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

Haswell Overclocking Guide [With Statistics] - Page 698

post #6971 of 19539
The link posted gives a very clear explanation as to the effects and concerns, so no need to repeat ... the other links clearly show a diversity of opinion on the matter. This opinion varies based upon what you do with your machine....and as I have been saying, what your goals are. This is no different than memory .... you may get a higher OC if you dial down ya memory speed timings and this may be well worthwhile on CPU intensive tasks but may be detrimental on memory intensive tasks. AutoDesk products in particularly are very "system wide" intensive and even puzzling .... things that I wouldn't have expected but actually improved / decreased performance over the years included:

-elimination of page file decreased performance even when there was oodles of RAM available
-was using a separate drive for page / temp file usage improved far more than I expected when that partition was changed to FAT32

I'm a bit busy with GFX testing at the moment to take on any additional testing .... what I am finding so far tho

-Alternate BIOS oft bring higher measured clocks, but all other things being equal, there are no performance gains associated therewith (that I have seen so far)
-Continually increasing clocks and memory at some point has diminishing returns and, while still stable, going further on, even has negative results ... I have only done 5 runs of each benchmark and I have been hampered / puzzled by the fact that having GPU-z open during the test can cause a Open GL error that I don't get when it's closed.
911 Carrera
(20 items)
 
  
CPUMotherboardGraphicsRAM
Intel w/ EK Supremacy Cooling Block - Clean P... Asus Maximus VI Formula 2 x Asus GTX 780 DCII w/ EK Full Cover Blocks a... Mushkin Red Line Ridgeback 2 x 8GB DDR302400 10... 
Hard DriveHard DriveOptical DriveCooling
2 x SAMSUNG 840 Pro Series MZ-7PD256BW 2.5" 256GB (2 x Seagate ST2000DX001 2TB 7200 RPM 64MB Cach... Asus BW-12B1ST/BLK/G/AS Blue Ray Writer 1 x Alphacool NexXxoS XT45-420 + 1 x Alphacool ... 
CoolingCoolingCoolingCooling
1 x Swiftech 35x2 w/ Heatsink and NB Fan EK -MultiOption RES X3 250 White Acetal Reservoir 5 x PH-F140SP_BK_BLED + 5 x PH-F140SP_BK E22 Rigid 10/12 Acrylic Tubing and Bitspower Ma... 
CoolingOSMonitorKeyboard
2 x Reeven Six Eyes Fan Controller Windows 7 64 Bit Professional OEM ASUS VG248QE Black 23" 144 Hz Monitor Logitech G19s 920-004985 
PowerCaseMouseAudio
Seasonic X-1250 Phanteks Enthoo Primo Ultimate Chassis Asus RoG GX950 Laser Mouse  Logitech Z5500 
  hide details  
Reply
911 Carrera
(20 items)
 
  
CPUMotherboardGraphicsRAM
Intel w/ EK Supremacy Cooling Block - Clean P... Asus Maximus VI Formula 2 x Asus GTX 780 DCII w/ EK Full Cover Blocks a... Mushkin Red Line Ridgeback 2 x 8GB DDR302400 10... 
Hard DriveHard DriveOptical DriveCooling
2 x SAMSUNG 840 Pro Series MZ-7PD256BW 2.5" 256GB (2 x Seagate ST2000DX001 2TB 7200 RPM 64MB Cach... Asus BW-12B1ST/BLK/G/AS Blue Ray Writer 1 x Alphacool NexXxoS XT45-420 + 1 x Alphacool ... 
CoolingCoolingCoolingCooling
1 x Swiftech 35x2 w/ Heatsink and NB Fan EK -MultiOption RES X3 250 White Acetal Reservoir 5 x PH-F140SP_BK_BLED + 5 x PH-F140SP_BK E22 Rigid 10/12 Acrylic Tubing and Bitspower Ma... 
CoolingOSMonitorKeyboard
2 x Reeven Six Eyes Fan Controller Windows 7 64 Bit Professional OEM ASUS VG248QE Black 23" 144 Hz Monitor Logitech G19s 920-004985 
PowerCaseMouseAudio
Seasonic X-1250 Phanteks Enthoo Primo Ultimate Chassis Asus RoG GX950 Laser Mouse  Logitech Z5500 
  hide details  
Reply
post #6972 of 19539
Username:holyking
CPU Model: i7-4770k
Core Multiplier:48x
CPU VID: 1.285v
Vcore: 1.296v
Input Voltage:
Uncore Multiplier: 43x
Uncore Voltage:1.180v
Cooling Solution: Custom Water cooling
Stability Test: Aida64 @12h
Batch Number:L315B347
Ram Speed: T11 ( will overclock later)



Edited by holyking - 12/16/13 at 4:44pm
post #6973 of 19539
Quote:
Originally Posted by holykingΒ View Post



OK. I increased voltage. It crash the problem much faster. What this lead to? thank you

Ahh that's interesting. Try dropping vcore down to where you had it before, and increase VRIN (VCCIN / eventual Input voltage) by 0.05v smile.gif

edit: I see in your screenshot above that VCCIN is only at 1.68v. Increase VCCIN to 1.85v and try x264 again smile.gif
post #6974 of 19539
Quote:
Originally Posted by fleetfeatherΒ View Post

Ahh that's interesting. Try dropping vcore down to where you had it before, and increase VRIN (VCCIN / eventual Input voltage) by 0.05v smile.gif

edit: I see in your screenshot above that VCCIN is only at 1.68v. Increase VCCIN to 1.85v and try x264 again smile.gif

I will work on that
post #6975 of 19539
Thread Starter 
Quote:
Originally Posted by JackNaylorPEΒ View Post

The link posted gives a very clear explanation as to the effects and concerns, so no need to repeat ... the other links clearly show a diversity of opinion on the matter. This opinion varies based upon what you do with your machine....and as I have been saying, what your goals are. This is no different than memory .... you may get a higher OC if you dial down ya memory speed timings and this may be well worthwhile on CPU intensive tasks but may be detrimental on memory intensive tasks. AutoDesk products in particularly are very "system wide" intensive and even puzzling .... things that I wouldn't have expected but actually improved / decreased performance over the years included:

-elimination of page file decreased performance even when there was oodles of RAM available
-was using a separate drive for page / temp file usage improved far more than I expected when that partition was changed to FAT32

I'm a bit busy with GFX testing at the moment to take on any additional testing .... what I am finding so far tho

-Alternate BIOS oft bring higher measured clocks, but all other things being equal, there are no performance gains associated therewith (that I have seen so far)
-Continually increasing clocks and memory at some point has diminishing returns and, while still stable, going further on, even has negative results ... I have only done 5 runs of each benchmark and I have been hampered / puzzled by the fact that having GPU-z open during the test can cause a Open GL error that I don't get when it's closed.

A diversity of opinion doesn't mean there isn't a right answer. There is a diversity of opinion on whether astrology is valid or not, doesn't give any merit to astrology. JJ from Asus has his own opinions. and everybody that follows his guide typically shares his opinion. There is diversity of opinion on Prime95 as well, since JJ said it's not 'certified' for Haswell. There is a lot of diversity of opinion. But that's irrelevant, I stick to hard data when I can. Every single person I've met that goes on and on about some bottleneck also have a striking tendency to show no evidence.

Β 

Now, I don't know what Β AutoDesk is. If AutoDesk is some application you use all the time on your machine, that's one thing. But unless you're going to knock me for my own results being flat out invalid, I've demonstrated for all the benchmarks/applications I used that uncore won't matter as much as core pretty much ever. And, even if indeed it is true that AutoDesk is something you use often and rely on, it doesn't prove 1:1 Cache ratio or the uncore being 3 under core bottleneck are valid. To demonstrate those claims, you have to demonstrate the performance taking a nosedive or giant speedup at the specified points. And no proof yet either that AutoDesk is indeed super uncore sensitive.

Β 

But for your original post to have merit, the app that is affected by uncore greatly needs to be something most people use. And most people don't use AutoDesk. So you're going to have to pick application(s) which most people probably use that is/are strongly affected by uncore. Otherwise you'd be talking about a niche program as Cyro mentioned. I'm sure we can design a benchmark that is specifically designed to detect uncore changes above all else, but that doesn't mean much if real world programs don't represent that level of sensitivity to uncore change.

Β 

Chess and x264, well, those are the two standard CPU benchmarks as far as I'm concerned. Rendering videos and brute force calculation, that's the most common intensive CPU use I think most people run into.

Β 

Quote:
Originally Posted by holykingΒ View Post

Username:holyking
CPU Model: i7-4770k
Core Multiplier:48x
CPU VID: 1.285v
Vcore: 1.296v
Input Voltage:
Uncore Multiplier: 43x
Uncore Voltage:1.180v
Cooling Solution: Custom Water cooling
Stability Test: Aida64 @12h
Batch Number:L315B347
Ram Speed: T11 ( will overclock later)


You will be charted, thank you.

Β 

Edit:

Picture verification accepted, thanks for following instructions and making myΒ life easier!

Quote:
Originally Posted byΒ fleetfeatherΒ View Post


Ahh that's interesting. Try dropping vcore down to where you had it before, and increase VRIN (VCCIN / eventual Input voltage) by 0.05vΒ smile.gif

edit: I see in your screenshot above that VCCIN is only at 1.68v. Increase VCCIN to 1.85v and try x264 againΒ smile.gif
Your new avatar is SWEET.Β :drool:

Edited by Darkwizzie - 12/16/13 at 8:18pm
Undelwalt (2017)
(29 items)
 
 
Celapaleis (2013)
(16 items)
 
CPUMotherboardGraphicsRAM
7600k @ 5.197/1.392 1.392v z170 Asus Hero EVGA 1080ti SC Black (2100/6250) x2 8gb GSkill Trident Z 3600 @ 3804 15-15-15-32-2T 
Hard DriveHard DriveHard DriveCooling
Samsung 950 Pro 512gb (Undelwalt) WD Red 2tb (Pack Yak II) Seagate Expansion Drive 5tb (Phoenix Down II) 3x560mm HardwareLabs Nemesis GTX 
CoolingCoolingCoolingCooling
XPSC Raystorm Pro Watercool Heatkiller 1080ti Full Cover Waterblock EK XTOP Revo Dual D5 (Serial, PWM, v4) EK ZMT (1/2 - 3/4 ID OD) 
CoolingCoolingCoolingCooling
14xEK ACF (Compression Fittings) EK x4 250 (v2) Distilled Water + PT Nuke (Copper Sulphate) 24x140mm Silent Wings 3 
OSMonitorKeyboardPower
Windows 10 Pro 64bit Catleap 27 Inch 2560x1440 IPS 60hz Display Coolermaster Storm Trigger (Brown Switches) EVGA P2 1000w 
CaseMouseMouse PadAudio
Corsair Air 540 Logitech G Pro Tek Syndicate "Raze the World" Desk Mat O2 + Odac by Mayflower Electronics (Rev A) 
AudioAudioAudioAudio
Sennheiser HD 800 2xKrk Rokit 6 G2 Blue Yeti (Grey) Rode PSA-1 Arm 
Other
Aquaero 6 LT + Aluminum Heatsink 
CPUMotherboardGraphicsRAM
i5 6600k Delid @ 4.848/4.848 Asus z170 Hero MSI GTX 980ti @ 1499/4082 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
Undelwalt (2017)
(29 items)
 
 
Celapaleis (2013)
(16 items)
 
CPUMotherboardGraphicsRAM
7600k @ 5.197/1.392 1.392v z170 Asus Hero EVGA 1080ti SC Black (2100/6250) x2 8gb GSkill Trident Z 3600 @ 3804 15-15-15-32-2T 
Hard DriveHard DriveHard DriveCooling
Samsung 950 Pro 512gb (Undelwalt) WD Red 2tb (Pack Yak II) Seagate Expansion Drive 5tb (Phoenix Down II) 3x560mm HardwareLabs Nemesis GTX 
CoolingCoolingCoolingCooling
XPSC Raystorm Pro Watercool Heatkiller 1080ti Full Cover Waterblock EK XTOP Revo Dual D5 (Serial, PWM, v4) EK ZMT (1/2 - 3/4 ID OD) 
CoolingCoolingCoolingCooling
14xEK ACF (Compression Fittings) EK x4 250 (v2) Distilled Water + PT Nuke (Copper Sulphate) 24x140mm Silent Wings 3 
OSMonitorKeyboardPower
Windows 10 Pro 64bit Catleap 27 Inch 2560x1440 IPS 60hz Display Coolermaster Storm Trigger (Brown Switches) EVGA P2 1000w 
CaseMouseMouse PadAudio
Corsair Air 540 Logitech G Pro Tek Syndicate "Raze the World" Desk Mat O2 + Odac by Mayflower Electronics (Rev A) 
AudioAudioAudioAudio
Sennheiser HD 800 2xKrk Rokit 6 G2 Blue Yeti (Grey) Rode PSA-1 Arm 
Other
Aquaero 6 LT + Aluminum Heatsink 
CPUMotherboardGraphicsRAM
i5 6600k Delid @ 4.848/4.848 Asus z170 Hero MSI GTX 980ti @ 1499/4082 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 #6976 of 19539
Ty for temp results Darkwizzie! What kinda ambients did you have? (10c or 30c)
Insert Name Here
(14 items)
 
  
CPUMotherboardGraphicsRAM
6700k Asus Maximus VIII Hero Gigabyte Aorus Xtreme 1080ti Corsair LPX 2x8GB 3200c16 
Hard DriveHard DriveCoolingOS
Old Seagate HDD Samsung 850 EVO Thermalright Silver Arrow SB-E SE Windows 7 Home Premium 64 bit 
MonitorKeyboardPowerCase
Asus PG258Q (240hz + Gsync) WASDKeyboards.com v1 semi custom w/ mx browns, ... Superflower Golden Green HX550 Air540 
MouseMouse Pad
Logitech G Pro Qck+ 
  hide details  
Reply
Insert Name Here
(14 items)
 
  
CPUMotherboardGraphicsRAM
6700k Asus Maximus VIII Hero Gigabyte Aorus Xtreme 1080ti Corsair LPX 2x8GB 3200c16 
Hard DriveHard DriveCoolingOS
Old Seagate HDD Samsung 850 EVO Thermalright Silver Arrow SB-E SE Windows 7 Home Premium 64 bit 
MonitorKeyboardPowerCase
Asus PG258Q (240hz + Gsync) WASDKeyboards.com v1 semi custom w/ mx browns, ... Superflower Golden Green HX550 Air540 
MouseMouse Pad
Logitech G Pro Qck+ 
  hide details  
Reply
post #6977 of 19539
Yep, delidding may be risky business... but it's certainly worth it lol. The value it adds to your chip in terms of performance is outrageous
post #6978 of 19539
I know that a lot of people hate on intel for it, but I for one am glad they aren't using soldered ihs' s direct die cooling is going to net me better results anf a soldered ihs ever will.
post #6979 of 19539
I appear to be getting ignored in the VI Hero thread, so I will post my question here.
Quote:
Does anyone know how to change the VCCIN or input voltage on this motherboard?
I need 1,355v to get a 4,4ghz stable OC on my 4670k. The temps don't go higher than 80C when stress testing, however a lot of people say that for this voltage I would have needed a custom water loop, not a Dark Rock Pro 2 (what I have now).
I opened up HWinfo and saw this: Be warned mobile users. (Click to show)
The screenshot was made during Prime95 with large FTT's.

And apparently ASUS loves making users confused because in the BIOS I set this:
Be warned once again mobile users. (Click to show)
This is the core voltage I set.
And again. (Click to show)


Yes, those were THREE options with what the VCCIN could be. However I would think none of them is, because of the earlier stated reasons.

Is HWinfo reading this out correctly? Did the VCCIN really mess up my OC and can I actually go higher? And how do I get the ACTUAL input voltage to be send to the CPU, also when the system booted up into windows?
post #6980 of 19539
On Asus ROG mobo's, VCCIN is called "Eventual CPU Input Voltage".

The reason your temps aren't breaking 80C with 1.35v is because you're using HWInfo64 to monitor core temperatures. HWInfo64 underestimates core temperatures by roughly 10C (in my experience). A much better estimate of core temperatures can be provided by HWMonitor. Once you inspect core voltages with HWMonitor, you'll probably see temperatures in the mid-80's, which is why those members in the ROG forum are suggesting you need a better cooling solution for 1.35v. However, 85C in a synthetic environment such as Prime95 isn't really a huge concern anyway, so I'm not sure why the ROG owners are stressing so much (pun intended).

Additionally, keep in mind that since you have a 4670k rather than a 4770k, you'll also experience lower temps when compared to a 4770k at the same Vcore, due to the lack of Hyper Threading (HT increases core temps). This may be another reason why those members in the ROG forum are surprised by your temps.

As far as the additional settings you've changed in your BIOS, you can:

- keep CPU Voltage Override as 1.35v (this is your Vcore value)
- return Initial CPU Input Voltage to "auto" (this voltage only needs tweaking if you fail to post under extreme environments)
- keep Eventual CPU Input Voltage as 1.90v (this is VCCIN as I previously mentioned)
- return VCCIN Shadow Voltage to "auto" (this voltage parameter only needs tweaking in unusual circumstances)

Hopefully that answers all your concerns smile.gif
Quote:
Originally Posted by DarkwizzieΒ View Post

snip

Sadly, the mods didn't like my avatar frown.gif A decent replacement has been sourced though smile.gif
Edited by fleetfeather - 12/17/13 at 8:30am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › Haswell Overclocking Guide [With Statistics]