AD_tears problems in Ironwail

Placo

New member
Jun 24, 2023
4
0
1
I have noticed several problems in the AD map Tears of the False God in Ironwail: I get stuck on some ramps, summoned knights and death knights are blocked behind the erected barrier in the (fake) last arena against the Judicator and in the last secret arena some fiends are stuck in the raising pillars, blocking them until their are finally crushed, but then ruining the pace of the battle.
I think it is linked to some unimplemented CSQC extensions as I can replicate the issues in vkQuake with pr_checkextension turned off.

Seeing as ad_tears was the poster child for Ironwail's first release, I wonder if there is a way to play the map "as intended" in that engine.
Also I'm curious, what kind of extensions does AD use that would explain those bugs? I thought the physics stuff was managed by the protocol (666, 999) implementations
 
The ramp sticking problem is something I've noticed on plenty of maps recently. Because I use Ironwail almost exclusively it didn't occur to me this might be a source-port-exclusive bug (I swear I've encountered it on and off historically across engines, mind).

If I'm identifying the same issue, it's a question of clipping on the surface joining a horizontal brush to a convex downward brush surface. Significantly, this bug comes up again and again on Zetabyts first map for Brutalist Jam 2: the big battle in the arena at the lowest point in the structure consists of a basin with inclines onto walkways either side — if you circle around the outer edge of the room you will consistently clip as you attempt to cross between the raised area and the slope. A jump is necessary to clear the edge.

I wonder, if this is indeed an engine bug, if it's related in some way to another symptom I've been encountering a lot recently: you'll be on a lift moving upwards and then seemingly without reason get squashed and go back down, even though there don't seem to be any nearby obstacles. This issue consistently prevented me from completing the final ascent on JoyModulos map on Ironwail, although the map had several playtesters and went through at least one post-release fix. I've encountered this thing before and since but it didn't occur to me it might be Ironwail-specific.
 
  • Like
Reactions: aoanla
If it helps, I also had zaratzara's experience both with Zetabyte's map in QBJ2 and with Joy%'s lift (although in that case, I think it was also that the lift was a weird shape and it was very easy to *just* clip something). I also had a lot of issues with the funicular in the same map doing the "I'm going to give you collision damage mid ascent" thing on the same map - but those have always been finicky, so I just thought it was good-old-Quake-physics-jank.
(I'm also ironwail 0.7.0)
 
  • Like
Reactions: zaratzara
I have noticed several problems in the AD map Tears of the False God in Ironwail...
Bal (the mapper for Tears) created, playtested and finalized the level using Quakespasm Spiked, which was the source-port of choice at the time. In general terms, Quakespasm Spiked changes the behavior of a lot of fundamental things in Quake, to the point where things that function on Quakespasm Spiked (specifically enemies being spawned in) may fail to work on any other engine. This is not a problem with vkQ, Ironwail, or any other engine. It's just a side-effect of using QSS as the main testing port.
As a result, you're better off just playing it in QSS and biting the performance bullet. At least the map will function.