Overclock.net › Forums › Intel › Intel CPUs › [Official] The Sandy Stable Club **Guides, Voltages, Temps & BIOS Templates** Inc SPREADSHEET
New Posts  All Forums:Forum Nav:

[Official] The Sandy Stable Club **Guides, Voltages, Temps & BIOS Templates** Inc SPREADSHEET - Page 646

post #6451 of 10702
Quote:
Originally Posted by shad0wfax View Post




The reason your GFlops is decreasing is that you're hitting a "silent" throttle at the CPU level. You're exceeding either the temperature, current, or power limitations and your CPU is cycling itself short (while keeping its full clock speed) in order to avoid harming itself. You can either decrease your voltage or decrease your overclock and you should see this change.

 

My guess is that it's a maximum current issue and not a temperature issue.

 

I managed to get my OC more efficient by reducing the Load Line Calibration, as that adds more current to the system for a given voltage, on my motherboard.

 

The reason that no other test shows this is probably because IBT stresses your system harder than any other floating point test that i know of. You're probably not hitting the power/current wall with other tests but IBT is enough to kick it up over the edge. Back down just a tiny bit and you should be rock-solid stable in the other tests and more efficient in IBT.

 


Are you certain about this? I just compared ITB with HT on and off and got similar results to error-id10t. I tried a couple things to get my current down and with every setting the non HT runs were had higher GFlops.
X299 Evolv'd
(13 items)
 
unRAID Server
(13 items)
 
 
CPUMotherboardGraphicsRAM
i7-7820x @ 4.8GHz Gigabyte Aorus X299 Gaming 7 EVGA 1080 ti FTW3 g.skill TridentZ RGB 3600 CL16 
Hard DriveHard DriveCoolingOS
2 x Samsung 960 EVO 500gb Raid 0 Samsung 850 EVO 1TB H115i w 4 x Noiseblocker B14 PS Push/Pull Windows 10 Pro 
MonitorKeyboardPowerCase
AOC Agon AG352UCG 3440x1440 100hz Ultra Wide Corsair K70 EVGA Supernova G2 850w Phanteks Evolv ATX Tempered Glass 
Mouse
Logitech G900 
  hide details  
Reply
X299 Evolv'd
(13 items)
 
unRAID Server
(13 items)
 
 
CPUMotherboardGraphicsRAM
i7-7820x @ 4.8GHz Gigabyte Aorus X299 Gaming 7 EVGA 1080 ti FTW3 g.skill TridentZ RGB 3600 CL16 
Hard DriveHard DriveCoolingOS
2 x Samsung 960 EVO 500gb Raid 0 Samsung 850 EVO 1TB H115i w 4 x Noiseblocker B14 PS Push/Pull Windows 10 Pro 
MonitorKeyboardPowerCase
AOC Agon AG352UCG 3440x1440 100hz Ultra Wide Corsair K70 EVGA Supernova G2 850w Phanteks Evolv ATX Tempered Glass 
Mouse
Logitech G900 
  hide details  
Reply
post #6452 of 10702
Quote:
Originally Posted by TahoeDust View Post

Are you certain about this? I just compared ITB with HT on and off and got similar results to error-id10t. I tried a couple things to get my current down and with every setting the non HT runs were had higher GFlops.



I don't know what the code for Linpack is like. (IBT uses Linpack). Let me do some digging on Linpack and see if it's optimized for Hyper-Threading to be 100% sure about this.

 

"*As of v2.1, Core i7 users that wish to test with HyperThreading enabled can now override the number of threads for Linpack to execute."  (emphasis added by me)

 

Make sure that you're doing that by clicking on the "All" button for the cores to use in IBT. It shouldn't be giving you such a performance hit as long as you're using IBT v 2.53 or better but there's a chance that it's just so poorly optimized that your GFlops will suck with HT enabled no matter what.

 

EDIT: Some users are reporting a 4% drop in GFlops with HT enabled in IBT. His drops are more like 20% which is a bit odd. What are your overclocks at?


Edited by shad0wfax - 1/2/12 at 1:00am
B2
(8 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core i7-5930K Haswell-E 6-Core 3.5GHz LGA... EVGA X99 Classified (151-HE-E999-KR) LGA 2011-v... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... 
RAMHard DriveCoolingCase
G.SKILL Ripjaws 4 series 16GB (4 x 4GB) 288-Pin... SAMSUNG 850 EVO MZ-75E1T0B/AM 2.5" 1TB SATA III... CORSAIR Hydro Series H100i Extreme Performance ... Phanteks Enthoo Luxe Series PH-ES614L_BK Black ... 
  hide details  
Reply
B2
(8 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core i7-5930K Haswell-E 6-Core 3.5GHz LGA... EVGA X99 Classified (151-HE-E999-KR) LGA 2011-v... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... 
RAMHard DriveCoolingCase
G.SKILL Ripjaws 4 series 16GB (4 x 4GB) 288-Pin... SAMSUNG 850 EVO MZ-75E1T0B/AM 2.5" 1TB SATA III... CORSAIR Hydro Series H100i Extreme Performance ... Phanteks Enthoo Luxe Series PH-ES614L_BK Black ... 
  hide details  
Reply
post #6453 of 10702
Quote:
Originally Posted by shad0wfax View Post




I don't know what the code for Linpack is like. (IBT uses Linpack). Let me do some digging on Linpack and see if it's optimized for Hyper-Threading to be 100% sure about this.

I just did two pulls at stock speed and default bios settings.

HT: 74 Gflops
non-HT: 88 Gflops

Pretty interesting.
X299 Evolv'd
(13 items)
 
unRAID Server
(13 items)
 
 
CPUMotherboardGraphicsRAM
i7-7820x @ 4.8GHz Gigabyte Aorus X299 Gaming 7 EVGA 1080 ti FTW3 g.skill TridentZ RGB 3600 CL16 
Hard DriveHard DriveCoolingOS
2 x Samsung 960 EVO 500gb Raid 0 Samsung 850 EVO 1TB H115i w 4 x Noiseblocker B14 PS Push/Pull Windows 10 Pro 
MonitorKeyboardPowerCase
AOC Agon AG352UCG 3440x1440 100hz Ultra Wide Corsair K70 EVGA Supernova G2 850w Phanteks Evolv ATX Tempered Glass 
Mouse
Logitech G900 
  hide details  
Reply
X299 Evolv'd
(13 items)
 
unRAID Server
(13 items)
 
 
CPUMotherboardGraphicsRAM
i7-7820x @ 4.8GHz Gigabyte Aorus X299 Gaming 7 EVGA 1080 ti FTW3 g.skill TridentZ RGB 3600 CL16 
Hard DriveHard DriveCoolingOS
2 x Samsung 960 EVO 500gb Raid 0 Samsung 850 EVO 1TB H115i w 4 x Noiseblocker B14 PS Push/Pull Windows 10 Pro 
MonitorKeyboardPowerCase
AOC Agon AG352UCG 3440x1440 100hz Ultra Wide Corsair K70 EVGA Supernova G2 850w Phanteks Evolv ATX Tempered Glass 
Mouse
Logitech G900 
  hide details  
Reply
post #6454 of 10702
Quote:
Originally Posted by TahoeDust View Post

Quote:
Originally Posted by shad0wfax View Post




I don't know what the code for Linpack is like. (IBT uses Linpack). Let me do some digging on Linpack and see if it's optimized for Hyper-Threading to be 100% sure about this.

I just did two pulls at stock speed and default bios settings.

HT: 74 Gflops
non-HT: 88 Gflops

Pretty interesting.


That sounds better. That's about an 8% drop. Compare that to the percent drop in your Overclock and I think that my assumption about throttling may still be valid.

 

My 4.7 GHz 2500K will IBT between 126.88 and 127.06 GFlops consistently. (Obviously I have no HT.)

B2
(8 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core i7-5930K Haswell-E 6-Core 3.5GHz LGA... EVGA X99 Classified (151-HE-E999-KR) LGA 2011-v... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... 
RAMHard DriveCoolingCase
G.SKILL Ripjaws 4 series 16GB (4 x 4GB) 288-Pin... SAMSUNG 850 EVO MZ-75E1T0B/AM 2.5" 1TB SATA III... CORSAIR Hydro Series H100i Extreme Performance ... Phanteks Enthoo Luxe Series PH-ES614L_BK Black ... 
  hide details  
Reply
B2
(8 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core i7-5930K Haswell-E 6-Core 3.5GHz LGA... EVGA X99 Classified (151-HE-E999-KR) LGA 2011-v... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... 
RAMHard DriveCoolingCase
G.SKILL Ripjaws 4 series 16GB (4 x 4GB) 288-Pin... SAMSUNG 850 EVO MZ-75E1T0B/AM 2.5" 1TB SATA III... CORSAIR Hydro Series H100i Extreme Performance ... Phanteks Enthoo Luxe Series PH-ES614L_BK Black ... 
  hide details  
Reply
post #6455 of 10702
Quote:
Originally Posted by shad0wfax View Post



That sounds better. That's about an 8% drop. Compare that to the percent drop in your Overclock and I think that my assumption about throttling may still be valid.

My 4.7 GHz 2500K will IBT between 126.88 and 127.06 GFlops consistently. (Obviously I have no HT.)

Why would it throttle at stock clock speed and bios settings? It still returned higher Gflops with HT disabled.
Edited by TahoeDust - 1/2/12 at 1:14am
X299 Evolv'd
(13 items)
 
unRAID Server
(13 items)
 
 
CPUMotherboardGraphicsRAM
i7-7820x @ 4.8GHz Gigabyte Aorus X299 Gaming 7 EVGA 1080 ti FTW3 g.skill TridentZ RGB 3600 CL16 
Hard DriveHard DriveCoolingOS
2 x Samsung 960 EVO 500gb Raid 0 Samsung 850 EVO 1TB H115i w 4 x Noiseblocker B14 PS Push/Pull Windows 10 Pro 
MonitorKeyboardPowerCase
AOC Agon AG352UCG 3440x1440 100hz Ultra Wide Corsair K70 EVGA Supernova G2 850w Phanteks Evolv ATX Tempered Glass 
Mouse
Logitech G900 
  hide details  
Reply
X299 Evolv'd
(13 items)
 
unRAID Server
(13 items)
 
 
CPUMotherboardGraphicsRAM
i7-7820x @ 4.8GHz Gigabyte Aorus X299 Gaming 7 EVGA 1080 ti FTW3 g.skill TridentZ RGB 3600 CL16 
Hard DriveHard DriveCoolingOS
2 x Samsung 960 EVO 500gb Raid 0 Samsung 850 EVO 1TB H115i w 4 x Noiseblocker B14 PS Push/Pull Windows 10 Pro 
MonitorKeyboardPowerCase
AOC Agon AG352UCG 3440x1440 100hz Ultra Wide Corsair K70 EVGA Supernova G2 850w Phanteks Evolv ATX Tempered Glass 
Mouse
Logitech G900 
  hide details  
Reply
post #6456 of 10702
Quote:
Originally Posted by TahoeDust View Post

Why would it throttle at stock clock speed and bios settings? It still returned higher Gflops with HT disabled.


Linpack isn't optimized for HT so losing about 4% of your GFlops with HT enabled is what other users have experienced. Where I was thinking that throttling occurred was when error-id10t said that he went from 118 GFlops down to 94 GFlops with HT enabled. That's far more of a performance loss than he should be seeing and that makes me wonder if he's hitting a wall somewhere in his settings.

 

HT is primarily for multi-tasking and Linpack was optimized to test physical cores not virtual cores. That's why many people leave HT off for Linpack. (Although having it on is a good way to test your thermal stability.) The same goes for Prime95, many will disable HT since it doesn't have much of an effect on clock stability, it simply consumes more power and adds more heat. So when you're testing an OC on a 2600K (or any other HT capable CPU) you can disable HT while you dial in your OC until it's stable and then test with HT on to make sure that your thermal output is controllable at that OC. Linpack (and thus IBT) is essentially the undisputed king of heating up your CPU.

 

Long story short, a 4-6% loss with HT makes sense and a 20.3% loss (118 GFlops down to 94 GFlops) tells me something's not quite right with the settings and some throttling is occurring for error-id10t.

 

 

B2
(8 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core i7-5930K Haswell-E 6-Core 3.5GHz LGA... EVGA X99 Classified (151-HE-E999-KR) LGA 2011-v... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... 
RAMHard DriveCoolingCase
G.SKILL Ripjaws 4 series 16GB (4 x 4GB) 288-Pin... SAMSUNG 850 EVO MZ-75E1T0B/AM 2.5" 1TB SATA III... CORSAIR Hydro Series H100i Extreme Performance ... Phanteks Enthoo Luxe Series PH-ES614L_BK Black ... 
  hide details  
Reply
B2
(8 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core i7-5930K Haswell-E 6-Core 3.5GHz LGA... EVGA X99 Classified (151-HE-E999-KR) LGA 2011-v... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... EVGA 04G-P4-3975-KR GeForce GTX 970 SSC ACX 2.0... 
RAMHard DriveCoolingCase
G.SKILL Ripjaws 4 series 16GB (4 x 4GB) 288-Pin... SAMSUNG 850 EVO MZ-75E1T0B/AM 2.5" 1TB SATA III... CORSAIR Hydro Series H100i Extreme Performance ... Phanteks Enthoo Luxe Series PH-ES614L_BK Black ... 
  hide details  
Reply
post #6457 of 10702
At stock, 88 Gflops to 74 Gflops is a ~16% drop. With the OC, 121 Gflops to 104 Glops is a 14% drop. So the performance difference is pretty consistent.
X299 Evolv'd
(13 items)
 
unRAID Server
(13 items)
 
 
CPUMotherboardGraphicsRAM
i7-7820x @ 4.8GHz Gigabyte Aorus X299 Gaming 7 EVGA 1080 ti FTW3 g.skill TridentZ RGB 3600 CL16 
Hard DriveHard DriveCoolingOS
2 x Samsung 960 EVO 500gb Raid 0 Samsung 850 EVO 1TB H115i w 4 x Noiseblocker B14 PS Push/Pull Windows 10 Pro 
MonitorKeyboardPowerCase
AOC Agon AG352UCG 3440x1440 100hz Ultra Wide Corsair K70 EVGA Supernova G2 850w Phanteks Evolv ATX Tempered Glass 
Mouse
Logitech G900 
  hide details  
Reply
X299 Evolv'd
(13 items)
 
unRAID Server
(13 items)
 
 
CPUMotherboardGraphicsRAM
i7-7820x @ 4.8GHz Gigabyte Aorus X299 Gaming 7 EVGA 1080 ti FTW3 g.skill TridentZ RGB 3600 CL16 
Hard DriveHard DriveCoolingOS
2 x Samsung 960 EVO 500gb Raid 0 Samsung 850 EVO 1TB H115i w 4 x Noiseblocker B14 PS Push/Pull Windows 10 Pro 
MonitorKeyboardPowerCase
AOC Agon AG352UCG 3440x1440 100hz Ultra Wide Corsair K70 EVGA Supernova G2 850w Phanteks Evolv ATX Tempered Glass 
Mouse
Logitech G900 
  hide details  
Reply
post #6458 of 10702
2 hours into 1.45v @ 4.8ghz testing...if this doesn't work I'm gonna have to lower the multiplier frown.gif don't wanna get any higher with the voltage.
DavidGamingPC
(14 items)
 
  
Reply
DavidGamingPC
(14 items)
 
  
Reply
post #6459 of 10702
Thread Starter 
Quote:
Originally Posted by MichaelZERO View Post

Here is my, 16 hours blend
Hope I didn't break any rules. biggrin.gif

938

Thank you for participating, the screenshot is perfect and the overclock is amazing!!! Yet another golden chip on the spreadsheet. +rep for your efforts and for sharing this amazing overclock with us. 5ghz with a 212+ is impressive. Welcome to OCN and the club thumb.gif

You beat Lord Xeb and taken the 5ghz golden entry smile.gif

Have you tested the max multiplier? With those kind of voltages I bet you could go much higher if you're on water. As for max multi might be a 57 or even a 58.
Edited by munaim1 - 1/2/12 at 6:21am
it looks alive
(17 items)
 
Sandy Transformed
(69 photos)
 
CPUMotherboardGraphicsRAM
i5 2500K [5ghz 1.44v 24/7] Asus Maximus IV Extreme SLI GTX 460 @ 920/1840/4100 [1.11v] 4GB Dominator GT @ 2133mhz CL8 [1.65v] 
Hard DriveHard DriveCoolingOS
Intel X25-M G2 80GB 3x Samsung F3 500GB RAID 0 EK Supreme HF Full Copper Windows 7 Ultimate 64bit 
MonitorPowerCaseAudio
LG42SL9500 42" 1080p [Surround Soon] XFX XXX Edition 650w Xigmatec Elysium JVC HA-RX700 [foam mod] 
  hide details  
Reply
it looks alive
(17 items)
 
Sandy Transformed
(69 photos)
 
CPUMotherboardGraphicsRAM
i5 2500K [5ghz 1.44v 24/7] Asus Maximus IV Extreme SLI GTX 460 @ 920/1840/4100 [1.11v] 4GB Dominator GT @ 2133mhz CL8 [1.65v] 
Hard DriveHard DriveCoolingOS
Intel X25-M G2 80GB 3x Samsung F3 500GB RAID 0 EK Supreme HF Full Copper Windows 7 Ultimate 64bit 
MonitorPowerCaseAudio
LG42SL9500 42" 1080p [Surround Soon] XFX XXX Edition 650w Xigmatec Elysium JVC HA-RX700 [foam mod] 
  hide details  
Reply
post #6460 of 10702
Quote:
Originally Posted by FPSDavid View Post

2 hours into 1.45v @ 4.8ghz testing...if this doesn't work I'm gonna have to lower the multiplier frown.gif don't wanna get any higher with the voltage.

Good-luck!thumb.gif
 
Blue Ice
(11 items)
 
 
CPUGraphicsHard Drive
3610QM GT 630M Crucial M4 
CPUMotherboardGraphicsRAM
i7 920 Asus Gigbyte 7950 3x2GB OCZ 
Hard DriveCoolingMonitorKeyboard
Seagate prolimatech megahalems HP 24" Razor 
PowerCaseMouse
Antec 650 Earth Watts Corsair Graphite Series 600T Special Edition WHITE Razor 
  hide details  
Reply
 
Blue Ice
(11 items)
 
 
CPUGraphicsHard Drive
3610QM GT 630M Crucial M4 
CPUMotherboardGraphicsRAM
i7 920 Asus Gigbyte 7950 3x2GB OCZ 
Hard DriveCoolingMonitorKeyboard
Seagate prolimatech megahalems HP 24" Razor 
PowerCaseMouse
Antec 650 Earth Watts Corsair Graphite Series 600T Special Edition WHITE Razor 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
  • [Official] The Sandy Stable Club **Guides, Voltages, Temps & BIOS Templates** Inc SPREADSHEET
Overclock.net › Forums › Intel › Intel CPUs › [Official] The Sandy Stable Club **Guides, Voltages, Temps & BIOS Templates** Inc SPREADSHEET