> the author claims that C limits what can be accomplished. No he doesn't! What he claims is that benchmarks are almost always chosen to be relatively easy to

> the author claims that C limits what can be accomplished. No he doesn't! What he claims is that benchmarks are almost always chosen to be relatively easy to implement in C. What's easy and what is possible are two very different things.