Overclock.net › Forums › Cooling › Water Cooling › Ever Wondered What The Heat Diffrence is with Hyper Threading Turned On or Off?
New Posts  All Forums:Forum Nav:

Ever Wondered What The Heat Diffrence is with Hyper Threading Turned On or Off?

post #1 of 13
Thread Starter 
Well I was playing about before...

Ambient is the blue line @ 16oC water temp at 20/21oC (there is a temp sensor on the Rad outlet, and another on the Res inlet)



That's QUITE a lot!!!! its staying turned off lol, but I think the temps are good...

... are the temps good?
Red Dwarf
(33 items)
 
CPUMotherboardGraphicsGraphics
i7 3770K Maxium v Formula Palit GTX670 Jetstream Palit GTX670 Jetstream 
RAMHard DriveHard DriveHard Drive
XMS Vertex 4 Seragate Barracuda Seragate Barracuda 
Optical DriveCoolingCoolingCooling
Pioneer Blu-ray XSPC XP240 Alphacool 5.25" Res D5 Pump 
CoolingCoolingCoolingCooling
EK Comp Fittings PrimoChill PrimoFlex Advanced LRT Red Koolance Drain Plug Bitspower temp sensor 
CoolingCoolingCoolingCooling
IC Diamond Liquid Metal Ultra Corsair SP120 Corsair SP120 
CoolingOSMonitorMonitor
Koolance CPU-380i Windows 8 Pro Epson TW2900 Acer GD245HQ 
MonitorKeyboardPowerCase
150" Projector Screen (Blackout Blind) CYBORGE 7 OCZ 1250w Thermaltake Level 10 GT 
MouseMouse PadAudioAudio
R.A.T 9 Razor IronClad Tritton Pro+ Denon 1610 
Audio
Energy Speakers 
  hide details  
Reply
Red Dwarf
(33 items)
 
CPUMotherboardGraphicsGraphics
i7 3770K Maxium v Formula Palit GTX670 Jetstream Palit GTX670 Jetstream 
RAMHard DriveHard DriveHard Drive
XMS Vertex 4 Seragate Barracuda Seragate Barracuda 
Optical DriveCoolingCoolingCooling
Pioneer Blu-ray XSPC XP240 Alphacool 5.25" Res D5 Pump 
CoolingCoolingCoolingCooling
EK Comp Fittings PrimoChill PrimoFlex Advanced LRT Red Koolance Drain Plug Bitspower temp sensor 
CoolingCoolingCoolingCooling
IC Diamond Liquid Metal Ultra Corsair SP120 Corsair SP120 
CoolingOSMonitorMonitor
Koolance CPU-380i Windows 8 Pro Epson TW2900 Acer GD245HQ 
MonitorKeyboardPowerCase
150" Projector Screen (Blackout Blind) CYBORGE 7 OCZ 1250w Thermaltake Level 10 GT 
MouseMouse PadAudioAudio
R.A.T 9 Razor IronClad Tritton Pro+ Denon 1610 
Audio
Energy Speakers 
  hide details  
Reply
post #2 of 13
One point you are not considering is processing power. For the temp comparisons to be significant, the CPUs need to be performing a similar amount of work. You need to be able to measure/control mips/flops or at least compare temps to a benchmark. redface.gif
post #3 of 13
Thread Starter 
Quote:
Originally Posted by pgdeaner View Post

One point you are not considering is processing power. For the temp comparisons to be significant, the CPUs need to be performing a similar amount of work. You need to be able to measure/control mips/flops or at least compare temps to a benchmark. redface.gif

Both runs were Prime95, is that what you mean? each run was like for like, the exact same test at the same amb temp

Edit...

Do you mean

CPU HT On : 250,000FLOPS / 64oC
CPU HT Off : 245,000FLOPS / 51oC

Type of comp, as in even though it wasn't producing heat, it was also not as fast?

But saying that, most stuff we do on the CPU does not use 8 threads anyway :/
Edited by Liqu1dShadow - 3/25/13 at 12:54pm
Red Dwarf
(33 items)
 
CPUMotherboardGraphicsGraphics
i7 3770K Maxium v Formula Palit GTX670 Jetstream Palit GTX670 Jetstream 
RAMHard DriveHard DriveHard Drive
XMS Vertex 4 Seragate Barracuda Seragate Barracuda 
Optical DriveCoolingCoolingCooling
Pioneer Blu-ray XSPC XP240 Alphacool 5.25" Res D5 Pump 
CoolingCoolingCoolingCooling
EK Comp Fittings PrimoChill PrimoFlex Advanced LRT Red Koolance Drain Plug Bitspower temp sensor 
CoolingCoolingCoolingCooling
IC Diamond Liquid Metal Ultra Corsair SP120 Corsair SP120 
CoolingOSMonitorMonitor
Koolance CPU-380i Windows 8 Pro Epson TW2900 Acer GD245HQ 
MonitorKeyboardPowerCase
150" Projector Screen (Blackout Blind) CYBORGE 7 OCZ 1250w Thermaltake Level 10 GT 
MouseMouse PadAudioAudio
R.A.T 9 Razor IronClad Tritton Pro+ Denon 1610 
Audio
Energy Speakers 
  hide details  
Reply
Red Dwarf
(33 items)
 
CPUMotherboardGraphicsGraphics
i7 3770K Maxium v Formula Palit GTX670 Jetstream Palit GTX670 Jetstream 
RAMHard DriveHard DriveHard Drive
XMS Vertex 4 Seragate Barracuda Seragate Barracuda 
Optical DriveCoolingCoolingCooling
Pioneer Blu-ray XSPC XP240 Alphacool 5.25" Res D5 Pump 
CoolingCoolingCoolingCooling
EK Comp Fittings PrimoChill PrimoFlex Advanced LRT Red Koolance Drain Plug Bitspower temp sensor 
CoolingCoolingCoolingCooling
IC Diamond Liquid Metal Ultra Corsair SP120 Corsair SP120 
CoolingOSMonitorMonitor
Koolance CPU-380i Windows 8 Pro Epson TW2900 Acer GD245HQ 
MonitorKeyboardPowerCase
150" Projector Screen (Blackout Blind) CYBORGE 7 OCZ 1250w Thermaltake Level 10 GT 
MouseMouse PadAudioAudio
R.A.T 9 Razor IronClad Tritton Pro+ Denon 1610 
Audio
Energy Speakers 
  hide details  
Reply
post #4 of 13
I get a 10-12c increase if I turn on HT. Seeing as though Im at 73c (Intelburntest) with it off, it stays off. biggrinsmiley.gif
post #5 of 13
does HT really give enough of a performance both to warrant using it? i might have to turn mine off till i get the volenti finished.
post #6 of 13
I'm not turning HT off because why would I want to cripple my 3770k down to a 3570k? (Yeah I know i7 has a little more L3 cache, but as far as I'm aware it makes no difference).
post #7 of 13
Thread Starter 
Quote:
Originally Posted by dr/owned View Post

I'm not turning HT off because why would I want to cripple my 3770k down to a 3570k? (Yeah I know i7 has a little more L3 cache, but as far as I'm aware it makes no difference).

think of the heat you will save lol thumb.gif but its not "crippling" it, as you prob will see no difference, but when you need it you have the option to turn it on smile.gif
Red Dwarf
(33 items)
 
CPUMotherboardGraphicsGraphics
i7 3770K Maxium v Formula Palit GTX670 Jetstream Palit GTX670 Jetstream 
RAMHard DriveHard DriveHard Drive
XMS Vertex 4 Seragate Barracuda Seragate Barracuda 
Optical DriveCoolingCoolingCooling
Pioneer Blu-ray XSPC XP240 Alphacool 5.25" Res D5 Pump 
CoolingCoolingCoolingCooling
EK Comp Fittings PrimoChill PrimoFlex Advanced LRT Red Koolance Drain Plug Bitspower temp sensor 
CoolingCoolingCoolingCooling
IC Diamond Liquid Metal Ultra Corsair SP120 Corsair SP120 
CoolingOSMonitorMonitor
Koolance CPU-380i Windows 8 Pro Epson TW2900 Acer GD245HQ 
MonitorKeyboardPowerCase
150" Projector Screen (Blackout Blind) CYBORGE 7 OCZ 1250w Thermaltake Level 10 GT 
MouseMouse PadAudioAudio
R.A.T 9 Razor IronClad Tritton Pro+ Denon 1610 
Audio
Energy Speakers 
  hide details  
Reply
Red Dwarf
(33 items)
 
CPUMotherboardGraphicsGraphics
i7 3770K Maxium v Formula Palit GTX670 Jetstream Palit GTX670 Jetstream 
RAMHard DriveHard DriveHard Drive
XMS Vertex 4 Seragate Barracuda Seragate Barracuda 
Optical DriveCoolingCoolingCooling
Pioneer Blu-ray XSPC XP240 Alphacool 5.25" Res D5 Pump 
CoolingCoolingCoolingCooling
EK Comp Fittings PrimoChill PrimoFlex Advanced LRT Red Koolance Drain Plug Bitspower temp sensor 
CoolingCoolingCoolingCooling
IC Diamond Liquid Metal Ultra Corsair SP120 Corsair SP120 
CoolingOSMonitorMonitor
Koolance CPU-380i Windows 8 Pro Epson TW2900 Acer GD245HQ 
MonitorKeyboardPowerCase
150" Projector Screen (Blackout Blind) CYBORGE 7 OCZ 1250w Thermaltake Level 10 GT 
MouseMouse PadAudioAudio
R.A.T 9 Razor IronClad Tritton Pro+ Denon 1610 
Audio
Energy Speakers 
  hide details  
Reply
post #8 of 13
Quote:
Originally Posted by Noskcaj View Post

does HT really give enough of a performance both to warrant using it? i might have to turn mine off till i get the volenti finished.

It really depends on what you use your computer for. Most things will see little to no benefit.
post #9 of 13
Depending on the task, HT can add 25-35% faster computational performance. You should benchmark your tasks with it enabled and disabled, and then decide if you want it enabled or not. Lets give you a real world example...

Say you do an H.264 encode that takes excellent advantage of Hyperthreading. It takes 20 minutes with it enabled, and 28 minutes with it disabled. If you could run 4.4 GHz overclocked with it enabled, and 4.8 GHz with it disabled ... you'd be better off running the 4.4 GHz and leaving it on because you've got a 40% increase from HT, but only a 10% increase from OC.

Gaming and operations with fewer threads are probably better with HT disabled and a higher OC, but I still think you should benchmark it to be sure. HT is not simply a gimmick, for heavily threaded tasks it makes a substantial difference when it's enabled.

Greg
post #10 of 13
Thread Starter 
Quote:
Originally Posted by hammong View Post

Depending on the task, HT can add 25-35% faster computational performance. You should benchmark your tasks with it enabled and disabled, and then decide if you want it enabled or not. Lets give you a real world example...

Say you do an H.264 encode that takes excellent advantage of Hyperthreading. It takes 20 minutes with it enabled, and 28 minutes with it disabled. If you could run 4.4 GHz overclocked with it enabled, and 4.8 GHz with it disabled ... you'd be better off running the 4.4 GHz and leaving it on because you've got a 40% increase from HT, but only a 10% increase from OC.

Gaming and operations with fewer threads are probably better with HT disabled and a higher OC, but I still think you should benchmark it to be sure. HT is not simply a gimmick, for heavily threaded tasks it makes a substantial difference when it's enabled.

Greg

doing this as we speak smile.gif

Heaven benchmark, Crysis 3, Borderlands 2 se no difference what so ever (that's loading the game up, then holding "W" and running in a straight line till I hit a wall or blocker and watching the FPS

just ripped a Blu-ray in 24minutes 34 seconds with it off.... its doing it now with it on...
Red Dwarf
(33 items)
 
CPUMotherboardGraphicsGraphics
i7 3770K Maxium v Formula Palit GTX670 Jetstream Palit GTX670 Jetstream 
RAMHard DriveHard DriveHard Drive
XMS Vertex 4 Seragate Barracuda Seragate Barracuda 
Optical DriveCoolingCoolingCooling
Pioneer Blu-ray XSPC XP240 Alphacool 5.25" Res D5 Pump 
CoolingCoolingCoolingCooling
EK Comp Fittings PrimoChill PrimoFlex Advanced LRT Red Koolance Drain Plug Bitspower temp sensor 
CoolingCoolingCoolingCooling
IC Diamond Liquid Metal Ultra Corsair SP120 Corsair SP120 
CoolingOSMonitorMonitor
Koolance CPU-380i Windows 8 Pro Epson TW2900 Acer GD245HQ 
MonitorKeyboardPowerCase
150" Projector Screen (Blackout Blind) CYBORGE 7 OCZ 1250w Thermaltake Level 10 GT 
MouseMouse PadAudioAudio
R.A.T 9 Razor IronClad Tritton Pro+ Denon 1610 
Audio
Energy Speakers 
  hide details  
Reply
Red Dwarf
(33 items)
 
CPUMotherboardGraphicsGraphics
i7 3770K Maxium v Formula Palit GTX670 Jetstream Palit GTX670 Jetstream 
RAMHard DriveHard DriveHard Drive
XMS Vertex 4 Seragate Barracuda Seragate Barracuda 
Optical DriveCoolingCoolingCooling
Pioneer Blu-ray XSPC XP240 Alphacool 5.25" Res D5 Pump 
CoolingCoolingCoolingCooling
EK Comp Fittings PrimoChill PrimoFlex Advanced LRT Red Koolance Drain Plug Bitspower temp sensor 
CoolingCoolingCoolingCooling
IC Diamond Liquid Metal Ultra Corsair SP120 Corsair SP120 
CoolingOSMonitorMonitor
Koolance CPU-380i Windows 8 Pro Epson TW2900 Acer GD245HQ 
MonitorKeyboardPowerCase
150" Projector Screen (Blackout Blind) CYBORGE 7 OCZ 1250w Thermaltake Level 10 GT 
MouseMouse PadAudioAudio
R.A.T 9 Razor IronClad Tritton Pro+ Denon 1610 
Audio
Energy Speakers 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › Ever Wondered What The Heat Diffrence is with Hyper Threading Turned On or Off?