• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 100 subscribers
  • Views 587 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Keystone Power Consumption

Morgan McBee
Prodigy 285 points

Can you help point me to the technical collateral refarding the spreadsheets and power brief outlined below? I can't find it on the website:

1.3 KeyStone Power Consumption
Your Keystone DSP can consume varying amounts of power depending heavily on
usage, implementation, component selection, and process variation. Refer to your
respective TI power application brief for early power estimations for the DSP, and the
respective power application note and spread sheet (when available) for more detailed
configurations and more accurate power figures. Your respective DSP application note
and power spreadsheet will be made available at some time after silicon becomes
available and characterization is completed.

Thanks,
Morgan

  • 0
    TI__Mastermind 23525 points

    Morgan,

     

    The power estimation spreadsheet is on the product folder page:

    http://www.ti.com/product/tms320c6678

     

    Search for C6678 Power Consumption Model

     

    Regards,

    Travis

     

  • 0 in reply to tscheck
    TI__Mastermind 30210 points

    Travis,

    How to get a realistic power consumption for C6671 and C6674 using the C6678 power consumption sheet?
    I guess there is still a big difference when cores are powered (even if not running) and when PSC keep the core unpowered.

    Thanks and best regards,

    Anthony

  • 0 in reply to AnBer
    TI__Mastermind 23525 points

    Anthony,

    When the PSC powers down a core, the clocks are gated saving power.  But, on the existing silicon in the powered down state, we don't have the cores on individual power islands, so there will be leakage.  This second issue will be addressed in rev 2.0. I'm not sure what is represented in the power spreadsheet with regards to this.  I'll have one of the experts respond.  As far as power consumption goes, what did you plug in for the activity factors?  I think this is the area that most customers misrepresent.  I think the percentage of time where all 8 functional units of a given core is active, is much lower than what you think.  Probably in the range of 20-30% is accurate.  We need to put together some guidelines for this.

     

    Regards,

    Travis