Overclock.net › Forums › Intel › Intel CPUs › Haswell Overclocking Guide [With Statistics]
New Posts  All Forums:Forum Nav:

Haswell Overclocking Guide [With Statistics] - Page 6

post #51 of 19296
Quote:
Originally Posted by Darkwizzie View Post

I was originally thinking of getting 4.7ghz... I'm at 1.36v for 4.6ghz on AIR. For 4.7ghz I might have to delid/watercool so I'll stay with 4.6ghz.

Maybe I missed your batch number? what you got? and what is your model 4770k or 4670k? Thank you.
post #52 of 19296
Hi,

CPU : 45X100 (more is doable but i like to keep the temps at 80-85 max when AVX)
Uncore : set to Auto to keep the idle function at 800MHz / Load 4000MHz (but it SPi32M at 1:1)
RAM : 2933MHz



post #53 of 19296
So I have been running at 4.6 (38x Uncore Min/Max) @ 1.215V for a few weeks and not a single issue. This clock is Linpack stability tested. Recently I decided to try 4.5 GHz at 1.180V and I'm stability testing it now. I noticed that it runs up to 10C cooler than 4.6 in most synthetic benchmarks and it allows me to increase the Uncore beyond 38 and is still stable. Now I may switch to the 45 for better thermals and overall longevity of the chipset. I mostly game and do a little video compression, so I am not beating it up by any means. I am just surprised at the dip in temps going from 1.180 to 1.215.

I know these are very low compared to some of the voltages posted here, but I just don't see a need to run mine at 4.6. To me it seems like every chip has a sweet spot that it really likes, and everything above that is just pushing it. What do you guys think? I am by no means uncomfortable with keeping it 4.6, it has been running fine. I am just surprised at the difference in temps going from one to the other. With large problems 4.5 only gets up to 80-82 in Linpack runs, while 4.6 can hit 90-92C easy. 4.5 just seems to be a better 24/7 clock, but maybe I am just being paranoid.
Sager NP6350
(9 items)
 
   
CPUMotherboardGraphicsRAM
Intel i7-3630QM Clevo W350ETQ Nvidia GTX 660M 12 GB Samsung DDR3 1600 MHz 
Hard DriveHard DriveOptical DriveOS
Crucial M4 256 GB WD Black 750 GB Toshiba DVD-RW Drive Windows 7 (64 bit) 
Monitor
LG Phillips 
  hide details  
Reply
Sager NP6350
(9 items)
 
   
CPUMotherboardGraphicsRAM
Intel i7-3630QM Clevo W350ETQ Nvidia GTX 660M 12 GB Samsung DDR3 1600 MHz 
Hard DriveHard DriveOptical DriveOS
Crucial M4 256 GB WD Black 750 GB Toshiba DVD-RW Drive Windows 7 (64 bit) 
Monitor
LG Phillips 
  hide details  
Reply
post #54 of 19296
Well, I read in owners topic, that linpack is the best for testing OC stability. Can you tell me how to setup it for the best testing? Thank you
Edited by Ponteral - 7/21/13 at 12:19pm
post #55 of 19296
Quote:
Originally Posted by BangBangPlay View Post

So I have been running at 4.6 (38x Uncore Min/Max) @ 1.215V for a few weeks and not a single issue. This clock is Linpack stability tested. Recently I decided to try 4.5 GHz at 1.180V and I'm stability testing it now. I noticed that it runs up to 10C cooler than 4.6 in most synthetic benchmarks and it allows me to increase the Uncore beyond 38 and is still stable. Now I may switch to the 45 for better thermals and overall longevity of the chipset. I mostly game and do a little video compression, so I am not beating it up by any means. I am just surprised at the dip in temps going from 1.180 to 1.215.

I know these are very low compared to some of the voltages posted here, but I just don't see a need to run mine at 4.6. To me it seems like every chip has a sweet spot that it really likes, and everything above that is just pushing it. What do you guys think? I am by no means uncomfortable with keeping it 4.6, it has been running fine. I am just surprised at the difference in temps going from one to the other. With large problems 4.5 only gets up to 80-82 in Linpack runs, while 4.6 can hit 90-92C easy. 4.5 just seems to be a better 24/7 clock, but maybe I am just being paranoid.


At your 4,6 voltage, my CPU hit 80-84 under LinX AVX (and it's really warm out there) ; the cooler is a part of the equation wink.gif

Then AVX stress tests are not representative of a daily usage. Just make a BF3 or whatever and report your temps @ 4,6GHz wink.gif
post #56 of 19296
Thankyou everyone for clarifying the uncore, I will post my results once I have time to do some testing smile.gif
    
CPUMotherboardGraphicsRAM
5820k MSI X99S SLI PLUS Gigabyte GTX 1060 Mini 6GB OC 16GB GSKILL 2400 Ripjaw 
RAMHard DriveHard DriveHard Drive
16GB Avexir Raiden Rog Certified 2600 DDR4 Samsung 840 EVO 750 Seagate 1TB Seagate 1TB 
CoolingOSMonitorKeyboard
Custome water loop, EK Supremecy MX CPU Block,... Windows 10 Pro 64Bit Benq XL2430T Keycool 87 II 
PowerCaseMouseMouse Pad
Seasonic Snow Silent 750 Platinum Cooltek W2 upgraded to tempered glass window pa... Func MS-2 MSI Thunderstorm Aluminium Mousepad 
Audio
Steelseries Siberia Elite 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
5820k MSI X99S SLI PLUS Gigabyte GTX 1060 Mini 6GB OC 16GB GSKILL 2400 Ripjaw 
RAMHard DriveHard DriveHard Drive
16GB Avexir Raiden Rog Certified 2600 DDR4 Samsung 840 EVO 750 Seagate 1TB Seagate 1TB 
CoolingOSMonitorKeyboard
Custome water loop, EK Supremecy MX CPU Block,... Windows 10 Pro 64Bit Benq XL2430T Keycool 87 II 
PowerCaseMouseMouse Pad
Seasonic Snow Silent 750 Platinum Cooltek W2 upgraded to tempered glass window pa... Func MS-2 MSI Thunderstorm Aluminium Mousepad 
Audio
Steelseries Siberia Elite 
  hide details  
Reply
post #57 of 19296
Thread Starter 
Quote:
Originally Posted by Ponteral View Post


Maybe I missed your batch number? what you got? and what is your model 4770k or 4670k? Thank you.

I posted it in the thread chart. 

 

Batch 330 Costa Rica 4670k.

 

The guide represents my opinion on Linpack. If that's your line between stable/unstable and hot/not too hot, your OC will be nerfed by the stress test because it will be excessively hot.

   
Celapaleis (2013)
(16 items)
 
RAMHard DriveHard DriveKeyboard
x2 8gb GSkill Trident Z 3600 @ 3733 15-14-14-31-2T Samsung 950 Pro 512gb (Undelwalt) WD Red 2tb (Pack Yak II) Kmac 2 (Brown Switch, Gon PCB, Red LED, Clear Top) 
CPUMotherboardGraphicsRAM
i5 6600k Delid @ 4.848/4.848 Asus z170 Hero MSI GTX 980ti @ 1499/8208 1.25v/134% Power 2 x 8gb Gskills Ripjaws V 3131 16-16-16-32 
Hard DriveHard DriveHard DriveHard Drive
Samsung Pro Series 840 256gb (Celapaleis) 1 TB Samsung Drive (Pack Yak) Samsung Pro Series 850 256gb (Celapaleis Reprise) Phoenix Down (External Backup) 
CoolingCoolingOSMonitor
Noctua D14 Kraken x61 Push-Pull for GPU Windows 10 Pro Catleap 2560x1440 @ 60hz 
KeyboardPowerCaseMouse
CM Storm Trigger Brown Switches EVGA Supernova 1000w P2 Corsair 540 Air Razer Abyssus 
Mouse PadAudioAudioAudio
Steelseries QcK Mini Blue Yeti Microphone Objective Dac + Amp Sescom AB Switch 
AudioAudioAudioAudio
Krk Rokit 6 G2 x2 Radioshack Ground Loop Isolator (For Rokits) Wooaudio Aluminum Headphone Stand Sennheiser HD 800 
AudioOther
Logitech c920 Webcam NZXT Sentry Mix 2 Fan Controller 
  hide details  
Reply
   
Celapaleis (2013)
(16 items)
 
RAMHard DriveHard DriveKeyboard
x2 8gb GSkill Trident Z 3600 @ 3733 15-14-14-31-2T Samsung 950 Pro 512gb (Undelwalt) WD Red 2tb (Pack Yak II) Kmac 2 (Brown Switch, Gon PCB, Red LED, Clear Top) 
CPUMotherboardGraphicsRAM
i5 6600k Delid @ 4.848/4.848 Asus z170 Hero MSI GTX 980ti @ 1499/8208 1.25v/134% Power 2 x 8gb Gskills Ripjaws V 3131 16-16-16-32 
Hard DriveHard DriveHard DriveHard Drive
Samsung Pro Series 840 256gb (Celapaleis) 1 TB Samsung Drive (Pack Yak) Samsung Pro Series 850 256gb (Celapaleis Reprise) Phoenix Down (External Backup) 
CoolingCoolingOSMonitor
Noctua D14 Kraken x61 Push-Pull for GPU Windows 10 Pro Catleap 2560x1440 @ 60hz 
KeyboardPowerCaseMouse
CM Storm Trigger Brown Switches EVGA Supernova 1000w P2 Corsair 540 Air Razer Abyssus 
Mouse PadAudioAudioAudio
Steelseries QcK Mini Blue Yeti Microphone Objective Dac + Amp Sescom AB Switch 
AudioAudioAudioAudio
Krk Rokit 6 G2 x2 Radioshack Ground Loop Isolator (For Rokits) Wooaudio Aluminum Headphone Stand Sennheiser HD 800 
AudioOther
Logitech c920 Webcam NZXT Sentry Mix 2 Fan Controller 
  hide details  
Reply
post #58 of 19296
Quote:
Originally Posted by The Real Deal View Post

At your 4,6 voltage, my CPU hit 80-84 under LinX AVX (and it's really warm out there) ; the cooler is a part of the equation wink.gif

Then AVX stress tests are not representative of a daily usage. Just make a BF3 or whatever and report your temps @ 4,6GHz wink.gif

Yeah, I had it at 4.5 for a few days and then switched back to 4.6. The difference in temps while not stress testing was minute. With that being said I do believe that 4.6 will be a sweet spot for some, and anything above that is just excess heat and wear and tear on your CPU. The last thing I want to do is have to swap this chip for another and renter the Haswell lottery. So maybe that is why I was cautious to begin with.
Sager NP6350
(9 items)
 
   
CPUMotherboardGraphicsRAM
Intel i7-3630QM Clevo W350ETQ Nvidia GTX 660M 12 GB Samsung DDR3 1600 MHz 
Hard DriveHard DriveOptical DriveOS
Crucial M4 256 GB WD Black 750 GB Toshiba DVD-RW Drive Windows 7 (64 bit) 
Monitor
LG Phillips 
  hide details  
Reply
Sager NP6350
(9 items)
 
   
CPUMotherboardGraphicsRAM
Intel i7-3630QM Clevo W350ETQ Nvidia GTX 660M 12 GB Samsung DDR3 1600 MHz 
Hard DriveHard DriveOptical DriveOS
Crucial M4 256 GB WD Black 750 GB Toshiba DVD-RW Drive Windows 7 (64 bit) 
Monitor
LG Phillips 
  hide details  
Reply
post #59 of 19296
Thank you for this guide Darkwizzie, like many others I had completely missed out on lowering Uncore that drastically and how little effect it has on performance. Will definitely play around some with this new information.

What I've been using before reading this post:

CPU: 4670k
Core Mult: 42
Uncore: Auto
Vcore: 1.19 adaptive
Uncore volt: Auto
Cooler: H80i (2x Gentle Typhoon 1850rpm)
Ram speed: 1333mhz (kept it on default thus far)
Batch #: L310B515 - Noticed The Real Deal has the same batch, which is just about the first person I see with this batchnumber besides myself.
Stability tested: Not very properly. Standard IBT and an hour of OCCT (~61-65avg, maxing around 72 I believe). Will do it more properly when I go for higher OC's, but it's been fine playing games with for now.

Hopefully I'll have better numbers posted in the coming days!

Edit: A question about the graphs from the OP. In the report from Skyrim, the 4.5ghz with 3.5 uncore is getting a better result than 4.6ghz with 3.4 uncore. Does this mean that for gaming, uncore affects performance more than for things like Cinebench for example?

Edit 2: Guess I might not be so lucky. Tried a 4.4ghz (3.4 uncore still) with 1.27 Manual Vcore / Auto uncore voltage. Started up prime95 blend and things were looking fine for the "long" tests, with temps around 66-70, maxing out at 76 for a second on one core. Then after 20min it switched to the Short test, and temps rose very fast to high-80s, hitting 93 as max and thus I immediately stop the test as this is simply running to hot at least for me to feel safe. Back to the drawing board I guess.. With vcore at 1.25 I would get blue screen when booting, with artifacts (which was odd to me because I've never seen that when it wasn't related to GPU before)

Of note in regards to this, 3 of the screw threads that fasten the h80i radiator to the chassi broke upon initial install yesterday even though I was being quite careful. I 'fixed' this by using screw thread tape (if that's the word in english) and it's decently fastened, but I can "pull" it some. Could this affect temps? Sorry if this is a stupid question, and probably also the wrong section for this.
Edited by katatoni - 7/23/13 at 7:40pm
post #60 of 19296
Thread Starter 
Quote:
Originally Posted by katatoni View Post

Thank you for this guide Darkwizzie, like many others I had completely missed out on lowering Uncore that drastically and how little effect it has on performance. Will definitely play around some with this new information.

What I've been using before reading this post:

CPU: 4670k
Core Mult: 42
Uncore: Auto
Vcore: 1.19 adaptive
Uncore volt: Auto
Cooler: H80i (2x Gentle Typhoon 1850rpm)
Ram speed: 1333mhz (kept it on default thus far)
Batch #: L310B515 - Noticed The Real Deal has the same batch, which is just about the first person I see with this batchnumber besides myself.
Stability tested: Not very properly. Standard IBT and an hour of OCCT (~61-65avg, maxing around 72 I believe). Will do it more properly when I go for higher OC's, but it's been fine playing games with for now.

Hopefully I'll have better numbers posted in the coming days!

Edit: A question about the graphs from the OP. In the report from Skyrim, the 4.5ghz with 3.5 uncore is getting a better result than 4.6ghz with 3.4 uncore. Does this mean that for gaming, uncore affects performance more than for things like Cinebench for example?

Edit 2: Guess I might not be so lucky. Tried a 4.4ghz (3.4 uncore still) with 1.27 Manual Vcore / Auto uncore voltage. Started up prime95 blend and things were looking fine for the "long" tests, with temps around 66-70, maxing out at 76 for a second on one core. Then after 20min it switched to the Short test, and temps rose very fast to high-80s, hitting 93 as max and thus I immediately stop the test as this is simply running to hot at least for me to feel safe. Back to the drawing board I guess.. With vcore at 1.25 I would get blue screen when booting, with artifacts (which was odd to me because I've never seen that when it wasn't related to GPU before)

Of note in regards to this, 3 of the screw threads that fasten the h80i radiator to the chassi broke upon initial install yesterday even though I was being quite careful. I 'fixed' this by using screw thread tape (if that's the word in english) and it's decently fastened, but I can "pull" it some. Could this affect temps? Sorry if this is a stupid question, and probably also the wrong section for this.

 

Hey, thanks for the info. Please keep me updated if you manage to wiggle out a better result! For some people uncore didn't help but for me and some others it made a world of difference. Haswell reeks of hit-or-miss-ness.

 

About Skyrim results: We have to keep in mind that CPU affects gaming less as a general rule. Even when we chug higher core clocks, the result won't be massive... Skyrim is a little bit more CPU dependent, that is why I chose that. I benchmarked it in a town to try to up the CPU usage (a benchmark while in combat is not practical). But uncore is less of a factor than core clock as well. So while the effects of uncore is already small, it's made smaller by the fact it's a game. I do expect uncore difference to be a little more noticeable in Cinebench or Chess as these are CPU benchmarks... But I'm not sure if uncore is making a larger difference compared to a change in core clock or if it's simply a result of CPU mattering more in a CPU benchmark.

 

Regarding your cooler issue: I'm not sure what you mean, and I'm not very experienced with water cooling. I've stuck with air so far, maybe somebody else knows more. And about your overclock... could you squeeze out  x43? From my personal experience and from word of mouth, it seems the effect of lower uncore on ability to adjust core clock tends to be more pronounced in higher GHZ ranges to begin with. Maybe if you had a better thermal solution... devil.gif

   
Celapaleis (2013)
(16 items)
 
RAMHard DriveHard DriveKeyboard
x2 8gb GSkill Trident Z 3600 @ 3733 15-14-14-31-2T Samsung 950 Pro 512gb (Undelwalt) WD Red 2tb (Pack Yak II) Kmac 2 (Brown Switch, Gon PCB, Red LED, Clear Top) 
CPUMotherboardGraphicsRAM
i5 6600k Delid @ 4.848/4.848 Asus z170 Hero MSI GTX 980ti @ 1499/8208 1.25v/134% Power 2 x 8gb Gskills Ripjaws V 3131 16-16-16-32 
Hard DriveHard DriveHard DriveHard Drive
Samsung Pro Series 840 256gb (Celapaleis) 1 TB Samsung Drive (Pack Yak) Samsung Pro Series 850 256gb (Celapaleis Reprise) Phoenix Down (External Backup) 
CoolingCoolingOSMonitor
Noctua D14 Kraken x61 Push-Pull for GPU Windows 10 Pro Catleap 2560x1440 @ 60hz 
KeyboardPowerCaseMouse
CM Storm Trigger Brown Switches EVGA Supernova 1000w P2 Corsair 540 Air Razer Abyssus 
Mouse PadAudioAudioAudio
Steelseries QcK Mini Blue Yeti Microphone Objective Dac + Amp Sescom AB Switch 
AudioAudioAudioAudio
Krk Rokit 6 G2 x2 Radioshack Ground Loop Isolator (For Rokits) Wooaudio Aluminum Headphone Stand Sennheiser HD 800 
AudioOther
Logitech c920 Webcam NZXT Sentry Mix 2 Fan Controller 
  hide details  
Reply
   
Celapaleis (2013)
(16 items)
 
RAMHard DriveHard DriveKeyboard
x2 8gb GSkill Trident Z 3600 @ 3733 15-14-14-31-2T Samsung 950 Pro 512gb (Undelwalt) WD Red 2tb (Pack Yak II) Kmac 2 (Brown Switch, Gon PCB, Red LED, Clear Top) 
CPUMotherboardGraphicsRAM
i5 6600k Delid @ 4.848/4.848 Asus z170 Hero MSI GTX 980ti @ 1499/8208 1.25v/134% Power 2 x 8gb Gskills Ripjaws V 3131 16-16-16-32 
Hard DriveHard DriveHard DriveHard Drive
Samsung Pro Series 840 256gb (Celapaleis) 1 TB Samsung Drive (Pack Yak) Samsung Pro Series 850 256gb (Celapaleis Reprise) Phoenix Down (External Backup) 
CoolingCoolingOSMonitor
Noctua D14 Kraken x61 Push-Pull for GPU Windows 10 Pro Catleap 2560x1440 @ 60hz 
KeyboardPowerCaseMouse
CM Storm Trigger Brown Switches EVGA Supernova 1000w P2 Corsair 540 Air Razer Abyssus 
Mouse PadAudioAudioAudio
Steelseries QcK Mini Blue Yeti Microphone Objective Dac + Amp Sescom AB Switch 
AudioAudioAudioAudio
Krk Rokit 6 G2 x2 Radioshack Ground Loop Isolator (For Rokits) Wooaudio Aluminum Headphone Stand Sennheiser HD 800 
AudioOther
Logitech c920 Webcam NZXT Sentry Mix 2 Fan Controller 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › Haswell Overclocking Guide [With Statistics]