5 ms·
People don't have an agenda against git — it simply doesn't work for them. It's not the holy grail of DVCS', and neither are Hg or bzr; different tools for diff
by inaequitas 17y ago
People don't have an agenda against git — it simply doesn't work for them. It's not the holy grail of DVCS', and neither are Hg or bzr; different tools for different people.
Where are you getting your performance results from? The footnote from Google's article[1] on choosing between Mercurial and git states that Hg was faster both over HTTP and when compared to git's wire protocol. But I'm willing to admit those numbers may be old, so is there a newer published benchmark anywhere?
[1]: http://code.google.com/p/support/wiki/DVCSAnalysis http://code.google.com/p/support/wiki/DVCSAnalysis
- freetard 17y agoWhere does it say that HG http is faster than git://? And second git local operations are way faster than hg's. But google only care about http obviously. Note that android and chromium both use git on google's servers and they employed some of the core git devs so it's only a matter of time before git becomes available on google code.
- inaequitas 17y agoHg - 8.1s/clone, 1.5s/pull git:// - 8.7s/clone, 2.8s/pull Again though, I was curious if you had some numbers. I'm not here to argue which is better, because, really, it's a subjective argument.
- freetard 17y agoI said git local operations are way faster, google them.