Logic Circuit with CUDA

I am new to CUDA. I essentially have a Logic Circuit in some code, that I need to evaluate repeatedly in a recursive algorithm. I have been looking at using a FPGA in order to speed this up.

Someone suggested using GPU instead. However I have only seen how to do massive parallel processing. Can I use CUDA to simulate a Logic Circuit which could effectively be evaluated instantly. If so this would be a lot easier and cheaper than using a FPGA.

I’m not entirely sure what you want to achieve (especially the combination of a recursive function and an FPGA), but maybe this article gives you some inspiration: High performance gate-level simulation with GP-GPU computing (direct pdf: http://www.date-conference.com/proceedings/PAPERS/2009/DATE09/PDFFILES/10.4_3.PDF).