what does TG think of this?

That anandtech article is very, very thorough. I don't see why M$ can't make a patch based on the registry fix they've already posted. To me, the problem appears to be an M$ one. So intel was right- it's a M$ driver issue.

I'm very surprised that it took this long for someone to detect it. This problem should have been noted way back when SP2 was released... I guess no one ever caught on until THG's article.

One quick comment: had THG never published that article in the first place, we'd all still be walking around oblivious to the USB 2.0/SP2 problem. Including anandtech.com.

-mpjesse

I am, well, not feeling happy, that's for sure. My natural tendency would be to sue them outright, but I am going to hold back because, I know how some people will twist it

This is my email to Anand regarding the issue. We are going to fight this crap every inch of the way.

I guess the major point of contention in their article are these statements, correct?

and...

and...

The title of the article, "Intel Core Duo USB Issue: A Mischaracterized Bug"

Otherwise I don't see anything else in the article that's overtly critical of THG. Or am I missing something else?

I discussed an article they wrote about the Athlon FX-60 in here once...

My point of contention was they used cheap generic DDR2-667 memory timed at CAS 5-5-5-15 timings to benchmark the Pentium 4 Extreme Edition 955, P4 EE 840 and Pentium D 820. Those timings are nearly the SLOWEST you can get for DDR2-667.

On the Athlon FX-60 based system they used premium OCZ PC3500 memory timed at CAS 2 (2-2-2-7). That's pretty much the fastest RAM you can get for an Athlon system (timing wise).

Seems a little unfair and biased, doesn't it? See for yourself:

http://www.anandtech.com/cpuchipse [...] i=2668&p=2

If super low memory timings are good for AMD then they should be good for Intel. That's my point I guess.

-mpjesse

It really makes you want to ......

Here's some ultra low timed DDR2-667 memory from OCZ, Geil, and Corsair:

http://www.ocztechnology.com/produ [...] 0_titanium CAS 4
http://www.corsair.com/corsair/pro [...] 5400UL.pdf CAS 3
http://www.geilusa.com/proddetail.asp?linenumber=64 CAS 3

So it ain't like lower timed DDR2-667 memory isn't made. Going from CAS5 to CAS3 is a big frigin jump and would make a noticable difference in benchmarking.

(for the record, THG used DDR2-667 timed at 4-4-4-8 in their FX-60 review)

-mpjesse

Quite frankly, we are quite stunned by the route Anandtech has chosen. We do not feel that those articles as well as previous pieces are in line with journalism ethics.

On a more factual basis, both articles basically repeat what we have said, with the emphasis that the problem shouldn't be blamed on Core Duo. Which we haven't done anyway.

The key here, however is, that Anandtech does not provide a solution for the user. Anandtech's article is incomplete in terms of the end result. We will be following up on our article, after having tested multiple notebooks with and without patch, with varying USB devices and with close participation of Intel and Microsoft. In the end we all can only speculate - only Intel and Microsoft know who is at fault.

Interestingly, Anand Shimpi stated in his blog that he has been working on the problem for quite a while, now we are reading that he was prompted by our article to do further testing. Quite frankly, I don't buy it that Anandtech knew about the problem before we wrote about it.


Our interest is a real solution for the problem; there was no follow article so far, as there was little new information to publish. Simply repeating that certain devices suffer and others don't does not help anyone here, with the exception of Anandtech. But we are glad that Anandtech found the Microsoft KB note with the help of our article, so the issue may get more exposure overall and may help contribute that people will be patching their notebooks.

However, the patch published is merely a band-aid. There is more information to the story. Stay tuned for a follow up article to be released on TG Daily soon.

Wolfgang

I sent this email to the author of the TechDaily article published. Read below, it includes the response I was given by TechDaily's editor in chief.

The sad part is, aside from all that, their facts were wrong. Granted I haven't read the Original THG article since it first came out, but I seem to remember that not only did they not say that it an Intel only issue, but the mentioned a pretty good case for it being an M$ Issue.

Now this article completely disregars that, and claims whatever it wants. Makes you wonder about what you are getting on the tech side.

For the record I did ask Anand why he tested the intel platform the way he did when the article was first published.

I didn't receive a response.

I was nice about it too! :-)

To be honest, those were exactly my thoughts when I intially read the article. I found the battery life part a mere curiosity. I was more concerned about Core Duo performance than battery life. Why? I don't give a rats ass about battery life. I'm almost always connected to an outlet in an airport, hotel, meeting, etc. The only time I use my battery is in an airplane... and even then my flights are relatively short (1-2hrs). Additionally, the nature of my work requires full use of the processor and video card- so battery life is already a joke in my arena.

That said, I'm sure there are plenty of others who do care about battery life.

:-)

-mpjesse

Ahh, so there may be truth to it that Dick was just fulfilling his obligation signed with satan, and as time draws nearer the disturbances in the cosmos fan the flames of old ....

That's disturbing in more ways than I can count.
But hey, a shotgun! You can't go wrong with a shotgun.