The summary contains both the QoR Assessment Score and Flow Guidance information.
The assessment score predicts how likely the design is to achieve timing goals
at the given point in the implementation flow. The earlier in the flow you run the
command, the greater the benefit as there is more compile time to save. There is a small
compromise on accuracy but the score should not be greater than one away from the final
post-route score. It is generated by analyzing a complex set of design metrics in areas
such as UltraFast methodology, device utilization,
control sets, clocking, setup slack, and hold slack. In addition, device-specific
characteristics are also considered and the design stage. For example, after
synth_design, clocking netlist structures are closely examined,
after place_design clock skew has a greater weighting as accuracy
increases and after route_design it will consider other new factors
such as if the design has been able to fully route. The scoring range is from 1 to 5.
When the score is less than 5, use report_qor_suggestions to improve the score.
Score details are provided in the following table:
| Score | Meaning |
|---|---|
| 1 | Design will likely not complete implementation. |
| 2 | Design will complete implementation but will not meet timing. |
| 3 | Design will likely not meet timing. |
| 4 | Design will likely meet timing. |
| 5 | Design will meet timing. |
Flow guidance is part of the overall assessment summary. It dynamically updates depending on the current status of the design. It provides information about:
- Whether you need to address methodology issues
- Whether the design improves using QoR suggestions
- Whether to use ML strategies or incremental compile
The following figure shows an example design with a QoR Assessment Score of 2.
