Well, I finally finished testing the XuLA2 prototype a few days ago and released the PCB to manufacturing. When I layed-out the prototype, I butted the microSD socket right up against the FPGA in order to keep the board at 2" x 1". I knew that wasn't going to fly for final production, but I wanted to get it done quick so I could find the other gotchas that typically occur when I move to using a different chip (in this case, the Spartan-6 FPGA).
It turns out there was a gotcha involving the strength of the Spartan-6 pullup resistors, but that was easily resolved once it was finally found. So the only change needed on the board was to find a way to fit the socket on there. I accomplished that by rotating it 90 degrees. While I don't think that looks as good, it was the only way I could find that didn't require making the board a little bit larger than the original XuLA. Here's what the final 4 x 2 array I sent to manufacturing looks like:
I should have the PCBs back around mid-August, meaning the XuLA2 should be available around the beginning of September.
So the XuLA2 ended up the same size as the XuLA, but with 6x the logic density, 3x more internal memory, 4x larger SDRAM, and a socket for holding a multi-GB memory card. You can find a lot more information on the XuLA2 product page, including the preliminary manual.
It's nice to be wrapping up the XuLA2. Now it's back to porting the XSTOOLs to Python and starting on XuLA3.