SARS really is very nasty

"SOMETIMES LETHAL VIRUS" ain't the right way to describe SARS. Try "highly contagious virus with no known treatment and a mortality rate of approximately 4%". There's a potential for a huge number of people dying, and total overload/paralysis of health care systems worldwide, and a lot of economic fallout..

Jim Roberts, MD, FRCPC

SARS death rate not the only problem. 75+% of those ill are in hospital for weeks. Percentage requiring intensive care will quickly outstrip supply and number of ventilators will not come close to filling the need.

Result: many more deaths, hospitals overwhelmed, AIDS patients will die off like never before. People will die of many other things because hospitals all over the world will be overwhelmed.

And, a few people may die in riots over stupid beliefs about what's going on.

If Toronto and Hong Kong hospitals were overwhelmed so easily, what are 3rd world hospitals going to experience?

V. Switzer, MSN, ARNP, COHN-S, Occupational Infectious Disease Specialist

Pictures of water cooled Opterons

Just wanted to say that if this subject line appeared in Inbox, I'd figure someone invented processor overclocker porn at last..."Oh baby, I am /so/ hot"...

And no, I am /not/ looking. Every night I go home to the greatest little Celeron in the world, we're very happy, and neither of us has any inclination whatever towards water-sports.

And anyway, I expect they want to have your credit card number before you get to see any real action :-)

(Strangely, more of the unsolicited mail I'm getting at the moment is about snoring cures.)

Robert

In you article entitled " Intel Invents Overclock Deterrent Method", you state with respect to the practice of computer integrator/vendors overclocking processors and selling them at a higher clock rate that the rating at which they were sold: "We're sure not many resellers do this kind of thing.". I can assure you it does happen, and it used to happen frequently. In the early '90's, I bought a computer from a white box maker in the Silicon Valley when I was living down there working for 3Com. Being the sort I am, I peeled off the heat sink just to see what I had and lo and behold, the processor I had bought in that 33 MHz machine was marked for 25 MHz! These guys didn't even bother to remark the thing, figuring that most of their customers wouldn't bother with messy heat sink grease and all to see what they had actually bought.

It really was a fairly common practice back when the patent you refer to in the article was filed. And, in fact, there were quite a few articles in the press of the day as I'm sure you are aware but have conveniently forgotten. Many of the articles mentioned the operations that actually remarked parts for a higher clock rate and then resold them. It is indicative of the conservatism with which Intel approaches spec-ing its parts that even though I paid for something that my vendor did not deliver, the computer worked flawlessly.

I can also tell you that Intel has many patents that never actually make it into products. I have no knowledge as to whether the technology covered by this patent, or any technology with the same purpose, has ever been put into an Intel product.

Your article is another clear example of the bias you bring to all things Intel. Its use of innuendo to ascribe anti-customer motives to an effort which was really quite pro-customer is something you should be ashamed of.

These are my personal views and not the views of my employer.

From an Intel email address

From the last paragraph of your article ( Congressman says Iraq should have American cell phones), I think that you are under the mistaken impression that the U.S. can magically turn GPS off for a specific country. This can not be done. (This may be possible for a region of the world such as the entire Middle East but doing so would effect GPS accuracy and availability for about one quarter of the planet and therefore, the U.S. has never done such a thing since the GPS system became operational).

Further, since U.S. military systems are heavy users of GPS, coalition forces would be effected more than Iraq's systems would if GPS were not available. Another good reason, we do not turn off GPS.

The semi-informed might assume since GPS has two frequencies sometimes referred to as 'military' and 'civilian' that perhaps the U.S. would just turn of the 'civilian' frequency near Iraq. This would be faulty reasoning for two reasons. First, their is that whole impacting civilian GPS reliability and accuracy for a large portion of the earth problem. Second, their really is no 'civilian' frequency. Both frequencies were originally intended for use by the military. More than 99% of 'military' GPS systems can not function without the C/A code which is broadcast over the frequency sometimes referred to as 'civilian'. This code/frequency must be acquired, locked onto, and decoded in order for the military GPS receivers to lock onto the second frequency that carries the P (or Y) code. If you turn off the C/A code, no more GPS for anyone.

In the future (and perhaps a very few systems have made it into the field already), military GPS receivers will have 'direct Y' capability and the C/A code/'civilian' frequency will no longer be required by the military but that day is well over a decade away.

Regardless of how you get to the second GPS frequency that provides the P (Y) code, military receivers still rely on the C/A code to account for ionospheric errors. This is what makes a military receiver more accurate than a civilian receiver. So taking away the C/A code would degrade the coalition's military receivers.

This too will change with the addition of a third and possibly fourth GPS frequency in the future. That will enable civilian and military frequencies to be completely separated but again, that is well over a decade away.

Their is one last thing that the U.S. could do with the 'civilian' frequency to hinder those evil Iraqi troops and that is to reinstate selective availability. This is the system that was used up to a few years ago that degraded civilian accuracy to 30 meters or more. This would have no effect on the Qualcomm/snaptrack geolocation technology as this uses an Assisted GPS (AGPS) technology that includes differential GPS capabilities. Differential GPS effectively eliminates the selective availability error sources as well as several others. In fact, Qualcomm's AGPS technology typically provides better position accuracy than a military GPS receiver since they do not generally have a differential base station available to them.

So go ahead and use your trusty hand held GPS receiver when visiting Iraq. Now that the Russian supplied jammers have been destroyed, it should work just fine.

Also I was confused by the last sentence in the second to last paragraph. You make it sound like it would be a bad thing for these phones to fall into the enemies hands. I can't think of anything better than to have an enemy stupid enough to carry around a cell phone that reports its position back to a friendly cellular network. If the enemy steps outside, the AGPS accuracy would certainly be good enough to drop the weapon of your choice on his head. If he is inside a building, depending on the building type, the accuracy may degrade to the point where you might need a larger weapon to assure his untimely demise but dead is dead.

Christopher

Hackers hit back at Al-Jazeera attack

I found this line in your article funny to the point of absurd. "According to the Wired report, individuals who attacked the Al Jazeera site are breaking the law and censorship is not part of the US way of life. In fact, quoting a member of Hacktivismo, part of the reason for Gulf War III was to fight for freedom." Hmmm, RIAA.......Any bells ringing yet? Does the word hypocrisy give voice to those bells?

Email address supplied

Your article " PC component failure-the problem that dare not speak its name" that was posted on theinquirer.net indicates that Abit is the only manufacture that is admitting to a problem with bad caps at this time (Feb, 2003). Abit is very proud of that fact because their Technical Support department has referred me to the article. However, I think there is more to the story...It's one thing to admit to a problem, but it's another to face up to the responsibility of that problem.

About a year ago I returned my Abit KA7-100, (along with $25) for repair/replacement for the leaky cap problem. By the time I received the "repaired" board back, I had already replaced it, so I had no need for it. I kept it (UNUSED) in the box, in the closet for about a year. This past weekend I thought I'd use it in another system I was building, only to find MORE LEAKY caps! Those caps leaked while in the box, in the closet! I promptly e-mailed Abit and their reply was "We're really sorry that our warranty for the replacement boards is 90 days from the date we sent it out". "There will be a $25 charge for another replacement board".

My issue is this; while they admit to a problem, I paid $25 to have them repair the board and all they did was replace the obviously leaking caps and they left the other BAD ones to leak at some future date! They KNEW which brand caps were bad, yet they did NOTHING with the ones that weren't leaking at the time of the repair.

Your article states "If there is a problem, hopefully these companies will buck the industry trend and follow Abit's example." In my opinion, I hope those other companies will buck Abit's example and deal with the problem in a responsible manner, and give the customer what they are paying for.

Until authors, who have a major internet presence (theinquirer.net), expose these kinds of companies, consumers will continue to be ripped off.

Thank you for letting me vent : )
Douglas

Serial ATA, more than a fashion statement

I like the Inquirer; it is mandatory reading for me every morning and thoroughly brightens my day, but I am getting increasingly irritated by the (total lack of) quality of the writing itself. The article referenced is just the latest example. There are whole paragraphs in there that I had to read three times before I understood what Michael Schuette was talking about, and even then I was only 75% sure:

"The new design is sleek and flexible and listens to the name of Serial ATA and mind me saying that the rainbow spectrum of the cables is just one aspect. "

What the hell does that mean? (He's not the worst, however: Fuad Abazovic comfortably takes that title.)

I realise that the some of your writers are clearly not native English speakers and that some of the articles are translated from other places. However, that cannot be an excuse for shoddy grammar and vocabulary in something that purports to be an English language publication. Please, get a copy editor or find writers who can actually write coherent sentences.

sincerely,
Jonathan Lucas

PS: obviously, don't correct anything in the flames of the week. Your
readership is always right ;-)

"When the original cables were designed, they were built to transfer 3MB/sec and now we are running at 133 MB/sec, 44 times faster than what the inventors thought would be reasonable and still using the same technology."

Come on now, where'd you get this guy? ZDNet? What kind of disk is he using to get 133 MB/sec? Just because the spec says 133 MB/sec doesn't imply that drives exist to fully utilize the bandwidth. In your next story you should laude Gigabit Ethernet as being ten times faster than Fast Ethernet. You can follow-up with a story about the 3 GHz Pentium 4 being three times faster than a 1 GHz Pentium III. This is what I call "objective" reporting.

"Jurassic Park taught us that Dinos were not just passive but Jurassic Park came out after Parallel ATA was invented."

It's important to not use controlled substances when writing advertorials to insure coherency.

"In some benchmarks we have already seen up to a doubling of the performance within the same drive just by switching from a parallel to a serial interface, most likely enabled by the reduced command overhead but this is only the beginning."

I'm willing to bet that these benchmarks don't involve sustained transfer rate.

Tony

Dear Tony

Any standard Maxtor UATA133 drive will suffice. Keep in mind that 133 MB/sec also includes the command overhead which, depending on the block transfer size, can make between 10 and 90% of the total transfer. That means that in the best case you will see actual data transfers of 115-120 MB/s and that is exactly what we see with any standard nForce2 system / UATA133 drive as long as the SW drivers are installed.

Lost circuits article link

Caffeine is addictive, I admit.

Sustained transfers are as you are well aware a function of media area density, rpm and TPI (track per inch) and in so far, the Effective Host Transfer Rate (Thx ) as a funcion of the controller interface will have no bearing.

Please feel free to bother me with any technical questions you have about HDD technology.

Hope this helps
Best Regards
Michael Schuette, Ph.D. [the article's author]

You seem to be saying that you can get 115-120 MB/s of actual read/write throughput from one hard drive which I think is nonsense. It's my contention that most newer drives are only able to sustain transfers at around 30 to 50 MB/s. Occasionally, you can get burst transfers that may approach 100+ MB/s when you have a cache "hit", but rarely will you ever sustain those lofty numbers.

Article on Tom's Hardware

Regards,
Name, address supplied

Dear Tony

Yes 115-120 MB/sec is about what a good UATA 133 host transfer rate is, whether you believe it or not but of course, those are transfers from the cache. The cache in turn can get data by using speculative lookahead to retrieve the data from the media before a read command is issued.

Your original point was that there is no such thing as a 133 MB/sec transfer across the cable but now you evade into sustained media transfers a.k.a. sequential reads and writes. Unfortunately, whoever did the review on THG does not specify what benchmark was run, most likely it is HDTach and those are sequential reads / writes using a benchmark that was written for drives of 5 GB capacity and less. If you are interested, I will be glad to spell out all pitfalls and inaccuracies for you but you should first try to understand the differences between effective host and effective drive transfers along with internal performance and host performance which are yet two separate issues.

If you have any specific questions, I will be more than glad to answer them.

Best Regards
Michael

" New Celerons appear for sale before being announced "

This quote:
" Prices on the new processors start low with the 2.3GHz model selling from only US$119 at current exchange rates. The 2.4GHz is US $10 more."

Why all the LOVE for a LOW END NEUTERED processor that won't be available for a quarter or two. Is the LOVE because Intel gives KICKBACKS for these low end products.

A more balanced story would have stated that as of TODAY anyone can buy a high end AMD Athlon XP2400+ for $110. So why wait two quarters for Intel's low end parts?

Steve

µ