Phenom II 960t power consumption?

[[AKC]Odin]

Hello everyone, I have a question about the Phenom II 960t.
.
Is there anyone out there that can tell me how many watts the 960t uses after "UNLOCKING" to a 5th core? I have used CPUZ, CPUID HWmonitor, and a couple other monitors, and I get numbers that may or may not be correct.
W. = Watts
CPU-960t not unlocked 4.0 OC = TDP 107 W. & Powers package reads 116 W.
CPU-960t unlocked 5 cores 4.0 = TDP 161 W. & Powers package reads 174 W.
.
At stock 3.0 the CPU lists as a 95 W. TDP, and I understand that OC uses more power, but I haven't found any reliable information, or any concrete numbers that relate to actual power consumption after unlocking. I'm well aware that CPUID-HWmonitor bunks out on the core temps after unlocking, and that it doesn't support unlocked CPUs, "BUT" when I look at the change between stock 3.0 and OC 4.0 not unlocked the math seems to add up. Also when I unlock the 5th core on this CPU, the TDP goes from 107 to 161 watts in CPUZ regardless of whether or not I have "OVERCLOCKED" in the unlocked state or not, it reads 174 powers package in CPUID-HWmonitor. Is this information correct, am I running in the range of 174 watts after unlocking?
.
I have seen a lot of speculation that this is just an indication of what the CPU could use, but I would really like to see some accurate numbers, does anyone out there know for sure?
.
Specs
Asus 990FX Sabertooth
Phenom II 960t CPU @ 4.045 GHZ
8GB Corsair Dominator 1600 @1792 MHZ timings 7-8-7-21
750 watt Coorsair 750tx PSU 60 amp single rail
.
Clock
CPU 4.045 GHZ
CPU-VCORE @ 1.440 to 1.453 Volts
Multiplier x 18.0
Bus @ 224
CPU/NB @ 2696.8
RAM @ 1792 MHZ 1.66 volts
.
STABLE on 4 cores... 4.04 Stable on 5 cores as well..
I would really appreciate accurate numbers, i'm just a little concerned about possible VRM issues, and I know what you are going to say, "DUDE, Overclock away man..!! you have the potential with your setup", but i'm interested in real numbers.... HELP..!! LOLZ

 

[RGone]

I doubt you get a "real" answer no matter what. Noone has a real watt meter to test the cpu. Everything software is an educated guess based on calculations that are little more than a guess since the cpu is no longer the cpu it was before unlocking.

To top that off it does not matter anyway since an overclock in and of itself throws the rated TDP out the back door.

 

[[AKC]Odin]

I doubt you get a "real" answer no matter what. Noone has a real watt meter to test the cpu. Everything software is an educated guess based on calculations that are little more than a guess since the cpu is no longer the cpu it was before unlocking.

To top that off it does not matter anyway since an overclock in and of itself throws the rated TDP out the back door.

.
.
I "UNDERSTAND" all of that, and said so in my post, also I have seen real numbers on other CPUs, as far as power consumption, I was hoping someone had plugged a $30.00 meter into their rig to see what they could come up with (something closer than guessing), but it's obvious you didn't and "Gee Thanks" for the constructive reply, it was in no way helpful.... I also know every 960t "CHIP" is a little differant, so don't go there.... Does "SOMEONE ELSE" have any helpful information..?? I'm in no-way new to this, I was just hoping someone might have some actual "experience" with this CPU.

 

[RGone]

Gotcha...

For OP. Has anyone actually tested locked vs unlocked 960T for current draw with watt meter?

 

[trents]

Another issue is that locked cores sometimes are locked because the draw more power than specs call for.

 

[[AKC]Odin]

Answer Found After Testing.. Trents is right of course about the locked cores.
Phenom II 960t Power Consumption Unlocked to 5 cores, and Overclocked to 4.045 GHZ @ 1.45 V.
.
960t CPU "UNLOCKED" Power draw = 162 Watts at load 5 cores 4.045 GHZ overclocked
960t x4 "NOT UNLOCKED" Power draw = 135 Watts at load 4.045 GHZ overclocked
.
Phenom II 1100t x6 clocked to 4.2 GHZ @ 1.465 V. power draw = 159.6 Watts.
.
Ran tests both Locked and unlocked for the 960t and this is what you can most likely expect unlocking to 5 cores, I have no info for 6 cores because my 960t won't unlock the 6th core.
.
Tests were run on my Asus Saberttoh 990fx with Phenom II 960t, and an Asus Crosshair V with a Phenom II 1100t, and both sets of numbers are most likely "SPOT ON".
The 960t test was run both "LOCKED" at 4 cores, and "UNLOCKED" to 5 cores and the differances in power draw was compared, and calculated, not in my head mind you, but by Power benchmark/monitoring software... Better than guessing at this point.
.
Tested at V-CORE of 1.45 volts Load Line Calibration set to 110%.
Will test again "TOMMOROW" adjusting CPU V-CORE to 1.48 with Load Line Calibration bumped to 120%, and I suspect power draw to increase, and I may post detailed results for all of those wondering what to expect from unlocking to 5 cores.
"DISCLAIMER": Results will most likely be differant from Chip to Chip and Mobo. ALSO, these readings could be totally wrong, but this is the best anything "ANYWHERE" that I could find would read..
.
What do you guys think..?? Should people be worried about 1000MHZ overclock possibly juicing up to 162 watts..?? TDP anyone?

 

[tcclaviger]

Odin,
I am also a 960T owner. I am using a customized Rasa kit for cooling and started my batch of testing today. Thankfully Martin, from martinsliquidlab.org has done testing on my radiators, block, and the pump was tested at another site. I have tested, credible data for all of my components to compare to.

I was running some tests @ 4.3ghz, 5 cores, 1.584 volts, and came up with some weird numbers for power usage, I think due to the GPU being massively overclocked (6870 @ 1130 core / 1140 mem / 1.3 volts). I have trimmed it down so everything is back at stock speed and am going to do some testing on the chip.

Since I know the C/W of my radiators/block, and I know how much heat my pump adds to the loop, using the same fans as martin, and I know their RPM and my flow rate, I can find the CPU wattage pretty accurately.

Once I have stock testing down, I will do 4.0ghz and 4.3ghz both with 5 cores on higher voltages. Hopefully a formula of voltage vs speed vs core count emerges and we can then extrapolate heat wattage for any point in the graph.

I am sure others are curious, as there is NO way that CPUZ is right saying I'm at 161 watts pushing up to 1.6 volts @ 4.4ghz, it just seems too low. Anyway, in a couple days I will post up my spreadsheet and graphs.

 

[stickey]

This is a good read if you are interested in TDP, it helped me decide that anything over 3.4 ghz 6 core is taking a rough chance at blowing your mobo.

http://www.overclock.net/t/943109/about-vrms-mosfets-motherboard-safety-with-125w-tdp-processors

If you get a chance could you tell me how many watts it would run at 3.0Ghz 5 core? I also own this chip and was wondering whether i was burning my TDP limit.

 

[tcclaviger]

@stickey: That article is pretty spot on, just scanned it now. I actually completely agree with him that 4+1 or 3+1 boards have ABSOLUTELY NO BUSINESS pushing modern CPUs anywhere near their actual limits (not the OEM limits, the OC limits). It IS dangerous, and if you insist on doing so, at least actively cool the VRM area with a fan!


Well my initial results show only a laughable 51watt heat output by the CPU at stock settings after 20 or so hours crunching on BOINC Moowrapper (RC5 Distributed Computing Project).

Keep in mind please that these values are based upon GPUz correctly reading the voltage and amperage off my 6870, if those readings are so will the thermal loading after calculations by the cpu.

The more I conduct testing the more the CPUz "wattage" value of 161 seems to hold true!
161 @ 1.584 volts, 5 cores, 4.2ghz.

What would be helpful is if someone with an actual hex core @ 4.2ghz could disable 1 core, push the same voltage and boot to see what cpuz gives for wattage!

 

[tcclaviger]

Ever wonder why ASUS boards cost a bit more, and generally seem to run VERY well:

Over Current Protection (OCP) is something I have recently been examining. Protection features exist against VRM overheating/overloading depending on motherboard model and brand. I believe it is a crucial feature on motherboards today, because this is the function that will protect your VRMs from a catastrophic failure. This is why I have never seen ASUS boards fail even if people take a lowly 3+1 ASUS boards and try to overclock a Phenom II x6 on it; ASUS boards feature this technology, it is a part of the PWM controller design. OCP can work in various ways; one of the ways it works is it downclocks the CPU speed & voltage - via cool'n'quiet or it's own function - if the VRM temperatures are detected as too high (similar to if CPU temps are too high), until they can recuperate and lower in temperature. As a result, it can reduce performance during a full load scenario. It is also how ASUS gets away with rating a few select 3+1 phase AMD motherboards at 125W, though at times the OCP may kick in too often at load even at stock speed/stock cooler and the rating would've been slightly improper for the board (there is no 3+1 phase board ready for 125W processors). Another common way is a full board shutdown; if MOSFETs are overloaded suddenly to the point where immediate shutdown is needed for protection (i.e. beginning an OCCT run on a 3+1 power phase on a Phenom II x6 OC'ed and at 1.5V), then OCP will kick in and the board will shut down to protect itself. ASRock boards and some Gigabyte boards are known for this. Some boards will do this past a certain point. Others don't. OCN members and I have found that most recent MSI AMD boards feature NO protection of any sort against VRM failure/over current/over temperature, and this is likely why a majority of the catastrophic failures in the horror stories list are MSI boards.

From that article, it clearly demonstrates why spending 10-20 more for an ASUS vs other brands with the same feature list is completely worth it for peace of mind!

 

[trents]

I doubt you get a "real" answer no matter what. Noone has a real watt meter to test the cpu. Everything software is an educated guess based on calculations that are little more than a guess since the cpu is no longer the cpu it was before unlocking.

To top that off it does not matter anyway since an overclock in and of itself throws the rated TDP out the back door.

Not so. You can buy a watt meter on ebay for $15-$20. I've got one. All you have to do is plug it into the wall and then plug the PSU power cable into it. Load the system with cores locked and then repeat with them unlocked. The differential is how much unlocking the core or cores increases the wattage.

 

[Shubri]

Hi all, just to report my experience with 960t. I have same readings, like 107w 4 cores no matter clocked or not. 161 when i unlock 6 cores.
But i have one question, i messed a little bit on stock cooler to see what i can get from 4 cores(6 cores on stock goes way too hot) and i managed to get it stable at 3.8ghz whitout raising any voltages and with good ram specs (raised both multy and FSB).
Hitted 5000 in passmark and was happy. But what i noticed: when i bought cpu voltages on auto for vcore was 1.3v , suddenly one night cooler went crazy and i turned off PC so i can sleep, in the morning powered it up and i see that amd software is loading smth new for processor. Checked voltages and temps, everything was good except that my processor isnt on 1.3v on auto, its on 1.475v in every configuration, even when i load safe defaults.

Did I damaged procc so it now needs more juice to function? Tnx for answer, btw mobo is 970a-ud3 gigabyte

 

[trents]

If you were only running 1.3 vcore I wouldn't think you damaged the CPU unless a VRM on the motherboard malfunctioned.

Can you take the CPU voltage control off of "Auto" and set it manually?

 

[Shubri]

@trents yes i can set it up on whatever voltage i want, i tried to set it on 1.3 and it worked, stress tested and etc. Im unsure how motherboard calculate auto vcore. AS i know on same cpu models arent same vcore voltages, some of 960t would run on 1.3 while otheres need 1.45 or smth like that.

Any ideas why auto voltage went up for a lot ?

 

[trents]

Nope.

 

[keny]

Any ideas why auto voltage went up for a lot ?

This sounds like your'e settings were not bootable when you turned back on and your'e board has had to load "fail safe" settings this ramps up the vcore to ensure it will boot to your'e OS. It happens all the time when your'e settings are on the edge of stability

 

[Scout]

These Chips pull alot of watts when unlocked and overclocked please run some type of active cooling on your vrm's or they will not be happy.

Edit: I also run one of the most notorious MSI boards for VRM failure it takes 2 or 3 40mm fans on the vrm heat sink to keep it cool.