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

Haswell Overclocking Guide [With Statistics] - Page 1895

post #18941 of 19540
Quote:
Originally Posted by QuacK View Post

Warning: Spoiler! (Click to show)


This is a realbench screenshot of my score at 4.9GHz.
At this core speed I still needed only 1.8v input voltage to pass the benchmark.
If i lower it or set it higher by even just 0.010v it will fail. If I change cache voltage just a little bit, it will fail also.

Now here I finally have a screenshot for Realbench @ 5GHz



For 5GHz to pass the benchmark I had to raise input all the way up to 2.270 in the bios which gave me 2.304v under full load.
Also I had to raise IOD /IOA by 0.24 which somehow translates to a 1.265 system agent voltage (dont know if anyone else has this kind of thing on a Gigabyte SOC Force)

I tested my input voltage in 0.010 increments from 1.780 set in bios.
Took me a ****load of time to get from 4.9GHz to 5, but im trying to find out why it needs so much more input voltage at 5GHz. So I can try to lower it, because I dont really wanna keep it at 2.3v and I dont believe this should be necessary for a 5GHz overclock since 4.9GHz still runs fine with 1.8v input tongue.gif

I tried:

Lowering cache multiplier
lowering cache voltage
higher cache multiplier,
higher cache voltage
Lowering Ram speed and voltages

But it will not pass with anything less than 2.304v input voltage along with the other voltage settings you can see in the screenshot.


Am I missing something here or is this just some strange chip I got here? lol rolleyes.gif
And also, anyone might know why my score result at 4.9GHz is like 40000 higher than the 5GHz benchmark run? thumbsdownsmileyanim.gif



You need over 1.4V for 4.9 GHz? That will kill your chip right?
The 1.35V I needed was too much I thought.
Arachnid
(26 items)
 
Lil bro's
(9 items)
 
 
CPUMotherboardGraphicsRAM
6700k Asus Maximus VIII Hero Gigabyte GeForce GTX 980Ti 6GB Windforce 3X OC ... Gskill Trident Z 16 GB 
Hard DriveHard DriveHard DriveCooling
Samsung SM951 Seagate Barracuda Western Digital Blue EK Supremacy MX 
CoolingCoolingCoolingCooling
EK-FC980 GTX Ti WF3 – Nickel Barrow YKD5G65 130 mm Barrow PD5BTZ-V1 Barrow LD5HK-V1 
CoolingCoolingCoolingOS
Alphacool VPP665 Alphacool Nexxxos XT45 480 mm Scythe Gentle Typhoon 1850rpm Windows 10 
MonitorKeyboardPowerCase
LG 29UM55 Logitech G710+ Cooler Master V700 Phanteks Eclipse P400 
MouseAudioAudioOther
Logitech G502 Logitech Z506 HyperX Cloud II Fanatec GT3 RS V2 EU 
OtherOther
Fanatec Clubsport V2 EU Xbox One Wireless Controller with Windows Adapter 
CPUMotherboardGraphicsRAM
E5450 Asus Commando Palit GTX 770 Hynix HYMP125U64CP8-S6 AB 
Hard DriveCoolingOSPower
WD Caviar Raijintek Triton Windows 10 BeQuiet BQT E6-500W 
Case
Golden Field V5S 
  hide details  
Reply
Arachnid
(26 items)
 
Lil bro's
(9 items)
 
 
CPUMotherboardGraphicsRAM
6700k Asus Maximus VIII Hero Gigabyte GeForce GTX 980Ti 6GB Windforce 3X OC ... Gskill Trident Z 16 GB 
Hard DriveHard DriveHard DriveCooling
Samsung SM951 Seagate Barracuda Western Digital Blue EK Supremacy MX 
CoolingCoolingCoolingCooling
EK-FC980 GTX Ti WF3 – Nickel Barrow YKD5G65 130 mm Barrow PD5BTZ-V1 Barrow LD5HK-V1 
CoolingCoolingCoolingOS
Alphacool VPP665 Alphacool Nexxxos XT45 480 mm Scythe Gentle Typhoon 1850rpm Windows 10 
MonitorKeyboardPowerCase
LG 29UM55 Logitech G710+ Cooler Master V700 Phanteks Eclipse P400 
MouseAudioAudioOther
Logitech G502 Logitech Z506 HyperX Cloud II Fanatec GT3 RS V2 EU 
OtherOther
Fanatec Clubsport V2 EU Xbox One Wireless Controller with Windows Adapter 
CPUMotherboardGraphicsRAM
E5450 Asus Commando Palit GTX 770 Hynix HYMP125U64CP8-S6 AB 
Hard DriveCoolingOSPower
WD Caviar Raijintek Triton Windows 10 BeQuiet BQT E6-500W 
Case
Golden Field V5S 
  hide details  
Reply
post #18942 of 19540
Quote:
Originally Posted by jdorje View Post

That looks about 10 degrees too hot to me. Does pushing down (gently) on the water block/heat sink lower temps? Are fans running on normal mode or silent mode? Is the AIO on intake? Is the airflow blocked in any way? What are the water temps (hwinfo can read this via corsair link)?
Why are those too high given how far away it is from the maximum per Intel? Is it because it is only a 4-core? My temperatures on a Noctua NH-D15S are high 60s under any stress testing I've tried (I've not messed with small FFTs), with infrequent peaks in the low 70s, but I have a 6-core with hyperthreading enabled when I test. Keep in mind he has an AIO watercooler, so temperatures won't be as low as a custom loop and will be similar to the top-end air coolers.

Edit: Or is he delidded and I missed it?
Edited by stargate125645 - 4/4/16 at 8:29am
BladeRunner v3.0
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7-5930K @ 4.6GHz Core, 4.4GHz Cache ASUS X99 Sabertooth Sapphire R9 380 Dual-X OC G.Skill TridentZ 32GB DDR4 @ 13-15-13-33-1T 320... 
Hard DriveCoolingOSKeyboard
Samsung 850 Pro 512GB Noctua NH-D15S Windows 10 Home 64-bit Logitech G910 Orion Spark 
PowerCaseMouse
EVGA SuperNova 1000W T2 NZXT Phantom 820 Black Logitech G5 
  hide details  
Reply
BladeRunner v3.0
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7-5930K @ 4.6GHz Core, 4.4GHz Cache ASUS X99 Sabertooth Sapphire R9 380 Dual-X OC G.Skill TridentZ 32GB DDR4 @ 13-15-13-33-1T 320... 
Hard DriveCoolingOSKeyboard
Samsung 850 Pro 512GB Noctua NH-D15S Windows 10 Home 64-bit Logitech G910 Orion Spark 
PowerCaseMouse
EVGA SuperNova 1000W T2 NZXT Phantom 820 Black Logitech G5 
  hide details  
Reply
post #18943 of 19540
Quote:
Originally Posted by mrgnex View Post

[/spoiler]

You need over 1.4V for 4.9 GHz? That will kill your chip right?
The 1.35V I needed was too much I thought.

If I would use Prime or IBT at this voltage then yes, that would kill my chip.
Thats why I switched to using Realbench to get stable for my use, also because I noticed if it can pass a realbench run, I rarely get any bsod's with gaming or anything else.

With the voltages im using now (1.44v vcore and 2.3 input) im still a few degrees under 80 degrees during gaming, and usually I just game for 2 hours or 3 at most so shouldnt be much of a problem I think.
But I do want to try to lower mainly the input voltage,and if possible vcore. When gaming, Vcore doesn't goes over 1.428v at these settings, but Realbench spikes it up to 1.44v.

Im just really wondering why at 4.9GHz I can use 1.8v input but at 5GHz I need so much more for just 100mhz more. Im thinking it can't be right but ill have to test more I guess. Or maybe im at the end and this is as good as it can get for 5GHz passing realbench, I dunno lol rolleyes.gif
post #18944 of 19540
Quote:
Originally Posted by stargate125645 View Post

Why are those too high given how far away it is from the maximum per Intel? Is it because it is only a 4-core? My temperatures on a Noctua NH-D15S are high 60s under any stress testing I've tried (I've not messed with small FFTs), with infrequent peaks in the low 70s, but I have a 6-core with hyperthreading enabled when I test. Keep in mind he has an AIO watercooler, so temperatures won't be as low as a custom loop and will be similar to the top-end air coolers.

Edit: Or is he delidded and I missed it?
I didn't delid, I may in the future though. I think my temps are just how my chip runs to be honest. With total default settings in bios when I stress on x264 my temps only hit the high 40'school to very low 50'speed and ove been told that normal. It's just once I put an overclock on the chip I get higher temps (know thatso normal but sounds like I may be a tad higher than normal expected temp increase for the volts and clock.) At 4.5 ghz 1.21v I get in the mid 60's to hottest core being around 70c.
post #18945 of 19540
Quote:
Originally Posted by stargate125645 View Post

Why are those too high given how far away it is from the maximum per Intel? Is it because it is only a 4-core? My temperatures on a Noctua NH-D15S are high 60s under any stress testing I've tried (I've not messed with small FFTs), with infrequent peaks in the low 70s, but I have a 6-core with hyperthreading enabled when I test. Keep in mind he has an AIO watercooler, so temperatures won't be as low as a custom loop and will be similar to the top-end air coolers.

Edit: Or is he delidded and I missed it?

Well the temps are fine if you don't overclock. I just guess they are hotter than they should be from his setup.
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 #18946 of 19540
Quote:
Originally Posted by jdorje View Post

Quote:
Originally Posted by stargate125645 View Post

Why are those too high given how far away it is from the maximum per Intel? Is it because it is only a 4-core? My temperatures on a Noctua NH-D15S are high 60s under any stress testing I've tried (I've not messed with small FFTs), with infrequent peaks in the low 70s, but I have a 6-core with hyperthreading enabled when I test. Keep in mind he has an AIO watercooler, so temperatures won't be as low as a custom loop and will be similar to the top-end air coolers.

Edit: Or is he delidded and I missed it?

Well the temps are fine if you don't overclock. I just guess they are hotter than they should be from his setup.
So would you call mine high, then? What part of his setup makes you consider them high? That's what I was asking before.
BladeRunner v3.0
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7-5930K @ 4.6GHz Core, 4.4GHz Cache ASUS X99 Sabertooth Sapphire R9 380 Dual-X OC G.Skill TridentZ 32GB DDR4 @ 13-15-13-33-1T 320... 
Hard DriveCoolingOSKeyboard
Samsung 850 Pro 512GB Noctua NH-D15S Windows 10 Home 64-bit Logitech G910 Orion Spark 
PowerCaseMouse
EVGA SuperNova 1000W T2 NZXT Phantom 820 Black Logitech G5 
  hide details  
Reply
BladeRunner v3.0
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7-5930K @ 4.6GHz Core, 4.4GHz Cache ASUS X99 Sabertooth Sapphire R9 380 Dual-X OC G.Skill TridentZ 32GB DDR4 @ 13-15-13-33-1T 320... 
Hard DriveCoolingOSKeyboard
Samsung 850 Pro 512GB Noctua NH-D15S Windows 10 Home 64-bit Logitech G910 Orion Spark 
PowerCaseMouse
EVGA SuperNova 1000W T2 NZXT Phantom 820 Black Logitech G5 
  hide details  
Reply
post #18947 of 19540
Quote:
Originally Posted by stargate125645 View Post

So would you call mine high, then? What part of his setup makes you consider them high? That's what I was asking before.

I have no idea your voltage or clock so no clue on that. And I don't really know 5820k (?) cooling.

He said he was running a 4690k at 4.5@1.21 with an h110i and was maxing around 70C under x264. Based on my own 4690k that I've run at 4.5@1.22 with an h80i (pre-delid) I said I thought that looked high (can't compare that any more, but post delid I'm about 55C on those settings). But he then dropped it down to stock and the temps looked a lot more like what I'd expect so I figured I must be wrong.

Small things that you don't think to mention can make a big difference on cooling obviously. AIO on intake versus exhaust makes a huge difference in gaming temps (you always want intake basically). Change in ambient can make a proportional difference. Any time I write down a number I try to normalize it to a 20C ambient first. But some people have 40C ambients and that changes their 70C to 90C.
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 #18948 of 19540
Quote:
Originally Posted by jdorje View Post

Quote:
Originally Posted by stargate125645 View Post

So would you call mine high, then? What part of his setup makes you consider them high? That's what I was asking before.

I have no idea your voltage or clock so no clue on that. And I don't really know 5820k (?) cooling.

He said he was running a 4690k at 4.5@1.21 with an h110i and was maxing around 70C under x264. Based on my own 4690k that I've run at 4.5@1.22 with an h80i (pre-delid) I said I thought that looked high (can't compare that any more, but post delid I'm about 55C on those settings). But he then dropped it down to stock and the temps looked a lot more like what I'd expect so I figured I must be wrong.

Small things that you don't think to mention can make a big difference on cooling obviously. AIO on intake versus exhaust makes a huge difference in gaming temps (you always want intake basically). Change in ambient can make a proportional difference. Any time I write down a number I try to normalize it to a 20C ambient first. But some people have 40C ambients and that changes their 70C to 90C.
That explanation is more what I was getting at. I didn't know the basis for your comment, so it's hard to gauge how to apply it to my situation (or if I need to). I didn't think those temperatures seemed overly bad, personally, so I wanted more information from your perspective.

(As for my rig, all the info is in my signature rig setup. To save you some clicks, I am using 1.23V for core. Given my 6-core processor vs. your/his 4-core, that I am using high-end air [roughly equivalent to AIO water], and am at higher frequencies, I would presume my temperatures are fine.)
Edited by stargate125645 - 4/4/16 at 1:27pm
BladeRunner v3.0
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7-5930K @ 4.6GHz Core, 4.4GHz Cache ASUS X99 Sabertooth Sapphire R9 380 Dual-X OC G.Skill TridentZ 32GB DDR4 @ 13-15-13-33-1T 320... 
Hard DriveCoolingOSKeyboard
Samsung 850 Pro 512GB Noctua NH-D15S Windows 10 Home 64-bit Logitech G910 Orion Spark 
PowerCaseMouse
EVGA SuperNova 1000W T2 NZXT Phantom 820 Black Logitech G5 
  hide details  
Reply
BladeRunner v3.0
(11 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7-5930K @ 4.6GHz Core, 4.4GHz Cache ASUS X99 Sabertooth Sapphire R9 380 Dual-X OC G.Skill TridentZ 32GB DDR4 @ 13-15-13-33-1T 320... 
Hard DriveCoolingOSKeyboard
Samsung 850 Pro 512GB Noctua NH-D15S Windows 10 Home 64-bit Logitech G910 Orion Spark 
PowerCaseMouse
EVGA SuperNova 1000W T2 NZXT Phantom 820 Black Logitech G5 
  hide details  
Reply
post #18949 of 19540
Ok so i think there's an issue with my overclock. Even though I can run the stability test just fine at 4.5 at 1.21v I noticed that when I go into a game or something my graphics card down clocks itself at that voltage. But if I up them to say like 1.25v it don't seem to do that. Can not enough volts but being at the cusp cause something like this where the processor acts up?
post #18950 of 19540
That would usually show in benchmarks.
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
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › Haswell Overclocking Guide [With Statistics]