All of the technical papers and tutorials seem to be very informative and useful for learning the challenges faced for High Speed Design and possible sollution to those challenges. I have downloaded the paper "Effects of Temperature and Relative Humidity in Transmission Systems Using Differential Signaling." and started scanning through...very interesting!!
First time before I downloaded I had to register on the DesignCon site.
I downloaded the paper about jitter. This has one of my most interests, the Challenges and Solutions in Characterizing a 10 Gb Device. Here in our lab we try to solve things with old (but affordable) HP equipment: We have an old TDR and network analyzer HP8753C+85046A testset at hand, single ended, but it is possible to do differential measurements with those old beasts. We made some UNBAL devices, broad band transformers, actually with a 50R to 100R pad. (See patent US 5,379,006 or anything that comes close to UNBAL and matching 50R to 100R differential) Not the greatest dynamic range left (35~40dB) but for signals <= 3GHz this worked out fine for us. For jitter measurements we use(d) an old LeCroy scope with their original OS. Wonderful peace of gear, and much lighter to work with in comparison to a scope with WIN-PC frontend. Our scope is a true WYSIWYG.
You might wonder why we do this playing around with old gear? Our year turnover probably is smaller than 1 or 2 new pieces of equipment (TDR or NWA) , in other words, we simply cannot afford those. But we *can* solve high-speed problems for our customers, we have proven that. (see http://ppg.teldevice.co.jp/m_board/images/wb_ug172.pdf as an example, in 2003 I guess. LVDS was rather unknown by then) Today, if signals go higher than 6GHz I throw my hands in the air.... Cannot solve everything.... We find RF and analog big fun !
Blog Doing Math in FPGAs Tom Burke 24 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 ...