- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
May I know how implemented frequency is calculated here and why the FMax is calculated in that way.
I am seeing the difference in calculating Fmax for the design.
Highlighted the considered values from reports and attached snapshots.
I have attached the project folder for your reference.
Can you help me to identify the issue.
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
I had tried to extract the folder and open the design. However can't find the related files in screenshot:
Could you provide the missing files as well?
Thanks,
Regards,
Sheng
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
May I know do you have any further concern or consideration?
Thanks,
Regards,
Sheng
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
your design fails timing analysis because Arria 10 maximum core clock frequency of 644 MHz is exceeded (timing analyser uses rounded value of 645.16MHz = 1/1.55ns). This causes a "minimum pulsewidth" violation for clk and a respective restricted Fmax value.
Possible contradicting reported values in timing analysis are referred. You need to solve the minimum pulsewidth problem first.
Regards
Frank
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
FvM is right. You may check this document link https://cdrdv2-public.intel.com/666844/a10_datasheet-683771-666844.pdf (page 36):
Parameter | Performance (All Speed Grades) | Unit |
Global clock, regional clock, and small periphery clock | 644 | MHz |
Large periphery clock | 525 | MHz |
As for setup slack, you may go to Assignments -> Settings -> Compiler Settings -> choose Performance
Thanks,
Regards,
Sheng
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
this .sdc passes timing analysis
create_clock -name clk -period 1.56ns -waveform {0.0ns 0.78ns} [get_ports {clk}]
derive_clock_uncertainty
set_false_path -to [all_outputs]
set_false_path -from [all_inputs]
Presume NR_symbol_modulator is planned to be used inside an Arria 10 design with some kind of high speed interface. Consider that FPGA fabric side of the interface can't run above 644 MHz clock. If intended 780 MHz clock is required by interface throughput, you'll need more bits for the parallel data stream.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
May I know do you have any further update or concern?
Thanks,
Regards,
Sheng
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page