Trying to tune 16 core 62xx ES

T

thinklet

Limp Gawd
Joined
Jul 4, 2012
Messages
142
I am trying to tune a new 4P with ES chips. Currently folding a 8101 @ TPF of 11:02
PPD 455k 712 watts Some readings are:
thinklet@SuperMicro5:~$ sudo tpc -l | grep "pstate 1"
core 0 pstate 1 (pb1) - En:1 VID:35 FID:12 DID:0.00 Freq:2800 VCore:1.1125
thinklet@SuperMicro5:~$ sudo tpc -l | grep "Maximum allowed"
Minimum allowed VID: 123 (0.0125V) - Maximum allowed VID 0 (1.5500V)
I am using : sudo tpc -psmax 1
These chips are supposed to be able to go past 3.2 GHz and 1.5500 vcore
Is it better to raise frequency or vcore or both?.
I would appreciate any tuning techniques to get the best performance out of these chips.:D
Best regards, Charlie
 
Last edited:
11:02 for 8101 is awesome. It's very closed to the highest score of OC'ed 4P 61xx!
Could you show me the full result of "tpc -l"?
 
Last edited:
Got it.
I think this is the world's first 4P ES Opteron 6200 rig. Could anybody confirm it?
I'm eager to know what the highest OC frequency you could get.
If you can OC it to above 3.6GHz, it would beat 4P 4650 and be the fastest 4P on earth!
 
Last edited:
quickz said:
Got it.
I think this is the world's first 4P ES Opteron 6200 rig. Could anybody confirm it?
I'm eager to know what the highest OC frequency you could get.
If you can OC it to above 3.6GHz, it would beat 4P 4650 and be the fastest 4P on earth!
Click to expand...

Thanks quickz,
After I receive the new PSU this afternoon I will try it.:D
Best regards, Charlie
 
You don't specify what motherboard you are using, but if you are going to have the vcore pushed up over ~1.25V, you will need to add heatsinks to the VRMs. I am using Enzotech MOS-C1s.

This is what I use on a 2P:

Code: 
!#/bin/bash
/usr/bin/TurionPowerControl -set core all pstate 1 frequency 3600 vcore 1.3275
/usr/bin/TurionPowerControl -htctemplimit 75
/usr/bin/TurionPowerControl -boostdisable
/usr/bin/TurionPowerControl -bst 1
/usr/bin/TurionPowerControl -nbfid 8

1. I choose PState1 as you need to change PStates in order to make any changes active
2. HTCTempLimit 75. The Supermicro mainboard will automatically shut the computer down at the critical temperature. On my processors both "high" and "critical" were set to 70. By setting critical to 75, the computer gives me an annoying alert at 70, but does not shut down.
3. We are already overclocking the processor, I don't want it doing anything which may cause instabilities.
4. Set the number of boost states to 1. By default the 6274 has 2xBoost states. Since we edited PState1 in line 1, this makes the highest non-boosted PState 1.
5. Changes the frequency of the North Bridge from 2000 to 2400.

Good luck.

I'm interested in seeing your results.

Code: 
TurionPowerControl -l
Turion Power States Optimization and Control - by blackshard - v0.425

Main processor is Family 15h (Bulldozer/Interlagos/Valencia) Processor
        Family: 0xf             Model: 0x1              Stepping: 0x0
        Extended Family: 0x15   Extended Model: 0x1
        Package Type: 0x3       BrandId: 0x0
Machine has 4 nodes
Processor has 8 cores
Processor has 7 p-states
Processor has 1 boost states

Power States table:
-- Node: 0 Core 0
core 0 pstate 0 (pb0) - Boost PState Disabled
core 0 pstate 1 (p0) - En:1 VID:18 FID:20 DID:0.00 Freq:3600 VCore:1.3250
core 0 pstate 2 (p1) - En:1 VID:43 FID:6 DID:0.00 Freq:2200 VCore:1.0125
core 0 pstate 3 (p2) - En:1 VID:45 FID:4 DID:0.00 Freq:2000 VCore:0.9875
core 0 pstate 4 (p3) - En:1 VID:47 FID:2 DID:0.00 Freq:1800 VCore:0.9625
core 0 pstate 5 (p4) - En:1 VID:48 FID:0 DID:0.00 Freq:1600 VCore:0.9500
core 0 pstate 6 (p5) - En:1 VID:50 FID:12 DID:1.00 Freq:1400 VCore:0.9250

Code: 
TurionPowerControl -spec
Turion Power States Optimization and Control - by blackshard - v0.425

Northbridge Power States table:
------ Node 0 of 3
NbVid 33 (1.1375V) NbDid 0 NbFid 8 NbCOF 2400 MHz
Northbridge Maximum frequency: no maximum frequency, unlocked NB multiplier

Not configured for 15h
Core 0 C1E CMP halt bit is disabled
Not configured for 15h
Core 1 C1E CMP halt bit is disabled
Not configured for 15h
Core 2 C1E CMP halt bit is disabled
Not configured for 15h
Core 3 C1E CMP halt bit is disabled
Not configured for 15h
Core 4 C1E CMP halt bit is disabled
Not configured for 15h
Core 5 C1E CMP halt bit is disabled
Not configured for 15h
Core 6 C1E CMP halt bit is disabled
Not configured for 15h
Core 7 C1E CMP halt bit is disabled

Voltage Regulator Slamming time register: 0
Voltage Regulator Step Up Ramp Time: 3
Voltage Regulator Step Down Ramp Time: 3
Processor PState Identifier: 0x0
Processor is using Serial VID Interface (probably Dual Plane mode)
PSI_L bit not enabled
------ Node 1 of 3
NbVid 33 (1.1375V) NbDid 0 NbFid 8 NbCOF 2400 MHz
Northbridge Maximum frequency: no maximum frequency, unlocked NB multiplier

Code: 
cat /proc/cpuinfo 
processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 21
model           : 1
model name      : AMD Eng Sample, ZS222445TGG44_32/22/2_2/16    
stepping        : 0
microcode       : 0x6000425
cpu MHz         : 2200.107
cache size      : 2048 KB
physical id     : 0
siblings        : 16
core id         : 0
cpu cores       : 8
apicid          : 32
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 13
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid amd_dcm aperfmperf pni pclmulqdq monitor ssse3 cx16 sse4_1 sse4_2 popcnt aes xsave avx lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 nodeid_msr topoext perfctr_core arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold
bogomips        : 4400.21
TLB size        : 1536 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 48 bits physical, 48 bits virtual
power management: ts ttp tm 100mhzsteps hwpstate cpb
 
I've had IL ES running for a longer while now (~4 months) -- here's what I use (top level):

Code: 
FREQ=3000
VCORE=1.1750
TurionPowerControl -boostdisable
TurionPowerControl -fo 1
TurionPowerControl -set ps 2 vcore $VCORE freq $FREQ
TurionPowerControl -set ps 1 vcore $VCORE freq $FREQ
TurionPowerControl -set ps 0 vcore $VCORE freq $FREQ
sleep 1
TurionPowerControl -fo 0

PowerNow and CPB/TurboBoost are turned off in the BIOS as well.

8101 TPF is ~10:30-~10:40 w/premium memory (reflashed).

EDIT: as quickz noticed, your performance numbers are very, very good, thinklet
 
Last edited:
tear said:
I've had IL ES running for a longer while now (~4 months) -- here's what I use (top level):

Code: 
FREQ=3000
VCORE=1.1750
TurionPowerControl -boostdisable
TurionPowerControl -fo 1
TurionPowerControl -set ps 2 vcore $VCORE freq $FREQ
TurionPowerControl -set ps 1 vcore $VCORE freq $FREQ
TurionPowerControl -set ps 0 vcore $VCORE freq $FREQ
TurionPowerControl -fo 0

PowerNow and CPB/TurboBoost are turned off in the BIOS as well.

8101 TPF is ~10:30-~10:40 w/premium memory (reflashed).
Click to expand...

EDIT: as quickz noticed, your performance numbers are very, very good, thinklet
Click to expand...

Hi tear,
Thanks for all the information, I am using Crucial DDR3 1600 8-8-8-24, is this good or do I need different memory? Best regards, Charlie
 
It's "good" as in "it's functional" so you don't really _need_ different memory.

Your memory runs at 9-9-9 DDR3-1600 (see TurionPowerControl -dram) so
there's only so much to gain by flashing (8-8-8).

At this time there's a bit more to extract from the CPUs so I'd focus on them first.

EDIT: I use G.Skill F3-12800CL6D-4GBXH (6-8-6-24 http://www.newegg.com/Product/Product.aspx?Item=N82E16820231444)
but it's ~3 times more expensive than your Crucials. You may wish to touch base with firedfly as he's looking into disposing
of his F3-12800CL7D-4GBXM (I think) sticks.
 
Last edited:
Hi Brainbug,
You must be a fan of "Starship Troopers", thanks for all the great information.
You don't specify what motherboard you are using
Click to expand...
I am using a Supermicro 4P H8QGi+-F motherboard.

I found the Enzotech MOS-C1s at Newegg and have ordered them. This is the first time I have needed a heat sink on VRM. Are any other heat sinks / cooling necessary? I should receive my new PSU today, I might have wrongly thought that the original one was bad due to my overclocking effort. Probably was overheating the motherboard. What is the power draw when folding on your unit? Memory type and configuration? Best regards, Charlie Thanks again for ALL the timely information.
 
I'm not using any other specific cooling, but the case does have good airflow.

The 2P draws approx 600W from the wall (according to the UPS LCD)

I'd go with whatever tear recommends, he by far the most knowledgeable on memory.
 
Hi, thinklet.
Looks like your tpf doesn't scale well from 2.8GHz to 3.0GHz.
Your tpf @2.8GHz (by -psmax 1) is 11:02, if it scales well the [email protected] should be ~10:18.
But you told me your real tpf @3.0GHz (by -set ps) was only 10:43, that is 4% slower than expected.
Have you found any reasons for it?
And have you got your new PSU? I'm still waiting to see if you could finally beat the 4P E5-4650!
 
Last edited:
tear said:
I've had IL ES running for a longer while now (~4 months) -- here's what I use (top level):

Code: 
FREQ=3000
VCORE=1.1750
TurionPowerControl -boostdisable
TurionPowerControl -fo 1
TurionPowerControl -set ps 2 vcore $VCORE freq $FREQ
TurionPowerControl -set ps 1 vcore $VCORE freq $FREQ
TurionPowerControl -set ps 0 vcore $VCORE freq $FREQ
sleep 1
TurionPowerControl -fo 0

PowerNow and CPB/TurboBoost are turned off in the BIOS as well.

8101 TPF is ~10:30-~10:40 w/premium memory (reflashed).

EDIT: as quickz noticed, your performance numbers are very, very good, thinklet
Click to expand...

What temperatures do you have with that settings? I get about 58 degrees using CM Evo cooler. Is there need for water cooling?
 
Last edited:
Hmmm you should not be that warm with the 212 evos, I am running MC ES chips at 3Ghz with quite a bit more voltage and running in the neighbourhood of 53 to 55 are you running push pull or just push, I am running push pull.
 
120mm Noctuas, push-pull. Stock fans except for one CPU (Nodes 4/5) -- that one has CM 212 fan on the push side.
EDIT: that's S8812, btw

Code: 
$ sudo TurionPowerControl -l | grep 0.pstate.2
core 0 pstate 2 (p0) - En:1 VID:32 FID:14 DID:0.00 NBPS:0 Freq:3000 VCore:1.1500
core 0 pstate 2 (p0) - En:1 VID:32 FID:14 DID:0.00 NBPS:0 Freq:3000 VCore:1.1500
core 0 pstate 2 (p0) - En:1 VID:30 FID:14 DID:0.00 NBPS:0 Freq:3000 VCore:1.1750
core 0 pstate 2 (p0) - En:1 VID:30 FID:14 DID:0.00 NBPS:0 Freq:3000 VCore:1.1750
core 0 pstate 2 (p0) - En:1 VID:33 FID:14 DID:0.00 NBPS:0 Freq:3000 VCore:1.1375
core 0 pstate 2 (p0) - En:1 VID:33 FID:14 DID:0.00 NBPS:0 Freq:3000 VCore:1.1375
core 0 pstate 2 (p0) - En:1 VID:30 FID:14 DID:0.00 NBPS:0 Freq:3000 VCore:1.1750
core 0 pstate 2 (p0) - En:1 VID:30 FID:14 DID:0.00 NBPS:0 Freq:3000 VCore:1.1750
$ sudo TurionPowerControl -temp
Turion Power States Optimization and Control - by blackshard - v0.43

Detected processor: Family 15h (Bulldozer/Interlagos/Valencia) Processor
Machine has 8 nodes
Processor has 8 cores
Processor has 7 p-states
Processor has 2 boost states
Processor temperature slew rate:9.0°C

Temperature table:
Node 0  C0:39   C1:39   C2:39   C3:39   C4:39   C5:39   C6:39   C7:39   
Node 1  C0:38   C1:38   C2:38   C3:38   C4:38   C5:38   C6:38   C7:38   
Node 2  C0:39   C1:39   C2:39   C3:39   C4:39   C5:39   C6:39   C7:39   
Node 3  C0:38   C1:38   C2:38   C3:38   C4:38   C5:38   C6:38   C7:38   
Node 4  C0:45   C1:45   C2:45   C3:45   C4:45   C5:45   C6:45   C7:45   
Node 5  C0:45   C1:45   C2:45   C3:45   C4:45   C5:45   C6:45   C7:45   
Node 6  C0:43   C1:43   C2:43   C3:43   C4:43   C5:43   C6:43   C7:43   
Node 7  C0:43   C1:43   C2:43   C3:43   C4:43   C5:43   C6:43   C7:43   

Done.
$
 
Last edited:
Grandpa_01 said:
Hmmm you should not be that warm with the 212 evos, I am running MC ES chips at 3Ghz with quite a bit more voltage and running in the neighbourhood of 53 to 55 are you running push pull or just push, I am running push pull.
Click to expand...

Just push on no 3 and 4 and pull on no 1 and 2. It is possible to add a fan on no 3 and 4 but not on 1 and 2. Number 1 is the hottest where air is pulled down...
 
How is the OC going thinklet?

My set has been stable at 3.0GHz at 1.1625V. At this setting, TPF with 8101 has been around 10:50-11:05, average at 11:00.

Two days ago I set to 3.1GHz, still at 1.1625V (1.175 was causing the system to crash). It's been running fine since then with TPF 10:30-10:40, average at 10:35. Nice boost I think and it's along the expected improvement. Temps range from mid 40's to low 50's, depending on the location of the CPU.

I think VRM temps are causing the system crashes and I ordered Enzo heatsinks, and I'm gonna add more fans. I think the CPUs are well capable to run at or beyond 3.2GHz. Let's see...

By the way, I'm running Crucial RAM, flashed to XMP: 1600 at 8-8-8.
 
DrFredrik said:
Just push on no 3 and 4 and pull on no 1 and 2. It is possible to add a fan on no 3 and 4 but not on 1 and 2. Number 1 is the hottest where air is pulled down...
Click to expand...
Er... what board are you using?
 
What model of ES chips are you guys using? I have a set of 1.6's that I bought here a while ago, but I haven't had a chance to get them running yet.
 
Are these IL ES that you have?

Mine are 2.2G, Rev. B0, ZS222445TGG44.
 
Your [email protected] with 8101 (11:00) is a little slow, it was just as fast as thinklet's [email protected]. And 1.175V is not a too high value for the vcore, I think it should work without problems.

theGryphon said:
How is the OC going thinklet?
My set has been stable at 3.0GHz at 1.1625V. At this setting, TPF with 8101 has been around 10:50-11:05, average at 11:00.
Two days ago I set to 3.1GHz, still at 1.1625V (1.175 was causing the system to crash). It's been running fine since then with TPF 10:30-10:40, average at 10:35. Nice boost I think and it's along the expected improvement. Temps range from mid 40's to low 50's, depending on the location of the CPU.
I think VRM temps are causing the system crashes and I ordered Enzo heatsinks, and I'm gonna add more fans. I think the CPUs are well capable to run at or beyond 3.2GHz. Let's see...
By the way, I'm running Crucial RAM, flashed to XMP: 1600 at 8-8-8.
Click to expand...
 
Posidon42 said:
What model of ES chips are you guys using? I have a set of 1.6's that I bought here a while ago, but I haven't had a chance to get them running yet.
Click to expand...

I have not been able to get my 1.6 GHz MC ES chips to work for folding. Problems with HT-retries and NB Watch Dog Error made me finally give up. I believe others have had similar problems too. If you manage to solve the problem, please let us know!

see this post for example: http://hardforum.com/showthread.php?t=1713779
 
Last edited:
tear said:
Er... what board are you using?
Click to expand...

I'm using the Supermicro H8QGL-IF+ board in a HAF 932 case. CPUs 1 and 2 are close together on that board. When the board is mounted in the case (standing up) it is difficult to solve the problem of hot air from CPU 1 heating up CPU2. I have tried adding a third fan on the right hand side blowing air inbetween 1 & 2 and using fans that pull air through the cooling flanges. This works well when the board is mounted flat (case lying down) but when standing up the CPU 1 reaches 58 degrees C. I have an extra fan below CPU 1 that blows hot air out from the case. A picture might help to explain perhaps... see this post in another forum:
http://www.overclock.net/t/1212613/...with-ppd-benchmarks-gallery/460#post_18341482

I am probably going to install water cooling for CPU 1 to get the temperatures down (that would affect the other three CPUs a bit as well).
 
DrFredrik, concerning the CPU socket spacing on the H8QGL-IF+ and cooling: Put the heatsink fan on the upper CPU (CPU2) in pull configuration and put the fan on the bottom CPU (CPU 1) in push configuration. This will allow ample space between the heatsinks to enable the top heatsink to pull in cool air blowing in from the front case fan. I have configured my H8QGL-IF+ that way and it makes a big difference.
 
As Linden explains, unidirectional flow does work better for H8QGL.

You can also take it step further and put a fan between the heatsinks and consider
upgrading the push fan (FBA12G12H1BX here) --
img_2115_4.jpg
 
tear said:
As Linden explains, unidirectional flow does work better for H8QGL.

You can also take it step further and put a fan between the heatsinks and consider
upgrading the push fan (FBA12G12H1BX here) --
Click to expand...

Thanks for your help. I will try that "triple fan wind tunnel" and see what happens.
I notice that you have no heat sinks on the VRMs in that picture. Have you measured the temperatures? (I bet that some are close to 100 deg C)
 
50 C on avg, one hotter spot at 65 C (measured with IR sensor).
 
You can also take it step further and put a fan between the heatsinks
Click to expand...
I will try that "triple fan wind tunnel" and see what happens.
Click to expand...
Sure try it, but keep in mind that such a three-fan configuration will only feed hot air through the top heatsink, sealing it off from cool air.

After you try that, try the configuration I recommended, with NO fans between the two heatsinks, which allows cooling air from the side fans/front case fan.
 
Last edited:
MCs at 3.0, no complaints :p

Nodes 0-1 -- first CPU in the path
Nodes 2-3 -- second CPU in the path

Top that!

Code: 
$ sudo TurionPowerControl -temp 
Turion Power States Optimization and Control - by blackshard - v0.43

Detected processor: Family 10h Processor
Machine has 8 nodes
Processor has 6 cores
Processor has 5 p-states
Processor has 0 boost states
Processor temperature slew rate:9.0°C

Temperature table:
Node 0	C0:50	C1:50	C2:50	C3:50	C4:50	C5:50	
Node 1	C0:50	C1:50	C2:50	C3:50	C4:50	C5:50	
Node 2	C0:51	C1:51	C2:51	C3:51	C4:51	C5:51	
Node 3	C0:50	C1:50	C2:50	C3:50	C4:50	C5:50	
Node 4	C0:52	C1:52	C2:52	C3:52	C4:52	C5:52	
Node 5	C0:51	C1:51	C2:51	C3:51	C4:51	C5:51	
Node 6	C0:49	C1:49	C2:49	C3:49	C4:49	C5:49	
Node 7	C0:49	C1:49	C2:49	C3:49	C4:49	C5:49	

Done.
$
 
It turned out that I had already both fans in the same direction (up) for CPU 1 and 2, and this gave the high temperature for CPU 1 (58 deg C). A 120 mm fan is also blowing from the side to cool the VRMs.
I tested adding another fan inbetween CPU1 and 2 - this resulted in a 5 degree lower temperature for CPU 1 (53 deg C) and CPU 2 stayed about the same.
The temperature makes sense comparing to the 50 deg C that tear got for MCs at 3.0 GHz as ILs should run a bit hotter.
 
Hi tear, could you tell me the vcore setting, the PPD (for 8101 or 8102), and the power draw number for your 4p MC rig at 3.0GHz? Thank you!
It seems 3.0GHz is not very difficult for your rig, did you ever give a try for higher frequencies?

tear said:
MCs at 3.0, no complaints :p

Nodes 0-1 -- first CPU in the path
Nodes 2-3 -- second CPU in the path

Top that!

Code: 
$ sudo TurionPowerControl -temp 
Turion Power States Optimization and Control - by blackshard - v0.43

Detected processor: Family 10h Processor
Machine has 8 nodes
Processor has 6 cores
Processor has 5 p-states
Processor has 0 boost states
Processor temperature slew rate:9.0°C

Temperature table:
Node 0	C0:50	C1:50	C2:50	C3:50	C4:50	C5:50	
Node 1	C0:50	C1:50	C2:50	C3:50	C4:50	C5:50	
Node 2	C0:51	C1:51	C2:51	C3:51	C4:51	C5:51	
Node 3	C0:50	C1:50	C2:50	C3:50	C4:50	C5:50	
Node 4	C0:52	C1:52	C2:52	C3:52	C4:52	C5:52	
Node 5	C0:51	C1:51	C2:51	C3:51	C4:51	C5:51	
Node 6	C0:49	C1:49	C2:49	C3:49	C4:49	C5:49	
Node 7	C0:49	C1:49	C2:49	C3:49	C4:49	C5:49	

Done.
$
Click to expand...
 
Last edited:
quickz said:
Your [email protected] with 8101 (11:00) is a little slow, it was just as fast as thinklet's [email protected]. And 1.175V is not a too high value for the vcore, I think it should work without problems.
Click to expand...

Yeah, I know [email protected] is a bit slow. Funny thing is it was so much better in the first WU (around 10:30), then it settled to around 11:00 in the other WUs.
It's doing much better at 3.1 GHz though. The average is settling at around 10:30.

The problem with increasing the Vcore is not that the CPUs cannot handle it. The VRM's get too hot for the task I think and the system shots hard-resets under load. I'll give it another shot with VRM heatsinks ;)
 
DrFredrik said:
It turned out that I had already both fans in the same direction (up) for CPU 1 and 2, and this gave the high temperature for CPU 1 (58 deg C). A 120 mm fan is also blowing from the side to cool the VRMs.
I tested adding another fan inbetween CPU1 and 2 - this resulted in a 5 degree lower temperature for CPU 1 (53 deg C) and CPU 2 stayed about the same.
The temperature makes sense comparing to the 50 deg C that tear got for MCs at 3.0 GHz as ILs should run a bit hotter.
Click to expand...
Still seems a bit hot to me -- my ILs in a different board: http://hardforum.com/showpost.php?p=1039246693&postcount=17 (folding an 8102) with noctuas though so not exactly apples-to-apples... but still, overall power draw of the IL system is lower (800W) than the MC system (900W), both at 3.0.

Ambient is about 70F.

Are you still running stock or OC? Can you post output of: sudo TurionPowerControl -l | grep 0.pstate.[0-2]
 
quickz said:
Hi tear, could you tell me the vcore setting, the PPD (for 8101 or 8102), and the power draw number for your 4p MC rig at 3.0GHz? Thank you!
It seems 3.0GHz is not very difficult for your rig, did you ever give a try for higher frequencies?
Click to expand...
PM sent. I feel I may have polluted this thread too much already :-(
 
Did anyone notice a strange phenomenon that the power draw number of an Opteron rig (MC or IL) is linearly dependent on the CPU temperature? And I find that every degree is corresponding to about 5 watts for a 4P rig (while the speed of all system/CPU fans remains unchanged)
Your IL rig's temperature is much lower than the MC rig, so it might be a bit unfair for the MC rig in the power draw comparison. :)

tear said:
Still seems a bit hot to me -- my ILs in a different board: http://hardforum.com/showpost.php?p=1039246693&postcount=17 (folding an 8102) with noctuas though so not exactly apples-to-apples... but still, overall power draw of the IL system is lower (800W) than the MC system (900W), both at 3.0.

Ambient is about 70F.

Are you still running stock or OC? Can you post output of: sudo TurionPowerControl -l | grep 0.pstate.[0-2]
Click to expand...
 
Yes. There is a dependency there (I never investigated it but I think someone in anandtech forum did some extensive testing in that department, backed by scientific models and all).

So yeah, it is possible that Noctuas are putting my IL chips in sort of a... 'upward spiral' (positive).
Noctua fans don't have PWM either so it would take a some effort to raise the temps here...
 
You must log in or register to reply here.
Back
Top