Stats

  • Locked Locked
  • Replies 1
  • Subscribers 62
  • Views 13152
  • 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

Any comments on the RC Physical Timing co-relation with EDI?

sureshm

 Dear All,

I have the database, which is physical optimized using the EDI flow & I want to further optimizations to reduce the slack, which i think is difficult in EDI,

I thought of picking up RTL physical flow for further optimization. I tried to compare the timing of the same design in RC Physical. But i see a humoungus difference between the EDI & RC-Physical (RC200 license).

In EDI WNS i noticed as -98ps.

In RC WNS is -735ps 

the load cap seen on the first few comparisions of Cells are as follows 6.1 fF & 2.5fF & the corresponding caps for the cells are 3.4 fF & 0.783fF.  

Realized that the Net Length Calculations  in EDI & RC are different. How to find out the Net Lengths in RC ? 

 

Thanks

suresh 

 

 

 

Thanks

suresh 

  •  Hi Suresh,

     there are indeed differences in the way RCP handles some of the nets and EDI. These differences are mostly related with signals such as reset, scan enable, high fanout nets, etc. In your case, the concerning aspect is that when things do not correlate, you would normally expect more optimism in synthesis and not the other way round as described in your example. The following report subcommands can provide you additional detail on the calculation

    net_cap_calculation    - reports how the capacitance of the net is calculated
    net_delay_calculation  - reports how the net delay is calculated
    net_res_calculation    - reports how the resistance of the net is calculated
    nets                   - prints a nets report

    The first thing I would look at are

    (1) Version of RC
    (2) Version of EDI
    (3) Are u using same netlist ?
    (4) Are u using same SDC ?
    (5) Are all constraints accepted cleanly ?
    (6) Are u looking at same corner ?

    Remember when using RCP, the GUI can be very handy. 

     report timing -gui. . .

    will show you the path in the GUI. Is it reasonably similar to what EDI is showing ? If not, why ?

    Last but not least, be careful when you assumeRCP will reduce slack. RCP will determine the physical dependencies more accurately hence produce a better netlist (better util, TNS, power, etc.) That being said, if your critical path is a flop to a flop without logic in between, there is not much RCP can do to make that better. Hope this is useful.

     regards,

    gh-

     

     

     

     

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.