause you have no idea what you're talking about!

Rob Ohlee [ohlee2000@hotmail.com]

Chat with friends online, try MSN Messenger: http://messenger.msn.com

Another of those intelligent and cogent remarks that makes one really want to write about the object of their enthusiasm. If all this reminds one of the time when I said the 128K Mac was a decent operating system attached to a toy computer, I can't help it.

 

 

 

TOP

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

read book now

TOP

Tuesday,  March 5, 2002

Well, I had thought I was done with the odd mail triggered by the AMD column, but I can't resist this one:

Dear Jerry, in regards to the article you published on the web, and speciffically this quote:

"I have a minor ethical dilemma here: I believe in competition, and I certainly want Intel to have at least one major competitor. At the same time, I just can't recommend any single-processor AMD system. It's not the AMD Athlon chips, which are so far as I can tell quite stable and solid; it's the support chip sets, which aren't."

I just have a few comments about the egregious article you recently wrote (i.e., that is, AMD Athlon single CPU chipsets are unstable). As one PhD to another, I find it utterly incredulous that you made such unwarranted statements based on, what...? Hearsay? Where's the proof? I don't know what your degree was based on, but at the universities I frequent - evidence is required before conclusions are passed.

I think it prudent that you retract your article until you provide the sources for your statements. As it stands, innocent readers are being done a disservice by reading this fetid diatribe. Someone may actually believe (by false appeal to authority, I might add) that what you are saying is actually true! Sincerely, J.

Air Athlon [thiazolium@hotmail.com]

Now I have no idea who J. is, but Air Athlon is suggestive. The language is impressive.

I will deal with all this next Monday. However, there is this:

If you haven't already found this, You might appreciate TecChannel's discussion of the problem.

http://www.tecchannel.de/hardware/817/index.html 

Also George Breese's patch:

http://bbs.pcstats.com/viahardware/
messageview.cfm?catid=19&threadid=10977
 

I've managed to duplicate some of the issue's they discussed doing video capture/compression (to U2W scsi raid stripes). I see portions of scan lines being dropped (grrrr).

I've taken the direct approach, a new Leadtek motherboard with SIS735 chipset (with no VIA chips on it).

-- -Monty Walls (mwalls@castor.oktax.state.ok.us) - MIS, Oklahoma Tax Commission - - My opinions are my own, my employer knows nothing about it.

What I would do is build a dual Athlon on a board with all Athlon chipset...

 

 

Now for more relevant subjects.

You have said that when you finally get DSL you will put up a Linux box to handle your mail. Don't wait!

You can route all your mail through a Linux box, without a DSL line, and you will gain some big benefits from doing so. An obvious one is that you can add one or even two layers of spam filtering.

There is a mail tool called fetchmail that was specifically created to let a Linux box handle mail without being connected 24x7 to the net. You could use that.

http://www.tuxedo.org/~esr/fetchmail/ 

Your time is valuable. Invest some time getting a proper mail server up, and you will win it back many times.

P.S. My brother has found a drastic, but very effective, solution to spam. I did a web search and I believe this is the one he is using:

http://www-2.cs.cmu.edu/~sleator/blowback/ 

If you would like me to find out exactly what he is using, and get a report on how it is working for him, give the word and I'll do it. When I spoke to him on the phone last night, he said he hasn't seen a single piece of spam in weeks now. -- Steve R. Hastings "Vita est" steve@ hastings.org http://www. blarg.net/~steveha

Well most of my stuff does come through the Netwinder, but it doesn't do much filtering. I use Outlook on a dual processor system for that. It's hard to make proper filtration rules because I do get broadcast mail from PR companies that I want.

I am managing. The AMD enthusiasts cost me an hour or so, but it was instructive, and much of that time was used sending things to spamcop and getting hotmail accounts closed. I may even have done some good with some of the Fresh Spam...

On The Skeptical Environmentalist:

The original message:

Jerry,

There was a convincing (to me) broadside leveled at this book in the January issue of Scientific American which questioned Mr. Lomborg's methods and conclusions. The article is not available online, but Mr. Lomborg's rebuttal at http://www.greenspirit.com/lomborg/ quotes extensively from it.

Tad

Theodore A.D. Slawecki mailto:tslawecki@limno.com Computer Manager

My reply:

Well I am glad it convinced you but isn't it interesting that they took the trouble to find all those people to try to destroy a book?

Incidentally have you read it?

Which solicited:

I have to admit that I come to this unbiased by any contact with the item in question, in part because of the SciAm commentary. If there was indeed an interesting conspiracy headed up by Scientific American, it was successful in convincing me not to invest the time. Of course, I am somewhat of an intellectual butterfly, alighting only on that which won't change my world view.

Your comment did prompt me to consider my lack of interest. A quick visit to the newsgroups found a lot of interesting discussions about Mr. Lomborg and a few intersting links, including http://members.aol.com/jimn469897/lomborg.htm  - "Correcting myths from Bjørn Lomborg." One of the links on this page is to a discussion of his treatment of well-known environmentalist claims about stormy weather made by ... Isaac Asimov? ( http://members.aol.com/jimn469897/stormy.htm )

If you have time to check these out, please let me know what you think.

So I went to http://members.aol.com/jimn469897/lomborg.htm and found:

I am still wading through The Skeptical Environmentalist:  Measuring the Real State of the World by Bjørn Lomborg (Cambridge University Press, 2001) but I am already disappointed by it.  (Contrary to what fans of the book are saying, it is boring.  I predict that many copies of this book are going to go unread.)  Instead of getting actual skepticism we get misleading arguments including attacks of strawmen, out of context quotes, hasty generalizations and omitted evidence. The main point of the book is that the environmental movement is driven by the Litany (always capitalized) which according to Lomborg is the claim that the environment is getting worse and worse. But the author's main examples of the Litany come not from the environmental movement but newsmagazines and a writer of science fiction.

This looks well referenced, but in fact the references are not to instances in the book but to definitions of the terms "misleading arguments" and "strawmen."

Lomborg is hardly of my political persuasion. The fact that he gathers many of his statements from news magazines ought not be surprising: that's where the political fight is. And the anti-Lomborg crowd have yet to correct these examples. If people are saying the wrong things in their name, surely they ought to be concerned? But I haven't seen much concern.

If the environmentalists have come to the conclusion that things are not in fact getting worse, then Lomborg's book is a work of supererogation; but in fact I don't find he has distorted what the environmentalists say in the press. What they say in scientific journals may be different, but the divergence between public and private positions is itself interesting. 

In any event I wasn't at all impressed by http://members.aol.com/jimn469897/stormy.htm but then I wasn't impressed by everything Isaac wrote: according to his timetable we ought to be in severe trouble right now. Isaac for a while bought the Paul Ehrlich thesis. Considering this chap's contempt for science fiction writers on his first page, he seems now to find them models of science as opposed to Lomborg.

Fred Pohl was a USAF weather charting sergeant in Italy in WW II. That gives him a lot more authority to talk about the weather than Isaac. But really the point is, are hurricanes increasing in severity due to global warming? As speculation it's spectacular. As science it is speculation, which is really what Lomborg is trying to say.

Lomborg isn't a fascinating writer. Fred and Isaac could write rings around him. But he does ask the forbidden questions: and like the AMD people who put me on porn mailing lists, one sees a bit of this in the "environmentalist" community, terror that if their conclusions are questioned the world will come to an end. They work at getting people not to read the opposition.

Lomborg's major point is that people are living longer and healthier lives: far from the EcoDistasters predicted from 1976 on (Carter and the national malaise and the Era of Limits, Ehrlich with his giant die-off by 1996, the various Models of Doom from Forrestor and the Meadows) things seem to be getting better. The sources of the doomsaying are almost all long forgotten and are now brought up mostly in news magazines.

But you can judge the accuracy of this web site by looking at

http://members.aol.com/jimn469897/limits.htm 

and then actually reading the Limits To Growth book. Sure there was the conclusion that things might be made better: they were summarized in the World Future Society magazine in an article entitled "Why We Have To Get Poor Quick," and the title wasn't misleading. Good heavens! The Models of Doom are now put forth as refuting the premise that the environmentalist predictions have been for things to get worse?  Really?

If this is an example of why Lomborg's book is so bad that it ought not be read, it's interesting...  But it does seem to have been successful.

In any event the links are all given in your letter. Following them hasn't changed my views much. The best of the links I came across was

http://ecsp.si.edu/lomborg.htm 

and one might do worse than to go there. The key point is that the Kyoto Accords won't do much good and will cost a lot. And see below.

Cold fusion has surfaced again.

Press reports have the principles downplaying any potential for an energy source here. It seems very real though. They've got neutrons and tritium. http://www.sciencemag.org/
feature/data/hottopics/bubble/index.shtml
 

-- Michael Juergens, mikejuer@attbi.com on 03/05/2002 at 12:43:32 PM

I hadn't heard they had neutrons and tritium. If they can make a few neutrons then there's no reason they can't make a lot of them given time and engineering. And that would change the world.

And we have this answer from Michael:

See Research Article in Science. They definitely claim neutrons. A burst, a palpable burst of neutrons! http://www.sciencemag.org/feature/data
/hottopics/bubble/1067589.pdf
 

 I was burned by the last cold fusion flames. I hope it is true but I'll wait for come confirmations before leaping about the room in exultation. At least I think I will. Making neutrons is a big deal... 

g

 

 

 

TOP

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

read book now

TOP

Wednesday, March 6, 2002

On Fusion:

Dear Mr. Pournelle, I would take the latest tabletop fusion claims with a large grain of salt unless and until they are replicated. The main problem is that they called in two experts in neutron detection from another group at Oak Ridge who proceeded to use their own neutron detector to measure zero extra neutrons with the same experimental apparatus. Shapira and Saltmarsh concluded 'We conclude that there is no evidence of any real coincidences in this experiment. ... Any excess neutron production was at least three orders of magnitude less than that required to explain the tritium production rate reported in reference 1 as being due to d-d fusion.'

The original authors have contested the Shapira/Saltmarsh comment and stand by their results. However, until someone else replicates it, I won't believe it. I can send you pdf files of the various articles, comments and responses in case your readers want to judge for themselves. (I didn't want to send you an unsolicited 1 MB of stuff.)

Sincerely, Larry Weinstein

----------------------- Lawrence Weinstein Associate Professor of Physics Old Dominion University Norfolk, VA 23529  weinstei@physics.odu.edu http://www.physics.odu.edu/~weinstei cumulus-stratus-nimbus (smoke signal)

That was about my conclusion: we wait for replication. Sure would be nice, though...

Every time I think I have found the limits of the AMD Zone community's maturity index, I get something else.

<<Jerry Pournelle, Ph.D., is a science-fiction writer and Byte.com's senior contributing editor.>>

I have built several AMD single chip set computers, and, my god, I even use them in a business.

Oh my, if I were to listen to your science fiction about AMD not being stable as a computing choice, well lets just say that its a good thing I live in the real world and not in your Intel drool.

William Johnson [ RaceHillFarmsRacing@worldnet.att.net ]

Now think on the logic here. Ah well. At least this one wasn't obscene. Does this mean that AMD enthusiasts have to resort to this sort of thing because they are desperate? 


Tired of seeing that ugly router or switch with it's gaggle of network cables? There is now have a decorative alternative: Teddy Borg!

http://draco.mit.edu/teddyborg/  

Some folks really have way too much time on their hands...

Ken McIntire Atlanta, GA

Heh.

Now for something more serious:

Jerry,

Thank you for taking the time to check out my suggestions [regarding Lomborg's book], and thanks also for posting all of the links on your site. I appreciate your willingness to look at both sides of the issue -- and to let your readers know that there is another side.

On a related note, I am curious whether you subscribe to the Skeptical Inquirer? Paraphrasing your original reply, isn't it interesting that they take the trouble to find all those people to try to destroy so many beliefs? If so, how are the investigators at the Skeptical Inquirer any different than the reviewers at Scientific American? Is their effort an honest expression of scientific methods applied to pseudoscientific ideas, or a nefarious attempt to cover up the truths which are out there? Although I do have to admit that I am often (bothered, irritated, annoyed - pick one or more) by the frequently supercilious tone of the articles, I personally land on the scientific side of the fence when looking at SI, and also on the Lomborg issue. In these cases, I am willing to trust the authority and integrity of the authors, editors, and journals, and not invest the time to read something which won't parse because I am predisposed to disagree. I'm trusting that these entities are accurate in their assessments of what is fraudulent, nonsensical, or wrong.

Although ... I do really dislike the new Scientific American magazine format, which makes the magazine seem less scientific and therefore less credible. I understand that their changes bring SA further into the modern information age by making it easier and more rewarding to skim (which I sometimes really appreciate), but there's a loss of gravitas. Maybe I should believe it possible that SA is pursuing a hidden agenda ...

I won't, of course, but I try to appreciate that there is always more than one side.

Tad

Theodore A.D. Slawecki 

I read the Skeptical Inquirer from time to time and at one time I used to meet with The Amazing Randi at various meetings, and sometimes have dinner with him.

Skeptical Inquirer has a view which it seeks to prove. It tries to be fair, and sometimes succeeds. And, in any given controversy, they are going to be on the correct side just about all the time: if there are any "psi" phenomena they are not repeatable, and sometimes those who try to claim respectability for PSI go to ridiculous lengths: as with Backster who communicated with yogurt, and the "plant sensitivity" people. (Andy Offutt once wrote a science fiction story based on the notion that plants were aware and could communicate. Good fantasy.)

But since psi phenomena are so elusive, it may not be necessary to bring out quite such large cannon against it. I have been involved with some off the wall projects -- see my account of the Dean Drive, and some of what went on around The Velikovsky Affair -- and I have tried to keep "an open mind" while being quite skeptical of "new age" science claims. One hopes there are odd things going on that we can discover and use, and I needn't trot out the examples of penicillin, or for that matter radio: cell phones would probably be considered magic by Sir Isaac Newton. One hopes, but if one has to bet a very great deal, one bets against them and with Skeptical Inquirer.

In my judgment the Skeptical Inquirer has the sign wrong: there is no point in spending lots of time and energy in showing that most psychic power claims are false. What we ought to be doing is looking for at least one success. Of course that way lies Dr. Rhine and his tens of thousands of repetitions of the Rhine Cards experiment -- do it enough times and you have to find a 'success'. The question is, is that success repeatable. But we are dealing with an inherently rare and unstable phenomenon, so that's hard to do.

All of which is the long way of saying that I find most of the information in The Skeptical Inquirer to have no information value in the Shannon sense -- I can predict they will have once again shown that some poor schlubb's claim to talk to his yogurt is based on either bad equipment or fraud or both.

Scientific American long ago fell into the hands of people with a particular political view. This means that most of what they publish, which isn't part of a political agenda, is correct and often fascinating. Niven and I got the terror birds that figure largely in The Burning Tower (work in progress) from a Scientific American article. But when there are political implications, most of what you will see in Scientific American is so colored by its political expectations that it isn't very useful, and other sources are better.

The current flap about Lomborg would be more interesting if there were more attempts to answer his arguments and present counter data than to go back and argue endlessly over what Asimov and Pohl had to say about hurricanes. The fact is that what Lomborg calls "The Litany" is taught in most of the colleges in America, and its defenders seem to resort to about the same tactics as the AMDzone people have regarding the inamorata. So it goes.

I owe my readers another essay on Bayes and the value of information; perhaps it will get into the column this month.

Hello

I just wanted to say that I showed that file [the assasination mpeg] to a coworker and he told me that it has been floating around the web for a few years now and is just renamed to fit the current situation. I have not confirmed this but thought I would bring it up incase you wanted to check up on that.

Thank you, George Zawalonka Tech Support Rep.

Thank you. I have no idea. I don't even seem to be able to find the original letter pointing it out to me. From time to time I put up pointers to things I am not myself much interested in because others may be. I did try to warn people.

And in commenting on the draft of my upcoming column Dan Spisak says: 

Some general information I wanted to throw out there for you regarding the AMD motherboard situation:

1. Any motherboard using the Nvidia NForce chipset will be devoid of the VIA problems, however it is itself a very new chipset (boards are out from Abit, ASUS, and MSI) 2. Any motherboard using the SiS 735 chipset is safe (however I can only seem to find two manufactures using this chip, ECS and Leadtek but they have good reviews) 3. Even AMD has had a problem or two with its own chipsets:

Taken from Tyan's FAQ for the Tiger MPX at http://www.tyan.com/support/html/f_tg_mpx.html  :

1. Why can't I browse Network Neighborhood with my 3Com 3C996 Gigabit Ethernet Adapter?

This is a driver issue with the 3c996 Gigabit Ethernet card and the AMD chipset. 3Com has released drivers to resolve this issue. The drivers are dated 12-18-01 & can be downloaded from ftp://ftp.3com.com/pub/nic/3c996/3c996bx5.exe  or search for this driver at www.3com.com.

3. Why did I receive a USB card with my Tiger MPX (S2466)? Doesn't the onboard USB work?

Due to an issue regarding USB on the chipset, the USB will not function on the Tiger MPX (S2466). Both the stacked USB ports, and the onboard headers, will not function.

The USB card has been provided in the box for your convenience, in case you would like to use USB devices. In addition, we have given you the I/O shield to help differentiate between onboard USB and the USB card, when plugging in USB devices.

8. Why don't my USB ports on the motherboard work? OR Why does the I/O shield included with the S2466 block my USB ports.

Due to abrupt changes in the AMD 760 MPX chipset, onboard USB function on the S2466 is disabled. However, we have included a USB 1.1 4-port controller for your convenience. Not only do you now have the same USB functionality as expected, but the 2466 now has 2 extra USB ports in case your needs expand. To install, simply insert the supplied USB 1.1 4-port controller into any of the 32-bit/33MHz PCI slots (short PCI slots).

4. A majority of the enthusiast sites and overclockers and speed freaks keep recommending the KT266A via based motherboards. They do work well and fast. However it has been known that certain combinations of hardware in particular kind of duty cycles can cause problems. In the past this was being blamed on peripheral makers (there was a long standing issue with Creative Soundblaster Live! cards having choppy sounds for a small subset of people/configurations). However I do not know if you are aware of this or not but someone in Germany went so far as to actually get out their test gear and check the actual protocol signalling the VIA chipsets were doing and found some very odd goings on: http://www.tecchannel.de/hardware/817/index.html  along with some potential fixes http://www.tech-report.com/news_reply.x/3303/1/ 

Just some info for you to all digest.

-Dan S.

Yes: I did have in hand some of that. Good to have it collected in one place. My bottom line is still that AMD CPU chips are good, the boards that their single-processor systems came in were less so. And in any event we aren't dealing with the vast majority of systems. Most work well. We're dealing with what Bill Godbout called error rates: if an error rate is high enough to measure, it's too high. Particularly in hardware if your business is testing or using advanced software that pushes things to the edge.

And enthusiast web sites are predictably enthusiastic...

Then there is this:

> Be, the failed maker of a computer operating system once considered a rival to Microsoft's Windows, <

Er, huh? Considered a rival by *whom*, pray tell? Be, NeXT, the two or three different flavors of BSD, have all been touted by their author(s) as being better than Windows, but I don't think anyone with any sense has ever seriously considered Be or any of them to be a serious rival to Microsoft Windows.

Even the current darling Linux is still admitted to "not quite being there", despite improvements in the desktops available for it and despite the fact that Microsoft has been doing everything it can (product activation, smarmy "We're doing it for you" corporate licensing changes, recovery disks, and so forth) to shoot itself in the corporate foot.

Seriously, where do these places get their writers? That "competitor to Microsoft Windows" bit sounds like it was lifted straight from a Be press release and passed right along to the Web site without any critical thinking or even knowledge of the computer business at all, posing as a "news" item. I might expect that, as taken straight from Reuters (who, themselves, should have *someone* on staff to scan these computer news reports for sanity), but for zdnet to publish it verbatim is crazy.

If you're going to publish a press release, publish the press release. If you're going to publish a news article, you've got to invoke at least some meagre journalistic talent, I would have thought.

William Harris

Mr. Harris wants it to be clear he is speaking about the journalists where that was posted. Perhaps they were serious. I put it down to irony, myself. The BE system had its good points, but I never suspected it would really be a rival to you-know-who.


 

And here is a man who has led a sheltered life:

Mr. Pournelle:

I take serious umbrage at your claim that all Athlon chipsets are unstable. Granted, the original Irongate and KX133 chipsets had some issues, but that was *three* years ago. If you properly configure any of the recent VIA, AMD, nVidia, SiS, or Ali chipsets, you’ll have a stable system. Shoot, I’ve got a board based on the VIA KT133 (just one generation after the very first KX133) and it’s even stable. Granted, you can make all these chipsets unstable by tweaking the BIOS, but you can do that on Intel systems too. I don’t know what hole you’ve been living in for the last three years. What were you thinking when you wrote that Byte article? That is quite possibly the most blatant example of irresponsible journalism I’ve ever seen in my life. I can only hope that most people have enough sense to see through your drivel. I’m loathe to make the comment, but there are only two explanations for your comment: (1) you really are that out of touch with the state of modern computer hardware (at which point you need to stop writing NOW and get yourself caught up) or (2) you have some personal interest in seeing AMD’s competitors do well. I’m not accusing you of anything more than bad journalism, but the implication could easily be made.

Sincerely,

Matthew Rognstad

 [Emphasis mine.] One wonders: does he live at Superlatives R Us?

One also wonders: does AMD need this kind of defense?

 

 

 

TOP

 

 

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

read book now

TOP

Thursday, March 7, 2002

Column deadline day. Some mail is really fun to get:

Hello Jerry,

Thank you for your wonderful web site, http://www.jerrypournelle.com. Taking time to "smell the roses" of your site was a terrific and necessary excursion.

I will likely subscribe, finances permitting, as soon as possible. I look forward to spending more time delving into your home on the net.

Cheers!

Robert W. Bectel 

To which I can only say Thanks!

And a question:

Hi Jerry,

Love your column and the expertise you provide. Just wanted to bounce a question off of you about file management. I have a total of 120 megs HD across about four networked machines and needless to say alot of files.

Organizing and locating files is a chore. I am sick of organizing files, why doesn' t my computer do this for me? The search tools in Windows XP are a great start but are there any tools that are more robust.

Thanks,

Toney

Which is in fact a pretty good question. I save stuff in all directions onto all kinds of machines, and I am lately spending too much time looking for where things went.  I had hoped that Franklin One-Place indexing would do the job, but I had real problems with it: anyone else have better experience?  I sure need some way to organize all this stuff.

Jerry wrote:

"Which is in fact a pretty good question. I save stuff in all directions onto all kinds of machines, and I am lately spending too much time looking for where things went. I had hoped that Franklin One-Place indexing would do the job, but I had real problems with it: anyone else have better experience? I sure need some way to organize all this stuff."

While you can, with a lot of effort, do what you want with MS IIS and some scripts, Odyssey Software's ISYS does a great job in return for some dollars.

http://www.isys.com.au 

Download the time-limited evaluation here:

http://www.isys.com.au/products/downloaddesktop.shtml 

Jonathan Sturm www.sturmsoft.com

Thanks!

Jerry wrote in Wednesday's Current Mail:

"One also wonders: does ATI need this kind of defense?"

I suspect you meant to write AMD and of course you are entirely correct. The Mac, the Amiga, Linux, AMD etc seem to be substitutes for God in some people's lives. It's all a bit of a storm in a teacup, really.

Perhaps these people are also illiterate. I don't have a machine younger than 6 years with an Intel CPU and have found nothing to take umbrage with in what you have written about my preferred choice of AMD CPUs. You have mentioned that any real-world differences between the two platforms are more often than not swamped by the software running on them.

What you haven't mentioned is that board design and construction quality can also swamp any putated differences. A "Yum Cha" board with passive voltage regulator and electrolytic rather than tantalum capacitors will not run an Intel chipset and CPU as stable as an AMD with VIA chipset on an ASUS, AOpen or other well-designed motherboard.

Interestingly, it was a shortcoming of Intel chipsets that originally drove me into building AMD/VIA machines. Back in the Socket 7 days, Intel gave me a choice of a maximum of 64 MB of cached SDRAM, or use older, slower, DRAM technology to build a machine with 128 MB. That machine with VIA chipset and 400 MHz AMD K6-2 is still working 24/7 as my server while the Intel machine with the TX chipset is long gone. It was stable, but far too slow due to the RAM bottleneck. Adding more RAM slowed it down as the OS was no longer cached.

One wonders how many more potentially great, but crippled chipsets would Intel have made without VIA and AMD as rivals.

Your home page is much improved BTW.

Jonathan Sturm www.sturmsoft.com

I meant AMD not ATI and that is fixed.

Your point is a good one, and indeed I used to use AMD systems a lot. I sent an AMD K6-2 Socket 7 system off to Somalia, where it worked just fine.

One goes with the flow. I wasn't a fan of the TX chipset. But then Intel brought out other Intel motherboards. I didn't recommend some of the earlier ones, but they got better -- while AMD didn't bring out a comparable line and boards. That's a pity. And the thrust of my column was that now there are all-AMD motherboards, but of course the, uh, enthusiasts at AMDZONE didn't seem to have read past the first paragraph. 

It's a moving target, always.

Dear Mr. Pournelle,

I have read your comment on chipsets intended for AMD CPUs and would like to bring your attention to few issues that are worthwhile keeping in mind:

1) Currently majority of chipsets for AMD CPUs are not only unstable, but are also severely flawed and buggy. For example, USB 2.0 part of 760MPX does NOT work. VIA chipsets suffer from number of issues that affect PCI and PCI cards and result from way below standard I/O performance to flawed functioning of devices.

2) All benchmarks I have seen so far indicate that dual AMD CPU configurations seem to have problem scaling. In other words, adding 2nd CPU creates performance delta that is much lower than expected (I do not know is this due to CPU, chipset, or both), and is not significant enough to justify going with 2 CPU solution at all. On the other side, Intel CPUs get much bigger boost than AMD's when going from 1 to 2 CPUs in otherwise equal configuration.

With best regards,

Zoran Cvijetic

I need to look more into that. Thanks!

The real danger with Windows XP and its automatic update feature is that Microsoft's servers can be compromised. Microsoft is probably using code signing to reduce that danger, but what security precautions are they taking with their private key? In addition, we know that they have had various worms and Trojans enter their network -- the QAZ worm reportedly could have allowed access to their source code control system -- and the code for a security patch can be altered by such a worm before it is compiled, signed, and distributed to the public.

Note that I am also assuming that the update system will only accept updates and patches signed by one carefully controlled private key. If this is not the case, a bogus certificate could be created by an attacker and used to sign "updates". We know that VeriSign has sold at least two certificates labeled "Microsoft Corporation" in the past, and although those certificates have been revoked, this type of incident could happen again.

And someone could always try to factor the public signing key -- does anyone really know how many computers on the Internet have been compromised, or what they are doing? This type of attack would probably be more difficult and expensive even than bribing a Microsoft employee with access to the key, but it is theoretically possible, given enough time.

Robert Ransom

Thanks. For those who don't know, Robert did much of the programming for Mrs. Pournelle's reading program, so we have a fairly high regard for his opinions...

 

 

 

TOP

 

 

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

read book now

TOP

Friday, March 8, 2002

The column is in, but the aftermath cleanup is now on. 

An economist at Cal State Fullerton has been studying the economics of Everquest. The following article from The Toronto Star is available until mid-March:

http://www.torontostar.com/NASApp/cs/
ContentServer?pagename=thestar/Layout/
Article_Type1&c=Article&cid=
1015023694760&call_page=TS_News&call_pageid
=968332188492&call_pagepath=News/News
 

S.Todd

And indeed it's an interesting article.

The subject of the following letter 

Everquest is Evil

http://www.christiananswers.net/
spotlight/games/2000/everquest
 

 

Maybe not my morals but this game has wrecked my life... I've got a level 60 druid and no career to speak of. It sure is insideous.

-Andy

Says a good bit. Alas, it sure can eat time.

Joel Rosenberg continues to collect stories of the Crazy Years:

http://www.washtimes.com/
commentary/20020308-1097847.htm
 

"In Michigan, an 8-year-old boy is being prosecuted for pointing a toy gun at three other youngsters and threatening to shoot them.... Tommy Davis was 7 at the time he committed three counts of 'assault with a dangerous weapon' by pointing his toy pistol....

 Four men broke into Ronald Biggs' Goldsboro, N.C., residence and assaulted him with a baseball bat. He broke up the attack by shooting one of the assailants. His assailants were charged with misdemeanors, and Mr. Biggs was charged with assault with a deadly weapon...

"Police are helpless, hapless and hopeless. This is why they arrest children." 

And it's hard to think of what to say when people put rules ahead of the purpose the rules were to accomplish. It is the stuff from which dictatorships arise.

And Joe Zeff sends this:

Check out http://www.cnn.com/2002/LAW/03/08/
hit.and.rundeath.ap/index.html
,

  especially the defense's claims. To claim that killing somebody while committing a felony (hit and run) isn't murder is nonsense, and the story of what happened after the accident disgusts me. If the story is even partially true, the woman deserves life without parole, if the death penalty isn't available.

I heard that on the news last night while fixing my dinner (Roberta is down at the beach but I had to finish the column). I didn't hear where it happened, and I thought it was here in Los Angeles. At least it wasn't one of my neighbors.

The son wonders why, reminding me of a character from The Groves of Academe...

Several from Roland:

Ah, so.

http://straitstimes.asia1.com.sg/primenews/
story/0,1870,106236,00.html
 

Great Leap Forward

Do you think the ChiComs are attempting a pun?

http://www.newscientist.com/news/news.jsp?id=ns99992012 

And Theropods

http://www.cnn.com/2002/TECH/science
/03/06/feathered.dinosaur/index.html
 

I still don't see where he gets the time to find all these things even with a T3 line...

On the military power of dragons:

http://www.pvv.ntnu.no/~madsb/dany/dragons/military/

Of course there's this big question the Ancients must have asked: Where do we get One?

Joey

Indeed

And A WARNING of

"An unusually effective virus"

Dear Jerry,

As the wife of a devoted fan (John Bartley), I'm sending you this "Security Update" which arrived in my email this morning. I consider myself at least reasonably competent as an Internet user, but my BS filters didn't engage until I noticed that the 'fix' was included as an attachment. The closing phrase was an ironic addition: 'With friendly greetings". :P

Thanks for your time, and thanks for all the good work you do. :)

Lea Rush

>Return-Path: <rdquest12@microsoft.com> >Received: from fep02-svc.swip.net ([130.244.199.130]) by ; Fri, 08 Mar >2002 08:49:19 -0600 >Received: from pfuckie ([213.101.33.129]) by fep02-svc.swip.net with SMTP > id <20020308144830.KYQA1227.fep02-svc.swip.net@pfuckie>; > Fri, 8 Mar 2002 15:48:30 +0100 >From: "Microsoft Corporation Security Center" <rdquest12@microsoft.com> >To: "Microsoft Customer" <'customer@yourdomain.com'> >Subject: Internet Security Update >Reply-To: <rdquest12@microsoft.com> >MIME-Version: 1.0 >Content-Type: multipart/mixed; > boundary="NextPart_000235" >Message-Id: <20020308144830.KYQA1227.fep02-svc.swip.net@pfuckie> >Date: Fri, 8 Mar 2002 15:49:16 +0100 >X-Rcpt-To: <lea@  acific.com> >X-DPOP: DPOP Version 2.4a >X-UIDL: 1015606294.005 >Status: U > > >Microsoft Customer, >

 > this is the latest version of security update, the > "4 Mar 2002 Cumulative Patch" update which eliminates all >known security vulnerabilities affecting Internet Explorer and >MS Outlook/Express as well as six new vulnerabilities, and is >discussed in Microsoft Security Bulletin MS02-005. Install now to >protect your computer from these vulnerabilities, the most serious of which >could allow an attacker to run code on your computer. > > >Description of several well-know vulnerabilities: > >- "Incorrect MIME Header Can Cause IE to Execute E-mail Attachment" >vulnerability. >If a malicious user sends an affected HTML e-mail or hosts an affected >e-mail on a Web site, and a user opens the e-mail or visits the Web site, >Internet Explorer automatically runs the executable on the user's computer. > >- A vulnerability that could allow an unauthorized user to learn the location >of cached content on your computer. This could enable the unauthorized >user to launch compiled HTML Help (.chm) files that contain shortcuts to >executables, thereby enabling the unauthorized user to run the executables >on your computer. > >- A new variant of the "Frame Domain Verification" vulnerability could >enable a >malicious Web site operator to open two browser windows, one in the Web >site's >domain and the other on your local file system, and to pass information from >your computer to the Web site. > 

>- CLSID extension vulnerability. Attachments which end with a CLSID file >extension >do not show the actual full extension of the file when saved and viewed with >Windows Explorer. This allows dangerous file types to look as though they >are simple, >harmless files - such as JPG or WAV files - that do not need to be blocked. > > >System requirements: >Versions of Windows no earlier than Windows 95. > >This update applies to: >Versions of Internet Explorer no earlier than 4.01 >Versions of MS Outlook no earlier than 8.00 >Versions of MS Outlook Express no earlier than 4.01 

> >How to install >Run attached file q216309.exe > >How to use >You don't need to do anything after installing this item. > >

 >For more information about these issues, read Microsoft Security Bulletin >MS02-005, or visit link below. >  http://www.microsoft.com/
windows/ie/downloads/critical/default.asp
   

>If you have some questions about this article contact us at >rdquest12@microsoft.com >

>Thank you for using Microsoft products. > >With friendly greetings, >MS Internet Security Center. >---------------------------------------- 

> >Microsoft is registered trademark of Microsoft Corporation. >Windows and Outlook are trademarks of Microsoft Corporation.

Insidious. Good catch. Thanks. [Emphasis in above added.]

But BUGTRAQ apparently thinks I was not clear, so I am adding this here and below:

My apologies to Bugtraq. I actually thought my readers were intelligent enough to figure this out. In case anyone doesn't realize it, DO NOT RUN THE DARNED MAIL ATTACHMENT, but then I think I have made that clear many times. Ye gods. And see below.

 

 

 

TOP

 

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

read book now

TOP

Saturday,

Roland sent this several days ago but I missed it. It is quite disturbing:

Untrusted computing.

http://www.theregus.com/content/4/24206.html 

One would think there would be more out there about it. I haven't seen a fix yet, either.

The Cold War is over, and Roland sends this strange story of one aftermath:

http://www.washingtonpost.com/
wp-dyn/articles/A63646-2002Mar8.html
 

And another, on US policy, which is written to be as disturbing as possible; you will have to draw your own conclusions.

The New Triad.

http://www.latimes.com/news/opinion/la-op-arkinmar10.story

Note that Triad is a word used both for the US nuclear deterrent policy (bombers, land-based missiles, and submarines) and for Chinese secret societies which have had considerable influence in history.

There is another proposal for "A new Triad" by Stansfield Turner, which is also disturbing in its own way.

http://www.unc.edu/depts/diplomat/
AD_Issues/amdipl_8/turner.html
 

The issue isn't going away.

 

 

 

 

 

TOP

 

 

This week:

Monday
Tuesday
Wednesday
Thursday
Friday
Saturday
Sunday

read book now

TOP

Sunday, March 10, 2002

And sometimes I am merely flabbergasted:


To: Jerry Pournelle
Subject: Microsoft Security Bulletin -- Alleged

Jerry,

 

The Microsoft Security Bulletin you posted is not real.

 

This is actually a worm. And is you read it you’ll know it’s not Microsoft. The mail originated at:

 

fep02-svc.swip.net@pfuckie

 

Gates doesn’t use pfuckie.

 

Larry 

Cambridge, MA

Windows NTBugtraq Mailing List

--------------------------------------------------

 

Date: Wed, 6 Mar 2002 13:08:44 -0500

Reply-To: Russ 

Sender: Windows NTBugtraq Mailing List <NTBUGTRAQ@LISTSERV.NTBUGTRAQ.COM>

From: Russ 

Subject: Alert: Worm posing as IE cumulative patch

Content-Type: text/plain; charset="us-ascii"

Lots of people have emailed me about the email making the rounds which states its from "Microsoft Corporation Security Center" and has a subject line of "Internet Security Update".

The email has an attachment which the message claims to be the "1 Mar 2002 Cumulative Patch" for IE.

In case you don't already know, there isn't such a patch and its not from Microsoft. Microsoft never emails out patches.

The bogus email is actually the GIBE worm, a description for which can be seen at;

http://www.f-secure.com/v-descs/gibe.shtml 

 http://www.sarc.com/avcenter
/venc/data/w32.gibe@mm.html
 

 http://vil.nai.com/vil/content/v_99377.htm 

 http://www.antivirus.com/cgi-bin
/vinfo.pl?OneVirus=WORM_GIBE.DR
 

Just to provide a few references (check the home page of your anti-virus vendor for details)

Cheers, Russ - NTBugtraq Editor

 

My apologies to Bugtraq. I actually thought my readers were intelligent enough to figure this out. In case anyone doesn't realize it, DO NOT RUN THE DARNED MAIL ATTACHMENT, but then I think I have made that clear many times. Ye gods.

Do I have one single reader who thought I was telling people to run that mail attachment?

(It turns out, I find from later mail, that someone sent the reader a part of what I had posted, so he replied to me without reading what I had said. There seems to be a lot of that going around.)

Then we have:

Dear Dr. Pournelle,

Take a look at this piece of insanity:

http://www.uniontrib.com/news/state/
20020309-9999_1n9spudguns.html
 

Yup, if you have any PVC tubing and a can of hairspray, you could become a felon.

Lest you think I'm being far-fetched, take a look at the way BATF already enforces the [unconstitutional] firearms laws already on the books. Possession of a single part is often used to charge people with possession of "machine guns". Fender washers (like you can get at Home Depot) are used to charge people with possession of "silencers".

No, not everyone who has washers is so charged. Just the ones the government wants to nail. Make everyone a criminal, and cherry pick the ones you want.

"There's no way to rule innocent men. The only power any government has is the power to crack down on criminals. Well, when there aren't enough criminals, one makes them. One declares so many things to be a crime that it becomes impossible for men to live without breaking laws. Who wants a nation of law-abiding citizens? What's there in that for anyone? But just pass the kind of laws that can neither be observed nor enforced nor objectively interpreted--and you create a nation of law-breakers--and then you cash in on the guilt." -- Ayn Rand, "Atlas Shrugged"

Notice that burglary, assault, robbery, rape, murder, and arson are already crimes -- and properly so. The only interest served by banning inatimate objects is the government's desire to force people to "obey us or else" -- it is a manifestation of power-lust.

Gordon Runkle

-- "Far and away the best prize that life has to offer is the chance to work hard at work worth doing." -- Theodore Roosevelt

Never ascribe to malice that which is adequately explained by incompetence. Napoleon Bonaparte.

I suspect the intention of the legislation was not quite as evil as you postulate, and no one ever went br0ke underestimating the intelligence of politicians in all matters other than retaining their offices.

But we were born free.

Now this:

 

Hello my name is Taliah and I was just doing a little research on on an e-mail that I recieved with the subject cooperation. It was from some guy named Dr.Emmanuel Tarfa. He offered me a business proposal. He was going on about the Nigerian railways and how they had a contract of (US$175,000,000.00) to refurbish the railway stations. Just like they told you the transaction was 100% risk free and my decion was needed promptley. I e-mailed him back and asked how he got my real name and e-mail address. It took him 4 days to respond and he said from an international business journal. He said that it was not convenient to tell me everythnig by e-mail and asked for my number and gave me his. What do you think about that?

What I think about that is that you should not answer that kind of mail at all, and never let them know who you are or where you live. These people can be dangerous. The likelihood that someone will send you millions of dollars is exceedingly low, and the protons will decay before you actually get any money from these people. Con games are con games. Incidentally, "you can't cheat an honest man" is a principle of most con artists: they appeal to the larceny in all our hearts. "Help us steal this money and we'll cut you in" is a component of most con games...

The principle is, if it sounds too good to be true, it is.

Here is a recent example:

DEPARTMENT OF PETROLEUM RESOURCES PLOT 225 KOFO ABAYOMI STREET VICTORIA ISLAND,LAGOS, NIGERIA. DIRECT FAX: 234 1 7590904. TEL; 234 –1- 7591519

ATTENTION: THE PRESIDENT/C.E.O/OWNER

RE: URGENT & CONFIDENTIAL BUSINESS PROPOSAL

Dear Sir/Madam,

I am DR. Mekudi Rangim (JP)Member Contract Award Committee of the above Department

Terms of Reference

My term of reference involves the award of contracts to multinational companies.

My office is saddled with the responsibility of contract award, screening, categorization and prioritization of projects embarked upon by Department of Petroleum Resources (DPR) as well as feasibility studies for selected projects and supervising the project consultants involved. A breakdown of the fiscal expenditure by this office as at the end of last fiscal quarter of 2000 indicates that DPR paid out a whooping sum of US$736M(Seven Hundred And Thirty Six Million, United States Dollars) to successful contract beneficiaries. The DPR is now compiling beneficiaries to be paid for the fourth Quarter of 2001.

The crux of this letter is that the finance/contract department of the DPR deliberately over –invoiced the contract value of the various contracts awarded. In the course of disbursements, this department has been able to accumulate the sum of US$38.2M(Thirty-eight Million, two hundred Thousand U.S Dollars) as the over-invoiced sum. This money is currently in a suspense account of the DPR account with the Debt Reconciliation Committee (DRC). We now seek to process the transfer of this fund officially as contract payment to you as a foreign contractor, who will be fronting for us as the beneficiary of the fund. In this way we can facilitate these funds into your nominated account for possible investment abroad. We are not allowed as a matter of government policy to operate any foreign account to transfer this fund into.

However, for your involvement in assisting us with this transfer into your nominated account we have evolved a sharing formula as follows: (1) 20% for you as the foreign partner (2) 75% for I and my colleagues (3) 5% will be set aside to defray all incidental expenses both Locally and Internationally during the course of this transaction.

We shall be relying on your advice as regard investment of our share in any business in your country. Be informed that this business is genuine and 100% safe considering the high-power government officials involved. Send your private fax/telephone numbers. Upon your response we shall provide you with further information on the procedures. Feel free to send response by Fax: 234-1-7590904 / TEL: 234-1-7591519 expecting your response urgently. All enquiries should be directed to the undersigned by FAX OR PHONE. Looking forward to a good business relationship with you.

Sincerely, DR. Mekudi Rangim (JP)

------------------------------------------------- Yahoo! Movies - Vote for your nominees in our online Oscars pool.

Now there is no reason whatever that I should be singled out to receive this mail or invest money for those people, none of whom I have ever met. They claim to have stolen a lot of money, and they are telling me about it to enlist my aid. They don't know me but they are confessing a crime against their own government, and they want me to help.

 

 

 

  TOP

 

 

birdline.gif (1428 bytes)