Page 2 of 3 FirstFirst 1 2 3 LastLast
Results 11 to 20 of 24
  1. #11
    Tech Marketing Manager HQ Array Raja@ASUS's Avatar
    Join Date
    Apr 2011
    Reputation
    161
    Posts
    7,454

    Quote Originally Posted by red454 View Post
    I do cooler (and case) reviews for a website, and I started using a FLIR camera a couple of weeks ago. The cooler I just finished (FrostFlow 240L) showed roughly a 5 degree C difference between the inlet and outlet temps. I can trace the temperature drop visually (using the FLIR camera) along the path of the coolant through the radiator. This is using a 4770K clocked at 4.2 under a Prime95 load for 1 hour. Ambient temp was 23 C.
    Yeah, that's another way of tracing it. If there is a perfect equilibrium between IO, the rad isn't doing anything, heh...

  2. #12
    ROG 師傅 Array Arne Saknussemm PC Specs
    Arne Saknussemm PC Specs
    Laptop (Model)Laptop?...No way! (Model?...Jun Amaki...yes way!)
    MotherboardROG ZENITH EXTREME
    ProcessorTHREADRIPPER 1920X
    Memory (part number)F4-3200C14Q-32GVK
    Graphics Card #1GTX Titan X
    Graphics Card #2SLI is dead to me
    Graphics Card #3Tri SLI is even dead to Nvidia
    Graphics Card #4Quad SLI is dead to everybody especially my credit card
    Sound CardXonar Essence STX
    MonitorASUS ROG Swift PG279Q
    Storage #1Samsung 950 PRO
    Storage #22x OCZ VERTEX 3/2x WD Caviar Black 500GB / 2x WD RED 2TB
    CPU CoolerCustom Loop: Dual D5s, Dual Alphacool Monsta 480s, XSPC Raystorm Neo TR4, EK TitanX WaterBlock
    CaseNope!...Phobya Bench...let it all hang out man!
    Power SupplySeasonic SS-1200XP3
    Keyboard Logitech G810 or Fender Rhodes Electric Piano (MKI 73)
    Mouse G502 or Speedy Gonzalez...not Mickey...don't do Disney!
    Headset ...firmly on neck
    Mouse Pad Mouse don't got his own pad man...lives with me
    Headset/Speakers Edifier Spinnaker...or you mean the speakers in my head...man too many voices to name them all
    OS Win XP, 7, 8, 8.1 and Windows 10 Spyware Edition
    Network Router56k modem
    Accessory #1 Umm...nice tie?
    Accessory #2 Err...belt?
    Accessory #3 3 accessories?! I'm not a girl!!
    Arne Saknussemm's Avatar
    Join Date
    Apr 2012
    Reputation
    481
    Posts
    13,406

    I'm shocked...just goes to show how crappy AIO cooling is then...of course there will be a difference across any "block" but this should tend to fractions of a degree as flow is optimised...

    5 degrees difference is terrible...with that kind performance loop order would start mattering...

    I will no longer consider AIO cooling watercooling

  3. #13
    ROG Guru: Orange Belt Array red454 PC Specs
    red454 PC Specs
    MotherboardRampage V Extreme BIOS 3801
    Processor5820K
    Memory (part number)32GB - G.Skill F4-2800C16Q-32GRK
    Graphics Card #1GTX Titan X
    MonitorDual 23"
    Storage #1Samsung 850 EVO 500GB SSD
    Storage #2Samsung 850 EVO 500GB SSD
    CPU CoolerCooler Master Nepton 280L
    CaseThermaltake Core X9
    Power SupplyDARK POWER PRO 11 1000W
    Keyboard Logitech G19S
    Mouse Logitech G600
    OS Win 10 Pro 64bit
    Network RouterLinksys EA9500
    Accessory #1 FreeNAS Server with 8TB
    red454's Avatar
    Join Date
    Nov 2011
    Reputation
    32
    Posts
    429

    Quote Originally Posted by Raja@ASUS View Post
    Yeah, that's another way of tracing it. If there is a perfect equilibrium between IO, the rad isn't doing anything, heh...
    Right - or the system is just idle. Any heat going in is dissipated by the hardware (pump / base, mounting components, hoses, etc.) before it ever gets to the radiator.
    ASUS Rampage V Extreme BIOS 3801 | i7-5820K | Thermaltake Core X9 | G.Skill F4-2800C16Q-32GRK | Cooler Master Nepton 280L | Dual Samsung 850EVO 500GB SSD | PSU: DARK POWER PRO 11 1000W | 3TB & 4TB HDD | NVIDIA GTX Titan X | ASUS 24x DVD±RW Drive | Win10 Pro

  4. #14
    ROG Guru: Orange Belt Array red454 PC Specs
    red454 PC Specs
    MotherboardRampage V Extreme BIOS 3801
    Processor5820K
    Memory (part number)32GB - G.Skill F4-2800C16Q-32GRK
    Graphics Card #1GTX Titan X
    MonitorDual 23"
    Storage #1Samsung 850 EVO 500GB SSD
    Storage #2Samsung 850 EVO 500GB SSD
    CPU CoolerCooler Master Nepton 280L
    CaseThermaltake Core X9
    Power SupplyDARK POWER PRO 11 1000W
    Keyboard Logitech G19S
    Mouse Logitech G600
    OS Win 10 Pro 64bit
    Network RouterLinksys EA9500
    Accessory #1 FreeNAS Server with 8TB
    red454's Avatar
    Join Date
    Nov 2011
    Reputation
    32
    Posts
    429

    Quote Originally Posted by Arne Saknussemm View Post
    I'm shocked...just goes to show how crappy AIO cooling is then...of course there will be a difference across any "block" but this should tend to fractions of a degree as flow is optimised...

    5 degrees difference is terrible...with that kind performance loop order would start mattering...

    I will no longer consider AIO cooling watercooling
    I don't follow - why is that a problem?

    Scale it up to an automotive application - the temp going into the radiator is much higher than what comes out and is returned to the engine. The efficiency is driven by the surface area exposed to the heat source, flow rate of the coolant, and finally by the surface area of the heat sink and airflow rate through the radiator.
    The delta T stabilizes when the heat going into the system is = to the heat rejected (to the room). All that stuff from my Thermo classes 30 years ago is coming back to me - oh, the pain!
    Last edited by red454; 10-07-2016 at 02:01 PM.
    ASUS Rampage V Extreme BIOS 3801 | i7-5820K | Thermaltake Core X9 | G.Skill F4-2800C16Q-32GRK | Cooler Master Nepton 280L | Dual Samsung 850EVO 500GB SSD | PSU: DARK POWER PRO 11 1000W | 3TB & 4TB HDD | NVIDIA GTX Titan X | ASUS 24x DVD±RW Drive | Win10 Pro

  5. #15
    New ROGer Array DWgia PC Specs
    DWgia PC Specs
    MotherboardAsus Rampage V Extreme
    ProcessorI7 5960x
    Memory (part number)Gskil Ripjaws 4 Series 16 (4x4) Gb 3000 Mhz
    Graphics Card #1EVGA Titan X Hybrid (Maxwell)
    Graphics Card #2EVGA Titan X Hybrid (Maxwell)
    MonitorAsus VQ 248QE
    Storage #1Samsung 850 evo 512gb
    CPU CoolerCorsair H110i GTX
    CaseCorsair C70
    Power SupplyEVGA Supernova 1600 P2
    OS Windows 7 Ultimate 64bit

    Join Date
    Sep 2016
    Reputation
    10
    Posts
    4

    Hi guys. Sorry for the confusion. Yes the testing does include shutting the system down between reboots to allow the temps to drop back to the same state as initial power on.

    Just two days ago I turned my system off (without unplugging anything) to test how temp goes again. I Let it rest over night as I go to bed. Then the morning after, CPU's temp got to 63C just after 30 mins playing BF4 comparing to 53C after 3hrs of BF4 in the night before. And it will reach 70s on next boot if i dont unplug and press power button for sure.

    For the cooler's coolant temp i mentioned in post #6, that 38.5C was about 1hr in BF4. The max coolant's temp I ever see was about 42-43 (after 4-5hrs in BF4).

    Below is a screen capture of my ideal idle temp just 3 mins after boot (the boot after unplugging).
    https://flic.kr/p/M1yvqn

    Thanks.

  6. #16
    Tech Marketing Manager HQ Array Raja@ASUS's Avatar
    Join Date
    Apr 2011
    Reputation
    161
    Posts
    7,454

    Quote Originally Posted by Arne Saknussemm View Post
    I'm shocked...just goes to show how crappy AIO cooling is then...of course there will be a difference across any "block" but this should tend to fractions of a degree as flow is optimised...

    5 degrees difference is terrible...with that kind performance loop order would start mattering...

    I will no longer consider AIO cooling watercooling
    The bigger the IO temp disparity, the better the rad.

  7. #17
    ROG 師傅 Array Arne Saknussemm PC Specs
    Arne Saknussemm PC Specs
    Laptop (Model)Laptop?...No way! (Model?...Jun Amaki...yes way!)
    MotherboardROG ZENITH EXTREME
    ProcessorTHREADRIPPER 1920X
    Memory (part number)F4-3200C14Q-32GVK
    Graphics Card #1GTX Titan X
    Graphics Card #2SLI is dead to me
    Graphics Card #3Tri SLI is even dead to Nvidia
    Graphics Card #4Quad SLI is dead to everybody especially my credit card
    Sound CardXonar Essence STX
    MonitorASUS ROG Swift PG279Q
    Storage #1Samsung 950 PRO
    Storage #22x OCZ VERTEX 3/2x WD Caviar Black 500GB / 2x WD RED 2TB
    CPU CoolerCustom Loop: Dual D5s, Dual Alphacool Monsta 480s, XSPC Raystorm Neo TR4, EK TitanX WaterBlock
    CaseNope!...Phobya Bench...let it all hang out man!
    Power SupplySeasonic SS-1200XP3
    Keyboard Logitech G810 or Fender Rhodes Electric Piano (MKI 73)
    Mouse G502 or Speedy Gonzalez...not Mickey...don't do Disney!
    Headset ...firmly on neck
    Mouse Pad Mouse don't got his own pad man...lives with me
    Headset/Speakers Edifier Spinnaker...or you mean the speakers in my head...man too many voices to name them all
    OS Win XP, 7, 8, 8.1 and Windows 10 Spyware Edition
    Network Router56k modem
    Accessory #1 Umm...nice tie?
    Accessory #2 Err...belt?
    Accessory #3 3 accessories?! I'm not a girl!!
    Arne Saknussemm's Avatar
    Join Date
    Apr 2012
    Reputation
    481
    Posts
    13,406

    Quote Originally Posted by red454 View Post
    I don't follow - why is that a problem?
    Well, it is and it isn't

    In this case since you are only cooling the CPU and nothing else it makes little difference (though I shouldn't wonder that for a set airflow through radiator higher flow through loop would give lower temps...

    However I'm extrapolating that to a loop with more than one block...where all of a sudden 5 degree temps across a block would be critical for OCing a CPU for example...in this scenario you would have to make an effort to make sure CPU followed radiator and simplest loop order could be compromised by need to introduce less clean, more restrictive loop to cater for order...and it would get worse and worse...

    Computer loop is not a car...a rise in 5 or more degrees over the GPUs for example, if fed then to CPU, would lead to less OC headroom on your CPU in a loop with this bad flow. Car engine is much more rough and ready in it's cooling needs. Performance there would hardly be affected.

    For a cooling loop with various components (blocks) you would ideally like the smallest rise or fall in temps across a block that you can manage. For this you want good flow/head pressure from your pumps, low restriction blocks and loop as a whole (hence parallel GPU blocks and the like). This way your only worry really is transferring heat from loop as a whole to air and no worries about doing this per block and worrying about the temp of water your CPU is receiving.

    In short, that kind of temperature difference over blocks is "unacceptable" in a high performance water loop with multiple blocks

  8. #18
    Administrator Array Silent Scone@ASUS's Avatar
    Join Date
    Mar 2016
    Reputation
    124
    Posts
    1,805

    Quote Originally Posted by Arne Saknussemm View Post
    I'm shocked...just goes to show how crappy AIO cooling is then...of course there will be a difference across any "block" but this should tend to fractions of a degree as flow is optimised...

    5 degrees difference is terrible...with that kind performance loop order would start mattering...

    I will no longer consider AIO cooling watercooling

    Do you do your testing with the system off? lol

  9. #19
    Tech Marketing Manager HQ Array Raja@ASUS's Avatar
    Join Date
    Apr 2011
    Reputation
    161
    Posts
    7,454

    Quote Originally Posted by Arne Saknussemm View Post
    Well, it is and it isn't

    In this case since you are only cooling the CPU and nothing else it makes little difference (though I shouldn't wonder that for a set airflow through radiator higher flow through loop would give lower temps...

    However I'm extrapolating that to a loop with more than one block...where all of a sudden 5 degree temps across a block would be critical for OCing a CPU for example...in this scenario you would have to make an effort to make sure CPU followed radiator and simplest loop order could be compromised by need to introduce less clean, more restrictive loop to cater for order...and it would get worse and worse...

    Computer loop is not a car...a rise in 5 or more degrees over the GPUs for example, if fed then to CPU, would lead to less OC headroom on your CPU in a loop with this bad flow. Car engine is much more rough and ready in it's cooling needs. Performance there would hardly be affected.

    For a cooling loop with various components (blocks) you would ideally like the smallest rise or fall in temps across a block that you can manage. For this you want good flow/head pressure from your pumps, low restriction blocks and loop as a whole (hence parallel GPU blocks and the like). This way your only worry really is transferring heat from loop as a whole to air and no worries about doing this per block and worrying about the temp of water your CPU is receiving.
    In short, that kind of temperature difference over blocks is "unacceptable" in a high performance water loop with multiple blocks

    You were confused about equilibrium; a good loop is never perfectly equal in temps across all areas.

  10. #20
    ROG 師傅 Array Arne Saknussemm PC Specs
    Arne Saknussemm PC Specs
    Laptop (Model)Laptop?...No way! (Model?...Jun Amaki...yes way!)
    MotherboardROG ZENITH EXTREME
    ProcessorTHREADRIPPER 1920X
    Memory (part number)F4-3200C14Q-32GVK
    Graphics Card #1GTX Titan X
    Graphics Card #2SLI is dead to me
    Graphics Card #3Tri SLI is even dead to Nvidia
    Graphics Card #4Quad SLI is dead to everybody especially my credit card
    Sound CardXonar Essence STX
    MonitorASUS ROG Swift PG279Q
    Storage #1Samsung 950 PRO
    Storage #22x OCZ VERTEX 3/2x WD Caviar Black 500GB / 2x WD RED 2TB
    CPU CoolerCustom Loop: Dual D5s, Dual Alphacool Monsta 480s, XSPC Raystorm Neo TR4, EK TitanX WaterBlock
    CaseNope!...Phobya Bench...let it all hang out man!
    Power SupplySeasonic SS-1200XP3
    Keyboard Logitech G810 or Fender Rhodes Electric Piano (MKI 73)
    Mouse G502 or Speedy Gonzalez...not Mickey...don't do Disney!
    Headset ...firmly on neck
    Mouse Pad Mouse don't got his own pad man...lives with me
    Headset/Speakers Edifier Spinnaker...or you mean the speakers in my head...man too many voices to name them all
    OS Win XP, 7, 8, 8.1 and Windows 10 Spyware Edition
    Network Router56k modem
    Accessory #1 Umm...nice tie?
    Accessory #2 Err...belt?
    Accessory #3 3 accessories?! I'm not a girl!!
    Arne Saknussemm's Avatar
    Join Date
    Apr 2012
    Reputation
    481
    Posts
    13,406

    Quote Originally Posted by Raja@ASUS View Post
    perfectly equal in temps across all areas.
    Hey come on! where did I say that? that would be perfect friction-less flow.

    A good loop with good flow will show only slight difference across block...something you would need very good sensors to detect.

    If your loop order matters you have bad flow and a bad loop.

    Quote Originally Posted by Raja@ASUS View Post
    The bigger the IO temp disparity, the better the rad.
    This kind of gives me a headache....I can imagine a radiator is ideally efficient if it dissipates all the heat it receives to the atmosphere and that with perfect friction-less flow it could achieve this with very little difference between intake and out. It would simply depend on heat transfer efficiency to atmosphere? and have nothing to do with fall of temp from one end to another? Or do I need a lie down
    Last edited by Arne Saknussemm; 10-07-2016 at 03:36 PM.

Page 2 of 3 FirstFirst 1 2 3 LastLast

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •