Phenom II 6 GHz+ OVERCLOCK

[Jazzman]

Forgive me if this has been posted but I thought you would like to know. SOURCE

AMD has reportedly slated two models of its Phenom II series processors for January, 2009. According to Greek website hwbox.gr, the company will be launching two high-end processors from the Phenom II X4 series, on the 9th of January. Details of these processors available at this point in time are:

AMD Phenom II X4 940 Black Edition (HDZ940XCGIBOX), 3.00 GHz, Socket AM2+, 8 MB cache (2 MB total L2 + 6 MB L3), 125W TDP.
AMD Phenom II X4 920 (HDX920XCGIBOX), 2.80 GHz, Socket AM2+, 8 MB cache (2 MB total L2 + 6 MB L3), 125W TDP.
There is no word on the prices, at this point, though they are expected to be "much lower" than those of the Intel Core i7.

[Particle]

I don't remember who said it or where, but somebody was rambling on about "we'll see what nehalem does when Intel does LN2" in response to the good AMD results. Well, it appears that actually happened and they got:

"One quick LN2 Nehalem setup on the spot managed to get to 5.02GHz, but not for long. At 4.8GHz it lasted long enough to churn out some benchmarketing numbers."

The frequencies aren't as high, but the game is still early on both sides to try to keep things in perspective. There's no reason to go product bashing anything.

[G0ldBr1ck]

I don't remember who said it or where, but somebody was rambling on about "we'll see what nehalem does when Intel does LN2" in response to the good AMD results. Well, it appears that actually happened and they got:

"One quick LN2 Nehalem setup on the spot managed to get to 5.02GHz, but not for long. At 4.8GHz it lasted long enough to churn out some benchmarketing numbers."

The frequencies aren't as high, but the game is still early on both sides to try to keep things in perspective. There's no reason to go product bashing anything.

from INQ. so no telling for sure.

[Rammsteiner]

I'm getting teh upgrade bug already AM3 and Deneb can't come soon enough! Actually though, it taking a couple more months is a good thing, so I can get past the holidays and have enough money saved again to actually buy it

Now you say, after canceling my order Im constantly looking around to order things lol

[Particle]

While INQ does like to exaggerate things, this doesn't strike me as one of those times. They have pictures of the event that they took and the author isn't Charlie. This also isn't the kind of conspiracy theory speculation that INQ is famous for. These are some actual numbers that they themselves were there to observe first-hand.

[Throwed]

January 2009 is not so bad. Pell Grants and student loans are due then so I can use some of my Pell money to buy this CPU without actually coming out of my own pocket.

CSCI FTW!

[G0ldBr1ck]

January 2009 is not so bad. Pell Grants and student loans are due then so I can use some of my Pell money to buy this CPU without actually coming out of my own pocket.

CSCI FTW!

our tax money at work.....

[Sparky]

our tax money at work.....

Yeah, that's what came to my mind but I was keeping my mouth shut

[Jazzman]

From a "salty" site:

[Morais]

why they hidden the multiplier without hidden the bus speed?

[Frodin]

I see... so they used the infamous 29.6 multiplier for that overclock.

[AbelJemka]

Moreover 5920Mhz frequency and 200Mhz bus speed, 29.6 mutiplier?
Very fishy shot for me...

edit : Frodin speedy writer!

[Rammsteiner]

That's indeed a bit... weird

I know AMD has the possibility to actually divide the multiplier, but AFAIk highest multi you can get is 57? If not even less though

[Jazzman]

I see... so they used the infamous 29.6 multiplier for that overclock.

While I make no claims for validity of these results, please note the cpu-z version however. This version has given wrong voltage in the past. Could be there is an error with clock speed to multiplier too.

I think Informal first showed us this site. Now he thinks it's salty too.

[informal]

It's fishy but could be done with 29.5 multi and clock variance (many boards do this).But apart from being possible it is still fishy.
Thanks Jazzman,i forgot about that site .

[herderien]

looks like a 3 digit though ...even if blured

[G0ldBr1ck]

My 790fx-DQ6 has a variance and usualy sits around 203 when set to 200.

[possessed]

even so, what happened to ht link value? vacation?

edit:
yay, post 69, now I have to aim for post 666

[Booj]

In that pic it's running 5.9ghz passing Orthos?
Nah I can't believe that...

[Sparky]

even so, what happened to ht link value? vacation?

CPU-Z not reading it perhaps? I know this has happened before on new chips that it couldn't fully read. An older version did that on my 9600BE, the HT link was blank.

[AbelJemka]

While I make no claims for validity of these results, please note the cpu-z version however. This version has given wrong voltage in the past. Could be there is an error with clock speed to multiplier too.

I think Informal first showed us this site. Now he thinks it's salty too.

same version and good info in box so i doubt this screen is legit

[Final8ty]

I would think a bencher would us an up to date CPU-Z.

[G0ldBr1ck]

Ya, but I remember when everyone just knew beyond doubt that that 4ghz screen was a fake and now look........

[villa1n]

It's fishy but could be done with 29.5 multi and clock variance (many boards do this).But apart from being possible it is still fishy.
Thanks Jazzman,i forgot about that site .

Yeah, that high a multi.. plus Orthos looping is hard to buy into.

[Particle]

Clock variance can be large. I've seen my HT reference clock spike enough to cause a periodic 70MHz jump at a 15x multi. General day-to-day I get +/- 5-10MHz while running.