- fixed: The shader code for handling special fixed colormaps did not use the color vertex attribute which was most evident with the 'shadow' render style on the spectre.
- Added A_KillTarget(damagetype, int flags).
- Added A_KillTracer(damagetype, int flags).
- Added A_RemoveTarget.
- Added A_RemoveTracer.
A_Kill (Master/Target/Tracer/Children/Siblings):
- KILS_FOILINVUL: foils invulnerability.
- KILS_KILLMISSILES: destroys projectiles. Does not work on invulnerable
projectiles without KILS_FOILINVUL, and doesn't work at all on missiles
with NODAMAGE flag.
- KILS_NOMONSTERS: actors that are monsters will not be killed.
A_Damage (Self/Target/Master/Tracer/Children/Siblings):
- DMSS_FOILINVUL: foils invulnerability.
- DMSS_AFFECTARMOR: damages the actor's armor instead of bypassing it
entirely.
- DMSS_KILL: damages the actor by its remaining health (useful for
modular DECORATE programming).
- Added A_SpawnItemEx flags:
- SXF_SETTARGET: sets the calling actor as the target.
- SXF_SETTRACER: sets the calling actor as the tracer.
Both of these functions take priority similar to SXF_SETMASTER over
SXF_TRANSFERPOINTERS.
In the function R_RebuildViewInterpolation, the pointer 'iview' was not initialized when the player or its camera were NULL, hence 'iview == NULL' was garbage. Also, the function FindPastViewer does not return NULL, hence the mentioned check is not needed at all. Just return early if the player camera does not exist.
- Extratic changed to server var, that can be changed at any time
(net_extratic)
- Added net_extratic 2 which resends all unconfirmed tics
- Corrected bad variable typo in delay reporting
- Added delay times of all players to the scoreboard
- Removed balancing from packet-server (tried it, didn't work)
- Calculations remove an extra tic to account for possible bias
- Guests can now attempt to match latency with the arbitrator.
(net_loadbalance)
- Added althud feature to show arbitrator and local latency.
(hud_showlag 1 is on for netgames, 2 is always on)
Since rendering is asynchronous, the camera texture scene may not be finished once the main scene's lights get filled in. Unfortunately forcing a synchronization with glFinisg has bad side effects on performance the only remaining option is to use separate parts of the buffer for both scenes, which in extreme cases may increase the light buffer's size - but on modern hardware that shouldn't be a problem.
