Wish there would be an actual push to “finish” LPV or provide something better. Static lighting is just so **** for gameplay purposes (as in using graphics to enhance gameplay in an interactive medium). Not to mention the ******* file sizes are ridiculous for anything other than really small(map) game (as SQEenix found out).
I work on distributed swarm agents & coordinator for lightmassing and it’s very common see iddle nodes while others are still computing at 50% 60% total calculations due bad distributed calculations ( big meshes vs. little meshes balance)
I don’t know how algorithms could do the same without very deep learning…
Good God!!!.. Tell me this is real:) (as in going to exist out of beta), this is a game changer (excuse the pun) on so many levels(more unintended puns).I’ve applied for the Beta and need to be accepted:).The idea of this is fundamental to where I’m going with future proposals…I’m giddy with the thought of testing it.
Well, I’m not sure what scale you’re talking about. We have run a 25,000 object, 100M polygon scene through Datasmith (data came from CityEngine) and it worked. It just took awhile. This is the demo we ran on the SIGGRAPH stage (video sometime next week).
Love the enthusiasm! I’m pretty confident you’ll mostly have a rewarding experience. We already have the alpha users pretty excited about it, and they represent major architectural firms and some of the best agencies out there. Big projects are getting processed and we’re finding lots of ways to speed up the process.
Sounds impressive but I’m not fully aware of the structure coming from CityEngine.
If it’s procedural buildings chances are pretty high it’s already way more lightmap friendly than a 60 story skyscraper modeled by a noob who booled the windows and balconies and made the whole thing into a scruffy monolith mesh.
What happens in cases like that?
Another example would be Evermotion scenes. They can look quite lovely but are typically modelled on a disaster scale for UE4. Right up to the point where remodelling would actually be faster than trying to patch things up.
There’s tons more like these. It’s actually rare for us to get usable meshes that can be unwrapped for lightmass without some significant cutting and retopo.
I’m pretty sure you guys are aware too. Just hoping to see some support for cases like that cause that would be ace
Well, we need to get more people into beta, that’s for sure. However, I haven’t seen these complaints in the early beta and we have pretty much all the top architectural and agencies in there. Doesn’t mean that they have all had time to thoroughly thrash things. Hopefully, you’ll be in the beta and test these issues out.