Sounds like it will be a while until the update and when v2 comes out, the thing will just work better and faster.
I understand about Unity is in fluke, but what about the nested prefab issues.
Any advice on how to work with nested prefab in the meantime?
It is more like workflow issues rather than performance.
Please read my last post above for more details.
Thanks
I am having a bit of trouble. I followed the tutorials and set up a test scene with a hallway. It seems to stream in fine going one way but the sections donât seem to sit right next to each other on every piece. So when you walk back through the hallway it doesnât spawn in at a certain point. Essentially I am trying to make it stream in a few rooms around the player but the rooms could be all sorts of different sizes. Screenshot - 49267ed7a1b9077efb2edbee54fdcc07 - Gyazo
Check objects pivots in relation to streamer grid.
I think such construction is rather non usefull. For such cases you got colider streamer. nIt would be pretty hard to manage all this scenes in such way. As we stream scenes we have nothing to prefabs at all currently. It will work with everything. Anyway as ecs comes we will see what we can do.
Is there a tutorial that better explains using collider streaming in combination with Layer streaming? I can get either one to work by itself but not combined together. My whole world is comprised of rooms and sometimes loading a room with a collider slows it down too much so I would like to have the room start like 3 or for layers to start loading.
Why it slowdown? Too much data or?
It is for VR so the loads have to be pretty small to not experience the steam compositor loading screen.
Hi, did you take a look at MegaCity? They release the asset at GDC 2019. One of the interesting things is that they support sub-scene streaming. It supports streaming low level LOD first and it also talks about low-level LOD can stay once itâs loaded without being destroyed if desired.
If we use the same setup, we donât have to have so many scene files. We can have one Scene file for one Grid element and stream in part of the scene, such as Small, Medium, Large, and also different LODs separately. It will defienely make the Scene management easier and I hope there are some peformance improvements as well. What do you think? Is there something we can take an advantage here?
Itâs really exciting and I really curious where World Streamer is headed.
Let we release r.a.m 2.0 first and we will sit into it:) for sure we will check possibilities from that project.
Haha â I came here to ask that exact question. 
I was playing with the World Streamer demo yesterday, and to be honest, I found the demo a little underwhelming performance-wise (even before the release of the ECS + texture/scene streaming features. )
It got me thinking â I have yet to seriously integrate World Streamer into any projects because I keep running into the same need to see long distances. And if performance stutters even with such a low draw distances (and heavy fog) like it does in the demo, I canât really justify using it in a serious open-world project. 
Using the âCurved Worldsâ plugin to avoid the crippling low draw-distances in the airplane video was clever, but I canât do that in any of my use-cases.
I really do think itâs a good idea to use ECS streaming like in the Mega City demo.
That being said â I too am excited and curious as to the future of World Streamer.
I would really like to put World Streamer to good use for once!
Truth is if we would release ws on 2018.3 we would have to rebuild it for 2019.1 
I feel your painâŚ
Lots of developers (including myself) are having to rewrite (and redesign!) many core things (in my case, nearly my entire asset!) from the ground up thanks to a new feature in 2019.1!
IMO â This is a good problem to have! â Thanks to these new engine/editor capabilities, I can provide the ultimate value to people who use and trust my tools to work when/where/how they need them! â Those very same people talk about my tools, recommending them to their like-minded friends! And since nothing ever goes away on the internet, a good name gets more valuable with time. Providing exceptional value (even when it costs, and you donât have to) speeds this process up! â The simple act of a âsurprise giftâ just naturally makes people happy enough to talk about its source.
Just my two-cents.
well still waiting to buy if it would be good with new unity versions. i hope you wont postpone for 2019.1 & 2019.2 too?
great work. that would be amazing for a multiplayer setup!!
Now that RAM v2 (2019) has been released, are you going to be working on WS full steam? I really hope so and we can take advantage of what MegaCity is doing in WS soon.
Subscene streaming will simplify quite a lot of thing and improve performance quite a bit, I hope.
Cheers!
Is this something youâre looking into doing, @NatureManufacture ?
I feel like it wouldnât be such a bad idea to get some official input on thisâŚ
Yes, we currently work on WS 2019:)
The awesomeness!!
So one MAJOR request I have is to give the ability to increase the draw distance (by plugging-in either Amplify Imposters (paid) or IMP (which is free) to let us procedurally bake imposters for large cells of land/details. These can be streamed in later before the individual (smaller) scenes/cells within the imposterâs loading zone (or the HLOD itself) gets loaded.
Generating these giant imposter cells manually alongside your system is a pretty huge task because we will eventually have to hook into WS, which means learning the source code and how it works.
Loading the impostors and HLODs during the streaming process is necessary for the large draw distances of todayâs games. The old system really wonât work for this.
So to make that loading process more simple, letting WS request the actual generation of those impostors and HLODs procedurally (at the right moments) and then taking the resulting data and making it streamable via ECS would make the whole process a breeze for the user and suddenly make true open-worlds for Unity accessible.
Add onto WS the texture-streaming for imposters (that games like Fortnite require for their incredible view distances), and weâd be golden.
Any chance of any of this happening? D:
Its good time to gather all ideas and requests:)