I think unity is working on something bigger ECS relevant and want to announce it with a big bang. ECS was introduced very early on, and then there was very early hype about a product that would still take years to develop. After the “bang” from Unreal Nanite / Lumen, they probably want something like that too.
I dont think speculating is against the rules nor in bad health I am sure many people lately are thinking the same thing, so what is wrong about finding out how everyone as a whole feels?
I think the correct response to this is to vote, not to try and stop others from expressing their views. Even if maybe they are not what you are hoping for
It’s not looking too good for DOTS, everything else is improving though, so we shouldn’t be too harsh on that end. 2021.2 and 2022.1 is looking good!
But DOTS is going abysmal. No release this whole year, stuck in 2020.3. I don’t get it, really, they surely have a good reason but they haven’t communicated it in any way shape or form.
Probably they did a complete rewrite of Entities and due to that, everything else is on hold because it’s unclear how modules on top of Entities have to be rewritten. If it weren’t for that, I can’t imagine taking such a long break.
I’m still hopeful, but the long, undefined wait is really killing my vibe.
Yes there are several key features which are in a rough state, but if you’re holding off on using them then Unity is actually in a pretty good state.
What I think has happened, is Unity poorly estimated the time and effort needed to get some of these new features into a complete and stable state. Give it a year IMO.
Thats nothing to do with Blender, the Flex/token licensing doesnt even make sense for anyone else then large companies. Token licensing has been available for some time now already ( called Flex licensing )
The typical blender user has no use for such a licensing scheme, as
A) you have to buy/spend tokens worth a minimum of $1.5K ( as far i remember )
B) those tokens become invalid within a year
Where’s the benefit in that, except if you are a big company/enterprise, which would otherwise have to pay full subscriptions for all packages, even if the product would only be used occasionaly
For small studios the flex/token licensing does not provide anything
What if Unity brought out UnityDOTS a DOTS only version of Unity, dropping all the legacy API’s and even maybe adopting a more DOTS focused programming language e.g. F#?
Could splitting Unity up into different products actually make for a better Unity (pun not intended)?
In theory a set of dedicated Unity products could massively reduce the complexity of each one, but what then would be the best way to divide and conquer Unity?
lol like project Tiny? Not F# but it’s a whole thing of its own and nobody uses it afaik.
I think a big thing holding dots back is how different it is to get animations working and transitioning from 1 texture to the next. There are a ton of examples with thousands of spheres/capsules moving around, I saw 1 asset where the guy had a character controller and he spent a ton of time developing it.
Honestly the only reason I haven’t bought that is because it only shows 1 character animated…if you could render a 1000 animated characters wouldn’t you show that in the demo?!? The lack of showing that implies it doesn’t. Though it is on sale now.
The example I played around with from Joachim only had 1 animation per model and no transitions. I could render 120mil skinned mesh verts at a decent framerate but no idea how you transition to another animation. And it seems like a research project to figure it out. Or just wait until they figure their shit out…