• Skip to main content
  • Skip to search
  • Skip to footer
Cadence Home
  • This search text may be transcribed, used, stored, or accessed by our third-party service providers per our Cookie Policy and Privacy Policy.

  1. Community Forums
  2. Digital Implementation
  3. power estimation by RTL-Compiler

Stats

  • Locked Locked
  • Replies 4
  • Subscribers 91
  • Views 15580
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

power estimation by RTL-Compiler

naderi
naderi over 16 years ago

Hello all,

In RTL-Compiler, I used "report power" command after synthesis, and got following report. Could someone please tell me what activity factor and clock frequency are considered by RC to generate that report?

my target clock period was 500ps and timing analysis shows a negetive slack time of -450ps.

Thanks,

Ali 

 ============================================================
  Generated by:           Encounter(r) RTL Compiler v06.20-s027_1
  Generated on:           Nov 21 2008  01:45:55 PM
  Module:                 decoder648
  Technology libraries:   IO90GPHVT_2V5_3V3_50A_7M2T
                          IO90GPHVT_2V5_50A_7M2T
                          IO90GPHVT_3V3_50A_7M2T
                          IO90GPHVT_ANA_50A_7M2T
                          IO90GPHVT_BASIC_50A_7M2T
                          IO90GPHVT_FILLERCUT_50A_7M2T
                          IO90GPHVT_REF_COMPENSATION_2V5_50A
                          IO90GPHVT_REF_COMPENSATION_3V3_50A
                          PR90M7
                          CORE90GPSVT
                          CORX90GPSVT
                          CLOCK90GPSVT
  Operating conditions:   wc_0.90V_125C (balanced_tree)
  Wireload mode:          top
============================================================

                             Leakage       Internal         Net         Switching   
     Instance       Cells   Power(nW)      Power(nW)     Power(nW)      Power(nW)   
------------------------------------------------------------------------------------
decoder648         480027 567216066.450 1140456537.747 407240261.752 1547696799.500

 

  • Cancel
  • grasshopper
    grasshopper over 16 years ago

     Hi Ali,

     

    I believe this is some of the information you are looking for

     

    rc:/> get_attr -h lp*default* *

    Usage: get_attribute <string> [<object>+]

        <string>:
            attribute name
        [<object>+]:
            object of interest (must be unique)

        attribute categories: lp_tcf tim

        attribute name: lp_default_toggle_percentage
              category: tim (controls timing optimization and returns timing information)
           object type: clock
           access type: read-write
             data type: fixed point
         default value: no_value
                  help: Default toggle-rate scale-factor of data signals based on clocks.

        attribute name: lp_default_probability
              category: lp_tcf (returns switching activity information)
           object type: instance
           access type: read-write
             data type: fixed point
         default value: 0.50000
                  help: Default probability for all nets within the hierarchical instance to be high.

        attribute name: lp_default_toggle_rate
              category: lp_tcf (returns switching activity information)
           object type: instance
           access type: read-write
             data type: fixed point
         default value: 0.02000
                 units: /ns
                  help: Default signal toggle rate for all nets within the hierarchical instance.

        attribute name: lp_default_probability
              category: lp_tcf (returns switching activity information)
           object type: design
           access type: read-write
             data type: fixed point
         default value: 0.50000
                  help: Default probability for all nets in current design to be high.

        attribute name: lp_default_toggle_rate
              category: lp_tcf (returns switching activity information)
           object type: design
           access type: read-write
             data type: fixed point
         default value: 0.02000
                 units: /ns
                  help: Default signal toggle rate for all top level nets in current design.

    rc:/> get_attr program_version /
    v08.10-s125_1

    Note that I included the RC version I was using since this may have changed in recent release. I strongly encourage you to look at the low power user guide since there is ample information there, not only about what you are asking, but also as to how activity is propagated and other information you may find of value such as looking at power vs. time dynamic activity and other useful low power features.

     

    good luck,

    gh-

     

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • grasshopper
    grasshopper over 16 years ago

     BTW, I meant the RTL Compiler Low Power User Guide

     

    gh-

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • naderi
    naderi over 16 years ago

     Thanks,

    if the default toggle rate is 0.02 (1/ns), then is it changed by only setting clock period? or toggle rate should be set by setting its attribute?

    I have a clock period of 500ps and default probability is 0.5 and I didn't tuch toggle rate attribute. In this case can I say that the power is reported for toggle rate of 1/250 ps == 4G, although timing shows a neg. slack time of -450?

    Thanks,

    Ali 

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel
  • grasshopper
    grasshopper over 16 years ago

     Hi Ali,

     the answer to your question is no. toggle activity is independant from clock definitions on SDC. In general, dynamic power analysis with default rates is not terribly accurate. I suggest you use VCD, TCF, or SAIF from simulation to produce more accurate results. VCD is nice in that it allows you to look at time vs. power profile which further enable you to address the block responsible for peak power surges. If you have access to Palladium emulation and power is a hot ticket item for your team, I would certainly inquire about DPA.

     gh-

     PS: Please post RC questions in Logic Design forums. I think you will get more replies there

    • Cancel
    • Vote Up 0 Vote Down
    • Cancel

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.

© 2025 Cadence Design Systems, Inc. All Rights Reserved.

  • Terms of Use
  • Privacy
  • Cookie Policy
  • US Trademarks
  • Do Not Sell or Share My Personal Information