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

Haswell Overclocking Guide [With Statistics] - Page 549

post #5481 of 19539
Quote:
Originally Posted by [CyGnus] View Post

creos7 you will not notice the difference with less those 100MHz even if you were at 4.2 the games would have the same fps they are very GPU bound now a days...

Emulators still benefit from an overclocked CPU, especially Dolphin. Many games on Dolphin run stable or great at stock settings, but some games need a pretty good OC to remove stuttering frames or audio. There's also a reason why people use BF3 and BF4 to determine stable overclocks. Most games are much more reliant on the GPU by a significantly greater degree, but not all are. It also has very little to do with FPS.
Edited by Transylvania - 11/16/13 at 8:02am
post #5482 of 19539
Edit- wrote that I was stable at 4.3 noticed my cpu multi was still set at 34 -.-, retesting, cant get 4.4 at 1.25-1.30, tried 1.35 for the hell of it, crashed at the end of loop 5 temps 65-70
Edited by Germx3 - 11/16/13 at 8:20am
post #5483 of 19539
Quote:
Originally Posted by Germx3 View Post

I have my Cpu stable at 4.3 multi 1.25 vcore 1.8 vccin 34 uncore 1.15v ran 87.4% vdroop offset 10 loops of x264 no bsod, ran 4.3 on 1.24 vcore did only 2 loops of x264 will test stability at 1.24 again. Tried my luck with 4.4, 1.25-1.30v bsod 101 error 2-5 min in x264. I decided to throw 1.35v at it to see what happens ran 5 loops and it crashed at the end of loop 5 with temps 65-72c mad.gif . Any suggestions on how to get 4.4 stable? No 3b error yet.

Are you lowering your ram speed and cache/uncore when trying higher multipliers? Simply raising the vcore by itself isn't going to really give you a good picture if you are.

I'm still fine tuning my own settings (currently working on Vccin), but my last stable setting at 4.4 (x264 x 20 and 8 hours with XTU) was:

Vcore: 1.35v (x44)
Uncore: 1.34v (x39) < seems off, but I can't go above x39 at that voltage, nor can I lower the voltage at that multiplier.
Vccin: 1.93 < set it higher initially to make sure that wasn't going to be the problem. Currently testing it at 1.90.
Ram: 1366mhz until everything else is stable.
Also using a Kraken x60, so with those two tests, my temps don't go above 72C on the hottest core. I don't even bother with Prime now because my temps go up to 86C, and that's not a temp I'm comfortable leaving the PC unmonitored.

I'm also not sure what's up with x264, but almost every time I ended up with a BSOD using that, it was always at the end of the very last loop. Every single time. It didn't matter if it was 5 loops or 20 either. I also seem to have a single core that's 6C lower than the hottest core, while the rest are 2C-3C within the same temp. I haven't noticed such a disparity with any other stress testing program either.
Edited by Transylvania - 11/16/13 at 8:34am
post #5484 of 19539
Quote:
Originally Posted by Transylvania View Post

I also seem to have a single core that's 6C lower than the hottest core, while the rest are 2C-3C within the same temp. I haven't noticed such a disparity with any other stress testing program either.

I think this is a function of the lack of solder and the cheap thermal compound Intel is using it -- it's really a matter of luck how it's distributed in the factory. My hottest core is on average 8C (and sometimes as much as 11C!!) hotter than my coolest core. Very similar boat as you, two cores are in the middle, one is about 3-4C cooler and one is 4-5C hotter. I'm sure if I delid, I could bring everything down the coolest core temps and be able to push to 4.7Ghz but that's just too much risk for me.
post #5485 of 19539
Quote:
Originally Posted by Transylvania View Post

Are you lowering your ram speed and cache/uncore when trying higher multipliers? Simply raising the vcore by itself isn't going to really give you a good picture if you are.

I'm still fine tuning my own settings (currently working on Vccin), but my last stable setting at 4.4 (x264 x 20 and 8 hours with XTU) was:

Vcore: 1.35v (x44)
Uncore: 1.34v (x39) < seems off, but I can't go above x39 at that voltage, nor can I lower the voltage at that multiplier.
Vccin: 1.93 < set it higher initially to make sure that wasn't going to be the problem. Currently testing it at 1.90.
Ram: 1366mhz until everything else is stable.
Also using a Kraken x60, so with those two tests, my temps don't go above 72C on the hottest core. I don't even bother with Prime now because my temps go up to 86C, and that's not a temp I'm comfortable leaving the PC unmonitored.

I'm also not sure what's up with x264, but almost every time I ended up with a BSOD using that, it was always at the end of the very last loop. Every single time. It didn't matter if it was 5 loops or 20 either. I also seem to have a single core that's 6C lower than the hottest core, while the rest are 2C-3C within the same temp. I haven't noticed such a disparity with any other stress testing program either.

Yah 34 uncore 1600mhz ram now using 1333, 1.15vring, 1.8 vccin, tested 10 loops of x264 no bsod on 4.3, 1.25 vcore, 1.24vcore bsod. Going to leave it at these settings for now. Also using 87.4% vdroop offset, Should i start increasing vccin and vring to get 4.4 stable, or just vccin?

I've experienced the crash at the last loop at 4.4 1.35vcore thats interesting...
post #5486 of 19539
Why are you adding 0.1v for 100mhz, and which crash?
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 #5487 of 19539
Quote:
Originally Posted by Darkwizzie View Post

Did you read the first post?

Yes, I did. In its entirety.

I noticed that very few members have the Costa Rica chips as opposed to the Malaysia chips. Which is why I'm asking.
Project X99 mATX
(13 items)
 
   
CPUMotherboardGraphicsRAM
Intel Core i7-5820k EVGA X99 Micro mATX EVGA GTX 980 SC ACX 2.0 Corsair Vengeance LPX 16GB DDR4 
Hard DriveCoolingOSMonitor
Samsung 850 Pro 512GB SSD Corsair H110 Windows 8.1 Pro ASUS ROG Swift PG278Q 
KeyboardPowerCaseMouse
Corsair K70 RGB Corsair HX850i Corsair 350D Logitech G700s 
Audio
Logitech G930 
CPUMotherboardGraphicsRAM
Intel Core i7-4770k @ 4.5 GHz ASUS Maximus VI Formula 2 x EVGA GTX 780 Ti in SLI 16GB G.SKILL Trident X DDR3-2400 
Hard DriveCoolingCoolingCooling
Samsung 840 256GB SSD EK-Supremacy Clean CSQ 2 x EK-FC780 GTX Ti - Acrylic EK-FC Terminal Dual Parallel 3-Slot 
CoolingCoolingCoolingOS
XSPC Photon 170 w/ D5 Vario Pump Alphacool NexXxoS UT60 280mm Alphacool NexXxoS XT45 360mm Windows 7 Professional 64-bit 
MonitorKeyboardPowerCase
ASUS PB278Q 1440p LED Logitech G510 Corsair AX1200i Corsair Obsidian 750D 
Mouse
Logitech G700s 
  hide details  
Reply
Project X99 mATX
(13 items)
 
   
CPUMotherboardGraphicsRAM
Intel Core i7-5820k EVGA X99 Micro mATX EVGA GTX 980 SC ACX 2.0 Corsair Vengeance LPX 16GB DDR4 
Hard DriveCoolingOSMonitor
Samsung 850 Pro 512GB SSD Corsair H110 Windows 8.1 Pro ASUS ROG Swift PG278Q 
KeyboardPowerCaseMouse
Corsair K70 RGB Corsair HX850i Corsair 350D Logitech G700s 
Audio
Logitech G930 
CPUMotherboardGraphicsRAM
Intel Core i7-4770k @ 4.5 GHz ASUS Maximus VI Formula 2 x EVGA GTX 780 Ti in SLI 16GB G.SKILL Trident X DDR3-2400 
Hard DriveCoolingCoolingCooling
Samsung 840 256GB SSD EK-Supremacy Clean CSQ 2 x EK-FC780 GTX Ti - Acrylic EK-FC Terminal Dual Parallel 3-Slot 
CoolingCoolingCoolingOS
XSPC Photon 170 w/ D5 Vario Pump Alphacool NexXxoS UT60 280mm Alphacool NexXxoS XT45 360mm Windows 7 Professional 64-bit 
MonitorKeyboardPowerCase
ASUS PB278Q 1440p LED Logitech G510 Corsair AX1200i Corsair Obsidian 750D 
Mouse
Logitech G700s 
  hide details  
Reply
post #5488 of 19539
Quote:
Originally Posted by Cyro999 View Post

Why are you adding 0.1v for 100mhz, and which crash?

Is 100mhz not worth the 0.1v? Im just trying to get a good idea of what the chip can do...and i was talking about x264 crashing at the end of the final loop.
post #5489 of 19539
is there an update on the costume loop on the x264 sorry if i missed it tongue.gif
post #5490 of 19539
Quote:
Originally Posted by idahosurge View Post


For people who do not have experience with Haswell and OCCT you should download and install OCCT then run the OCCT CPU:OCCT 1 hour test for Large Data Set 64 bits. It is harder to pass than the OCCT CPU:Linpack AVX 64 bits test.
http://www.ocbase.com/index.php/download
Yeah thats how I was stress testing my OC, I didn't had BSOD with x264 but OCCT gave me plenty tongue.gif
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › Haswell Overclocking Guide [With Statistics]