Understanding DPTF ... Smarter Throttling

Discussion in 'Microsoft Surface Pro 3' started by GreyFox7, Aug 21, 2015.

  1. GreyFox7

    GreyFox7 Super Moderator Staff Member

    Joined:
    Jul 27, 2014
    Messages:
    6,334
    Likes Received:
    1,279
    Trophy Points:
    113
    Ref: Understanding Intel's Dynamic Power and Thermal Framework 8.1: Smarter Throttling

    In mobile, thermal throttling is effectively a fact of life. Corrected :)

    Although Intel’s Dynamic Platform and Thermal Framework (DPTF) 8.1.x has been out for months now, these features haven’t really received much attention so far. For those that are probably unaware of what Intel’s DPTF is, it’s effectively Intel’s solution to managing throttling in a smart manner according to the TDP limits of the device based upon thermal sensors and power monitoring for x86 tablets, 2-in-1s, and PCs in general. If you think this sounds a lot like ARM’s Intelligent Power Allocation in some ways, you’d be right.

    For the most part, previous iterations of DPTF have been pretty standard in the sense that they rely on a fixed correlation between temperature sensors and critical values like Tskin max and Tjunction max o all chips on the board. ...

    However in the case of DPTF 8.1, this system has changed. Instead of a fixed correlation, the system is now adaptive depending upon a number of factors. ...

    With previous iterations of DPTF, this worst-case style setup was what was used to determine how to correlate temperature sensors with skin temperatures.

    The problem with that approach was that when the device was placed in a situation where cooling was better, such as held vertically in the air or held in a dock with a circulation fan, DPTF wouldn’t change the temperature sensor correlations to skin temperature. This meant that in long run TDP-gated situations that the device was throttled to a greater extent than truly necessary.

    It turns out this one change has enormous effects on performance in these thermally limited situations. With a vertical orientation, heat dissipation and thereby power headroom increases by 66%. With an active cooling dock, power headroom increases by 97%. As Intel reasons and as their data backs up, there are clear benefits in not being conservative with throttling in situations where physics says cooling performance is better than the worst case scenario.

    Adaptive_Performance_Bench2_575px.png

    So do we think Microsoft will move to an active cooling dock?
     
    Last edited: Aug 21, 2015
  2. GreyFox7

    GreyFox7 Super Moderator Staff Member

    Joined:
    Jul 27, 2014
    Messages:
    6,334
    Likes Received:
    1,279
    Trophy Points:
    113
    In a poor mans implementation simply standing the Surface with Kickstand in the airflow of any household fan has it's merits.

    However until the new DPTF is implemented in the Surface (assuming that it's possible and I assume that it is) we still have the Fixed throttling solution so benefits will be muted. Clear benefits have been shown even with the older fixed DPTF implementation it just wont be as good.
     
  3. GreyFox7

    GreyFox7 Super Moderator Staff Member

    Joined:
    Jul 27, 2014
    Messages:
    6,334
    Likes Received:
    1,279
    Trophy Points:
    113
    I would expect the SP4 to utilize the new adaptive DPTF along with a Skylake CPU. The goodness factor will make for the best Surface Pro ever.
     
  4. Cothek

    Cothek Active Member

    Joined:
    Jul 26, 2014
    Messages:
    130
    Likes Received:
    31
    Trophy Points:
    28
    This is good news! I'm assuming this would be at the firmware level for the SP3?
     
  5. Haldi

    Haldi Member

    Joined:
    Nov 27, 2014
    Messages:
    199
    Likes Received:
    15
    Trophy Points:
    18
    My Device:
    SPro3
    Would be the only valid choice for their continuum 2in1 approach.
    6W TDP Passive cooled tablet for on the go, and an active cooled dock which allows up to 35W TDP when docked.

    The CPU will probably be way bigger than now, maybe there will be special SoC which have like ARM big.LITTLE design with 2 Atom Cores and 4 core i Cores which dynamically can be activated or deactivated.
     
  6. GreyFox7

    GreyFox7 Super Moderator Staff Member

    Joined:
    Jul 27, 2014
    Messages:
    6,334
    Likes Received:
    1,279
    Trophy Points:
    113
    35w wont happen. Active cooling would allow the full 15W maybe a little more but fitting a higher power CPU in is another wont happen. Other exotic designs are not in the cards either unless you see at least a 10-15x increase in market share... nope still not happening.
     
  7. Haldi

    Haldi Member

    Joined:
    Nov 27, 2014
    Messages:
    199
    Likes Received:
    15
    Trophy Points:
    18
    My Device:
    SPro3
    Have you seen the new ASUS GX700?
    we need that in waaaaayy smaller :)
    The Future is now!
     
  8. GreyFox7

    GreyFox7 Super Moderator Staff Member

    Joined:
    Jul 27, 2014
    Messages:
    6,334
    Likes Received:
    1,279
    Trophy Points:
    113
    It just can't be done, if someone could they would... well maybe with liquid nitrogen cooling but then you'd need a whole other design.
    In 20 years Intel might have the same processing power in a 4.5W SoC but then a gaming laptop will be 200x more powerful and we will have the same issue. :)
     

Share This Page

Search tags for this page

dptf adaptive cooling

,

dptf cooling

,

intel dynamic platform and thermal framework interrupts

,

intel dynamic platform and thermal framework power level forum

,

throttling dptf