Is there anything in the book about naming conventions?
I'm recalling Altera's total disregard of system issues, but I did get the impression that with intelligent naming of signals, one's job might be greatly simplified through the magic of wildcards... Maybe. My brain just shut down, and I was fortunately able to keep using the classic timing analyzer. Might not be so lucky next time.
@Michael - I hear what you are saying and as I was reading the book I found myself asking - isn't there a better way to do this. However, that is the very question I should be asking if I were still an EDA tool developer, but as writers of a book they are attempting to help the people who have to use the system that is in place. So - is the book that long because they move too slowly or is the book that long because that is the mess that engineers have to deal with and they are trying to help them navigate. I think it is more of the latter.
I've been doing digital design for 30+ years, but have never been so confused as when I encountered SDC via Altera Quartus.
Confusion stemmed from top as well as bottom. At the system level, I found it impossible to grasp what was required in a complex design. I've got a bunch of my own HDL, a piece of 3rd-party IP, and a few Altera IP blocks in a design. Now what?
At the bottom, the docs just couldn't seem to explain what it all meant. A very simple concept that should have taken one page to explain was spread out over ten pages! A strange case of OVER-documentation that ended up losing the simple point it was trying to make in a tangle of pointless verbiage and formulae.
I've yet to look at Xilinx's Vivado dox to see if they explain SDC understandably.
Looking at this book's chapters, I wonder if it doesn't fall into the same traps. I mean...253 pages to explain basic timing specifications? What's wrong with this picture? Is the emperor naked?
Blog Doing Math in FPGAs Tom Burke 16 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...