I totally agree with you. It was mentioned that stability is needed, because people are investing money. I meant it in the context of long term investments, the Blender Game Engine is likely not the first choice at this point, because it is going to change radically.
This is exaclty what, in my previous post, I was suggesting Ton should have stated clearly. “GE will change a lot, will be (re)written for blender purpose, go trust on it because it will sure last for <n> more years”, etc. That is, sharing his vision.
But remember that I don’t have a complete history of Blender/BGE from the very beginning, maybe I missed something along the way.
Something like this:
Exactly, a roadmap. I could not find something more recent than 2013 (related to GE). For a project like Blender, it would be a ‘must’, with an update once/twice a year.
This is why I have been surprised the initiative has been taken by sdfgeoff and not by Ton (afaik).
There are plans announced from time to time on code.blender.org. But, they are usually rather unreliable and since the game engine is pretty low on the priority list at this point, nothing is really happening. If nothing is happening, I don’t see a reason for any announcement, especially not if the plans have not been changed.
personally
next for me is UPBGE get forward+ light cuts using octree from bullet.
Well, yes and no. There are reasons even to state quiescent periods, like some gossip that my circulate (see the title of this topic), or to keep update newcomers. Also, no communications doesn’t strictly imply there is nothing new.
But maybe I am drifting a little OT, let’s see what will happen with IMBGE, I’m really curious.:yes:
“IMBGE”…what did I miss?
In the world of software development, silence usually means that nothing changed. There is so much gossip in this forum that it would require a full time communication person from the Blender Foundation to tackle it all. When it comes to Blender development, just ignore Gossip or at least check the sources and you should usually be fine.
He means the “Interactive Mode” with that.
upbge evoles daily
ge_normals branch has the command
KX_meshProxy.recalcNormals()
mathfu is tests to update the mathsystem
no modifiers was merged recently, (applies all modifiers at game start)
[allowing the removal of ‘fallback render code’ I think]
youle tinkers with eevee and bugfixes and adds features
panzergame has been doing major upgrades and changes to physics.
TwisterGe is working on a gui system for games
lordLoki has not been committing as much, but has some huuuuuge commits.
I know: The UPBGE is an evolving project - but the way it is developed does not match what Ton wants as the future engine (the IMBGE, I call it.) He does not want the old codebase at all, no matter how much it evolves. He just want the features a game engine has inside of blenders features that already exist.
Seriously: Read the original post, word for word. Read it carefully.
Very, VERY carefully!
Now, it isn’t that UPBGE is a bad engine. It is just that the developers of it treat the old codebase as the definition of what it is as an engine itself. Not bad way to think of it, however - due to it being a tangible product that already has a fanbase, of course it attracts people to use it and even improve it.
It’s kind of funny, if you think about it: The Foundation Developers keep telling people the reasons why the official BF BGE has not been improved upon for several years. And yet there is a nicely improved version of it already that, if they wanted to, to review the changes made by the UPBGE, and merge the progress. However, then they complain that there is no one yet willing to create their ‘IMBGE’, and from that it isn’t a real priority to even develop it - nor merge such nice progress of the existing UPBGE.
Sounds like 'My way or the Highway" on each side of the coin, and even more so about the foundation…
I mean, I’d say that the fact that they’re going to have to pay someone out of nowhere to build the IMBGE - as opposed to getting true voluntary support first - speaks a lot to me on how they’re going about it. Most blender projects I’ve seen started as voluntary wishes of developers to just improve Blender, even before payment comes about. This IMBGE… Not so much, if you ask me.
it’s not old as they have been updating it constantly.
they have been updating and replacing old bad code with new good code for quite some time.
lordloki upgraded to sdl2 for example
many many old bugs have died
bottlenecks cracked wide open
you have no idea what the code base looks like.
That’s beyond my entire point. This is my point from reading what the OP pointed out:
I know: The UPBGE is an evolving project - but the way it is developed does not match what Ton wants as the future engine (the IMBGE, I call it.) He does not want the old codebase at all, no matter how much it evolves. He just wants the features a game engine has inside of blenders features that already exist.
yeah and upbge is free to meanwhile pull eevee out and use it if they see fit 
if Ton wants a to not merge I am sure upbge will grow on without his help.
I personally see upbge adding support for forward+ lamp cuts, and for grids of prerendered cube map reflections,
with
reflect stencil = metalness
cube map = mix of mipmap levels of cube map based on roughness
global illumination done by probes
we have pbr anyway with or without eevee.
if this were all pakaged up neatly in editable glsl node setup and shared, or even a ubber material replacement for mpans easygame people who don’t code can have pbr.
If so, then it sounds like the UPBGE would go the “BforArtists” route… Only much more politically correct. 
I have been looking a UPBGE’s approach to improving the engine, and it is opposite of what Ton wants: It sounds more like the Engine assimilating Blender’s features, as opposed to Blender assimilating the Engine’s features. It’s the “who’s consuming who” that tells me the different ambitions of the two coding groups. Nither one would be ideal to me, but I’m just reading this from the OP of this thread.
If the UPBGE wishes to really commit to that route, I’d recommend them doing a full fork that bends the whole code base of blender to the bidding of the engine. It would a crazy task to do, but it would really improve the experience of the engine. After all, this was the greatest part of using it: Having the modeling pipeline integrated into game creation.
I mean, just imagine how nice the engine experience could be if there was a full workflow between all aspects of game creation - to a point that you never have to use external software at all just to make good ones. There could be improvements to the text editor, a sound editor integrated, a workflow-based interface designed primarily for assets, etc. It would be a very neat project, and it would be a great way to revolutionize how we can make games.
upbge is a fork that has been a fork for years**
If we speak in terms of that, it is only half a fork: Only the Game Engine was really forked here - not blender as a whole. It doesn’t really mess with the rest of the codebase much at all. If they decided at some point to modify that part to a good degree, that is what I mean by a “full fork”
PLEASE KEEP THEORY BGE. DO NOT REMOVE BGE. PLEASE. I’m planning take a video game with it since it’s open source and free!
Do read the original post: It won’t be removed until this Interactive Mode (I call IMBGE) is finished. When it will be finished, though… Well, for all I know, it hasn’t even been started yet, so I’d guess the older code base will stay around in Blender even when 2.8 gets released.
@cflow Okay I now read the original post (and pardon me for not reading it before).
I’m concerned about whether Ton’s artist-vision includes a Blender capable of making and selling video games…
I don’t have lots of money nor do I wanna give my rights to someone else for developing a game…
Will/Does Ton consider making a competitive video game “art”? There certainly is a creative aspect to it. But also, I have to admit that my goal isn’t the same as my Dad’s artistic goals… There is overlap. But I wouldn’t say it’s 100% overlap…
Maybe someone will join in to lead developing a game engine for video games??
Or should I be looking for other open source game-making alternatives?
But hey: if Ton made Blender open source, then I have to thank Ton for making my back-up plan affordable: animation!
In short: I HOPE THE FUTURE WILL BRING AND SUPPORT A GOOD OPEN-SOURCE GAME-MAKER!!! >o<