09:18 <sf-slack> <tmichalak> @litghost @mithro: So looking at the discussion regarding VREF (and all other properties that can be defined in xdc) and your comments in the gdoc, do I get it right that we want to go with the xdc/internal format and internal format/fasm option?

09:21 <sf-slack> <tmichalak> If that's the case then I wonder whether we we want to merge the internal format with the genfasm output via fasm2frames. Since we do have fasm features for VREF it would be cleaner to make genfasm generate a fasm that already contains these features.

13:17 <mithro> tmichalak: Option 4 is currently the preferred solution

19:24 <hackerfoo> https://ieeexplore.ieee.org/abstract/document/8649665

19:24 <tpb> Title: ParRA: A Shared Memory Parallel FPGA Router using Hybrid Partitioning Approach - IEEE Journals & Magazine (at ieeexplore.ieee.org)

19:24 <hackerfoo> 24.3x faster than VPR 7.0 using 16 threads.

19:27 <hackerfoo> Another method for parallel routing: https://arxiv.org/pdf/1803.03885.pdf

19:27 <hackerfoo> 3x with 4 threads.