Overclock.net › Forums › Intel › Intel CPUs › [Official] The Sandy Bridge 5ghz + Club
New Posts  All Forums:Forum Nav:

[Official] The Sandy Bridge 5ghz + Club - Page 170  

Poll Results: Should we use blend only or small fft too?

Poll expired: Feb 25, 2011  
  • 53% (37)
    1 hr Blend only
  • 46% (32)
    1 hr Small Fft, 1 hr Blend
69 Total Votes  
post #1691 of 2120
Quote:
Originally Posted by Jasonn20 View Post
So the Costa Ricas are indeed having a high success rate on 5ghz...??
after reading about all this Malay,Costa Rica chips i think this is appropriate to post;
http://forums.anandtech.com/showthread.php?t=750461
"I posted it this in the mammoth Northwood thread, and Wingz posted it in GH. I thought I'd post it once more just for the extra exposure.


Just to clear something up, the markings on the back of the CPU refer to the packaging site - not the manufacturing site of the actual silicon chip itself. Intel has no fabs in either Malaysia or Costa Rica, they are packaging facilities. The silicon die/chips are manufactured elsewhere in the world and are shipped to either of these two packaging sites. In addition, Intel manufacturing has a goal of running a "virtual fab" - meaning that, among other things, products from one fab are statistically indistinguishable from those manufactured at another fab. So even if, for example using fake names, Malaysian packages used chips only from fab #1 and packages marked Costa Rica use chips from fab #2, there should be no difference statistically between these two.

I saw these discussions back in the Celeron days, and I commented back then, but this time around it seems a little different. I have started seeing some online retailers charging more for parts from a specific packaging site and this disturbs me. There is no difference between parts from these two packaging sites. Just as you can have 6 head/tails coin tosses come out heads, there may seem to be a correlation that heads is more likely than tails, but there isn't. In reality the odds are still approximately 50/50.

The silicon is what defines the speed of a CPU, not the package. And the silicon comes from multiple fabs scattered all over the place that are all supposed to be identical anyway.

If you are considering spending more, or buying from a shadier vendor, in order to get a specific package, I would urge you to reconsider. There is no difference and you are only wasting money, and or risking getting ripped off.

Patrick Mahoney
Microprocessor Design Engineer
Intel Corp."
    
CPUMotherboardGraphicsRAM
i7 4930K Rampage IV Gene SLI Evga GTX 780 Classy 16Gb TeamX2400c9 
Hard DriveCoolingOSMonitor
Samsung 840 Pro/ Hitachi Deskstar 1TB  H20 (NexXxoS 420+280) w7x64 VG248QE 
KeyboardPowerMouseAudio
Logitech G110 SuperNOVA 1300 G2 G502 Proteus Core Foobar2K/WASAPI >NuForce Icon HDP >Crack with S... 
  hide details  
    
CPUMotherboardGraphicsRAM
i7 4930K Rampage IV Gene SLI Evga GTX 780 Classy 16Gb TeamX2400c9 
Hard DriveCoolingOSMonitor
Samsung 840 Pro/ Hitachi Deskstar 1TB  H20 (NexXxoS 420+280) w7x64 VG248QE 
KeyboardPowerMouseAudio
Logitech G110 SuperNOVA 1300 G2 G502 Proteus Core Foobar2K/WASAPI >NuForce Icon HDP >Crack with S... 
  hide details  
post #1692 of 2120
Quote:
Originally Posted by catcherintherye View Post
Hi Falkentyne, yes I did. I have been on the F4H bios the whole time, but thanks for the tip. I'm not sure if that's glitchy but it seems to be pretty stable for me compared to the other bios's. Although I admit I am not using LLC1, haha. Perhaps I'll try again later with a lower LLC.



I ended up going to Microcenter and hunting for an A batch Costa Rica chip, and finally broke the barrier. Thanks for the support, and if I can do it you can do it too.
One person who went back to F3H nticed that he needed 0.020v less voltage for stability, as well as less vdimm. The problem with both F3 and f4a is that the vcore undershoots/overshoots at really strange times when it shouldn't. For example, 1.440v idle LLC6 might suddenly become 1.464v during gaming, and then drop to 1.392v during a certain pass of prime, while f3h would be (with LLC1) 1.440v idle, 1.416v load)

The ten LLC levels are nice, but at the moment, they are too flaky. I mean, LLC#1 (out of the 10) has MORE vdroop than LLC turned off (standard/auto).
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
post #1693 of 2120
Quote:
Originally Posted by coolhandluke41 View Post
after reading about all this Malay,Costa Rica chips i think this is appropriate to post;
http://forums.anandtech.com/showthread.php?t=750461
"I posted it this in the mammoth Northwood thread, and Wingz posted it in GH. I thought I'd post it once more just for the extra exposure.


Just to clear something up, the markings on the back of the CPU refer to the packaging site - not the manufacturing site of the actual silicon chip itself. Intel has no fabs in either Malaysia or Costa Rica, they are packaging facilities. The silicon die/chips are manufactured elsewhere in the world and are shipped to either of these two packaging sites. In addition, Intel manufacturing has a goal of running a "virtual fab" - meaning that, among other things, products from one fab are statistically indistinguishable from those manufactured at another fab. So even if, for example using fake names, Malaysian packages used chips only from fab #1 and packages marked Costa Rica use chips from fab #2, there should be no difference statistically between these two.

I saw these discussions back in the Celeron days, and I commented back then, but this time around it seems a little different. I have started seeing some online retailers charging more for parts from a specific packaging site and this disturbs me. There is no difference between parts from these two packaging sites. Just as you can have 6 head/tails coin tosses come out heads, there may seem to be a correlation that heads is more likely than tails, but there isn't. In reality the odds are still approximately 50/50.

The silicon is what defines the speed of a CPU, not the package. And the silicon comes from multiple fabs scattered all over the place that are all supposed to be identical anyway.

If you are considering spending more, or buying from a shadier vendor, in order to get a specific package, I would urge you to reconsider. There is no difference and you are only wasting money, and or risking getting ripped off.

Patrick Mahoney
Microprocessor Design Engineer
Intel Corp."
Thanks for the explanation! This was really insightful.
Buena Vista
(18 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 4790K [4.6GHz] [1.25V] Asus Maximus VII Formula [BIOS 3003] NVIDIA GTX 1080 Ti FE [2050MHz] [12GHz QDR] Samsung MV-3V4G3 16GB (4x4GB) DDR3 2133 
Hard DriveHard DriveHard DriveHard Drive
Intel 535 Series 240GB [RAID 1] Crucial M500 960GB Seagate Barracuda 3TB Western Digital VelociRaptor 500GB [RAID 0] 
CoolingCoolingOSMonitor
Corsair H110i GT (Qty 2) Phanteks PH-F140XP Microsoft Windows 10 Pro [64-bit] Dell S2716DG [G-Sync] 
KeyboardPowerCaseMouse
Razer BlackWidow Chroma EVGA Supernova G2 1300W NZXT H440 [Black/Red] Logitech G900 
Mouse PadAudio
Razer Firefly Logitech G933 
  hide details  
Buena Vista
(18 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 4790K [4.6GHz] [1.25V] Asus Maximus VII Formula [BIOS 3003] NVIDIA GTX 1080 Ti FE [2050MHz] [12GHz QDR] Samsung MV-3V4G3 16GB (4x4GB) DDR3 2133 
Hard DriveHard DriveHard DriveHard Drive
Intel 535 Series 240GB [RAID 1] Crucial M500 960GB Seagate Barracuda 3TB Western Digital VelociRaptor 500GB [RAID 0] 
CoolingCoolingOSMonitor
Corsair H110i GT (Qty 2) Phanteks PH-F140XP Microsoft Windows 10 Pro [64-bit] Dell S2716DG [G-Sync] 
KeyboardPowerCaseMouse
Razer BlackWidow Chroma EVGA Supernova G2 1300W NZXT H440 [Black/Red] Logitech G900 
Mouse PadAudio
Razer Firefly Logitech G933 
  hide details  
post #1694 of 2120
Quote:
Originally Posted by Falkentyne View Post
One person who went back to F3H nticed that he needed 0.020v less voltage for stability, as well as less vdimm. The problem with both F3 and f4a is that the vcore undershoots/overshoots at really strange times when it shouldn't. For example, 1.440v idle LLC6 might suddenly become 1.464v during gaming, and then drop to 1.392v during a certain pass of prime, while f3h would be (with LLC1) 1.440v idle, 1.416v load)

The ten LLC levels are nice, but at the moment, they are too flaky. I mean, LLC#1 (out of the 10) has MORE vdroop than LLC turned off (standard/auto).
I just flashed back to F3H and tried stress testing it. At LLC Level 1 it would BSOD as soon as I started P95 at the same settings I had under F4A. At LLC2, P95 would run for a while but the voltages shot up to like 1.59, and I quickly terminated it. Seems like F3H isn't for everyone, but thanks anyway, worth a shot.
Monster
(13 items)
 
  
CPUMotherboardGraphicsRAM
2600K Z77MA-G45 GTX 580 G.Skill Ripjaws 1600 4x4GB 
Hard DriveMonitorKeyboardPower
Vertex 4 ASUS VE276Q CM Storm Quickfire Cherry Red MX AX 860i 
CaseMouse
SG09 Logitech G9X 
  hide details  
Monster
(13 items)
 
  
CPUMotherboardGraphicsRAM
2600K Z77MA-G45 GTX 580 G.Skill Ripjaws 1600 4x4GB 
Hard DriveMonitorKeyboardPower
Vertex 4 ASUS VE276Q CM Storm Quickfire Cherry Red MX AX 860i 
CaseMouse
SG09 Logitech G9X 
  hide details  
post #1695 of 2120
Quote:
Originally Posted by coolhandluke41 View Post


RIP..sorry to hear that ..1.53 hmm interesting
Edit ; you should be proud ..i think you the first one (beside Sin) on OCN to successfully do it JK
offset or fixed ??
Offset

And it wasn't truly being stressed, not per say dedicated f@h or boinc.
The load came from supporting the gpu crunching, occasional gaming/benchmarking.
The hardest it was stressed was maybe 5, 5 pass IBT runs, 2 hours of prime95.

I'm not 100% sure it's the proc, waiting on some ram to test.


I blame Intel for making a proc that could be stabilized and ran with decent temps at 5ghz
Mighty-iTX
(12 items)
 
One foot in..
(17 items)
 
mATX
(12 items)
 
CPUMotherboardGraphicsRAM
i7-4770K M7-Impact Strix GTX980 Samsung 
Hard DriveCoolingOSMonitor
Crucial M4 256 Glacer 240L W8.1 K272HUL 
KeyboardPowerCaseMouse
Saitek Seasonic X-650 Prodigy MX518 
CPUMotherboardGraphicsGraphics
4770K/45/43 Maximus VI Extreme R290X R290X 
RAMHard DriveHard DriveOptical Drive
Samsung Crucial M4 64GB RAIDR LSSWM 
CoolingOSMonitorKeyboard
Cooler Master Glacer 7 Ultimate 64 SyncMaster Eclipse 
PowerCaseMouseMouse Pad
XFX 1050 BE 600T Silver MX518 Desk 
CPUMotherboardGraphicsRAM
i5-4330 Z87 Gryphon GTX690 G.Skill RipJawsX 
Hard DriveCoolingOSMonitor
WD Blacks Hyper 212 W8.1 ACER 
KeyboardPowerCaseMouse
Logitech Ultra X3 1000w Corsair 230T Orange Razer 
  hide details  
Mighty-iTX
(12 items)
 
One foot in..
(17 items)
 
mATX
(12 items)
 
CPUMotherboardGraphicsRAM
i7-4770K M7-Impact Strix GTX980 Samsung 
Hard DriveCoolingOSMonitor
Crucial M4 256 Glacer 240L W8.1 K272HUL 
KeyboardPowerCaseMouse
Saitek Seasonic X-650 Prodigy MX518 
CPUMotherboardGraphicsGraphics
4770K/45/43 Maximus VI Extreme R290X R290X 
RAMHard DriveHard DriveOptical Drive
Samsung Crucial M4 64GB RAIDR LSSWM 
CoolingOSMonitorKeyboard
Cooler Master Glacer 7 Ultimate 64 SyncMaster Eclipse 
PowerCaseMouseMouse Pad
XFX 1050 BE 600T Silver MX518 Desk 
CPUMotherboardGraphicsRAM
i5-4330 Z87 Gryphon GTX690 G.Skill RipJawsX 
Hard DriveCoolingOSMonitor
WD Blacks Hyper 212 W8.1 ACER 
KeyboardPowerCaseMouse
Logitech Ultra X3 1000w Corsair 230T Orange Razer 
  hide details  
post #1696 of 2120
Yeah ..like i said 1.49 on offset here for around two months no problem..keep us updated
hope she's OK
    
CPUMotherboardGraphicsRAM
i7 4930K Rampage IV Gene SLI Evga GTX 780 Classy 16Gb TeamX2400c9 
Hard DriveCoolingOSMonitor
Samsung 840 Pro/ Hitachi Deskstar 1TB  H20 (NexXxoS 420+280) w7x64 VG248QE 
KeyboardPowerMouseAudio
Logitech G110 SuperNOVA 1300 G2 G502 Proteus Core Foobar2K/WASAPI >NuForce Icon HDP >Crack with S... 
  hide details  
    
CPUMotherboardGraphicsRAM
i7 4930K Rampage IV Gene SLI Evga GTX 780 Classy 16Gb TeamX2400c9 
Hard DriveCoolingOSMonitor
Samsung 840 Pro/ Hitachi Deskstar 1TB  H20 (NexXxoS 420+280) w7x64 VG248QE 
KeyboardPowerMouseAudio
Logitech G110 SuperNOVA 1300 G2 G502 Proteus Core Foobar2K/WASAPI >NuForce Icon HDP >Crack with S... 
  hide details  
post #1697 of 2120
Here's my chip, hit 5GH/Z this evening.. I'll see how far I can push it next

Oh yea I have no idea what my batch number is, excuse my ignorance but if you can show me how to get it that'd be much appreciated!

edit: the rig is on water too

Renegard - H20
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500K @ 5Ghz ASUS P67 Sabertooth Gigabyte 4890 @ 1000mhz/4000mhz 2x2GB G.Skill PI 1600 
Hard DriveOSMonitorKeyboard
WD 320gb 32mb cache Windows 7 64Bit BenQ 22" Widescreen Logitech G15 V2 
PowerCaseMouse
Superflower 650watt HAF 932 Logitech G5 
  hide details  
Renegard - H20
(13 items)
 
  
CPUMotherboardGraphicsRAM
2500K @ 5Ghz ASUS P67 Sabertooth Gigabyte 4890 @ 1000mhz/4000mhz 2x2GB G.Skill PI 1600 
Hard DriveOSMonitorKeyboard
WD 320gb 32mb cache Windows 7 64Bit BenQ 22" Widescreen Logitech G15 V2 
PowerCaseMouse
Superflower 650watt HAF 932 Logitech G5 
  hide details  
post #1698 of 2120
Quote:
Originally Posted by catcherintherye View Post
I just flashed back to F3H and tried stress testing it. At LLC Level 1 it would BSOD as soon as I started P95 at the same settings I had under F4A. At LLC2, P95 would run for a while but the voltages shot up to like 1.59, and I quickly terminated it. Seems like F3H isn't for everyone, but thanks anyway, worth a shot.
1.59v? for LLC2? What was your BIOS voltage set to?

Did you clear the CMOS after flashing the BIOS?
(Proper procedure for cmos clear is
clear the CMOS by doing the following:
Flash the BIOS (i use flashspi),
after the flash is done, power off the computer, turn off the PSU, wait 60 seconds so the +5 vsb is cleared from the board.
Clear cmos by pressing button for 15-30 seconds
(A full and proper clear will erase all saved profiles).
Go back into BIOS and set everything.

But anyway:

What were your exact settings you used with f4a?
You can't correlate settings on such different BIOSes; you have to know what your load voltage needs to be.

If you needed 1.476v for stability at load on f4a, there is NO way that F3h is going to be MORE unstable. It's your settings, and I don't know what LLC setting you used on f4a.

But anyway,
Try 1.525v BIOS on f3h, and LLC1. This should droop down to about 1.47v during prime.
This should NOT BSOD prime right away. Is this what you tried?
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
SB Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-2600k Gigabyte P67A-UD5 B3 R9 290X 16GB 
Hard DriveOptical DriveOSMonitor
WD 2 TB, WD 1 TB, 250gb Liteon ihbs212 blu-ray burner/reader XP+W7 Benq XL2720Z 
KeyboardPowerCaseMouse
Ducky Shine 69/Year of the Goat/Shine 5 Seasonic Platinum X1000 Corsair 760T Logitech G502 with r0ach approved ™ sensor 
Mouse Pad
Puretrak Talent 
  hide details  
post #1699 of 2120
Quote:
Originally Posted by jcniest5 View Post
You are not alone, buddy. I have gone through close to 8 CPUs now and can't find any that runs above 5Ghz stable. I finally gave up and just sold my last 2600K. I'm on 2500K now. The heck with HT and the 1MB additional L3 cache on the 2600K.



Congrats...at least it runs stable for you. Vcore is a little too high for the speed, but not unimagineable.

Found something interesting..... setting my memory timings back to stock (I had tightened them up) has allowed me to drop the vcore a little.
Think I'm going to continue experimenting and see if lowering it from 2133 to 1866MHz would allow me to reduce even further....
Seems to me really no need for running at 2133...
post #1700 of 2120
Quote:
Originally Posted by Falkentyne View Post
1.59v? for LLC2? What was your BIOS voltage set to?

Did you clear the CMOS after flashing the BIOS?
(Proper procedure for cmos clear is
clear the CMOS by doing the following:
Flash the BIOS (i use flashspi),
after the flash is done, power off the computer, turn off the PSU, wait 60 seconds so the +5 vsb is cleared from the board.
Clear cmos by pressing button for 15-30 seconds
(A full and proper clear will erase all saved profiles).
Go back into BIOS and set everything.

But anyway:

What were your exact settings you used with f4a?
You can't correlate settings on such different BIOSes; you have to know what your load voltage needs to be.

If you needed 1.476v for stability at load on f4a, there is NO way that F3h is going to be MORE unstable. It's your settings, and I don't know what LLC setting you used on f4a.

But anyway,
Try 1.525v BIOS on f3h, and LLC1. This should droop down to about 1.47v during prime.
This should NOT BSOD prime right away. Is this what you tried?
I did not clear CMOS after flashing bios, will try. But everything seemed to be gone so I assumed it had cleared itself. I had it set on LLC 9 under F4A to get 1.476 stability. Accordingly, I set 1.48 Bios for F3H and LLC 2 and the voltage shot up like I described. But, I will try your recommendations and get back later.

edit: Just tried, with 1.525 bios and then 1.54, with LLC1. Then tried 1.45 with LLC2. All BSODed after a few minutes. Any more suggestions? It's interesting how the voltage drops at load when the settings are set like this, care to explain the reason behind this?
Edited by catcherintherye - 4/17/11 at 10:27am
Monster
(13 items)
 
  
CPUMotherboardGraphicsRAM
2600K Z77MA-G45 GTX 580 G.Skill Ripjaws 1600 4x4GB 
Hard DriveMonitorKeyboardPower
Vertex 4 ASUS VE276Q CM Storm Quickfire Cherry Red MX AX 860i 
CaseMouse
SG09 Logitech G9X 
  hide details  
Monster
(13 items)
 
  
CPUMotherboardGraphicsRAM
2600K Z77MA-G45 GTX 580 G.Skill Ripjaws 1600 4x4GB 
Hard DriveMonitorKeyboardPower
Vertex 4 ASUS VE276Q CM Storm Quickfire Cherry Red MX AX 860i 
CaseMouse
SG09 Logitech G9X 
  hide details  
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
This thread is locked  
Overclock.net › Forums › Intel › Intel CPUs › [Official] The Sandy Bridge 5ghz + Club