Runtimes in CPU hours for generation of 100 designs using RECON, both by fixed backbone (FBB) and backbone minimized (BBM) methods, and MPI_MSD algorithms. Comparison of CPU runtimes for multi-specifi
Shown is time in seconds until convergence of each optimization method, averaged across ten independent runs of the method. Runs that did not converge are excluded from reporting. MUR, multiplicative