are new benchmarks needed?

[Charles Kerr]

Luca Barbato wrote:
>>On 12/29/2010 03:21 AM, Charles Kerr wrote:
>> Oh come on... "A is faster than B" is *clearly* not the same as "No
>> practical difference between A and B."  Saying they're equivalent just
>> because they're both true doesn't even pass the laugh test. :)
>
> libev is faster in many scenarios due to its api and, depending on the
> version, due some data structures in use, what's exactly false in that?

In a word, nothing. I said the same thing earlier today when I linked
to the 2010 benchmarks! :) As I've said, I appreciate Marc's work on
libev. I'm also *very* glad that there are two well-maintained event
loop libraries to choose from.

...if you think I'm here to bash on libev maybe I'm not explaining
myself well. Let me try again. My complaint is simple and has nothing
to do with the library, only its marketing:

(1) When libev compares itself to libevent, it should use reasonably
up-to-date information. If libevent2 is better than libevent1, as it
seems to be, then continuing to omit libevent2 /is/ misleading,
whether intentional or not.

(2) If there are no practical performance difference between libev and
libevent, as libev's author says, then when libev compares itself to
libevent in the README and in other places, it should *say that*
instead of saying "faster" or "much faster" as it does now. It's
simply self-evident that "faster" is not equivalent to "no practical
difference."

The rest of this thread is just an offshoot from Marc's claim that
neither he nor his writeup of the 2008 benchmarks say that there's a
significant difference between libev and libevent. I find that
incomprehensible, but maybe I should have let it slide by since it's
tangental. If the two suggestions above were accepted, I would be
happy and have no complaints.

cheers,
Charles