Martijn van Beurden

I've tested with an Intel Xeon E-2224G and an Intel Core i5 7200U. Results for the latter are graphed [here](https://github.com/xiph/flac/files/15196500/very.large.set.of.tracks.-.doublesumpatch.pdf). No difference in speed, at the cost of a bigger...

To be honest, I'm not yet convinced. I think that a gain of 1.7% with non-standard compiler options, non-standard compression settings and somewhat uncommon hardware is not something worth integrating....

Diving into this, I see now that CMake Doxygen installing is broken anyway, but this patch would make installing from a tarball even more broken. It would probably be better...

Thanks for your suggestions @H2Swine, it indeed seems long overdue. Could I perhaps persuade you to try to make a PR out of this? Makes comparing and commenting a whole...

Yes, that looks like what I had in mind. Note that after you've created a PR (step 7), you can still add more commits to that branch. So, creating a...

Maybe the explain text should be removed and the -H option should display the file https://github.com/xiph/flac/tree/master/man/flac.md?plain=1 verbatim? Maybe with a few tweaks like removing emphasis or backslashes? That will add...

> If the executable is to be kept slim, could the "man page" (whether it will stay that way) be distributed as a separate file? That is already the case....

The problem with removing the backslashes is that it renders wrong when someone views it with an actual markdown editor or viewer. Also, as we already use pandoc to convert...

I don't know the answers yet, I'll have to find some time to check the code. The [manpage](https://xiph.org/flac/documentation_tools_flac.html) says the following > --serial-number=# When used with --ogg, specifies the serial...

Thanks! I can't think of anything bad that will happen by creating a pull request, so go ahead 😄