Toms new article on asus springdale pat

Freezermug · Jun 3, 2003

It appears Toms hardware has taken a further look into how asus is gaining canterwood like speeds with thier springdale. May want to take it with a grain of salt. It IS toms hardware after all.

What I gathered from the article ( I could be wrong) is
Apparently, Asus does NOT enable pat on thier springdale. You cant, as its a hardware thing. However, Asus has found a work around from being locked out of pat. It essentionally tricks the board and goes around to achieve the same result as pat.

So, the numbers are not fake in any way. They simply found a new way to achieve the same result.

Kind of like someone saying, this exit off the highway is closed, go to the next exit. Then the driver saying screw that, and jumping over the medium and grass to get to the bottom of the ramp instead.

JKeefe · Jun 3, 2003

Article here, BTW.

Freezermug said:
What I gathered from the article ( I could be wrong) is
Apparently, Asus does NOT enable pat on thier springdale. You cant, as its a hardware thing. However, Asus has found a work around from being locked out of pat. It essentionally tricks the board and goes around to achieve the same result as pat.

Not quite; PAT is enabled on the P4P800. From THG:

Note that with the 875P chipset (Canterwood), PAT is only activated with a CPU with FSB 800 is detected by the Northbridge of the chipset. This involves the signal BSEL [1:0]. ... However, this function is deactivated with the 865 chipset (Springdale). Asus found a sort of "back door" for this. Although the state [1:0] is blocked on the 865 chipset (Springdale), Asus has still found out that the fast memory access PAT is kept open for the states [0:0] and [1:0], or FSB 400 and FSB 533 respectively. ... Because the BSEL signals are normally used for controlling the clock generator as well, Asus came up with a trick: with the help of the Super I/O unit, which is integrated on all boards of the P4P800 series, the BSEL signals are intercepted and manipulated. This allows the clock generator to be programmed with the correct FSB clock frequency of the CPU as before, while the chipset is led to believe that a CPU with FSB 533 is involved.

I for one think that is damn cool.

I'm sure Intel does not agree.

So, if you beleive Tom's (they've done some real work on this, so I would), PAT is enabled on the P4P800.

Freezermug · Jun 4, 2003

Thanks, It is alot more clear now.

That tiny quote is easier to understand then the whole article. Just whats revelent

JKeefe · Jun 4, 2003

Hmmmmmm. Tom's pulled the article...

Something's up.

Ec]-[oMaN · Jun 4, 2003

JKeefe said:
Hmmmmmm. Tom's pulled the article...

Something's up.

yeah cause for once TOM told the truth and a little too much lol,many people are trying to figure what out what he said and put it to use

JKeefe · Jun 4, 2003

Article is back up, slightly edited.

The good part is still true:

the manufacturer has succeeded in activating the PAT optimization with the Intel 865PE chipset. However, after Intel intervened, Asus is no longer allowed to call it "PAT" and now calls it "Hyper Path" instead.

Big news also for Abit:

Update, June 4, 2003: a day after this article was posted, the BIOS version 1.3 Beta for the Abit IS7-G, a rival product, arrived. The initial tests show that Abit has made the same leap in performance and is now at the same level as Asus. Thus, the frame rates in Quake 3 have increased from 366 to 418 fps - this may be a sign that Abit has also activated PAT.

martindemon · Jun 5, 2003

There is something i fear here. The p4p800 is very tempting for me but in the article, they say that intel test their chipsets and only those not passing certain tests are used to make springdale chipsets. So by enabling PAT on failed tested chipsets, it's a little bit like overclocking... I wonder about long term stability... I need a Very stable system for scientific simulations...

JKeefe · Jun 5, 2003

If you are running scientific simulations you should probably be running a Canterwood board with ECC memory - then your PAT problem is solved, because PAT is validated on all 875 boards.

bmxers21 · Jun 5, 2003

man that almost makes me wanna go buy it.

martindemon · Jun 5, 2003

JKeefe said:
If you are running scientific simulations you should probably be running a Centerwood board with ECC memory - then your PAT problem is solved, because PAT is validated on all 875 boards.

I want to use dual channel OCZ ddr400 memory (it is ordered...); i don't think i need to be that stable, with ECC memory.
Yes there is PAT on the p4c800 but no real good sata raid 0. I know i'm crazy to want stability and security and still run a raid 0 array but i will be using 2 raptors; i figured that if they have a 5 years warranty, it should mean that they are long lasting...
I could always buy a p4c800 and only one raptor but i want to try raid with these great little beasts

JKeefe · Jun 5, 2003

Well, if you're concerned, you have the option of not enabling it...

I know i'm crazy to want stability and security and still run a raid 0 array but i will be using 2 raptors; i figured that if they have a 5 years warranty, it should mean that they are long lasting...

Yes, you are crazy.

Seriously, the warranty doesn't mean the drives will last - all it means is that WD will replace them for free if they break (which may or may not happen, and probably has a lower chance of happening than an ATA drive). But, by the time they've broken, you're SOL.

martindemon · Jun 5, 2003

JKeefe said:
Well, if you're concerned, you have the option of not enabling it...

Yes, you are crazy.

Seriously, the warranty doesn't mean the drives will last - all it means is that WD will replace them for free if they break (which may or may not happen, and probably has a lower chance of happening than an ATA drive). But, by the time they've broken, you're SOL. [/B]

You know i just sent back my two 80gig western digital hard drives for repair because they both died. I was in raid0 with them at the beginning. When the first bagan to die, i recognised the symptoms and quickly backuped (without powering off the computer!) Then it promptly died. I put the other one alone and it died (clic clik clickkk...) two months later. Now i'm buying two more WD drives

I AM crazy. But i know that it was just bad luck and that WD usually make good drives and that my experiences are statistically insignificant. Now must wait for those hungry raptors to come my way

Dan87951 · Jun 11, 2003

My buddy just bought a WD 120GB and WD80GB from Circuit not to long ago, both drives are clicking already... wonder if WD made a bad batch run as my other buddy just go a 80GB and his is clicking as well. They need to pay more attention to quality in my opinion! Luck of the draw I guesS??

Toms new article on asus springdale pat

Freezermug

Member

JKeefe

Member

Freezermug

Member

JKeefe

Member

Ec]-[oMaN

Member

JKeefe

Member

martindemon

Member

JKeefe

Member

bmxers21

Registered

martindemon

Member

JKeefe

Member

martindemon

Member

Dan87951

Registered

Similar threads