The design used the unique instabilities and errors of that one FPGA.

YR write to me:
This reminds me of an FPGA that was programed via evolution and not design. They wanted to program an FPGA to reconize a certain frequency. They started with random designs and used a computer to select the better designs and breed them with other better designs. A type of natural selection. It took thousands of iterations for it to get anywhere, but the end result was really weird.

Out of the 100 gates avaliable, only over 30 were used (can’t remember the exact amount). However the gates were not useing binary, but more this sort of metastate. Even weirder was the fact that 8 gates were being used, but inderpendantly of the rest of the circuit. If they turned these gates off the circuit would stop working, even though it seemed like they were doing nothing. They concluded that these gates created the exact interfeance needed for the rest of the circuit to work.

Not surprising was the fact that this was not able to be transplanted into another FPGA. The design used the unique instabilities and errors of that one FPGA.

Leave a comment

Your email address will not be published. Required fields are marked *


seven × 2 =

Leave a Reply