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

Haswell Overclocking Guide [With Statistics] - Page 518

post #5171 of 19540
Doug2507, those lower temp number is super tempting. I also toy with the idea of delid but just not comfortable enough yet risking any damage to the chip and the wrath of the wife redface.gif . I am hoping a custom water cool set up alone might give me a little over 10c lower temp since under stress test I am hitting upper 80s and mid 90s. Rendering, encoding and editing I can hit up to mid 70s.
ThermalTron
(16 items)
 
  
CPUMotherboardGraphicsRAM
4790k MSI Z87 Mpower MSI 7870 Team Xtreem Dark 
Hard DriveHard DriveCoolingOS
Western digital Samsung 840 pro H100 Windows 8 64bit enterprise 
MonitorMonitorMonitorKeyboard
Apple HCD Dell LG Corsair K90 
PowerCaseMouseMouse Pad
Kingwin 850 Platinum Thermaltake Chaser MK1 TT Theron Roccat Sense Glacier Blue 
  hide details  
Reply
ThermalTron
(16 items)
 
  
CPUMotherboardGraphicsRAM
4790k MSI Z87 Mpower MSI 7870 Team Xtreem Dark 
Hard DriveHard DriveCoolingOS
Western digital Samsung 840 pro H100 Windows 8 64bit enterprise 
MonitorMonitorMonitorKeyboard
Apple HCD Dell LG Corsair K90 
PowerCaseMouseMouse Pad
Kingwin 850 Platinum Thermaltake Chaser MK1 TT Theron Roccat Sense Glacier Blue 
  hide details  
Reply
post #5172 of 19540
Other thing to remember is what you're doing on the rig day to day. From the start i had intened to bench mine so knew i'd be pushing high load a lot of the time. Before delid i was hitting 90deg at 1.25v with 27.9, XTU would have been less, and that was with a Phanteks (i reckon they're quite good!). Delid dropped things down to 70deg with 27.9. Add on custom loop (3x180mm {FT02}) and i'm hitting that with IBT on max. Normal usage i'm around 40-60deg on load.

Defo worth it as lower temps=longer life but as far as delidding goes, if there's not the voltage headroom to OC further then i don't see there being any point.
post #5173 of 19540
Quote:
Originally Posted by Ali Man View Post

There's no way for it not to be stable at x40 with 1.200V, so I stick to my sentence.

To be fair, my uncore/ring multi is a bit sketchy at voltages close to that. Doesn't work on 1.15v and i'm pretty sure i got occasional issues on 1.175
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 #5174 of 19540
Quote:
Originally Posted by Cyro999 View Post

To be fair, my uncore/ring multi is a bit sketchy at voltages close to that. Doesn't work on 1.15v and i'm pretty sure i got occasional issues on 1.175

This isn't the first time that you've mentioned this, I think you've mentioned this before too and the only thing that comes to mind is your unique case that just representative of itself, rather than comparing it with the general population of values.

What I'd say is that it's not exactly possible to gain unstability by raising the voltage, especially at those lower volts.
Quote:
Originally Posted by Doug2507 View Post

Maybe so but it can have a knock on effect with vrin. Iif it was stable with 1.2v@x43 then 1.2v@x40 is overvolting for that multi. The whole idea of dropping uncore down (& RAM) is to isolate them whilst OC'ing core. As core OC increases, so can the required vring (and possibly vrin) for uncore, never mind SA/IOA/IOD.

Each to their own then.thumb.gif

I wouldn't necessary call it 'overvolting' as the end outcome is to get the VCore stable and that would more like just be a guarantee of keeping the things towards the uncore side, rather solid. Later on, of course, it can be decreased to find the lowest for that too, if you see what I'm trying to say.

In other words, the most obvious fact of matter is that if he 'doesn't BSOD with a lower uncore and that same 1.200V voltage, then it means that he was actually suffering from a lower uncore value. It would make no sense to change the Uncore and its voltage value 'together', unless one has solid evidence that how much each uncore multiplier takes.

Definitely to each his own, but some things are just common knowledge to overclocking.
post #5175 of 19540
Quote:
This isn't the first time that you've mentioned this, I think you've mentioned this before too and the only thing that comes to mind is your unique case that just representative of itself, rather than comparing it with the general population of values.

What I'd say is that it's not exactly possible to gain unstability by raising the voltage, especially at those lower volts.

No i don't mean that (ring volts for core stability) i mean i need like ~1.18 ring to stabilize 40x uncore (which i'm using because i can do that and have drop to 800mhz at idle) and ~1.26 for 44x AFAIK

^at least, on my 46x which is rock solid stable
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 #5176 of 19540
Quote:
Originally Posted by Ali Man View Post

This isn't the first time that you've mentioned this, I think you've mentioned this before too and the only thing that comes to mind is your unique case that just representative of itself, rather than comparing it with the general population of values.

What I'd say is that it's not exactly possible to gain unstability by raising the voltage, especially at those lower volts.
I wouldn't necessary call it 'overvolting' as the end outcome is to get the VCore stable and that would more like just be a guarantee of keeping the things towards the uncore side, rather solid. Later on, of course, it can be decreased to find the lowest for that too, if you see what I'm trying to say.

In other words, the most obvious fact of matter is that if he 'doesn't BSOD with a lower uncore and that same 1.200V voltage, then it means that he was actually suffering from a lower uncore value. It would make no sense to change the Uncore and its voltage value 'together', unless one has solid evidence that how much each uncore multiplier takes.

Definitely to each his own, but some things are just common knowledge to overclocking.

I see where you're coming from but dropping it down will completely remove it from the equation to determine if core is 100% stable 1st which i would question in this instance. I guess both ways are just 2 of many in skinning a cat.thumb.gif

I think it would also be preferred to increase voltage to find stability rather than decrease it. For one it'll take longer to decrease than increase and as a consequence, the side effect of having the system under 100% load for longer.
post #5177 of 19540
Quote:
Originally Posted by Cyro999 View Post

No i don't mean that (ring volts for core stability) i mean i need like ~1.18 ring to stabilize 40x uncore (which i'm using because i can do that and have drop to 800mhz at idle) and ~1.26 for 44x AFAIK

^at least, on my 46x which is rock solid stable

Well that's definitely good for you man, not sure what the comment was about then.
Not sure if I'm missing something out.
Quote:
Originally Posted by Doug2507 View Post

I see where you're coming from but dropping it down will completely remove it from the equation to determine if core is 100% stable 1st which i would question in this instance. I guess both ways are just 2 of many in skinning a cat.

He's getting 'clock watch dog errors', they occur with less VRin or less uncore voltage, hence, I said what I did.
Quote:
I think it would also be preferred to increase voltage to find stability rather than decrease it. For one it'll take longer to decrease than increase and as a consequence, the side effect of having the system under 100% load for longer.

Also what I said, but what you previously referred to as 'overvolting'.
post #5178 of 19540
I was quoting this
Quote:
There's no way for it not to be stable at x40 with 1.200V
because AFAIK it's within possibility for a chip significantly worse than mine to be unstable due to uncore w/ 40x uncore and 1.2 ring if mine is completely not-fine at 1.15
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 #5179 of 19540
Quote:
Originally Posted by Cyro999 View Post

I was quoting this
because AFAIK it's within possibility for a chip significantly worse than mine to be unstable due to uncore w/ 40x uncore and 1.2 ring if mine is completely not-fine at 1.15

Alright, while your point is sound, it would still be a unique case for that as I haven't seen any as yet.
post #5180 of 19540
Hm really? It's commonplace for people to run a decent core clock and 4ghz uncore on notably less than ~1.18 ring?
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
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › Haswell Overclocking Guide [With Statistics]