Re: cleaner/better zlib sources?

On Fri, 16 Mar 2007, Jeff Garzik wrote:

Although it sounds like zlib could indeed be optimized to reduce its startup
and shutdown overhead, I wonder if switching compression algorithms to a pure
Huffman or even RLE compression (with associated lower startup/shutdown costs)
would perform better in the face of all those small objects.

And another random thought, though it may be useless in this thread:  I bet
using a pre-built (compiled into git) static zlib dictionary for git commit
and tree objects might improve things a bit.

See my last post.  We'll do even better with special object 
encoding altogether.  Those representations are so dense that 
compression provides no gain at all making the point moot.


Nicolas