Overclock.net › Forums › Intel › Intel CPUs › [Official] The Sandy Stable Club **Guides, Voltages, Temps & BIOS Templates** Inc SPREADSHEET
New Posts  All Forums:Forum Nav:

[Official] The Sandy Stable Club **Guides, Voltages, Temps & BIOS Templates** Inc SPREADSHEET - Page 1031

post #10301 of 10699
Here's my 5.0ghz run
http://www.overclock.net/t/968053/official-the-sandy-stable-club-guides-voltages-temps-bios-templates-inc-spreadsheet/8510#post_17060755

Here's my 5.1ghz run
http://www.overclock.net/t/968053/official-the-sandy-stable-club-guides-voltages-temps-bios-templates-inc-spreadsheet/8780#post_17182815

@topet2k12001 & Durden
I did some testing using a lower LLC and higher offset. beyond the obvious increase in idle voltage, max load temps actually decreased even though CPU-Z reports the same voltage. I've never picked this up before but in the ASUS BIOS, description for LLC actually tells you that LLC will increase thermal.

Another thing I've observed is that at Zero LLC, voltage at max load is actually lower than the voltage at light to mid load. Perfect example of vdroop in action right there. This however is a no-go for me because my system will always be a light to mid load for everyday usage, and there's no reason to live with a higher than required voltage just because I need to boost the offset to accomodate for vdroop at max load so I can pass Prime95.

My conclusion is that my lowest acceptable LLC setting is High, any lower the load vs voltage curve will behave in an inefficient way. I haven't had time to really dive into this and do more testings I believe that at 5ghz or if the target Vcore is around 1.45v, then the optimal LLC level will be Ultra High.
Nebula
(33 items)
 
  
CPUCPUMotherboardGraphics
Intel Core i7 2700K Intrel Core i7 2700K Asus p8z68-v Pro Gen3 MSI N580GTX Lightning Xtreme 3GB Edition 
GraphicsRAMHard DriveHard Drive
MSI N580GTX Lightning Xtreme 3GB Edition G.Skill RipjawX F3-17000CL9D-8GBXM Western Digital Caviar Black 2TB 7200RPM Seagate Barracuda LP 2TB 5900RPM 
Optical DriveCoolingCoolingCooling
Pioneer BDR-206BK 12X Blu-Ray Writer Drive Koolance RP-452X2 Dual 5.25in Reservoir for 1-2... Koolance 450s Pump Koolance 450s Pump 
CoolingCoolingCoolingCooling
EK Supreme HF Bitspower Black Freezer Acrylic Top With Clear ... Bitspower Black Freezer Acrylic Top With Clear ... Black Ice GTX 240 Radiator 
CoolingCoolingCoolingCooling
Black Ice GTX 480 Radiator Koolance CTR-SPD24X2 Dual 24V Pump Controller Nidec Servo Scythe Gentle Typhoon 120mm Case Fa... Xigmatek Crystal Red LED 140mm Case Fan 1000RPM 
CoolingCoolingOSMonitor
Bitfenix Spectre PRO Red LED 230mm Case Fan 900RPM Silverstone Air-Penetrator UV-Blue 120mm Case F... Windows 7 Home Premium 64 Bits Benq XL2410T 120hz 
MonitorKeyboardPowerCase
Samsung SyncMaster 740N 17" Azio Levetron Mech 5 Mechanical Gaming Keyboard Cooler Master Silent Pro Hybrid 1300W Aerocool X-Predator Evil Black Edition 
MouseAudioOtherOther
Logitech G9X Gaming Mouse Auzentech X-Meridian 7.1 2G Lamptron FC5 V2 Fan Controller Bitspower Dual 20cm Cold Cathode Kit UV 
Other
Bitspower Dual 30cm Cold Cathode Kit UV 
  hide details  
Reply
Nebula
(33 items)
 
  
CPUCPUMotherboardGraphics
Intel Core i7 2700K Intrel Core i7 2700K Asus p8z68-v Pro Gen3 MSI N580GTX Lightning Xtreme 3GB Edition 
GraphicsRAMHard DriveHard Drive
MSI N580GTX Lightning Xtreme 3GB Edition G.Skill RipjawX F3-17000CL9D-8GBXM Western Digital Caviar Black 2TB 7200RPM Seagate Barracuda LP 2TB 5900RPM 
Optical DriveCoolingCoolingCooling
Pioneer BDR-206BK 12X Blu-Ray Writer Drive Koolance RP-452X2 Dual 5.25in Reservoir for 1-2... Koolance 450s Pump Koolance 450s Pump 
CoolingCoolingCoolingCooling
EK Supreme HF Bitspower Black Freezer Acrylic Top With Clear ... Bitspower Black Freezer Acrylic Top With Clear ... Black Ice GTX 240 Radiator 
CoolingCoolingCoolingCooling
Black Ice GTX 480 Radiator Koolance CTR-SPD24X2 Dual 24V Pump Controller Nidec Servo Scythe Gentle Typhoon 120mm Case Fa... Xigmatek Crystal Red LED 140mm Case Fan 1000RPM 
CoolingCoolingOSMonitor
Bitfenix Spectre PRO Red LED 230mm Case Fan 900RPM Silverstone Air-Penetrator UV-Blue 120mm Case F... Windows 7 Home Premium 64 Bits Benq XL2410T 120hz 
MonitorKeyboardPowerCase
Samsung SyncMaster 740N 17" Azio Levetron Mech 5 Mechanical Gaming Keyboard Cooler Master Silent Pro Hybrid 1300W Aerocool X-Predator Evil Black Edition 
MouseAudioOtherOther
Logitech G9X Gaming Mouse Auzentech X-Meridian 7.1 2G Lamptron FC5 V2 Fan Controller Bitspower Dual 20cm Cold Cathode Kit UV 
Other
Bitspower Dual 30cm Cold Cathode Kit UV 
  hide details  
Reply
post #10302 of 10699
Quote:
Originally Posted by Sashimi View Post

Here's my 5.0ghz run
http://www.overclock.net/t/968053/official-the-sandy-stable-club-guides-voltages-temps-bios-templates-inc-spreadsheet/8510#post_17060755

Here's my 5.1ghz run
http://www.overclock.net/t/968053/official-the-sandy-stable-club-guides-voltages-temps-bios-templates-inc-spreadsheet/8780#post_17182815

@topet2k12001 & Durden
I did some testing using a lower LLC and higher offset. beyond the obvious increase in idle voltage, max load temps actually decreased even though CPU-Z reports the same voltage. I've never picked this up before but in the ASUS BIOS, description for LLC actually tells you that LLC will increase thermal.

Another thing I've observed is that at Zero LLC, voltage at max load is actually lower than the voltage at light to mid load. Perfect example of vdroop in action right there. This however is a no-go for me because my system will always be a light to mid load for everyday usage, and there's no reason to live with a higher than required voltage just because I need to boost the offset to accomodate for vdroop at max load so I can pass Prime95.

My conclusion is that my lowest acceptable LLC setting is High, any lower the load vs voltage curve will behave in an inefficient way. I haven't had time to really dive into this and do more testings I believe that at 5ghz or if the target Vcore is around 1.45v, then the optimal LLC level will be Ultra High.

 

Nice...thanks for this information, Sashimi!

 

Just for my clarification so that we are on the same page, since we have different motherboards (I'm on Gigabyte and you and Durden are on Asus): "High" LLC Setting for Asus = less vdroop, yes/no?

 

For Gigabyte, what I have are levels of LLC, where: Level 1 is lowest and Level 10 is highest...as you change to a higher level, say Level 6, more "vrise" (if there is such a term) is applied.

 

Here is one article that I used as a guide when I started overclocking via manual method (I can't use LLC when using offset method in Gigabyte): http://forum.overclock3d.net/showthread.php?t=35531

 

From Sashimi's findings, what I can infer is that my observations may be somehow different because on Gigabyte, using offset method of overclocking disables LLC so I am unable to combine the two (in short, I can only do offset method at "zero LLC"). I have seen some threads where there was a "modified BIOS" for Gigabyte that enables both, but they don't seem to be available anymore (I get 404 errors when clicking on the download links). Also I am on the latest available BIOS for my Gigabyte board (version F7)...I can no longer "downgrade" my BIOS version, is that correct?


Edited by topet2k12001 - 5/17/13 at 8:14am
Switch 810 Rig
(20 items)
 
TJ07 Rig
(15 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Gigabyte GA-P67A-UD7-B3 Radeon HD 6970 Radeon HD 6970 
RAMRAMHard DriveCooling
Corsair CMP4GX3M2A 1600C9 Corsair CMP4GX3M2A 1600C9 WDC WD10EALX-009BA0 EK Supreme HF 
CoolingCoolingCoolingCooling
EK-RAM Dominator EK-FB Kit GA P67A EK-FC6970 MCP655-B 
OSOSMonitorKeyboard
MS Windows 8 64-bit Ubuntu 12.04 AOC T2242we Logitech K320 
PowerCaseMouseAudio
Corsair AX750 NZXT Switch 810 Gunmetal Edition Logitech M325 Altec Lansing 
CPUMotherboardGraphicsRAM
Intel Core i7-2700K Gigabye GA-P67A-UD7 B3 AMD HD6970 Corsair Dominator 2x2GB CL9 
Hard DriveOSMonitorKeyboard
Western Digital Caviar Blue Windows 8.1 and Ubuntu 12.04 AOC 21.5 inches Logitech K320 
PowerCaseMouseOther
Corsair AX750 Silverstone Temjin 07 Logitech M325 Wireless Mouse EK Supreme HF 
OtherOtherOther
EK FC-6970 EK-FB KIT GA P67A (UD7) EK RAM Dominator 
  hide details  
Reply
Switch 810 Rig
(20 items)
 
TJ07 Rig
(15 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Gigabyte GA-P67A-UD7-B3 Radeon HD 6970 Radeon HD 6970 
RAMRAMHard DriveCooling
Corsair CMP4GX3M2A 1600C9 Corsair CMP4GX3M2A 1600C9 WDC WD10EALX-009BA0 EK Supreme HF 
CoolingCoolingCoolingCooling
EK-RAM Dominator EK-FB Kit GA P67A EK-FC6970 MCP655-B 
OSOSMonitorKeyboard
MS Windows 8 64-bit Ubuntu 12.04 AOC T2242we Logitech K320 
PowerCaseMouseAudio
Corsair AX750 NZXT Switch 810 Gunmetal Edition Logitech M325 Altec Lansing 
CPUMotherboardGraphicsRAM
Intel Core i7-2700K Gigabye GA-P67A-UD7 B3 AMD HD6970 Corsair Dominator 2x2GB CL9 
Hard DriveOSMonitorKeyboard
Western Digital Caviar Blue Windows 8.1 and Ubuntu 12.04 AOC 21.5 inches Logitech K320 
PowerCaseMouseOther
Corsair AX750 Silverstone Temjin 07 Logitech M325 Wireless Mouse EK Supreme HF 
OtherOtherOther
EK FC-6970 EK-FB KIT GA P67A (UD7) EK RAM Dominator 
  hide details  
Reply
post #10303 of 10699
My offset is +.18 and I am running @ 102.4X46. Fast enough, stable & cool. Idle is about 1.04vc. Max is 1.46 or so.
Edited by Mule928 - 5/17/13 at 6:05pm
Beast II
(14 items)
 
  
CPUMotherboardGraphicsRAM
2500K @5.007 GIGABYTE Z68X-UD5 XFX ATI Radeon 5670 MUSHKIN REDLINE 996997@2226Mhz 9-11-10 
Hard DriveCoolingOSMonitor
SEAGATE ST95005620 THERMALRIGHT SILVER ARROW Win 8 64 Pro LG 50" Plasma 
PowerCaseMouse
Kingwin Laser 850 Azza HURRICAN MS Trackball 
  hide details  
Reply
Beast II
(14 items)
 
  
CPUMotherboardGraphicsRAM
2500K @5.007 GIGABYTE Z68X-UD5 XFX ATI Radeon 5670 MUSHKIN REDLINE 996997@2226Mhz 9-11-10 
Hard DriveCoolingOSMonitor
SEAGATE ST95005620 THERMALRIGHT SILVER ARROW Win 8 64 Pro LG 50" Plasma 
PowerCaseMouse
Kingwin Laser 850 Azza HURRICAN MS Trackball 
  hide details  
Reply
post #10304 of 10699
Quote:
Originally Posted by Sashimi View Post

Warning: Spoiler! (Click to show)
Here's my 5.0ghz run
http://www.overclock.net/t/968053/official-the-sandy-stable-club-guides-voltages-temps-bios-templates-inc-spreadsheet/8510#post_17060755

Here's my 5.1ghz run
http://www.overclock.net/t/968053/official-the-sandy-stable-club-guides-voltages-temps-bios-templates-inc-spreadsheet/8780#post_17182815

@topet2k12001 & Durden
I did some testing using a lower LLC and higher offset. beyond the obvious increase in idle voltage, max load temps actually decreased even though CPU-Z reports the same voltage. I've never picked this up before but in the ASUS BIOS, description for LLC actually tells you that LLC will increase thermal.

Another thing I've observed is that at Zero LLC, voltage at max load is actually lower than the voltage at light to mid load. Perfect example of vdroop in action right there. This however is a no-go for me because my system will always be a light to mid load for everyday usage, and there's no reason to live with a higher than required voltage just because I need to boost the offset to accomodate for vdroop at max load so I can pass Prime95.

My conclusion is that my lowest acceptable LLC setting is High, any lower the load vs voltage curve will behave in an inefficient way. I haven't had time to really dive into this and do more testings I believe that at 5ghz or if the target Vcore is around 1.45v, then the optimal LLC level will be Ultra High.

Great stuff! I've currently got my 4.5 clock using ultra high and will look to lower it and convert to offset (this weekends my first free weekend in a long time!!) will keep you posted on how it goes smile.gif

Quote:
Originally Posted by topet2k12001 View Post

Warning: Spoiler! (Click to show)
Nice...thanks for this information, Sashimi!

Just for my clarification so that we are on the same page, since we have different motherboards (I'm on Gigabyte and you and Durden are on Asus): "High" LLC Setting for Asus = less vdroop, yes/no?

For Gigabyte, what I have are levels of LLC, where: Level 1 is lowest and Level 10 is highest...as you change to a higher level, say Level 6, more "vrise" (if there is such a term) is applied.

Here is one article that I used as a guide when I started overclocking via manual method (I can't use LLC when using offset method in Gigabyte): http://forum.overclock3d.net/showthread.php?t=35531

From Sashimi's findings, what I can infer is that my observations may be somehow different because on Gigabyte, using offset method of overclocking disables LLC so I am unable to combine the two (in short, I can only do offset method at "zero LLC"). I have seen some threads where there was a "modified BIOS" for Gigabyte that enables both, but they don't seem to be available anymore (I get 404 errors when clicking on the download links). Also I am on the latest available BIOS for my Gigabyte board (version F7)...I can no longer "downgrade" my BIOS version, is that correct?
That's correct on the LLC front. High or ultra high would be around 6-8 on your board I think and would as Sashimi mentioned decrease vdroop but at the cost of higher temperatures.

On the downgrading of bios, it's not recommended to however I recently had to do a downgrade as the latest BIOS for my board caused problems! I'm sure if you search online you should find some instructions.. You basically have to force a lower BIOS onto the board and then "upgrade" to the bios you want. Do be warned however that any settings you do have saved (profiles!) will be lost in the process so take notes!
Quote:
Originally Posted by Mule928 View Post

My offset is +.18 and I am running @ 102.4X46. Fast enough, stable & cool. Idle is about 1.04vc. Max is 1.46 or so.

Nice work! What temps are you getting under load?!
post #10305 of 10699
Thread Starter 
Spreadsheet updated thumb.gif

Thank you all for participating. Sandy stable club has come quite far with just over 450 members sharing their overclocking experience and journey, I'm almost certain it has helped many and therefore I would like to take this opportunity to thank each and everyone of you for taking the time to help others and for keeping this thread alive, Sandy will never DIE!!!!!

CREATOR: gd-jpeg v1.0 (using IJG JPEG v62), quality = 100

thumb.gif
it looks alive
(17 items)
 
Sandy Transformed
(69 photos)
 
CPUMotherboardGraphicsRAM
i5 2500K [5ghz 1.44v 24/7] Asus Maximus IV Extreme SLI GTX 460 @ 920/1840/4100 [1.11v] 4GB Dominator GT @ 2133mhz CL8 [1.65v] 
Hard DriveHard DriveCoolingOS
Intel X25-M G2 80GB 3x Samsung F3 500GB RAID 0 EK Supreme HF Full Copper Windows 7 Ultimate 64bit 
MonitorPowerCaseAudio
LG42SL9500 42" 1080p [Surround Soon] XFX XXX Edition 650w Xigmatec Elysium JVC HA-RX700 [foam mod] 
  hide details  
Reply
it looks alive
(17 items)
 
Sandy Transformed
(69 photos)
 
CPUMotherboardGraphicsRAM
i5 2500K [5ghz 1.44v 24/7] Asus Maximus IV Extreme SLI GTX 460 @ 920/1840/4100 [1.11v] 4GB Dominator GT @ 2133mhz CL8 [1.65v] 
Hard DriveHard DriveCoolingOS
Intel X25-M G2 80GB 3x Samsung F3 500GB RAID 0 EK Supreme HF Full Copper Windows 7 Ultimate 64bit 
MonitorPowerCaseAudio
LG42SL9500 42" 1080p [Surround Soon] XFX XXX Edition 650w Xigmatec Elysium JVC HA-RX700 [foam mod] 
  hide details  
Reply
post #10306 of 10699

Hi Friends (esp. dedicated to Durden and Sashimi who have shared very good points for my learning),

 

Remember in my previous posts that I was collecting data by logging them while doing my stress tests? Here are some that I have, and my apologies for the delay.

 

My Overclocking Adventures and Experiences

by topet2k12001

 

Commercial: Help me gather more data to study!

 

  1. Install and run HWiNFO while you do your stress-testing (HWiNFO has a one-click button to enable logging).
  2. Drop your raw data here: http://www.mediafire.com/?ua6vcpey99d0w.
  3. PM me so that I can check my MediaFire Account and download your file.
  4. I will turn them to graphs and post them in this thread.

 

----------------------------------------------------------------------------------------------------

 

Having had this Intel Core i7-2700K for a little over a year, I have never imagined having to try out overclocking. As it is, these Sandy Bridge processors are very powerful even at "stock" (default) settings. It has "Turbo Boost Technology" which also automatically speeds up the processor on-demand, at which the default processor clocks can go up to 3.8 or 3.9 GHz. This amount of processing power is more than enough for my daily needs.

 

My interest in computer hardware started primarily with "aesthetics" (i.e. modding). At times, I got to glance upon threads related to tweaking/overclocking such as this thread. So I gave it a shot, out of curiosity. I have seen a lot of threads not only on this forum site that talks about sharing and comparing results and for most threads, the most common method is to share "proof of tweaking results" by screenshots. I have rarely seen this done in such a way that results are presented via "hard data" (for the lack of a better term).

 

Being the norm, I likewise followed the steps: run stress tests, let it finish, and post a screenshot. However, I noticed something. For example, if I am going to run a 12-hour Prime95 stress test, while at the end of the 12-hour run the software "RealTemp" will register the highest-marked temperatures, I was more interested in knowing: on the average, what are my "highest temps" and not just the "highest registered temperature?" The reason is because when I was watching Prime95 do its thing (and RealTemp registering the data), the "highest registered temperature" happened only on the 2nd hour of the entire 12-hour run. And when I looked at the raw data, this "highest registered temperature" only had one instance registered.

 

This prompted me to always run a software that collects/logs data throughout the entire 12-hour Prime95 activity, and then look at the data and sometime put them in graphs/charts so that I could visually inspect them.

 

Main Computer Components:

 

Motherboard: Gigabyte GA-P67A-UD7-B3, BIOS Version is F7 (latest available)

Processor: Intel Core i7-2700K

RAM: Corsair Dominator CMP4GX3M2A 1600C9

Video Card: 2 pcs. AMD Radeon HD 6970

Power Supply: Corsair AX750

Hard Disk: mechanical hard drive, Western Digital Caviar Blue, 1TB

 

Cooling Components:

 

Motherboard: EK-FB Kit GA P67A

Processor: EK Supreme HF (EK Supremacy is the new line of product for this)

RAM: EK RAM Domintoar (now renamed to EK-RAM Monarch)

Video Card: EK-FC6970, attached together via EK-FC Link R48X0/58X0 and EK-FC Bridge DUAL Parallel

Pump: MCP655-B (fixed speed, I believe this is Speed 4)

Pump Top/Cover: Bitspower mod kit and top

Reservoir: Bitspower Water Tank Z-Multi 150

 

Case: NZXT Switch 810 (the Gunmetal Edition)

 

Methodology:

 

  1. For data recording/logging, I use HWiNFO64. By default, HWiNFO logs data in intervals of 2000ms. You can change it if you want, but for consistency's sake I did not touch it.
  2. I still open the required apps such as CPU-Z, RealTemp, Prime95 etc. when stress testing (because I collect data while doing the 12-hour Prime95 for submissions to the Sandy Stable Club).
  3. For graphs/charts:
  • I only factored in 100% Usage data to eliminate outliers in my data (basically filter it in Excel).
  • I used the Core with the highest-registered temperature.

 

Testing Conditions:

 

  1. Non-airconditioned area.
  2. Summer months here in the Philippines.
  3. Average weather at this time is between 33-35 degrees Celsius (about 91 to 95 degrees Fahrenheit)

 

Disclaimer:

 

  1. I am not a Subject-Matter Expert in the field. I am very new to all of this. As such, I cannot claim expertise in the results that I will share.
  2. Nevertheless, the reason I am sharing this to the community is the hope that those Subject-Matter Experts can look at them and help in interpreting and validating them.
  3. I do not have enough tools and resources to claim that my methods are "scientific" enough, but what I can say is that, at least, they are data-driven.

 

So far, here is what I have accomplished for overclocking (and data-gathering):

 

  1. 4.0GHz Overclock, Manual Method
  2. 4.0GHz Overclock, Offset Method
  3. 4.4GHz Overclock, Offset Method
  4. 4.5GHz Overclock, Offset Method

 

I haven't finished putting all my data into graphs/charts, but I think it is already time to share what I have. Let's start with my 4.3GHz Overclock (sorry my data is not presented in order of Overclock attempt):

 

4.3GHz Overclock, Offset Method (-0.040V) Results

 

 

 

Without Data Labels

 

 

 

With Data Labels

 

The chart above shows:

1. For each VCore change (droop or rise), what were the temperatures registered (counting how many times they registered)?

2. I put data labels on the lowest VCore (1.212, the lables with blue background and black text) as well as the VCore with the highest number of counts (in this case, 1.236).

 

 

Data Arranged by Time

 

This chart shows the relationship of rise-and-fall of Temperatures versus the rise-and-fall of VCore.

 

This chart shows the relationship of rise-and-fall of Temperatures versus the rise-and-fall of CPU Package Power (in Watts).

 

This chart shows the relationship of rise-and-fall of VCore versus the rise-and-fall of CPU Package Power (in Watts).

 

Data Arranged/Sorted by CPU Package Power (W), Ascending Order

 

We take the last 3 charts above and arrange the data not by time, but instead in ascending or increasing order of CPU Package Power (in Watts).

 

 

 

 

Inferences:

 

Here is how I understand the data...and this is where I need help from the experts in reading and interpreting as well, since I am not an expert.

 

  1. I would read changes in temperature/VCore Voltage/Power (Watts) in this manner: "temperature increases as power (Watts) increase". The data above looks like it provides an explanation to some other threads on why people observed "lower VCore to have higher temperatures".
  2. "Drop in VCore Voltage (VDroop) is an effect or a countermeasure to protect the processor" instead of stating that "drop in VCore causes increase in temperature", since it drops as the power consumption increases (as it causes higher temperatures). In other words, it seems to be more accurate to say that the VCore drop is a reaction and not a cause (hence, most enthusiasts call VDroop as a "countermeasure").
  3. "Highest-recorded temperatures" when stress-testing (i.e. 12-hour Prime95) should be taken as-is, as does not automatically translate to telling us that this will be the temperature that we will get most often in regular usage. As seen on the first two (2) graphs, my "highest-recorded temperature" was 69 degrees Celsius, but that was for only about 1 "tick" (1 occurence, given that HWiNFO records at 2000ms intervals).
  4. Based on the first two (2) graphs, I would "statistically" be getting an average temperature of between 58 to 60 degrees Celsius on 100% load at my designed overclock, on the "hottest Core". Likewise, I would "statistically" be getting a VCore of 1.236V on 100% load at my designed overclock.
  5. At 1.224V of VCore (or when my processor drops to this VCore voltage) I would statistically be having an average temperature of between 64 to 66 degrees Celsius at 100% load on the hottest Core.
  6. Assuming long-term usage (since the data was collected via a 12-hour Prime95 stress-test), at my designed overclock, my processor would run at 1.236V of VCore for about 60% of the time, followed by 1.224V of VCore for about 30% of the time, while for the remaining 10% of the time my processor would run at 1.248V of VCore.

Edited by topet2k12001 - 5/18/13 at 9:59pm
Switch 810 Rig
(20 items)
 
TJ07 Rig
(15 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Gigabyte GA-P67A-UD7-B3 Radeon HD 6970 Radeon HD 6970 
RAMRAMHard DriveCooling
Corsair CMP4GX3M2A 1600C9 Corsair CMP4GX3M2A 1600C9 WDC WD10EALX-009BA0 EK Supreme HF 
CoolingCoolingCoolingCooling
EK-RAM Dominator EK-FB Kit GA P67A EK-FC6970 MCP655-B 
OSOSMonitorKeyboard
MS Windows 8 64-bit Ubuntu 12.04 AOC T2242we Logitech K320 
PowerCaseMouseAudio
Corsair AX750 NZXT Switch 810 Gunmetal Edition Logitech M325 Altec Lansing 
CPUMotherboardGraphicsRAM
Intel Core i7-2700K Gigabye GA-P67A-UD7 B3 AMD HD6970 Corsair Dominator 2x2GB CL9 
Hard DriveOSMonitorKeyboard
Western Digital Caviar Blue Windows 8.1 and Ubuntu 12.04 AOC 21.5 inches Logitech K320 
PowerCaseMouseOther
Corsair AX750 Silverstone Temjin 07 Logitech M325 Wireless Mouse EK Supreme HF 
OtherOtherOther
EK FC-6970 EK-FB KIT GA P67A (UD7) EK RAM Dominator 
  hide details  
Reply
Switch 810 Rig
(20 items)
 
TJ07 Rig
(15 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Gigabyte GA-P67A-UD7-B3 Radeon HD 6970 Radeon HD 6970 
RAMRAMHard DriveCooling
Corsair CMP4GX3M2A 1600C9 Corsair CMP4GX3M2A 1600C9 WDC WD10EALX-009BA0 EK Supreme HF 
CoolingCoolingCoolingCooling
EK-RAM Dominator EK-FB Kit GA P67A EK-FC6970 MCP655-B 
OSOSMonitorKeyboard
MS Windows 8 64-bit Ubuntu 12.04 AOC T2242we Logitech K320 
PowerCaseMouseAudio
Corsair AX750 NZXT Switch 810 Gunmetal Edition Logitech M325 Altec Lansing 
CPUMotherboardGraphicsRAM
Intel Core i7-2700K Gigabye GA-P67A-UD7 B3 AMD HD6970 Corsair Dominator 2x2GB CL9 
Hard DriveOSMonitorKeyboard
Western Digital Caviar Blue Windows 8.1 and Ubuntu 12.04 AOC 21.5 inches Logitech K320 
PowerCaseMouseOther
Corsair AX750 Silverstone Temjin 07 Logitech M325 Wireless Mouse EK Supreme HF 
OtherOtherOther
EK FC-6970 EK-FB KIT GA P67A (UD7) EK RAM Dominator 
  hide details  
Reply
post #10307 of 10699
Nice read. My overclocking journey had been one that was results orientated and I admit I have not really dived into the electronics science of the PC much. having said that, based on your data, seems what is happening is that vdroop effective decreases voltage while maintaing CPU wattage by increasing the amp. Since voltage is what creates temp and kills the chip, it is essentially a good thing. To take this further if CPU's total power is what determines stability, vdroop should actually lower the voltage requirement for a certain clock, which in turn improves thermal and longevity of the chip.


One problem I encounter in high OC is, vdroop only kicks in at high cpu load. At light to medium load, voltage actually jumped higher than when the chip is at max load and that really is not desirable. To me Vdroop seems to produce good results in prime95 but not ideal for everyday usage, not for offset overclocking at high clocks anyway.

It appears vdroop would be better suited for manual OC. Offset OC I believe it's still optimal to hv a certain level of LLC active.
Nebula
(33 items)
 
  
CPUCPUMotherboardGraphics
Intel Core i7 2700K Intrel Core i7 2700K Asus p8z68-v Pro Gen3 MSI N580GTX Lightning Xtreme 3GB Edition 
GraphicsRAMHard DriveHard Drive
MSI N580GTX Lightning Xtreme 3GB Edition G.Skill RipjawX F3-17000CL9D-8GBXM Western Digital Caviar Black 2TB 7200RPM Seagate Barracuda LP 2TB 5900RPM 
Optical DriveCoolingCoolingCooling
Pioneer BDR-206BK 12X Blu-Ray Writer Drive Koolance RP-452X2 Dual 5.25in Reservoir for 1-2... Koolance 450s Pump Koolance 450s Pump 
CoolingCoolingCoolingCooling
EK Supreme HF Bitspower Black Freezer Acrylic Top With Clear ... Bitspower Black Freezer Acrylic Top With Clear ... Black Ice GTX 240 Radiator 
CoolingCoolingCoolingCooling
Black Ice GTX 480 Radiator Koolance CTR-SPD24X2 Dual 24V Pump Controller Nidec Servo Scythe Gentle Typhoon 120mm Case Fa... Xigmatek Crystal Red LED 140mm Case Fan 1000RPM 
CoolingCoolingOSMonitor
Bitfenix Spectre PRO Red LED 230mm Case Fan 900RPM Silverstone Air-Penetrator UV-Blue 120mm Case F... Windows 7 Home Premium 64 Bits Benq XL2410T 120hz 
MonitorKeyboardPowerCase
Samsung SyncMaster 740N 17" Azio Levetron Mech 5 Mechanical Gaming Keyboard Cooler Master Silent Pro Hybrid 1300W Aerocool X-Predator Evil Black Edition 
MouseAudioOtherOther
Logitech G9X Gaming Mouse Auzentech X-Meridian 7.1 2G Lamptron FC5 V2 Fan Controller Bitspower Dual 20cm Cold Cathode Kit UV 
Other
Bitspower Dual 30cm Cold Cathode Kit UV 
  hide details  
Reply
Nebula
(33 items)
 
  
CPUCPUMotherboardGraphics
Intel Core i7 2700K Intrel Core i7 2700K Asus p8z68-v Pro Gen3 MSI N580GTX Lightning Xtreme 3GB Edition 
GraphicsRAMHard DriveHard Drive
MSI N580GTX Lightning Xtreme 3GB Edition G.Skill RipjawX F3-17000CL9D-8GBXM Western Digital Caviar Black 2TB 7200RPM Seagate Barracuda LP 2TB 5900RPM 
Optical DriveCoolingCoolingCooling
Pioneer BDR-206BK 12X Blu-Ray Writer Drive Koolance RP-452X2 Dual 5.25in Reservoir for 1-2... Koolance 450s Pump Koolance 450s Pump 
CoolingCoolingCoolingCooling
EK Supreme HF Bitspower Black Freezer Acrylic Top With Clear ... Bitspower Black Freezer Acrylic Top With Clear ... Black Ice GTX 240 Radiator 
CoolingCoolingCoolingCooling
Black Ice GTX 480 Radiator Koolance CTR-SPD24X2 Dual 24V Pump Controller Nidec Servo Scythe Gentle Typhoon 120mm Case Fa... Xigmatek Crystal Red LED 140mm Case Fan 1000RPM 
CoolingCoolingOSMonitor
Bitfenix Spectre PRO Red LED 230mm Case Fan 900RPM Silverstone Air-Penetrator UV-Blue 120mm Case F... Windows 7 Home Premium 64 Bits Benq XL2410T 120hz 
MonitorKeyboardPowerCase
Samsung SyncMaster 740N 17" Azio Levetron Mech 5 Mechanical Gaming Keyboard Cooler Master Silent Pro Hybrid 1300W Aerocool X-Predator Evil Black Edition 
MouseAudioOtherOther
Logitech G9X Gaming Mouse Auzentech X-Meridian 7.1 2G Lamptron FC5 V2 Fan Controller Bitspower Dual 20cm Cold Cathode Kit UV 
Other
Bitspower Dual 30cm Cold Cathode Kit UV 
  hide details  
Reply
post #10308 of 10699
Quote:
Originally Posted by Sashimi View Post

Nice read. My overclocking journey had been one that was results orientated and I admit I have not really dived into the electronics science of the PC much. having said that, based on your data, seems what is happening is that vdroop effective decreases voltage while maintaing CPU wattage by increasing the amp. Since voltage is what creates temp and kills the chip, it is essentially a good thing. To take this further if CPU's total power is what determines stability, vdroop should actually lower the voltage requirement for a certain clock, which in turn improves thermal and longevity of the chip.


One problem I encounter in high OC is, vdroop only kicks in at high cpu load. At light to medium load, voltage actually jumped higher than when the chip is at max load and that really is not desirable. To me Vdroop seems to produce good results in prime95 but not ideal for everyday usage, not for offset overclocking at high clocks anyway.

It appears vdroop would be better suited for manual OC. Offset OC I believe it's still optimal to hv a certain level of LLC active.

 

Thanks for your appreciation, Sashimi. :) Honestly, what I have done is something I can't even qualify as scientific (I think it was far from it)...it was more of just an attempt to arrive at a certain understanding using data that is collected. But I appreciate it. :) Your insights gave me confidence to carry on.

 

Unfortunately for my case, I am limited in providing more insights due to:

 

  1. My motherboard: as I have mentioned in previous posts, I can only use LLC if I am on Manual OC method (static/locked VCore even at idle). Conversely, if I am on Offset OC method (VCore scales down at idle), LLC settings are disabled so I am unable to combine the usage of LLC with Offset OC method.
  2. The climate in my location: very hot and humid! As such, I am currently afraid to try going beyond 4.5GHz. Um...maybe I'll try moving my computer to the bedroom where there is air-conditioning (if I get my wife to agree, hehehe...).
  3. Element of time: trying out different overclock settings can be time-consuming. :D

 

...which is the reason I am calling for help from the community. If I can get people to install HWiNFO (32 or 64 bit version, depending on their OS) to record/log their data and upload it to my MediaFire (link indicated in my previous post), I would much appreciate it. :)

Switch 810 Rig
(20 items)
 
TJ07 Rig
(15 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Gigabyte GA-P67A-UD7-B3 Radeon HD 6970 Radeon HD 6970 
RAMRAMHard DriveCooling
Corsair CMP4GX3M2A 1600C9 Corsair CMP4GX3M2A 1600C9 WDC WD10EALX-009BA0 EK Supreme HF 
CoolingCoolingCoolingCooling
EK-RAM Dominator EK-FB Kit GA P67A EK-FC6970 MCP655-B 
OSOSMonitorKeyboard
MS Windows 8 64-bit Ubuntu 12.04 AOC T2242we Logitech K320 
PowerCaseMouseAudio
Corsair AX750 NZXT Switch 810 Gunmetal Edition Logitech M325 Altec Lansing 
CPUMotherboardGraphicsRAM
Intel Core i7-2700K Gigabye GA-P67A-UD7 B3 AMD HD6970 Corsair Dominator 2x2GB CL9 
Hard DriveOSMonitorKeyboard
Western Digital Caviar Blue Windows 8.1 and Ubuntu 12.04 AOC 21.5 inches Logitech K320 
PowerCaseMouseOther
Corsair AX750 Silverstone Temjin 07 Logitech M325 Wireless Mouse EK Supreme HF 
OtherOtherOther
EK FC-6970 EK-FB KIT GA P67A (UD7) EK RAM Dominator 
  hide details  
Reply
Switch 810 Rig
(20 items)
 
TJ07 Rig
(15 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Gigabyte GA-P67A-UD7-B3 Radeon HD 6970 Radeon HD 6970 
RAMRAMHard DriveCooling
Corsair CMP4GX3M2A 1600C9 Corsair CMP4GX3M2A 1600C9 WDC WD10EALX-009BA0 EK Supreme HF 
CoolingCoolingCoolingCooling
EK-RAM Dominator EK-FB Kit GA P67A EK-FC6970 MCP655-B 
OSOSMonitorKeyboard
MS Windows 8 64-bit Ubuntu 12.04 AOC T2242we Logitech K320 
PowerCaseMouseAudio
Corsair AX750 NZXT Switch 810 Gunmetal Edition Logitech M325 Altec Lansing 
CPUMotherboardGraphicsRAM
Intel Core i7-2700K Gigabye GA-P67A-UD7 B3 AMD HD6970 Corsair Dominator 2x2GB CL9 
Hard DriveOSMonitorKeyboard
Western Digital Caviar Blue Windows 8.1 and Ubuntu 12.04 AOC 21.5 inches Logitech K320 
PowerCaseMouseOther
Corsair AX750 Silverstone Temjin 07 Logitech M325 Wireless Mouse EK Supreme HF 
OtherOtherOther
EK FC-6970 EK-FB KIT GA P67A (UD7) EK RAM Dominator 
  hide details  
Reply
post #10309 of 10699
Quote:
Originally Posted by topet2k12001 View Post

Warning: Spoiler! (Click to show)
Thanks for your appreciation, Sashimi. smile.gif Honestly, what I have done is something I can't even qualify as scientific (I think it was far from it)...it was more of just an attempt to arrive at a certain understanding using data that is collected. But I appreciate it. smile.gif Your insights gave me confidence to carry on.

Unfortunately for my case, I am limited in providing more insights due to:
  1. My motherboard: as I have mentioned in previous posts, I can only use LLC if I am on Manual OC method (static/locked VCore even at idle). Conversely, if I am on Offset OC method (VCore scales down at idle), LLC settings are disabled so I am unable to combine the usage of LLC with Offset OC method.
  2. The climate in my location: very hot and humid! As such, I am currently afraid to try going beyond 4.5GHz. Um...maybe I'll try moving my computer to the bedroom where there is air-conditioning (if I get my wife to agree, hehehe...).
  3. Element of time: trying out different overclock settings can be time-consuming. biggrin.gif

...which is the reason I am calling for help from the community. If I can get people to install HWiNFO (32 or 64 bit version, depending on their OS) to record/log their data and upload it to my MediaFire (link indicated in my previous post), I would much appreciate it. smile.gif
Great read my friend! I promised you some data and will try and find a time when my PC is not in use by me or my GF soon as I'd like to see the difference between your offset mode to my manual mode settings. I think that'd make an interesting comparison! smile.gif
post #10310 of 10699
Quote:
Originally Posted by Durden View Post


Great read my friend! I promised you some data and will try and find a time when my PC is not in use by me or my GF soon as I'd like to see the difference between your offset mode to my manual mode settings. I think that'd make an interesting comparison! smile.gif

 

That...is exactly what I have been doing. :) I have been overclocking on different clock frequencies and comparing them. This is the reason why I am encouraging forum members to use HWiNFO (I'm not an endorser, it's just that with HWiNFO all the data is in a single Excel file).

 

So far the only data that I have with both Offset and Manual Overclock Methods is my initial overclock attempt: the 4.0GHz overclock. I only get to do this on weekends, so imagine how much time it will take because I will only get to run Prime95 once in 12 hours, lol...

 

Data that I have with me through my own stress testing are:

 

1. 4.0GHz, Offset OC

2. 4.0GHz, Manual OC

3. 4.3GHz. Offset OC (the one in my post)

4. 4.4GHz, Offset OC (one of my official entries to this club, I think it's at page 1025)

5. 4.5GHz, Offset OC (one of my official entries to this club, I think it's at page 1026)

 

EDIT: I suddenly realized...if we compare our results (even with the same overclocks) we will most likely have different temperatures because of differences in our locations. That would be the biggest variable. But still please do if you can collect data, that will be much helpful. :)


Edited by topet2k12001 - 5/19/13 at 11:25am
Switch 810 Rig
(20 items)
 
TJ07 Rig
(15 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Gigabyte GA-P67A-UD7-B3 Radeon HD 6970 Radeon HD 6970 
RAMRAMHard DriveCooling
Corsair CMP4GX3M2A 1600C9 Corsair CMP4GX3M2A 1600C9 WDC WD10EALX-009BA0 EK Supreme HF 
CoolingCoolingCoolingCooling
EK-RAM Dominator EK-FB Kit GA P67A EK-FC6970 MCP655-B 
OSOSMonitorKeyboard
MS Windows 8 64-bit Ubuntu 12.04 AOC T2242we Logitech K320 
PowerCaseMouseAudio
Corsair AX750 NZXT Switch 810 Gunmetal Edition Logitech M325 Altec Lansing 
CPUMotherboardGraphicsRAM
Intel Core i7-2700K Gigabye GA-P67A-UD7 B3 AMD HD6970 Corsair Dominator 2x2GB CL9 
Hard DriveOSMonitorKeyboard
Western Digital Caviar Blue Windows 8.1 and Ubuntu 12.04 AOC 21.5 inches Logitech K320 
PowerCaseMouseOther
Corsair AX750 Silverstone Temjin 07 Logitech M325 Wireless Mouse EK Supreme HF 
OtherOtherOther
EK FC-6970 EK-FB KIT GA P67A (UD7) EK RAM Dominator 
  hide details  
Reply
Switch 810 Rig
(20 items)
 
TJ07 Rig
(15 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Gigabyte GA-P67A-UD7-B3 Radeon HD 6970 Radeon HD 6970 
RAMRAMHard DriveCooling
Corsair CMP4GX3M2A 1600C9 Corsair CMP4GX3M2A 1600C9 WDC WD10EALX-009BA0 EK Supreme HF 
CoolingCoolingCoolingCooling
EK-RAM Dominator EK-FB Kit GA P67A EK-FC6970 MCP655-B 
OSOSMonitorKeyboard
MS Windows 8 64-bit Ubuntu 12.04 AOC T2242we Logitech K320 
PowerCaseMouseAudio
Corsair AX750 NZXT Switch 810 Gunmetal Edition Logitech M325 Altec Lansing 
CPUMotherboardGraphicsRAM
Intel Core i7-2700K Gigabye GA-P67A-UD7 B3 AMD HD6970 Corsair Dominator 2x2GB CL9 
Hard DriveOSMonitorKeyboard
Western Digital Caviar Blue Windows 8.1 and Ubuntu 12.04 AOC 21.5 inches Logitech K320 
PowerCaseMouseOther
Corsair AX750 Silverstone Temjin 07 Logitech M325 Wireless Mouse EK Supreme HF 
OtherOtherOther
EK FC-6970 EK-FB KIT GA P67A (UD7) EK RAM Dominator 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
  • [Official] The Sandy Stable Club **Guides, Voltages, Temps & BIOS Templates** Inc SPREADSHEET
Overclock.net › Forums › Intel › Intel CPUs › [Official] The Sandy Stable Club **Guides, Voltages, Temps & BIOS Templates** Inc SPREADSHEET