VGOEmulator.net

Bug ID       : 225 - 1. invite toon_b into your gro... 
Bug Date     : 2015/07/12 03:13:05
Assigned To  : zippyzee
Priority     : Critical

Category     : VGClient
Sub-Category : Gameplay: Other
Severity     : Standard
Reproducible : Every time

Details:
1. invite toon_b into your group (optional step)
2. run a ways out, start attacking random npc_a
3. use .goto toon_b command before npc_a dies (may be optional)
4. toon_b will be invisible, nor can you target with F2
5. other local npcs will also be invis

Originated From World: Targonor (2)
Chunk                : Tanvu (14)
Location             : -36221 -50072 3005

Zippy/Xinux, wasn't sure which one of you would be interested in taking a look at this. Our hero, Moldew, has figured out a way to replicate the "I can't see shit" effect on Targonor. This makes it seem like it's related to Combat, which seems like a stretch, but look at the bug notes. It might also reflect what Valorith and Xinghao experience on Targonor with vanishing entities.

Please have a look, try and reproduce. If you need Targ's DB locally, I can provide that.

What happens if you just run normal to the other person and not use the .goto command?

I don't see how the combat code has a way of doing this unless players who have their actor staus as in-combat somehow don't show up. Otherwise nothing changes when a character enters combat other than the animations for damage and updating health bar, etc. There's no magic under the hood that would make me worry about side effects suddenly arising.

That's why you have me, zippy... to think outside the box

Combat, the art of killing something, despawns things. That's new. That isn't on New Telon. Far as I know, this problem doesn't exist on NT. So if despawning something now suddenly has the effect of a teleporting client not seeing things, I would suspect it is worth investigating.

The "nothing has changed" argument is out the window, buddy

I thought he said the behavior occurs before the mob is dead, and that the pc is invisible. That doesn't fit the despawn pattern in my view.

Pc's do not despawn with my code, at all, so that can't be an issue.

If an NPC despawns, it is gone from the chunk. Nobody should be able to interact or even see it afterward. When it respawns it does so like any other chunk spawn, and doesn't depend on the client.

The pattern is also that a number of npc's are invisible to the other player. That seems like another issue hard to explain by despawning. I could understand the npc that was in combat being a problem, but not others.

If there is some issue with the npc list being locked during processing, that could create weirdness, but I'm not sure that fits here.

I can think outside the box plenty, but having such an unusual set of circumstances to occur makes it that much harder to pinpoint. There needs to be something more direct to pin it down or I can't very well conjecture what could be happening. If I had to guess, there is something about the goto that has to do with targetting. Some of the combat operations depend on the offensive target. That may confuse some part of this.

The spawns not showing entirely to a player issue seems like a bigger stretch. Try turning the combat code off using the server rule and test the same scenarios. At least then we would know for sure that it is screwing something up. Also, we need to look in whatever parts of the code cycles through the spawns and feeds them to the Unreal engine for displaying to the player. It would be interesting to know if the whole process bails out if one spawn is unavailable, or if it will continue feeding more spawns and ignore erroneous ones. I am actually plaing dangerously near that chunk of code right now with aggression checks so I'll take a peek.

I will soon have a code update but don't think any of it directly pertains to this or would miraculously cure it. The respawn timer will go to the spawn table value which is closer to 5 minutes. That might make some things more obvious. I'll continue to dig.

[quote="zippyzee"]I can think outside the box plenty[/quote]
That is good news. I can finally stop wasting my time offering suggestions.

if you come up with any additional clues/suggestions let me know. I'm thinking out loud to stimulate thought, not to disagree with your assessment. I'm only going off the info I can read above and offering my candid opinion based on what I added programmtically.