Overclock.net › Forums › Intel › Intel CPUs › i7 3770K Core Tuning. Reasons to not set all core multipliers equal?
New Posts  All Forums:Forum Nav:

i7 3770K Core Tuning. Reasons to not set all core multipliers equal? - Page 2

post #11 of 25
Thread Starter 
Quote:
Originally Posted by y2kcamaross View Post

I'm pretty sure you can only set core 0 the highest, so even if core 1 was your coolest core, you couldnt set it for example at 47/50/47/47, the very first core has to be the highest multiplier

Do you know this from experimenting or have your read that somewhere?

I did notice that Core 0 is my coolest core, but it's like 8+ C cooler than the max core temp at max load so I was always assuming that the thermistor was just off. Core 1 and 2 seem to be within 1/2 C of each other and then Core 3 seems to be 3/4C cooler than the max core temp.

Example: Here is a typical temp spread for my cores at max load while running Prime 95:

Core 0 @ 70C
Core 1 @ 80C
Core 2 @ 80C
Core 3 @ 77C

Anyone know what physical location Core 0 is in an i3770K? I wonder if that is the reasoning behind the temp difference or if it's just simply a thermistor being off.
Edited by lampshade9909 - 10/4/12 at 12:12pm
post #12 of 25
Quote:
Originally Posted by y2kcamaross View Post

I'm pretty sure you can only set core 0 the highest, so even if core 1 was your coolest core, you couldnt set it for example at 47/50/47/47, the very first core has to be the highest multiplier

Fine with me my Core 0 is my coolest core. smile.gif
Rig 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k Asus P8Z77-V EVGA GTX780 SC ACX Samsung DDR 3 (2 DIMMS) MV-3V4G3D/US 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 SSD WD Caviar Black Asus DVD/RW Xigmatek s-1283 HDT (Air Cooling) 
CoolingCoolingOSMonitor
Scythe Slipstream 1200RPM (x2) Antec 140mm + Antec 120mm + Xigmatek 120mm (x2) Win 7 64bit Acer S243HL bmii - 24" 
PowerCaseMouseAudio
Corsair TX750 Corsair Carbide 300R MX 518 Auzentek X-Fi Forte 7.1 
  hide details  
Reply
Rig 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k Asus P8Z77-V EVGA GTX780 SC ACX Samsung DDR 3 (2 DIMMS) MV-3V4G3D/US 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 SSD WD Caviar Black Asus DVD/RW Xigmatek s-1283 HDT (Air Cooling) 
CoolingCoolingOSMonitor
Scythe Slipstream 1200RPM (x2) Antec 140mm + Antec 120mm + Xigmatek 120mm (x2) Win 7 64bit Acer S243HL bmii - 24" 
PowerCaseMouseAudio
Corsair TX750 Corsair Carbide 300R MX 518 Auzentek X-Fi Forte 7.1 
  hide details  
Reply
post #13 of 25
Quote:
Originally Posted by lampshade9909 View Post

Do you know this from experimenting or have your read that somewhere?
I did notice that Core 0 is my coolest core, but it's like 8+ C cooler than the max core temp at max load so I was always assuming that the thermistor was just off. Core 1 and 2 seem to be within 1/2 C of each other and then Core 3 seems to be 3/4C cooler than the max core temp.
Example: Here is a typical temp spread for my cores at max load while running Prime 95:
Core 0 @ 70C
Core 1 @ 80C
Core 2 @ 80C
Core 3 @ 77C
Anyone know what physical location Core 0 is in an i3770K? I wonder if that is the reasoning behind the temp difference or if it's just simply a thermistor being off.



I know in my bios when Im trying the change the multiplier, I can't change any of them any higher than core 0
ivybeech
(21 items)
 
4K-HTPC
(12 items)
 
 
CPUMotherboardGraphicsRAM
Intel i7 6700k@4.7ghz Gigabyte z170xp-sli Nvidia GTX 1080Ti 16gb Team Xtreem DDR4 @16/18/18/37 3333mhz 
Hard DriveHard DriveHard DriveHard Drive
256gb M4 SSD boot 4TB Seagate - Raid 0 4TB Seagate - Raid 0 2TB Western Digital Green for storage 
Hard DriveHard DriveHard DriveHard Drive
5TB Seagate 3 tb Seagate  2TB Seagate 4TB of various external storage 
CoolingOSMonitorMonitor
Corsair H100 in push/pull Windows 10 pro ASUS ROG Swift PG278Q 27" OC Catleap Q270SE IPS 2560x1440@120hz 
KeyboardPowerCaseMouse
Xarmor mechanical cherry mx-brown Hale90 1000w 80+gold EVGA DG-87 Logitech G502 
CPUMotherboardGraphicsRAM
Intel i5-2400 Intel DP67BG Nvidia GTX 950 8gb Gskill Ripjaws 1600mhz ddr3 
Hard DriveOptical DriveCoolingOS
256gb samsung SSD Asus Blu-ray Drive Model BC-12B1ST/BLK/B/AS Corsair H50 push/pull Windows 7 Professional 
MonitorPowerCase
Vizio m60-c3 - 3840x2160 OCZ ZT 750w Silverstone Grandia gd-05 
  hide details  
Reply
ivybeech
(21 items)
 
4K-HTPC
(12 items)
 
 
CPUMotherboardGraphicsRAM
Intel i7 6700k@4.7ghz Gigabyte z170xp-sli Nvidia GTX 1080Ti 16gb Team Xtreem DDR4 @16/18/18/37 3333mhz 
Hard DriveHard DriveHard DriveHard Drive
256gb M4 SSD boot 4TB Seagate - Raid 0 4TB Seagate - Raid 0 2TB Western Digital Green for storage 
Hard DriveHard DriveHard DriveHard Drive
5TB Seagate 3 tb Seagate  2TB Seagate 4TB of various external storage 
CoolingOSMonitorMonitor
Corsair H100 in push/pull Windows 10 pro ASUS ROG Swift PG278Q 27" OC Catleap Q270SE IPS 2560x1440@120hz 
KeyboardPowerCaseMouse
Xarmor mechanical cherry mx-brown Hale90 1000w 80+gold EVGA DG-87 Logitech G502 
CPUMotherboardGraphicsRAM
Intel i5-2400 Intel DP67BG Nvidia GTX 950 8gb Gskill Ripjaws 1600mhz ddr3 
Hard DriveOptical DriveCoolingOS
256gb samsung SSD Asus Blu-ray Drive Model BC-12B1ST/BLK/B/AS Corsair H50 push/pull Windows 7 Professional 
MonitorPowerCase
Vizio m60-c3 - 3840x2160 OCZ ZT 750w Silverstone Grandia gd-05 
  hide details  
Reply
post #14 of 25
Quote:
Originally Posted by lampshade9909 View Post

Anyone know what physical location Core 0 is in an i3770K? I wonder if that is the reasoning behind the temp difference or if it's just simply a thermistor being off.



I think core 0 is the far left. It only has 1 face with another core and more contact with the heat spreader.
Rig 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k Asus P8Z77-V EVGA GTX780 SC ACX Samsung DDR 3 (2 DIMMS) MV-3V4G3D/US 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 SSD WD Caviar Black Asus DVD/RW Xigmatek s-1283 HDT (Air Cooling) 
CoolingCoolingOSMonitor
Scythe Slipstream 1200RPM (x2) Antec 140mm + Antec 120mm + Xigmatek 120mm (x2) Win 7 64bit Acer S243HL bmii - 24" 
PowerCaseMouseAudio
Corsair TX750 Corsair Carbide 300R MX 518 Auzentek X-Fi Forte 7.1 
  hide details  
Reply
Rig 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k Asus P8Z77-V EVGA GTX780 SC ACX Samsung DDR 3 (2 DIMMS) MV-3V4G3D/US 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 SSD WD Caviar Black Asus DVD/RW Xigmatek s-1283 HDT (Air Cooling) 
CoolingCoolingOSMonitor
Scythe Slipstream 1200RPM (x2) Antec 140mm + Antec 120mm + Xigmatek 120mm (x2) Win 7 64bit Acer S243HL bmii - 24" 
PowerCaseMouseAudio
Corsair TX750 Corsair Carbide 300R MX 518 Auzentek X-Fi Forte 7.1 
  hide details  
Reply
post #15 of 25
Thread Starter 
Great find on the photo! My temperatures make perfect sense now.
post #16 of 25
You can not pick a single core to use the 50 multiplier. That's not how Intel Core i CPUs work.

When you set a multiplier of 50 and you select 1 core, that means that when 1 core is active, the active core in the CPU will try to use the 50 multiplier. It can be any one of the 4 cores that will be active at any point in time trying to use the 50 multiplier. The CPU decides. There is no way for a user to control what individual core will go to 50.
post #17 of 25
Ok so I tried it.

I set in order from core 0-3
Multi: 50, 47, 47, 46
Left it on auto voltage for offset just to see if it works.
Ran 1 thread in Prime95. No affinity set.

Here you go



YA 5 GIGAHERTZ
You will notice that it was only 1 thread so it was passed around the cores.
Additionally the cores rotated in order... 5, then 4.7, then 4.7, then 4.6, then 5... etc... so you can't pick the cores but the different turbo settings work, they did in fact change from different values as the core load was passed around the different cores.
Rig 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k Asus P8Z77-V EVGA GTX780 SC ACX Samsung DDR 3 (2 DIMMS) MV-3V4G3D/US 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 SSD WD Caviar Black Asus DVD/RW Xigmatek s-1283 HDT (Air Cooling) 
CoolingCoolingOSMonitor
Scythe Slipstream 1200RPM (x2) Antec 140mm + Antec 120mm + Xigmatek 120mm (x2) Win 7 64bit Acer S243HL bmii - 24" 
PowerCaseMouseAudio
Corsair TX750 Corsair Carbide 300R MX 518 Auzentek X-Fi Forte 7.1 
  hide details  
Reply
Rig 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k Asus P8Z77-V EVGA GTX780 SC ACX Samsung DDR 3 (2 DIMMS) MV-3V4G3D/US 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 SSD WD Caviar Black Asus DVD/RW Xigmatek s-1283 HDT (Air Cooling) 
CoolingCoolingOSMonitor
Scythe Slipstream 1200RPM (x2) Antec 140mm + Antec 120mm + Xigmatek 120mm (x2) Win 7 64bit Acer S243HL bmii - 24" 
PowerCaseMouseAudio
Corsair TX750 Corsair Carbide 300R MX 518 Auzentek X-Fi Forte 7.1 
  hide details  
Reply
post #18 of 25
Thread Starter 
Nice test Aparition, much appreciated. Later this evening I'll adjust my cores and run some benchmarks in Prime95 and post the results. I'll do a few tests, something like 48/48/48/48, 49/48/48/48, 50/48/48/48, 50/49/48/48, etc then compare and contrast benchmark results and temps/voltages (if any significant change occurs)
Edited by lampshade9909 - 10/5/12 at 6:59am
post #19 of 25
I messed around with affinity trying to isolate Prime95 to core 0, worked ok, still moved to other cores - 5.0 then 4.7, but largely remained at 5.0.
For the short time I ran it temperatures were very good for 5.0 and ~1.385-1.4 voltage. = ~70'c.
Mind you that was with Voltage on Auto because I had no idea how it would react. Finding a VID setting was rather challenging as well for setting an offset voltage. It like 1.385 a lot though.
Rig 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k Asus P8Z77-V EVGA GTX780 SC ACX Samsung DDR 3 (2 DIMMS) MV-3V4G3D/US 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 SSD WD Caviar Black Asus DVD/RW Xigmatek s-1283 HDT (Air Cooling) 
CoolingCoolingOSMonitor
Scythe Slipstream 1200RPM (x2) Antec 140mm + Antec 120mm + Xigmatek 120mm (x2) Win 7 64bit Acer S243HL bmii - 24" 
PowerCaseMouseAudio
Corsair TX750 Corsair Carbide 300R MX 518 Auzentek X-Fi Forte 7.1 
  hide details  
Reply
Rig 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k Asus P8Z77-V EVGA GTX780 SC ACX Samsung DDR 3 (2 DIMMS) MV-3V4G3D/US 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 SSD WD Caviar Black Asus DVD/RW Xigmatek s-1283 HDT (Air Cooling) 
CoolingCoolingOSMonitor
Scythe Slipstream 1200RPM (x2) Antec 140mm + Antec 120mm + Xigmatek 120mm (x2) Win 7 64bit Acer S243HL bmii - 24" 
PowerCaseMouseAudio
Corsair TX750 Corsair Carbide 300R MX 518 Auzentek X-Fi Forte 7.1 
  hide details  
Reply
post #20 of 25
Thread Starter 
I did a benchmark in Prime 95 of the following two settings, 50-48-48-48, and 48-48-48-48. Clearly the 50-48-48-48 outperformed the 48-48-48-48 the majority of the time when number of threads were small. Notice that when only 1 thread was running it was winning by a decent margin. As the number of threads became larger, the timings become closer together.



Benchmark timings with threads 1 through 4. 50-48-48-48 on the left and 48-48-48-48 on the right.



Benchmark timings with threads 5 through 8. 50-48-48-48 on the left and 48-48-48-48 on the right.
Edited by lampshade9909 - 10/5/12 at 9:14pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › i7 3770K Core Tuning. Reasons to not set all core multipliers equal?