AMD "Thuban" Core (Phenom II X6) XS Overclocking Charts

[Particle]

There are way too many factors to define stable...

You aren't providing any revelations here when it comes to stress tests not proving stability. If you start over from the start of this thread, you'll see quite quickly that I early on was mandating a minimal degree of stability testing that most people refused to do. You'll also find us all collectively bickering about how to define stable. There is no reasonable method of testing that I can mandate here that actually proves absolute stability. That also isn't the point.

What I require now is a compromise between what I had originally intended to be the bare minimum proof (24 hours) since an empty chart is wholly useless. What we've got now is a good indicator of "some reasonable degree of stability with some validation via testing having been performed". The entire point of making people test is not to define settings that are known to be 100% rock solid but rather it is to weed out the extremely poor quality of what normally passes for "stable" in a post-your-clocks thread like this, which is that someone says it is with absolutely no formal bar established nor proof required.

Unless your argument is we shouldn't require as good of information as we've got now, you don't really have much of a point to make about what we've done here. The bar has been raised to a minimum level where we have a minimum quality to our results not found elsewhere. Even as it stands, you'll note that half of all people STILL can't be bothered to do even such minimal testing. Mandating perfectly accurate data from users is absolutely and utterly impractical.

Any individual contributor is welcome to provide any superior amount of testing and more accurate temps or voltages they wish. It's also rather insulting to myself and others considering the amount of work we've all put into making this information an order of magnitude more organized, standardized, and useful than what is normally seen in efforts like this. I'm not going to apologize for the excellent work done by members of this board for the sole purpose of benefiting others like yourself.

[Manicdan]

the simple answer is the honesty code, if someone had a BSOD after posting results, we take it off the sheet and request them to not be on our forum anymore, right? lol

[BeepBeep2]

Any individual contributor is welcome to provide any superior amount of testing and more accurate temps or voltages they wish. It's also rather insulting to myself and others considering the amount of work we've all put into making this information an order of magnitude more organized, standardized, and useful than what is normally seen in efforts like this. I'm not going to apologize for the excellent work done by members of this board for the sole purpose of benefiting others like yourself.

Why so angry? I'd like to make this a stimulating conversation rather than a heated arguement.
I'll happily run your 8 hours at speeds I trust the board wont kill itself at.
I'm not looking for an apology, and I never insulted your hard work. I know how hard it is to gather up this much information and put it into graphs and charts as you have.

I just would have liked something a little more accurate as far as recorded voltages go.

1.45v does not equal 1.488v with LLC on...nor does a set voltage of 1.51v equal a load voltage of 1.45v when its off. People have been posting different numbers all over the place.

I would just like it where you cant get away with saying your 8-hour "AFAIK stable" overclock only needs 1.43v when CPU-Z is reporting 1.47 or a guy's 8-hour "AFAIK stable" run set at 1.49v is only giving 1.43v at load.

[Particle]

I see. The impression I'd gotten from you was that you were saying the data was all worthless, and intentionally or not that is rather insulting considering all the work put into this project. This is especially so considering that your complaint was that we hadn't worked very hard to standardize submission data which is absolutely not true. That has been and continues to be a major focus of this thread, and I believe it to be a large part of why this effort has been so successful. If that's not the case, all the better, but that is why your comments had provoked hostility.

I've always encouraged accurate data. You'll note many people report multiple voltages (BIOS, load, idle, DMM, etc) with their submissions. I always use whichever is highest. I've modified the original post to make what I'm after more clear to new contributors, but there's a strong chance that few will ever read all the instructions so we'll see what happens.

[almighty15]

Can't get 4.2Ghz on my 1075T

Seems it's heat that stooping me as I'm at 59c when stress testing at 4.1Ghz

As for the whole different Vcore argument I'm pretty sure the OP asked people to state there LOAD voltages.

Mines 1.452v idle and 1.488v load.

[Hell Hound]

CPU Model: AMD Phenom II X6 1100T
CPU Stepping: CCBBE CB 1105MPM
CPU Frequency: 4004.7 MHz
CPU VID: 1.475/1.520 VCORE
CPU Multiplier: 16x
CPU Turbo: Disabled
CPU NB Speed: 2403.2 MHz
HT Ref Speed: 267 MHz
RAM Speed: DDR2-1066
RAM Timings: 5-6-6-18-2T
RAM Configuration: 8 GB (4 x 2 GB)
RAM VCORE: 1.75/2.16 VTT
Motherboard: GIGABYTE GA790 X UD4P
Chipset/Socket: AM3, 790X + SB750
Cooling: 2 LITER WATERBATH WATER@AMBIENT W/240MM HEATERCORE
Temps: 23C Idle / 65C Load
Operating System: Windows VISTA
32/64-Bit: 64
Stable/Suicide/Untested:STABLE 8HR PRIME BLEND

[felix_w]

Seems we have the same batch...

[Hell Hound]

Did you test w/ 8gb @ 1ghz or higher,if so what was highest NB clock.

[felix_w]

I tested, but never run over 4hrs prime. The closest i have to what you said is this one.

It wasn't tested for stability, cause i changed to 4GB Blade, but ran for 5-6 days with no BSODs or any other trouble...maybe i was lucky...

I also have this test run, but as far as i remember, i ended up with giving close to 1.375V to CPU-NB for running over 4 hours....no ss though (changed to blade, for testing )

You're having trouble increasing the NB with 8GB ram ?

[Hell Hound]

Cpu-NB/Vid and Cpu Vid won't stick so can't boot higher than 2.4ghz.

[felix_w]

Cpu-NB/Vid and Cpu Vid won't stick so can't boot higher than 2.4ghz.

What do you mean by that ?

[Hell Hound]

Settings are not applied,tried all value's and used AOD for verification.

[felix_w]

AOD has never worked for me with this processor....try another tool...

[Hell Hound]

I'm talking about booting @ these settings not windows clocking.
K10 Stat shows no change as well.Cpu NB-Vid Bios option is not working.

[felix_w]

i never mentioned windows clocking....i΄m talking about what voltages they show...perhaps the settings that u enter cause the bios to auto load previous (working) settings? I have previous (SB600) version of this board (DS4) and i can recall a similar behaviour in such cases...maybe you can report in detail your bios settings, if possible?

[Hell Hound]

Are you saying voltage resets to default.

[felix_w]

Bios Defauts, or last known working setting...my M4A79DLX when a multi/voltage is not ok (does not boot at first) reverts to normal value...
It's that 'Boot Guard' or something like that ( i don't demember if ASRock named it that way or ASUS)...or just a normal BIOS function...

Give us an example of failing CPU-NB multi/voltage that does not succeed to be recognised...

(Have you run another bios ? did it do the same ? how about other cpu ?)

[thematrixhazune]

I'm talking about booting @ these settings not windows clocking.
K10 Stat shows no change as well.Cpu NB-Vid Bios option is not working.

Have you tried saving the settings with one stick of ram in the slot. It's weird but I had a board do that to me and all I had to do was take out one ram save then insert back in.

[Hell Hound]

Haven't tried that yet.

Here is lower volts though,I started hardware monitor after prime that's why min is high.Mobo is GA-MA790X-UD4P I forgot to put MA last time sorry.Also bios and hardware monitor shows memory @ 2.14v it jumps to 2.16v under load.

CPU Model: AMD Phenom II X6 1100T
CPU Stepping: CCBBE CB 1105MPM
CPU Frequency: 4004.7 MHz
CPU VID: 1.4625/1.504 VCORE
CPU Multiplier: 16x
CPU Turbo: Disabled
CPU NB Speed: 2403.2 MHz
HT Ref Speed: 267 MHz
RAM Speed: DDR2-1066
RAM Timings: 5-6-6-18-2T
RAM Configuration: 8 GB (4 x 2 GB)
RAM VCORE: 1.75/2.16 VTT
Motherboard: GIGABYTE GA-MA790X-UD4P
Chipset/Socket: AM3, 790X + SB750
Cooling: 2 LITER WATERBATH WATER@AMBIENT W/240MM HEATERCORE
Temps: 23C Idle / 65C Load
Operating System: Windows VISTA
32/64-Bit: 64
Stable/Suicide/Untested:STABLE 8HR PRIME BLEND

[Particle]

Updated

[saint19]

Here is the max that I can get without kill the chip and just using multiplier.

CPU Model: AMD Phenom II X6 1090T BE
CPU Stepping: ----------
CPU Frequency: 4630.82 MHz
CPU vCore: 1.6V
CPU Multiplier: 23x
CPU Turbo: Disabled
CPU NB Speed: 3000 MHz
HT Ref Speed: 2000 MHz
RAM Speed: DDR3-1600
RAM Timings: 8-8-8-20-1T
RAM Configuration: 4 GB (2 x 2 GB)
RAM vDIMM: 1.65V
Motherboard: Gigabyte 890FXA-UD5 Rev. 2.0
Chipset/Socket: AM3, 890FX + SB850
Cooling: Water (EK Supreme HF + MCR320-QP + MCP655-D)
Temps: 35ºC Idle / --C Load
Operating System: Windows 7
32/64-Bit: 64
Stable/Suicide/Untested: Unstable

Link to CPU-Z validation: http://valid.canardpc.com/show_oc.php?id=1628727

[TheBreezyBB]

I'm gonna start doing my 8 hour prime stress test I'll post my results after Its done.

4 hours & a shutdown. need more testing to get the most stable Clocks I'll be posting in a week or so I guess. Getting my self G.skill sniper 8GB kit @ 1600MHz

[felix_w]

Here is the max that I can get without kill the chip and just using multiplier.

CPU Model: AMD Phenom II X6 1090T BE
CPU Stepping: ----------
CPU Frequency: 4630.82 MHz
CPU vCore: 1.6V
CPU Multiplier: 23x
CPU Turbo: Disabled
CPU NB Speed: 3000 MHz
HT Ref Speed: 2000 MHz
RAM Speed: DDR3-1600
RAM Timings: 8-8-8-20-1T
RAM Configuration: 4 GB (2 x 2 GB)
RAM vDIMM: 1.65V
Motherboard: Gigabyte 890FXA-UD5 Rev. 2.0
Chipset/Socket: AM3, 890FX + SB850
Cooling: Water (EK Supreme HF + MCR320-QP + MCP655-D)
Temps: 35ºC Idle / --C Load
Operating System: Windows 7
32/64-Bit: 64
Stable/Suicide/Untested: Unstable

Link to CPU-Z validation: http://valid.canardpc.com/show_oc.php?id=1628727

...really curious to see what you can get out of this chip stable...

[uncle john]

Cpu-NB/Vid and Cpu Vid won't stick so can't boot higher than 2.4ghz.

That's odd, Hell Hound. I'm not experiencing those issues using same board, bios, and cpu.

[Hell Hound]

Huh thread filled w/ users with the same problem on all bioses.

http://www.overclock.net/amd-motherb...0x-t-ud4p.html