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 789

post #7881 of 10702
Quote:
Originally Posted by samwiches View Post

Just selecting custom produces that behavior? If you don't change anything in the boxes, it's running with the settings of whichever test was selected previously (maybe was on small FFT?)

Yeah, the default custom (8k-4096k) with 90% Ram usage creates a lower Vcore in CPU-Z than a standard blend. It's resulted in me having to increase the offset higher in order to compensate for the difference.
Ragnarok
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5-2500k Asus P8Z68-V PRO/GEN3 EVGA GeForce GTX 680 G.Skill Ares Series 8GB (2x4GB) 9-9-9-24 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 Series 256GB SSD Western Digital Caviar Black 1TB 7200 RPM Samsung SH-B123L Blu-Ray/DVD Burner/CD Burner XSPC RS240 Watercooling 
OSMonitorPowerCase
Windows 7 Home Premium 64-Bit Samsung LN55C630K 55" 1080p LCD HDTV Seasonic X-750 NZXT Phantom Black/Orange 
Audio
Samsung HW-D450ZA Soundbar 
  hide details  
Reply
Ragnarok
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5-2500k Asus P8Z68-V PRO/GEN3 EVGA GeForce GTX 680 G.Skill Ares Series 8GB (2x4GB) 9-9-9-24 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 Series 256GB SSD Western Digital Caviar Black 1TB 7200 RPM Samsung SH-B123L Blu-Ray/DVD Burner/CD Burner XSPC RS240 Watercooling 
OSMonitorPowerCase
Windows 7 Home Premium 64-Bit Samsung LN55C630K 55" 1080p LCD HDTV Seasonic X-750 NZXT Phantom Black/Orange 
Audio
Samsung HW-D450ZA Soundbar 
  hide details  
Reply
post #7882 of 10702
Make sure that's 90% of available memory--not total memory, not free memory.

Or just keep watching it. Temps won't max out for at least an hour, maybe two. I never catch the moment when it happens but it probably has something to do with vcore.
COMPUTER
(16 items)
 
  
CPUMotherboardGraphicsRAM
3570K Gigabyte GA-Z77X-UD3H Gigabyte GTX 670 Windforce OC 2GB Samsung 30nm 
Hard DriveHard DriveCoolingOS
WD Caviar Blue WD Caviar Black NZXT Havik 140 Windows 7 64bit 
OSMonitorKeyboardPower
Windows 8 64bit ASUS GD235HZ (GD245HQ) 120Hz Dell OCZ GameXStream 850W 
CaseMouseMouse PadAudio
Corsair Carbide 300R Logitech G9 naked Allsop Black Cloth (low profile) Logitech G35 Surround 
  hide details  
Reply
COMPUTER
(16 items)
 
  
CPUMotherboardGraphicsRAM
3570K Gigabyte GA-Z77X-UD3H Gigabyte GTX 670 Windforce OC 2GB Samsung 30nm 
Hard DriveHard DriveCoolingOS
WD Caviar Blue WD Caviar Black NZXT Havik 140 Windows 7 64bit 
OSMonitorKeyboardPower
Windows 8 64bit ASUS GD235HZ (GD245HQ) 120Hz Dell OCZ GameXStream 850W 
CaseMouseMouse PadAudio
Corsair Carbide 300R Logitech G9 naked Allsop Black Cloth (low profile) Logitech G35 Surround 
  hide details  
Reply
post #7883 of 10702
What version of Prime95 should we use? Version 27 is supposed to make use of AVX instructions for SNB...but in my experience it stresses harder than version 25.x (i.e., higher temps and load vcore using the same settings), and it doesn't seem like a coincidence that the lowest-voltage OCs for each multiplier don't use version 27. The most I could find on it in this thread was a comment on version 26.x
http://www.overclock.net/t/968053/official-the-sandy-stable-club-guides-voltages-temps-bios-templates-inc-spreadsheet/5940

So, my questions are:

1). I should test with version 25.x if I want to compete for the lowest stable vcore for a given multiplier, correct?

2). Does using 27.4 give me a better test of stability? It takes my load vcore a little higher, so it at least gives a more accurate reading of how high that might go (OCCT 4.1.0 also gives a higher load vcore than p95 25.x).

Thanks for the input.
post #7884 of 10702
Last time i checked 26.6 was the latest NON-beta version of Prime95 so i'd suggest you use this one...if you additionally want to test your system using the AVX instructions then do a few runs with Linx AVX (or similar apps, 20 runs for me is more than enough to double check stability in conjunction with Prime)...
Edited by fommof - 3/20/12 at 12:25am
post #7885 of 10702
Similar apps like OCCT 4.1.0 wink.gif
COMPUTER
(16 items)
 
  
CPUMotherboardGraphicsRAM
3570K Gigabyte GA-Z77X-UD3H Gigabyte GTX 670 Windforce OC 2GB Samsung 30nm 
Hard DriveHard DriveCoolingOS
WD Caviar Blue WD Caviar Black NZXT Havik 140 Windows 7 64bit 
OSMonitorKeyboardPower
Windows 8 64bit ASUS GD235HZ (GD245HQ) 120Hz Dell OCZ GameXStream 850W 
CaseMouseMouse PadAudio
Corsair Carbide 300R Logitech G9 naked Allsop Black Cloth (low profile) Logitech G35 Surround 
  hide details  
Reply
COMPUTER
(16 items)
 
  
CPUMotherboardGraphicsRAM
3570K Gigabyte GA-Z77X-UD3H Gigabyte GTX 670 Windforce OC 2GB Samsung 30nm 
Hard DriveHard DriveCoolingOS
WD Caviar Blue WD Caviar Black NZXT Havik 140 Windows 7 64bit 
OSMonitorKeyboardPower
Windows 8 64bit ASUS GD235HZ (GD245HQ) 120Hz Dell OCZ GameXStream 850W 
CaseMouseMouse PadAudio
Corsair Carbide 300R Logitech G9 naked Allsop Black Cloth (low profile) Logitech G35 Surround 
  hide details  
Reply
post #7886 of 10702
Quote:
Originally Posted by fuloran1 View Post

Get to work!

lol I doubt he gets paid to update the spreadsheet.
post #7887 of 10702
I'm getting 0X124's on idle and coming out of load. LLC is set to Level 1. Any ideas?
XGR
(15 items)
 
  
CPUMotherboardGraphicsRAM
i5 2500K 4.8Ghz 1.368v ASUS P8P67 Deluxe MSI NVIDIA GeForce GTX 480 840/1680 G.Skill RipJaw X 8GB 
Hard DriveOptical DriveCoolingCooling
Patriot Wildfire 120GB LG 20x DVD-R/RW Ultra 120 eXtreme RT Rev. C H60 
OSMonitorKeyboardPower
Win7 x64 Ultimate SP1 32" LG S-IPS 1080P TV Eclipse II HX850 
CaseMouseMouse Pad
Corsair D800 DeathAdder Supermat 
  hide details  
Reply
XGR
(15 items)
 
  
CPUMotherboardGraphicsRAM
i5 2500K 4.8Ghz 1.368v ASUS P8P67 Deluxe MSI NVIDIA GeForce GTX 480 840/1680 G.Skill RipJaw X 8GB 
Hard DriveOptical DriveCoolingCooling
Patriot Wildfire 120GB LG 20x DVD-R/RW Ultra 120 eXtreme RT Rev. C H60 
OSMonitorKeyboardPower
Win7 x64 Ultimate SP1 32" LG S-IPS 1080P TV Eclipse II HX850 
CaseMouseMouse Pad
Corsair D800 DeathAdder Supermat 
  hide details  
Reply
post #7888 of 10702
Quote:
Originally Posted by ElDictator View Post

What version of Prime95 should we use? Version 27 is supposed to make use of AVX instructions for SNB...but in my experience it stresses harder than version 25.x (i.e., higher temps and load vcore using the same settings), and it doesn't seem like a coincidence that the lowest-voltage OCs for each multiplier don't use version 27. The most I could find on it in this thread was a comment on version 26.x
http://www.overclock.net/t/968053/official-the-sandy-stable-club-guides-voltages-temps-bios-templates-inc-spreadsheet/5940

So, my questions are:

1). I should test with version 25.x if I want to compete for the lowest stable vcore for a given multiplier, correct?

2). Does using 27.4 give me a better test of stability? It takes my load vcore a little higher, so it at least gives a more accurate reading of how high that might go (OCCT 4.1.0 also gives a higher load vcore than p95 25.x).

Thanks for the input.


You pose a great question and one I'm sure will be debated as we do with the question "What's best Prime or IBT/LinX for stability". Huge can of worms aside I'll try to answer you question as I see it.

 

First let's get a couple things out of the way.

  1. The Intel 2600k (sandy bridge) supports SSE4 & AVX instruction sets. see here
  2. Prime95 27.1 - 27.4 betas & IBT 2.51 both are fully optimized to exploit AVX instruction set

 

Given these two facts it would be counter productive to find stability any other way with a Sandy Bridge or Ivy Bridge CPU. Otherwise you are leaving out what separates them for previous architectures. If lower vcore or any other voltage adjustment is what blows your hair back that's fine, heck that's great, but I liken not using the AVX instruction to not using max memory at the same time we prime or linpack. It simply does you little good unless of course you must use them to qualify for a club here in these forums in which case you have no choice but to follow the rules.

 

Ask anyone, who gets and completes work for the Prime servers, if AVX produces results and you will be amazed to find a 25% speed increase in terms of how fast you can complete the work. Sandy Bridge AVX rocks! Nuff said.

 

Now in regards to Q1 I'm not sure I understand what you are asking but if you are competing to show proof that your processor is capable of using the lowest vcore for a given multi then I would say yes use a previous version of prime as the current 27.4 beta would not help in this regard. It mostly certainly adds loading & overhead to complete the same task.

 

For Q2 the answer is yes 27.4 would be a better test for stability for a Sandy Bridge or Ivy Bridge processor IMHO. Further more I see IBT & LinX as overkill and in no way repesent day to day system usage but, and I satand by my previous statements, there is no better test for testing your cooling or burning in TIM. In other words it has a rightful place in your toolbox. Just like we use stress test and benchmark apps for our GPUs P95 is not the be all to end all.

 

All that being said P95 27.4 is a beta therefore all bugs may not be worked out at this time but it is the 4th iteration and problem seems to be pointing to AMD and there implementation and not so much with Sandy Bridge. Still if you have trouble interperting errors consult us here or over in Prime Land forums and there will be a learmning curve which should be expected with a beta.

 

A brief look at the release note helps prove my point.

 

Quote:

 

New features in Version 27.3 of prime95.exe

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

 

1)  64-bit optimizations for AVX-capable computers.  Single core benchmarks on Intel Sandy

    Bridge computers should see a decent speed increase.  However, running on 4 cores

    runs into main memory bandwidth bottleneck.  Sandy Bridge-E CPUs with 4 memory channels

    should avoid this problem.

 

Prime95 beta version 27.4 is available.

This version fixes a number of bugs in 27.3 - mostly related to LLR and PFGW issues. There are no speed improvements over 27.3. Hopefully, Bulldozer processors now properly run SSE2 FFTs.

Unless you have a Sandy Bridge CPU, there is little reason to upgrade to this version.

 

New features in Version 27.1 of prime95.exe

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

 

1)  32-bit FFTs optimized for AVX-capable computers.  Intel Sandy Bridge computers should

    see a 25% speed increase.

 

New features in Version 26.6 of prime95.exe

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

1)  For rare cases where the program cannot figure out the number of cores and hyperthreading,
    the NumPhysicalCores option may help.  See undoc.txt.
2)  Faster FFT implementations are now selected for Core 2 CPUs with 1MB L2 cache or less
    (marketed under the Celeron and Pentium label).
3)  New, slightly higher, trial factoring breakeven points.

 

 

Finally I want to revisit a subject I've touched on here in these forums before. When I last mentioned it takes at least 18 hour to pass all available test using P95 blend, I should have mentioned, that was for version 26.6 using the standard 8k to 4096k iterations which produces 70 separate test. At the 18th hour I mentioned is the time to start looking to see if you've begun repeating iteration over again. If you use the default 15 min. per iteration * 70, do the math, it would take a minimum of 17.5 hours to complete all test.

 

With 27.4 (w/AVX optimizations) this changes a bit P95 blend adds 15 new test and deletes 3 for 82 total test or a minimum of 20.5 hrs. to complete all test. Using either version I would not consider my self stable without running the program for those time periods. Even at that, and for what ever reason, the actual time to complete each iteration can and will vary. It's been my observation that on average it takes at least 16.65 minutes when 15 minute iterations are selected, meaning it would take 19.5 & 22.75 hrs. minimum respectivly.

 

For Prime junkies ver. 27.4 adds 18, 21, 25, 864, 36, 60, 100, 120, 140, 35, 50, 168, 200, 280, & 4000K then deletes 56, 14, & 1680k iterations please correct me if I'm wrong.

 

 

P95 26.6 SB 70 test loop (Click to show)

 

  • [Feb 22 08:13] Self-test 640K passed!
  • [Feb 22 08:14] Self-test 8K passed!
  • [Feb 22 08:15] Self-test 720K passed!
  • [Feb 22 08:16] Self-test 12K passed!
  • [Feb 22 08:17] Self-test 800K passed!
  • [Feb 22 08:18] Self-test 16K passed!
  • [Feb 22 08:19] Self-test 960K passed!
  • [Feb 22 08:20] Self-test 24K passed!
  • [Feb 22 08:21] Self-test 1120K passed!
  • [Feb 22 08:22] Self-test 32K passed!
  • [Feb 22 08:24] Self-test 1200K passed!
  • [Feb 22 08:25] Self-test 48K passed!
  • [Feb 22 08:26] Self-test 1344K passed!
  • [Feb 22 08:27] Self-test 64K passed!
  • [Feb 22 08:28] Self-test 1536K passed!
  • [Feb 22 08:29] Self-test 80K passed!
  • [Feb 22 08:30] Self-test 1680K passed!
  • [Feb 22 08:31] Self-test 96K passed!
  • [Feb 22 08:32] Self-test 1792K passed!
  • [Feb 22 08:33] Self-test 128K passed!
  • [Feb 22 08:34] Self-test 2048K passed!
  • [Feb 22 08:35] Self-test 160K passed!
  • [Feb 22 08:36] Self-test 2304K passed!
  • [Feb 22 08:38] Self-test 224K passed!
  • [Feb 22 08:39] Self-test 2560K passed!
  • [Feb 22 08:40] Self-test 256K passed!
  • [Feb 22 08:41] Self-test 2800K passed!
  • [Feb 22 08:42] Self-test 320K passed!
  • [Feb 22 08:43] Self-test 3072K passed!
  • [Feb 22 08:44] Self-test 384K passed!
  • [Feb 22 08:45] Self-test 3360K passed!
  • [Feb 22 08:46] Self-test 448K passed!
  • [Feb 22 08:48] Self-test 3584K passed!
  • [Feb 22 08:49] Self-test 512K passed!
  • [Feb 22 08:50] Self-test 576K passed!
  • [Feb 22 08:51] Self-test 672K passed!
  • [Feb 22 08:52] Self-test 10K passed!
  • [Feb 22 08:53] Self-test 768K passed!
  • [Feb 22 08:54] Self-test 14K passed!
  • [Feb 22 08:56] Self-test 896K passed!
  • [Feb 22 08:57] Self-test 20K passed!
  • [Feb 22 08:58] Self-test 1024K passed!
  • [Feb 22 08:59] Self-test 28K passed!
  • [Feb 22 09:00] Self-test 1152K passed!
  • [Feb 22 09:01] Self-test 40K passed!
  • [Feb 22 09:02] Self-test 1280K passed!
  • [Feb 22 09:03] Self-test 56K passed!
  • [Feb 22 09:04] Self-test 1440K passed!
  • [Feb 22 09:06] Self-test 72K passed!
  • [Feb 22 09:07] Self-test 1600K passed!
  • [Feb 22 09:09] Self-test 84K passed!
  • [Feb 22 09:10] Self-test 1728K passed!
  • [Feb 22 09:11] Self-test 112K passed!
  • [Feb 22 09:12] Self-test 1920K passed!
  • [Feb 22 09:13] Self-test 144K passed!
  • [Feb 22 09:14] Self-test 2240K passed!
  • [Feb 22 09:15] Self-test 192K passed!
  • [Feb 22 09:16] Self-test 2400K passed!
  • [Feb 22 09:17] Self-test 240K passed!
  • [Feb 22 09:18] Self-test 2688K passed!
  • [Feb 22 09:20] Self-test 288K passed!
  • [Feb 22 09:21] Self-test 2880K passed!
  • [Feb 22 09:23] Self-test 336K passed!
  • [Feb 22 09:24] Self-test 3200K passed!
  • [Feb 22 09:25] Self-test 400K passed!
  • [Feb 22 09:26] Self-test 3456K passed!
  • [Feb 22 09:28] Self-test 480K passed!
  • [Feb 22 09:29] Self-test 3840K passed!
  • [Feb 22 09:30] Self-test 560K passed!
  • [Feb 22 09:31] Self-test 4096K passed!

 

 

P95 27.4 beta SB 83 test loop (Click to show)

 

  • [Mar 20 08:03] Self-test 448K passed!
  • [Mar 20 08:04] Self-test 8K passed!
  • [Mar 20 08:05] Self-test 512K passed!
  • [Mar 20 08:06] Self-test 12K passed!
  • [Mar 20 08:08] Self-test 576K passed!
  • [Mar 20 08:09] Self-test 18K passed!
  • [Mar 20 08:10] Self-test 672K passed!
  • [Mar 20 08:11] Self-test 21K passed!
  • [Mar 20 08:12] Self-test 768K passed!
  • [Mar 20 08:13] Self-test 25K passed!
  • [Mar 20 08:14] Self-test 864K passed!
  • [Mar 20 08:15] Self-test 32K passed!
  • [Mar 20 08:16] Self-test 960K passed!
  • [Mar 20 08:17] Self-test 36K passed!
  • [Mar 20 08:18] Self-test 1120K passed!
  • [Mar 20 08:19] Self-test 48K passed!
  • [Mar 20 08:21] Self-test 1200K passed!
  • [Mar 20 08:22] Self-test 60K passed!
  • [Mar 20 08:23] Self-test 1344K passed!
  • [Mar 20 08:24] Self-test 72K passed!
  • [Mar 20 08:25] Self-test 1536K passed!
  • [Mar 20 08:26] Self-test 84K passed!
  • [Mar 20 08:27] Self-test 1728K passed!
  • [Mar 20 08:29] Self-test 100K passed!
  • [Mar 20 08:30] Self-test 1920K passed!
  • [Mar 20 08:31] Self-test 120K passed!
  • [Mar 20 08:32] Self-test 2240K passed!
  • [Mar 20 08:33] Self-test 140K passed!
  • [Mar 20 08:34] Self-test 2400K passed!
  • [Mar 20 08:35] Self-test 160K passed!
  • [Mar 20 08:36] Self-test 2688K passed!
  • [Mar 20 08:38] Self-test 192K passed!
  • [Mar 20 08:39] Self-test 2880K passed!
  • [Mar 20 08:40] Self-test 224K passed!
  • [Mar 20 08:41] Self-test 3200K passed!
  • [Mar 20 08:42] Self-test 256K passed!
  • [Mar 20 08:43] Self-test 3456K passed!
  • [Mar 20 08:44] Self-test 288K passed!
  • [Mar 20 08:45] Self-test 3840K passed!
  • [Mar 20 08:46] Self-test 336K passed!
  • [Mar 20 08:48] Self-test 400K passed!
  • [Mar 20 08:49] Self-test 480K passed!
  • [Mar 20 08:50] Self-test 10K passed!
  • [Mar 20 08:51] Self-test 560K passed!
  • [Mar 20 08:52] Self-test 16K passed!
  • [Mar 20 08:53] Self-test 640K passed!
  • [Mar 20 08:54] Self-test 20K passed!
  • [Mar 20 08:55] Self-test 720K passed!
  • [Mar 20 08:57] Self-test 24K passed!
  • [Mar 20 08:58] Self-test 800K passed!
  • [Mar 20 08:59] Self-test 28K passed!
  • [Mar 20 09:00] Self-test 896K passed!
  • [Mar 20 09:01] Self-test 35K passed!
  • [Mar 20 09:02] Self-test 1024K passed!
  • [Mar 20 09:03] Self-test 40K passed!
  • [Mar 20 09:04] Self-test 1152K passed!
  • [Mar 20 09:05] Self-test 50K passed!
  • [Mar 20 09:06] Self-test 1280K passed!
  • [Mar 20 09:07] Self-test 64K passed!
  • [Mar 20 09:08] Self-test 1440K passed!
  • [Mar 20 09:10] Self-test 80K passed!
  • [Mar 20 09:11] Self-test 1600K passed!
  • [Mar 20 09:12] Self-test 96K passed!
  • [Mar 20 09:13] Self-test 1792K passed!                                                                                                               
  • [Mar 20 09:14] Self-test 112K passed!
  • [Mar 20 09:15] Self-test 2048K passed!
  • [Mar 20 09:17] Self-test 128K passed!
  • [Mar 20 09:18] Self-test 2304K passed!
  • [Mar 20 09:19] Self-test 144K passed!
  • [Mar 20 09:20] Self-test 2560K passed!
  • [Mar 20 09:21] Self-test 168K passed!
  • [Mar 20 09:22] Self-test 2800K passed!
  • [Mar 20 09:24] Self-test 200K passed!
  • [Mar 20 09:25] Self-test 3072K passed!
  • [Mar 20 09:26] Self-test 240K passed!
  • [Mar 20 09:27] Self-test 3360K passed!
  • [Mar 20 09:28] Self-test 280K passed!
  • [Mar 20 09:29] Self-test 3584K passed!
  • [Mar 20 09:31] Self-test 320K passed!
  • [Mar 20 09:32] Self-test 4000K passed!
  • [Mar 20 09:33] Self-test 384K passed!
  • [Mar 20 09:34] Self-test 4096K passed!
  • [Mar 20 09:35] Self-test 448K passed!

 

 


Edited by owcraftsman - 3/20/12 at 4:11pm
Luxe 1
(33 items)
 
Enthoo Primo
(28 items)
 
 
CPUMotherboardGraphicsRAM
i7 7700k MSI Z270 Gaming M7 Asus GTX 1080Ti FE G.SKILL 32GB (4 x 8GB) Ripjaws V Series DDR4 PC... 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 950 Pro M.2 WDC WD1003FZEX-00MK2A0 Seagate Backup+ Desk SCSI Disk Device asus bc-12b1st 
CoolingCoolingCoolingCooling
XSPC Raystorm Pro corsair sp120 high performance led tygon E-1000 1/2 ID x 3/4 OD plasticizer free kryographics Pascal for NVIDIA TITAN X and GTX ... 
CoolingCoolingCoolingCooling
Mayhem Pastel White UV Monsoon center compression fittings  Alphacool NexXxoS XT45 Full Copper 280mm radiator  Alphacool NexXxoS UT60 Full Copper 240mm radiator  
CoolingCoolingCoolingOS
Monsoon Series Two D5 Premium Dual Bay Reservoi... Phanteks PH-F140SP Case Fan Aquacomputer backplate for kryographics Pascal ... Windows 10 Pro x64 
MonitorMonitorKeyboardPower
Asus PG278Q Gsync (Primary) Asus PB278Q (Extended) Corsair Gaming K70 LUX RGB Mechanical Gaming Ke... CORSAIR HX Series HX750 750W ATX12V 2.3 / EPS12... 
CaseMouseMouse PadAudio
Phanteks Enthoo Luxe PH-ES614L  Func MS-3 rev2 Razer exactMat X speed Creative Sound Blaster Zx 116dB PCIe Gaming Sou... 
AudioAudioOtherOther
Logitech Z906 5.1 Surround Sound Speaker System... HyperX Cloud II Gaming Headset Darkside dimable LED strips  Modsmart Kobra Cables 
Other
EVGA PowerLink 
CPUMotherboardGraphicsGraphics
i7 4790k @ 4.8 MSI Z97 Gaming 7  Saphire R9 290x (hynix) Saphire R9 290x (Hynix) 
RAMHard DriveHard DriveHard Drive
GSkill Ripjaws Z F3-2400C10D-8GZH 2x Samsung 850 Pro Raid-0  AMD RAMDisk WD1001FALS (Data/Backup) 
Optical DriveCoolingOSMonitor
Asus BC-12B1ST BlueRay Custom Water Cooled Windows 10 Pro x64 Asus PG279Q 
MonitorKeyboardPowerCase
Asus PG278Q Corsair K70 EVGA 1000w G2 Phanteks Enthoo Primo 
MouseMouse PadAudioAudio
FUNC MS-3 rev2  XTRAC PADS Ripper Optical Mouse pad Kingston HyperX Cloud II  Creative Sound Blaster Z PCIe 116dB SNR Gaming ... 
OtherOtherOtherOther
Alphacool NexXxoS UT60 Full Copper 420mm EK Supremacy EVO Alphacool NexXxoS Monsta 360 Radiator EK-FC R9-290X Waterblock - Nickel Acetal (Rev.2.0) 
OtherOtherOtherOther
EK-FC R9-290X backplate Bitspower Dual / Single D5 Top Upgrade Kit 150 ... Bitspower D5 MOD TOP (Black "S" Model) PrimoFlex Advanced LRT 10ft Tubing - 1/2in. ID ... 
  hide details  
Reply
Luxe 1
(33 items)
 
Enthoo Primo
(28 items)
 
 
CPUMotherboardGraphicsRAM
i7 7700k MSI Z270 Gaming M7 Asus GTX 1080Ti FE G.SKILL 32GB (4 x 8GB) Ripjaws V Series DDR4 PC... 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 950 Pro M.2 WDC WD1003FZEX-00MK2A0 Seagate Backup+ Desk SCSI Disk Device asus bc-12b1st 
CoolingCoolingCoolingCooling
XSPC Raystorm Pro corsair sp120 high performance led tygon E-1000 1/2 ID x 3/4 OD plasticizer free kryographics Pascal for NVIDIA TITAN X and GTX ... 
CoolingCoolingCoolingCooling
Mayhem Pastel White UV Monsoon center compression fittings  Alphacool NexXxoS XT45 Full Copper 280mm radiator  Alphacool NexXxoS UT60 Full Copper 240mm radiator  
CoolingCoolingCoolingOS
Monsoon Series Two D5 Premium Dual Bay Reservoi... Phanteks PH-F140SP Case Fan Aquacomputer backplate for kryographics Pascal ... Windows 10 Pro x64 
MonitorMonitorKeyboardPower
Asus PG278Q Gsync (Primary) Asus PB278Q (Extended) Corsair Gaming K70 LUX RGB Mechanical Gaming Ke... CORSAIR HX Series HX750 750W ATX12V 2.3 / EPS12... 
CaseMouseMouse PadAudio
Phanteks Enthoo Luxe PH-ES614L  Func MS-3 rev2 Razer exactMat X speed Creative Sound Blaster Zx 116dB PCIe Gaming Sou... 
AudioAudioOtherOther
Logitech Z906 5.1 Surround Sound Speaker System... HyperX Cloud II Gaming Headset Darkside dimable LED strips  Modsmart Kobra Cables 
Other
EVGA PowerLink 
CPUMotherboardGraphicsGraphics
i7 4790k @ 4.8 MSI Z97 Gaming 7  Saphire R9 290x (hynix) Saphire R9 290x (Hynix) 
RAMHard DriveHard DriveHard Drive
GSkill Ripjaws Z F3-2400C10D-8GZH 2x Samsung 850 Pro Raid-0  AMD RAMDisk WD1001FALS (Data/Backup) 
Optical DriveCoolingOSMonitor
Asus BC-12B1ST BlueRay Custom Water Cooled Windows 10 Pro x64 Asus PG279Q 
MonitorKeyboardPowerCase
Asus PG278Q Corsair K70 EVGA 1000w G2 Phanteks Enthoo Primo 
MouseMouse PadAudioAudio
FUNC MS-3 rev2  XTRAC PADS Ripper Optical Mouse pad Kingston HyperX Cloud II  Creative Sound Blaster Z PCIe 116dB SNR Gaming ... 
OtherOtherOtherOther
Alphacool NexXxoS UT60 Full Copper 420mm EK Supremacy EVO Alphacool NexXxoS Monsta 360 Radiator EK-FC R9-290X Waterblock - Nickel Acetal (Rev.2.0) 
OtherOtherOtherOther
EK-FC R9-290X backplate Bitspower Dual / Single D5 Top Upgrade Kit 150 ... Bitspower D5 MOD TOP (Black "S" Model) PrimoFlex Advanced LRT 10ft Tubing - 1/2in. ID ... 
  hide details  
Reply
post #7889 of 10702
Thanks for the input owcraftsman; I can't say I'm very familiar with instruction sets or p95, so I wanted to check my reasoning. It might be worth using an asterisk or something to demarcate SSC entries involving AVX instructions. If one wants to do quantitative comparisons, then one could compute the average effect of using AVX instructions vs. not using AVX instructions (in my experience, at 4.5Ghz the difference is .024-.032 v).

I think I'll go for stable settings using 27.4, and then maybe see what I can get on 26.6. If it's notably lower and puts me in contention for a top spot, I'll re-run; otherwise I'll submit using 27.4.
post #7890 of 10702
Quote:
Originally Posted by owcraftsman View Post

  1. The Intel 2600k (sandy bridge) supports SSE4 & AVX instruction sets. see here
  2. Prime95 27.1 - 27.4 betas & IBT 2.51 both are fully optimized to exploit AVX instruction set

Thanks for the download links! Just moved from Lynnfield i5 to Sandy Bridge i7 and wanted to get some stability tests done before I started clocking.

Also, does 1.2V seem reasonable for a default loading voltage in CPU-Z when using default CPU settings (ie. C1E, EIST, Turbo, Asus TPU)?

EDIT: That was w/ an old version of Linx...here's what this crazy new version of Intel Burn did!!

2600kdefaults.jpg

Temps were considerably less w/ Linx, haha.
Edited by Jodiuh - 3/20/12 at 12:37pm
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