On Mar 11, 9:42 am, "Jon Skeet <sk...@pobox.com>" <sk...@pobox.com>
> There's still an optimisation I want to make in terms of unknown field
> sets, which made a difference to the C# code, but this result is
> really amazing...

Indeed, the Server JIT has a much wider set of optimisations at its
disposal. The Client JIT assumes that the application will be running
for a short period of time and is much less ambitious. Some
experimentation was done with a Tiered JIT[1] that would start with
Client, but move to Server if the application ran for long enough
(this is a simplification, but that's the basic idea). That was never
finished though and I am unsure if anyone is working on it at the


[1] http://blogs.sun.com/fatcatair/entry/tiered_compilation_update
You received this message because you are subscribed to the Google Groups 
"Protocol Buffers" group.
To post to this group, send email to protobuf@googlegroups.com
To unsubscribe from this group, send email to 
For more options, visit this group at 

Reply via email to