Overclock.net › Forums › Intel › Intel CPUs › The Intel Devil's Canyon Owners Club
New Posts  All Forums:Forum Nav:

The Intel Devil's Canyon Owners Club - Page 1985

post #19841 of 22136
Yes, 300mv is excessive. That VCCSA boost. that I needed for the 4770k was not much - it put it li 900 - 950 mv range. I guess I am saying I wouldn't rule out small bump like that. At the same time I would think that if he had 1.5v RAM there is something else wrong if he needs 1.65v.
post #19842 of 22136
Quote:
Originally Posted by GeneO View Post

Yes, 300mv is excessive. That VCCSA boost. that I needed for the 4770k was not much - it put it li 900 - 950 mv range. I guess I am saying I wouldn't rule out small bump like that. At the same time I would think that if he had 1.5v RAM there is something else wrong if he needs 1.65v.

I don't know if it came out like that, but I never meant he doesn't need ANY boost in VCCSA. I just said he doesn't need that much. thumb.gif

At this point, without knowing more details about his particular system, I can't make any more guesses.
post #19843 of 22136
Tolis626 - I use 1 kit 2x4GB DDR3 1600 CL9 1.5v Kingston PnP 9-9-9-27
and 1 kit 2x4GB DDR3 1600 CL9 1.5V kingston hyperx Savage red 9-9-9-27

Savage is XMP but i disable XMP profile and add timmings manual to 9-9-9-27. (This timmings add and MB with use XMP profile but i disable XMP - first kit kingston PnP is not a xmp.)

Today i test and only with 1 (first pnp kit) without different, this time i test with 1.2 cashe V, 1.800 input and + 0.100 Vccsa and I/Os

after 1 hour i give you the result, before with old settings system crash in 7 loop x264
Edited by madboy128 - 2/13/16 at 1:45pm
post #19844 of 22136
10 loops x264 4900mhz @ 1.30 vcore !!!! Thanks TOLIS ! More Cache V and low input ...
Max temp is 70C* after 10 loops x264.

Time to 5ghz ...

vtt 1.13v and 1.13v vccsa is safe ?
Edited by madboy128 - 2/13/16 at 2:53pm
post #19845 of 22136
my 4790K@ 5,1Ghz

and g3258@5,08GHz



graduated.gif
Edited by stjepanj - 2/13/16 at 3:57pm
    
CPUMotherboardGraphicsRAM
I7 4790K "Polished" Gigabyte G1 Sniper Z87 MSI Gtx 970 Gaming OC Edition @ 1647/2150mhz Kingston 
Hard DriveHard DriveHard DriveCooling
Barracuda  Kingston SSD V-300 Wd Black Noctua NH-D15 Polished 
OSKeyboardPowerCase
Windows 7 64bit BenQ Gw2760 Xfx 550w 80+ Zalman Z3+ 
MouseMouse Pad
Madcatz R.A.T. 3 White Cm Storm L 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
I7 4790K "Polished" Gigabyte G1 Sniper Z87 MSI Gtx 970 Gaming OC Edition @ 1647/2150mhz Kingston 
Hard DriveHard DriveHard DriveCooling
Barracuda  Kingston SSD V-300 Wd Black Noctua NH-D15 Polished 
OSKeyboardPowerCase
Windows 7 64bit BenQ Gw2760 Xfx 550w 80+ Zalman Z3+ 
MouseMouse Pad
Madcatz R.A.T. 3 White Cm Storm L 
  hide details  
Reply
post #19846 of 22136
Quote:
Originally Posted by madboy128 View Post

10 loops x264 4900mhz @ 1.30 vcore !!!! Thanks TOLIS ! More Cache V and low input ...
Max temp is 70C* after 10 loops x264.

Time to 5ghz ...

vtt 1.13v and 1.13v vccsa is safe ?

VTT at 1.13V should be safe. 1.13V for SA is kinda too much though, I wouldn't go over 1.05V for 24/7. Why don't you try them at stock and then work upwards from there? Also try running without the high cache voltage and see what happens.
post #19847 of 22136
Quote:
Originally Posted by stjepanj View Post

my 4790K@ 5,1Ghz

and g3258@5,08GHz



graduated.gif

And, are these stable OC's?

If they are, what stress program did you use, at what settings, and duration? proof.gif
Watson III
(31 items)
 
Watson V
(24 items)
 
 
CPUMotherboardGraphicsRAM
Intel i7 6700K ASUS Maximus VIII Gene Asus GeForce GTX 1080 Founders Edition Corsair Dominator Platinum 3000MHz C15 
Hard DriveOptical DriveCoolingCooling
Samsung 950 Pro NVMe 512GB Lite-On w/ Lite-Scribe Watercool D5 Bitspower Black Sparkle 
CoolingCoolingCoolingCooling
3 - Koolance QD3 Quiick Disconnects Hardware Labs Black Ice Nemesis 360 GTX Primochill Adavanced LTR 3/5 x 5/8 Bool Red Tubing 3 - Noctua NF-F12's Industrial 3000 RPM Push/In... 
CoolingCoolingOSMonitor
2 - Noctua NF-F12's 1500 RPM as front intake Swiftech Helix 1800 RPM PWM 120mm fan as exhaust Windows 7 Ultimate 64bit ASUS VC239H 
MonitorKeyboardPowerCase
ASUS VS228H-P HP Mechanical PS/2 Corsair AX860 Platinum CaseLabs S5 Two-Tone White/Gunmetal 
MouseMouse PadAudioAudio
Logitech M510 Staples Boston Acoustics Bowers & Wilkins P5 Series II Headphones 
  hide details  
Reply
Watson III
(31 items)
 
Watson V
(24 items)
 
 
CPUMotherboardGraphicsRAM
Intel i7 6700K ASUS Maximus VIII Gene Asus GeForce GTX 1080 Founders Edition Corsair Dominator Platinum 3000MHz C15 
Hard DriveOptical DriveCoolingCooling
Samsung 950 Pro NVMe 512GB Lite-On w/ Lite-Scribe Watercool D5 Bitspower Black Sparkle 
CoolingCoolingCoolingCooling
3 - Koolance QD3 Quiick Disconnects Hardware Labs Black Ice Nemesis 360 GTX Primochill Adavanced LTR 3/5 x 5/8 Bool Red Tubing 3 - Noctua NF-F12's Industrial 3000 RPM Push/In... 
CoolingCoolingOSMonitor
2 - Noctua NF-F12's 1500 RPM as front intake Swiftech Helix 1800 RPM PWM 120mm fan as exhaust Windows 7 Ultimate 64bit ASUS VC239H 
MonitorKeyboardPowerCase
ASUS VS228H-P HP Mechanical PS/2 Corsair AX860 Platinum CaseLabs S5 Two-Tone White/Gunmetal 
MouseMouse PadAudioAudio
Logitech M510 Staples Boston Acoustics Bowers & Wilkins P5 Series II Headphones 
  hide details  
Reply
post #19848 of 22136
I really do hope the board can cope with it. According to the actual VCore (1.312) , I'm beyond manual max input.(1.299) biggrin.gif
Whatever.. lol So far it works. run benches and stresses and daily gaming and it works. biggrin.gif


Melinda V 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core I7 7700K MSI Z270 GAMING PRO MSI GTX 1070 GAMING Z 16 GB CORSAIR DOMINATOR PLATINUM DDR4 3000 
Hard DriveHard DriveHard DriveCooling
Samsung 960 EVO NVMe 250 GB Samsung 850 EVO 1 TB Samsung 850 EVO 250 GB Corsair H110i GTX 
OSMonitorKeyboardPower
Windows 10 Pro x64 BENQ EW2400 Logitech mk 365 (Wireless Kit incl. Mouse) Cooler Master V750 750 Watts 80 Plus Gold 
CaseMouse PadOtherOther
Fractal Desgn R5 MSI Gaming 2x Noctua NF-F12 Case Fans 1x Fractal Design 140mm Case Fan 
  hide details  
Reply
Melinda V 2.0
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core I7 7700K MSI Z270 GAMING PRO MSI GTX 1070 GAMING Z 16 GB CORSAIR DOMINATOR PLATINUM DDR4 3000 
Hard DriveHard DriveHard DriveCooling
Samsung 960 EVO NVMe 250 GB Samsung 850 EVO 1 TB Samsung 850 EVO 250 GB Corsair H110i GTX 
OSMonitorKeyboardPower
Windows 10 Pro x64 BENQ EW2400 Logitech mk 365 (Wireless Kit incl. Mouse) Cooler Master V750 750 Watts 80 Plus Gold 
CaseMouse PadOtherOther
Fractal Desgn R5 MSI Gaming 2x Noctua NF-F12 Case Fans 1x Fractal Design 140mm Case Fan 
  hide details  
Reply
post #19849 of 22136
THE EFFING WOES OF PRIME 95

Background: I don't use p95 on higher overclocks, but it's pretty effective at low overclocks when not thermally limited. Or so I thought.

Someone or other on some overclocking forum or other recommended using prime95's actual prime-searching method for p95 stressing. A few weeks(?) ago I was trying to narrow down my uncore voltages needed, so I started using it. This prime search is usually cool as p95 goes, comparable to the 1344-1344 that's often recommended, though it can occasionally get up a fair bit hotter. I've never seen it anywhere near what you can get with a tiny FFT size though. So it seemed effective. p95 also seems to be more stressful on uncore and ram than other tests.

It takes ~10 days of running to check a prime number (seems strange to me, but hey) and I left it going overnight for some of my more serious tests. So by the time I was done with my uncore tests I was about 50% in on some of the threads (each of the 4 threads seems to be independent, and they go at slightly different speeds).

That's when all hell broke loose. The overclock I'd been using to test uncore, 4.5 ghz at 1.25V, started crashing. My summer overclock, 4.5 ghz at 1.22V, crashed. I raised input voltage and core voltage and cache voltage and it still crashed. I dropped back to stock and it didn't crash.

I thought maybe my chip was damaged somehow, but an x264 stress revealed the exact same stability levels as before - 4.5 ghz at 1.21V crashes after hours of x264, and 1.22V is stable.

So something in the variability of the p95 test is causing an absurd amount of stress. With settings that are stable for everyday use and x264 and p95 1344-1344, it typically crashes within 3 minutes. THREE MINUTES! The very regularity of it made me think it might be crashing (by which I mean BSOD of course) at the exact same point every time. I was reminded of the skylake AVX+HT bug that only shows up in certain prime sizes in p95. I can't be sure but getting 4.5 ghz stable needs around 1.27V, a good .05V more than I need for x264 stability and .06V more than I need for p95 1344-1344 stability. Temps are in the low 60s on average and ~never go over 70C with this level of overclock under p95.

But...only 4 days of computing left and I'm sure I'll find a new Mersenne prime! Can't stop now! I've dropped to 4.4 ghz though.
Dorje
(12 items)
 
  
CPUMotherboardGraphicsRAM
4690k Gigabyte z97x-sli MSI TF 280 G Skill Sniper 1600/9 
Hard DriveCoolingOSMonitor
Seagate 4 tb hybrid Corsair h80i Windows 8.1 SE39UY04 
PowerCaseMouseAudio
EVGA NEX-G 750W S340 (white) Bloody ZL5A Razer Blackshark 
  hide details  
Reply
Dorje
(12 items)
 
  
CPUMotherboardGraphicsRAM
4690k Gigabyte z97x-sli MSI TF 280 G Skill Sniper 1600/9 
Hard DriveCoolingOSMonitor
Seagate 4 tb hybrid Corsair h80i Windows 8.1 SE39UY04 
PowerCaseMouseAudio
EVGA NEX-G 750W S340 (white) Bloody ZL5A Razer Blackshark 
  hide details  
Reply
post #19850 of 22136
How does adaptive voltage work?

I have my 4690k stable @ 4.5 1.180V but it bugs me that using override does not lower the voltage at idle even though all C states are enable.
Is there a way to circumvent this using adaptive voltage? As in setting a max peak so it won't go too high?

Sent from my GT-I9505 using Tapatalk
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › The Intel Devil's Canyon Owners Club