Bert McComas & Van Smith
InQuest Market Research - May 2001

<A HREF="http://www.inqst.com/articles/athlon4/0516main.htm" target="_new">http://www.inqst.com/articles/athlon4/0516main.htm</A>

Highly interesting read:

The P4’s 128-byte cache line size can reduce bus efficiency under certain circumstances.

The P4’s actual maximum thermal dissipation is much higher than was commonly understood.

Power Expectations for the 2GHz Willamette P4 (110watts)

P4 Performance Throttling:
As mentioned above, Alex Behrens of Hardware-Unlimited identified what appeared to be CPU clock throttling in Quake III at 1.7GHz. The condition that he observed was instantly verifiable. Using a production 1.7GHz P4, an Intel Motherboard and a world class copper base heat sink, Quake3 would irreversibly throttle by approximately 30% after running for several minutes. We had to exit Quake 3 or reboot in order to restore normal operation.



<P ID="edit"><FONT SIZE=-1><EM>Edited by noko on 05/18/01 03:35 PM.</EM></FONT></P>

nice info...

=)



--call it what you wish, with this machine I can make mercury flow in 3 directions at once--

Isn't the 2ghz northwood using 1.3v only make 40w at 2ghz.

!!! Leader of the Anti Nvidia Army !!!

LOL ROTFLMFAO!!!

Heheh...pardon the outburst...

You see, AmdMeltdown linked to that article a few days ago in a thread trying to discredit InQuest as a hoax. He got told twice that he wasn't successfully discrediting it so much as drawing attention to its existence. And now look at this! LoL.

You have insulted AmdMeltdown's idol; you have wounded his pride! Prepare to be flamed by him....ah screw it, why prepare? I might as well say, "Prepare to be swatted by a 12-year-old twinkie kid wielding a rolled-up PostIt."

Kelledin

bash-2.04$ kill -9 1
init: Just what do you think you're doing, Dave?

Well maybe Tom will do some testing on this to verify what was found by two different sources.

LOL too.

HAHA...Funny you won't see Everett6 and AMDsmeltsWhines post anything about this....That damn throttling is what gets me the most. I can't wait till the desktop Palomino is out and these Intel Cultists will be eating crow!

It worked yesterday!

The clock throttling shown in that article is actually software based, on the motherboard. The throttling cuts the clockspeed by less than 50% and does not occur on other motherboards. Additionally, when they use diagnostic equipment to turn on the throttling, the system clearly slows down further than the previous throttled measurements, showing that indeed, the P4 hardware throttle had not been active. These tests are showing a motherboard or BIOS bug.

-Raystonn

= The views stated herein are my personal views, and not necessarily the views of my employer. =

you should go visit Aceshardware forums and watch how these two bozos get laughed out of the internet.

Inquest.com is a huge fraud, if they were inpartial or objective then:

where are their reports of cracked athlon cores before all the copper shims were available? did ppl know this? nope.

where is the investigation on ASUS who claim that a cpu without a fan will burn up +1 hour?, test it lets see the proof!

why are they only focusing on AMD and Intel why don't they broading their horizons and include Sun, Sgi etc?.,

It doesn't help much when Van Smith publicly states that Tom Pabst does not like him one bit and would spit on the ground while mentioning his name.

that's why I like the Inquire and the Register they make fun of both and some of it is just as hard to believe, but it's fun.

"AMD/VIA...you <i>still</i> are the weakest link, good bye!"

A detailed discussion on P4's architecture problems can also be found on
http://www.emulators.com/pentium4.htm

This is one of the reasons why Intel has problems selling P4 (the other reason is it uses RDRAM exclusively).

**Spin all you want, but we the paying consumers will have the final word**

I would not be surprised if Inquest <i>is</i> a hoax. I take it with a grain of salt in any event. It's just entertaining seeing you get so distraught about it all the time.

Kelledin

bash-2.04$ kill -9 1
init: Just what do you think you're doing, Dave?

Well you are maybe right, I don't know. Seems like more testing is required by more independent and objective testers. I expect this to occur because the claims are pretty excessive if proven false and very detrimental for Intel if believed. Which if proven false will immediately make the authors credibility worthless and they will have to either take new names or stop posting information on the web. Can you show us of any evidence or testing to refute those claims?? Just saying it is bogus is not the same. These are some very serious claims, peoples jobs are at stack as well as anyones credibility also. If the Pentium IV throttles its performance due to excessive heat buildup in the core at 100% cpu usage then the days are numbered for the Pentium 4 in its current form. Just my take on the matter.

<P ID="edit"><FONT SIZE=-1><EM>Edited by noko on 05/18/01 08:50 PM.</EM></FONT></P>

If I am not mistaken, Van Smith used to work for Tomshardware.com. He wrote a series of articles on Rambus RDRAM. See e.g.
http://www4.tomshardware.com/mainb [...] index.html

His articles are usually hard-hitting but insightful. Agreeing with him or not, it is not a hoax.

**Spin all you want, but we the paying consumers will have the final word**

Hmmm yes...I knew I remembered that name from somewhere.

Kelledin

bash-2.04$ kill -9 1
init: Just what do you think you're doing, Dave?

I've had short discussions with Van Smith over email when he worked at Tom's Hardware.
I like the guy.
seems he doesn't pull punches.
our discussions were about RDRAM, he agreed it had its applications, they simply weren't with the P3 and the 820.
*gasp*
he may have been right...

----------------------
why, oh WHY, is the world run by morons?