Thanks for the nice words! Greatly appreciated.
Leonard wrote:
I'm confused, the "string version" change was always used as the correct changeset since at least 2.0, why would that suddenly change now?
I do the "string version" change when we think the current version is ready for release, that's why that changeset usually is the one corresponding to the official binaries. That's no rule though. Sometimes it happens that we still sneak in a few changes after the version string was updated. The rule is that the tagged changeset is the correct one. Due the the imminent change to a new repository, some technical reasons prevent me from adding tags for 2.1 and 2.1.2. I'll do that once the switch is complete. When there is no tag yet, the changeset used for the Windows binaries should be considered to be the correct one.
otaru1921 wrote:
I am having tons of Rubberbanding every once in a while on a Server I didn't have rubberbanding before 2.1.2
Sounds like a problem with your connection to the server. 2.1.2 shouldn't have more rubberbanding than 2.1.
Watermelon wrote:
It's when you lag and the ticbuffer compensates
No, this is an effect of packet loss and client side prediction that also happens without a tic buffer (albeit it may behave a little differently with the buffer). Roughly speaking, it happens when the player position predicted by the client and the actual position on the server differ too much and is usually caused if too many client -> server packets with movement commands are lost.