binary 4x slower than cereal on my data type

I heard on IRC that this should never be the case, so I provided a benchmarked counterexample

https://github.com/ghorn/binary-counterexample

stats here: http://ghorn.github.io/binary-counterexample/

Start by reproducing the benchmark in the linked ghorn/binary-counterexample repository and reviewing the published stats. Compare the binary and cereal results for the reported data type, then trace the relevant serialization path in this repository. Done means identifying the cause of the slowdown and validating an agreed performance improvement against the counterexample.

Written by the indexing model from the issue text.