Honored Contributor
Honored ContributorThe two reasons to do post-fit on your logic with SignalTap are:
a) Faster compile times b) Preserving the placement and routing when there is a bug that comes and goes depending on the place and route. When you set a partition to post-fit, there is an option to Ignore Changes in Source File. If you do that, then changing the seed/.sdc/etc. will not get a new place-and-route. If you don't have this set, then Quartus will try to be smart and will start from source if it thinks it needs to, specifically if there is a change in the source code. I'm not sure about the other things. Generally for a SignalTap flow where you want a) and/or b), you would set all of your partitions to post-fit and would not change anything else in the design(your debugging the existing fit, not adding changes), and therefore there are no changes for it to figure out. If you are making changes to the design, then I generally don't set it to post-fit. Basically I don't ever rely on it knowing what to do, not that it doesn't work, but I know enough of whether I want everything locked down or not that I don't need it to be smart. That's the flow I've seen most used.
