| 1 Aug 2021 |
 cc9cii | I don't see our OpenCS providing that functionality when OpenMW doesn't use it? | 05:49:22 |
| cc9cii | That is why I mentioned my fork. | 05:49:31 |
 CMAugust#6575 | If it uses the same logic, the generated navmesh in CS should look the same as in game right? | 05:56:51 |
| cc9cii | yes, "in game" being Skyrim or CK | 06:02:18 |
| CMAugust#6575 | But not OpenMW's navmesh? | 06:06:25 |
| cc9cii | OpenMW does not use navmesh | 06:07:20 |
| cc9cii | not the way Skyrim uses them | 06:07:37 |
| cc9cii | you can see OpenMW's in-game rendering of the "tiles" | 06:08:56 |
| CMAugust#6575 | Ah, so you would rather CS not visualise OpenMW's navmesh when it's different to Skyrim's method. | 06:11:56 |
| cc9cii | In OpenCS it is more useful to see what OpenMW will actually use | 06:12:30 |
| CMAugust#6575 | Wait, I'm hopelessly confused. | 06:17:50 |
| cc9cii | OpenMW uses the output of recast library which is not the navmesh triangles | 06:18:29 |
| cc9cii | They are related, but not exactly the same | 06:18:38 |
| cc9cii | I was initially confused about that as well, until elsid explained it | 06:19:16 |
| CMAugust#6575 | Oh well. Anyway, I can see TR dev's point that it's hard to know how OpenMW nagivation will behave. | 06:21:44 |
| CMAugust#6575 | Until they actually load the level in game, that is. | 06:22:13 |
| cc9cii | In that case we just need to load elsid's cache to OpenCS and enable the debug rendering that shows the navmesh tiles | 06:22:58 |
| CMAugust#6575 | 👍 | 06:23:45 |
| cc9cii | Initially I was giving him gentle hints to use navemesh triangles as the common intermediate format but I don't think he took the hint (if he did, he wasn't too keen on the idea) | 06:24:14 |
| CMAugust#6575 | Ah, I see | 06:24:38 |
| cc9cii | I'm guessing that is because there will be extra processing (and possible resultant stutter) | 06:25:09 |
| cc9cii | But I think that is not right | 06:25:33 |
| cc9cii | I suspect any stutter is in the generation | 06:25:44 |
| cc9cii | not interpreting the navmesh | 06:25:55 |
| CMAugust#6575 | The stutter from generation is already pretty bad for lower-spec PCs, yeah. | 06:25:59 |
| CMAugust#6575 | Though that might be a regression, the report I heard today wasn't present in 0.46-0.47 apparently. | 06:27:43 |
|  Arbok joined the room. | 07:00:11 |
|  Sinscere joined the room. | 08:11:40 |
 Hemaris#9366 | for the record, I am the TR/PT dev who most recently asked about navmesh visualization within the OpenMW CS, but that idea came from elsid in this conversation: https://discord.com/channels/260439894298460160/260443579908882434/865316143739961344 | 16:53:36 |
 Quote#8404 | Levels have to be properly designed for working pathfinding. Navmesh is not a magic. I want to add navmesh generation to the editor so level designers can see what they're doing. Linked by Hemaris#9366 from #troubleshooting | 16:53:37 |
