[zuzun]

The benchmarks on the xxHash Github page were done on a 32-bit system using an old Visual C++ compiler. Of course you can beat hash functions that were explicitly designed for modern 64-bit architectures. The Core 2 lacks the CRC instruction introduced by the SSE 4.2 extension, which is the main reason Google wrote their hashes.

[loeg]

Look at XXH64.

[bla2]

xxHash has a 64-bit version.

[zuzun]

All modern hashes are designed for 64-bit architectures and some even rely on recent instruction sets. So a benchmark on an old 32-bit CPU isn't indicative of the performance you'll see on a modern processor, e.g. Metrohash will definitely not be 5x slower than xxHash.

[loeg]

Not sure what you're trying to say.

I'm asking for an apples-to-apples comparison here. Using one figure from a random CPU on github and comparing it to another figure from a different random CPU on github is not an apples to apples comparison.

I don't think anyone claimed xxhash would be 5x faster than metrohash. (Maybe you are getting the 5x number from metrohash's claim that it is 5x faster than siphash?)