Could B3 Phenom be the rebirth of AMD....+1000MHZ OC done!

[DoubleZero]

3125MHZ was fully stable at default vcore, no pics as im pushing 3250 now with a small bump in voltage.

Didn't saw that coming...

[wiak]

Loser777

That bench is in XP, I ran 2x ultra's on the same build of vista64 with the same video driver...how can my score be glitched?

Also I was running as close to stock FSB as I could to remain apples to apples, HTT makes no difference on Phenom so running 250 or 200 is no different.

funny, how low can you put HTT until it affects scores?

[Tony]

FSB on Intel and HTT on Phenom, I ran 340 to remain as close to 333FSB (stock) to keep cache speed near stock on the Intel. On the AMD HTT makes no difference to cache at all so running 200 or 300 would give the same score.

Im confused by what you are asking here?

[Cooper]

Thanks man
Unlike Intel 45nm, you can push high volts into these on air for a bit without trouble if you can keep the temp. below 60C - guide states 1.55v as maximum tolerance by core, so I assume 1.7v for 30 minutes won't be too bad

Trust me, I was at a wall with 1.44v - until I pushed 1.6v into it
Drops to 1.512v load anyway.

SocketMan, thanks for the links! Latest "errata" and other register info such as PVI codes will all be there, updated

Got my TRUE today morning. Still need to sand that thing
1.7V on AIR

@CooperCan you take a picture of the cpu when you change the coolers pls?thx

JAAFB AA 0810 APMW
Last 4 digits of SN 0063


I just flashed the DFI board with 331 BIOS and it refuses to post with 9850
Triple-core B3 works fine though. i don't get it

[davidletterboyz]

Wow...tripple core B3! Hope u can test that as well.

[Dagalidis]

Triple-core B3 works fine though. i don't get it

We want to test this baby too....

[Blacklash]

The only "glitch" in the thread is someone comparing an XP score to Vista x64. The CPU score for 06 under Vista will be lower. There is no user error on Tony's part. At 3.8GHz I lose up to 480pnts between the two in the CPU portion of 06 alone, that's WinXP SP2 vs Vista x64.

[justapost]

What have you done!?!
You've tasted the forbidden fruit of spi

He He, check the screenshot at least it's an intel chipset.

[SocketMan]

He He, check the screenshot at least it's an intel chipset.

Yea thats why the time is so high probably,
virtualization?
***************
Thanks Cooper

[justapost]

Yea thats why the time is so high probably,
virtualization?

:Yepp: Trying XEN 4.1 here. At 2.6GHz I get around 1000MB/s thruput in winrar with both the 9600BE and the 9850BE. That is a little odd. As if they apply the linux tlb workaround no matter what stepping the CPU has.
Had the 9850BE with it's stock cooler in an GBT780G mobo with 4x1GB G.Skills 6400HK's. Configured it to warn me if the core temps get over 80°C, that was an anoying benchmarking at 2.8/2.2GHz 1.325/1.3V with all that noisy beeps. I did not use thermal compound, just mounted a cleaned stock cooler, but i guess that will not bring temps in resonable regions if they are above 80°. Ocing that thing in an htpc will be tricky.
After the bench i had to clear the cmos to get the system bootup again.

[YukonTrooper]

Amd Is Back!!!!!

3.5Ghz at 1.5V?

[Mechromancer]

:Yepp: Trying XEN 4.1 here. At 2.6GHz I get around 1000MB/s thruput in winrar with both the 9600BE and the 9850BE. That is a little odd. As if they apply the linux tlb workaround no matter what stepping the CPU has.
Had the 9850BE with it's stock cooler in an GBT780G mobo with 4x1GB G.Skills 6400HK's. Configured it to warn me if the core temps get over 80°C, that was an anoying benchmarking at 2.8/2.2GHz 1.325/1.3V with all that noisy beeps. I did not use thermal compound, just mounted a cleaned stock cooler, but i guess that will not bring temps in resonable regions if they are above 80°. Ocing that thing in an htpc will be tricky.
After the bench i had to clear the cmos to get the system bootup again.

I was under the impression that no thermal compound = fiery death of a CPU. Is no thermal paste common around here?

[justapost]

I was under the impression that no thermal compound = fiery death of a CPU. Is no thermal paste common around here?

Don't think it's common.Was just husslin over here and only ran 30min benches.

[ownage]

3.5Ghz at 1.5V?

What is wrong with 1.5v?

[THE JEW (RaVeN)]

You can run cpu's without thermal compound--it's just not recommended. I do it all the time when testing parts. However, once I'm building the system for a more long term timeline, I use the paste. Using thermal compound will give you better thermal transfer.

[Mechromancer]

What is wrong with 1.5v?

If you're not joking then read this:

Electromigration is whats wrong. Voltage is like vodka, have a little bit and the party gets started, have a lot and something dies.
http://en.wikipedia.org/wiki/Electromigration Do a Ctrl-F search for "overclocking" on that page and read the explanation.

[ownage]

If you're not joking then read this:

Electromigration is whats wrong. Voltage is like vodka, have a little bit and the party gets started, have a lot and something dies.
http://en.wikipedia.org/wiki/Electromigration Do a Ctrl-F search for "overclocking" on that page and read the explanation.

LOL, that is old stuff man, everybody knows that. I don't have experience with Phenoms, but is 1.5v really that much for a Phenom?

[Calmatory]

No. CPUs last for decades, if not more. I have never witnessed a dead CPU, other than one which I broke with overvoltage (Duron 1.8GHz @ 2.2GHz @ 2.2V instead of 1.75 what it was supposed to get, stock cooling. ). They hardly fail if they seem stable with some voltage. Sure Phenom @ 2V @ LN2 won't last for years 24/7, but long enough for 99% of people.

Basically if it runs stable, don't worry. Don't worry about the temps either, ran my A64X2 @ +108C for both cores, multiple times. Still kicking and living well w/ OC's ofc.

[BrowncoatGR]

LOL, that is old stuff man, everybody knows that. I don't have experience with Phenoms, but is 1.5v really that much for a Phenom?

It's alot but iirc Phenom is rated for up to 1.55V.

[GenTarkin]

hey wiak, Im interested in getting the same DS5 board you just got. Did you try OC'in the 9850 yet? Cuz Im looking at that very combo and was wondering how its treating you. Also, how many phases is the DS5 is its PWM high enough quality to drive overclocking the 9850?


Thanks

[ownage]

It's alot but iirc Phenom is rated for up to 1.55V.

Doesn't surprise me. Phenom is 65nm. Conroe is 65nm and is also rated for up to 1.55v, so i guess Kentsfield also. If you're cooling is sufficient enough then why bother about 1.5v.

[Brother Esau]

If you have deep pockets the desire to blow up then......High Voltage is right for you!

Look at the Sexy Bananas^^^^^^^^^^

[BrowncoatGR]

Doesn't surprise me. Phenom is 65nm. Conroe is 65nm and is also rated for up to 1.55v, so i guess Kentsfield also. If you're cooling is sufficient enough then why bother about 1.5v.

Well because electromigration is real.

[knopflerbruce]

Well because electromigration is real.

How can 1.5v be so bad for daily use, when benchers use 2v on 45nm for hours and hours? If 1.5v would kill a chip in less than a couple of years, stock voltage would kill it within about 5 years I guess, and that's pretty insane.

[Calmatory]

I got 8 years old Duron 800, @ 996 MHz @ 2.01V for what, well over 3 years about 24/7, with load temps well over 90C. Still alive, though nowadays @ 900MHz @ 1.75V.

If they feel stable, they won't fail.