Future of BGE and Eevee

You are right that he did not start the discussion. However, Tristan did and clearly stated that the focus is to “only” integrating Eevee into the game engine. The person who does the actual work clearly defined the scope. Since the discussions in that list are always rather focused, everything outside of that scope is clearly noise or off topic.
You are free to call that as being an aggressive or negative move from my side to state it like that. It is the way in which this mailing list works.

As I wrote, there are very few exceptions. However, there are rarely messages from those people which are off topic.

This mailing list is for people who intend to do actual work on the Blender source code. Usually the focus is on the C side, because for Python there are other mailing lists too. But there are exceptions for this as well of course.
However, it is the same story. It is for people who intend to actually work on code that is supposed to be committed to Blender, or it has to be at least within the scope that the person who does the work is defining.

It is usually not a place for brainstorming and certainly not a feature request place.

Interestingly, I just saw that there is another post: https://lists.blender.org/pipermail/bf-committers/2017-July/048527.html

I am waiting for the developers to either ignore that completely or to point out that the mailing list is not for that kind of discussion. By the way, it wouldn’t be the first time…

I can’t be the only one who’s confused by all these names that are thrown around.

•Armory engine
•UPBGE
•Eeve

I have yet to try any of them out, which is the best one and which one is going to be the “future” of the blender game engine. Armory engine looks most impressive, UPBGE I don’t even know what is and eeve seems to just be better real time rendering, not even sure if it has anything to do with the game engine.

Armory engine allows you to set up materials and models within Blender. But it is not a part of Blender and it is also not freely available as you can see here: http://armory3d.org/download.html
UPBGE is a fork of the official BGE. UPBGE contains a lot of improvements and cleanups that are not in the BGE. It is planned to merge UPBGE into Blender 2.8 to make it the official BGE.
Eevee is the new real time renderer for the Blender 2.8 viewport. The proposal is to use Eevee also as the renderer for the BGE.

•I looked through the feature list on UPBGE and the most useful things I could find were subsurface scattering and a cleaner UI. Most of the features seems to be under the hood stuff that won’t help anyone make a better game.
•Armory engine looks impressive, 50 dollars seems a low prize to pay for a game engine that actually looks like it’s capable of creating a game, the logic nodes seems to open up possabilties for smoother gameplay. If anything this is the future of the bge.
•I tried eeve and I don’t get the hype, slightly better rendering? How exactly is that going to enable people to make better games. Even if combined with UPBGE it still looks like the game engine lacks basic features.

I think the choice is obvious, armory engine is the future of the game engine.

The game engine was not really maintainable anymore and the work they did was necessary to make it possible to continue making improvements at all without inevitably breaking other functionality. As far as I can see, they did a great job.

Besides the improved visuals due to physically based shading, they integrated lots of performance improvements. If you had to work with large scenes containing lots of objects previously, it pretty quickly became slow. Even more so when animating characters. It was unavoidable in many situations to use proxy characters. With Eevee, this is not anymore needed thanks to the improvements they made.
For games, this will mean that the performance is improved and that this can be used e.g. to have more detailed characters or more characters or a more immersive environment. It basically gives game designers more possibilities.
In an optimal case it will also mean that if e.g. they have to make performance improvements in the viewport due to a Blender Animation Studio project, the Blender Game Engine could directly profit from that as well and the other way around as well.

The Blender game engine is the thing inside of Blender, by definition. As such, it is not the future of the Blender game engine. Especially not because it is not a free open source project.
It certainly looks like an impressive project which can be a good choice for many games.

The Blender game engine is the thing inside of Blender, by definition. As such, it is not the future of the Blender game engine. Especially not because it is not a free open source project.
It certainly looks like an impressive project which can be a good choice for many games.

Personally I think they should remove the blender game engine completely from blender, either you do it right or not at all. The graphical limitations don’t really matter, games can be “ugly” and still playable. Armory engine is doing it right, why would anyone ever attempt to make a game in blender’s game engine when armory engine exists. Because it costs 50 dollars and blender is free, if you’re serious about making some game that’s going to be money worth spent.

just walk away.

its not you putting any effort on the line. the devs will spend their efforts as they choose, and there are many who are grateful.

THANK YOU BLENDER/UPBGE DEVS!!

You can say that about every game engine. If you find a game engine that suits your needs, there is no reason to use another one. You found your one, that’s great.
In my opinion the Blender game engine has a lot of potential. I think that the movie making parts can profit from the game engine and the other way around.
It is not that unlikely that someone implements a terrific camera solution for an easy and quick previsualization of movies at one point. In the sense of move the camera from here to here and while you do so, have the character in a certain section of the frame and at the end transition to another character smoothly. It would be fantastic to use this kind of stuff in the game engine too, without reinventing the wheel.
On the other hand, it would be fantastic to use the animation capabilities of the game engine for previsualization purposes. It would be amazing to be able to say when the character has to be where and the animation is automatically played back, depending on the speed and rotation speed. But maybe that is not enough for previsualization and it would be useful to transition between different walking styles and in the case they implement it, the game engine could profit as well.
At the same time this kind of previsualization tools could be used for cut scenes in the game engine.

Those are just examples of how they could profit from each other in the future. There is no plan for anything like that, but I strongly believe that there is a huge potential to bring Blender and the game engine closer to each other.

@Dantus: Tristan stated that his plan was letting bge use eevee…true…after it was mentioned that interactive mode should somehow be a part of that…he did not simply state ‘no’. He clearly stated something along the lines of yes, but first off I would like to get the game engine going(this is clearly not a quote and I do not speak for him)…I do not know what will happen…none of us really do.

I hope to get eevee+BGE or +UPBGE…something…I would even be ok with interactive mode…with bpy running the logic…so long as those scenes could be exported to a file that others can execute…tbh I would love the latter…the full arsenal of bpy may be better…it opens up more, but it is also not very gamer-centric…but from there we would have the basic building blocks to move forward…these are just random thoughts as I am writing this…I still wonder why you are so…negative…did anyone here make you feel threatened …or have we talked down to you?..something is clearly not right…or I am just reading into the things that you are saying…because from my POV I am getting a ‘I am right, everyone else is wrong…and you are idiots’ type of vibe…just saying. If I somehow irritated you…send me a pM, we can work it out…

everyone remembers endi right :wink: He even haunted the regular forums…

I’d like to think as game developers, or aspiring game developers…we are all amongst friends. I don’t think most of us here have people in our day to day lives that we can talk to about or game dev endeavors…tbh, you people are the only people that ‘get it’…and I want to keep things friendly.

UPBGE has the merit of having its own dev. team behind it, so it’s not taking much time away from the core developers (which is obviously better spent on the current things being tackled for 2.8). At least it doesn’t take much time outside of discussion on how to proceed with its integration as the drop in replacement for the current engine.

I agree that he didn’t write a clear “no”, but it was a clear “not now”. Which means for me it is off the table for now and no discussion is needed.

I am usually not a “I am right” kind of person. In this particular situation, it should be fairly clear just when looking at the mailing list, that its purpose is to have a communication channel for the kind of people I mentioned, which is thankfully not locked behind closed doors. There are usually no lengthy discussion or anything like that, because the purpose is to keep it on the point and short.
If someone joins the list and does not respect that and starts to talk about topics that are not actually being developed at that point, it is an unnecessary distraction and very disrespectful in my opinion.
This is nothing personal against anyone. This is how the mailing list works since years and Jacob know that, because he is the only person I am aware of who got temporarily banned because of distractions.

In all fairness, the tendency to create a tangent towards things of personal interest is probably more a result of his Autistic nature than anything else (which he has said before he has). I know I had issues with doing that before in the past (both online and in real life).

However, he is capable of learning how to control this for the most part and the BF tends to run a tight ship on both the developer website and the mailing list (they are far less tolerant of noise and random topic shifts than this forum). This forum is actually pretty lenient on this type of thing in some cases so there should be no surprise if you hit a wall doing the same thing there.

I am aware of that and in fact, it is the reason why I am trying to formulate it as clearly as possible.

I just went back to reread my posts as my intention was to write as objectively as possible. The first post was clearly not like this and that’s why I want to apologize to @JustinBarrett and @BluePrintRandom.

I actually can do all I need in logic right now in upbge.

The only thing I could use is a compiled fast node based A*

this is for people just starting not to hit a wall in logic as hard and for the raw speed compiled code can wield, yet remain flexible.

propose a better solution and we will talk.

I have studied game design and python and game engines for years,

once BP-player bidirectional streaming lands normal upbge(pre Eevee)
will already be a brute

Hi, I share a new video of our work in progress in upbge 2.8 branch: https://youtu.be/J_zt0JVNZcc
I’d like to modify what I said in the first thread, there was a misanderstanding. I thought that the main upbge dev had been hired to port upbge work into Blender official but in fact, there’s no definitive decision about the future of the official bge as far I know. I’m very sorry that I gave wrong infos about that (I didn’t understood correctly).

But we’re still working on upbge and for now, we try to make eevee’s features work in upbge. Thanks very much to panzergame (main upbge dev) for his work and to Blender/2.8 devs who are doing an awesome work!!

this looks amazing

nice work everyone!

To allow everyone to have a powerful overview of BGE/UPBGE/Armory/Cycles/EEVEE’s future, let me show yous how I organize their capabilities.

  1. Render — Covers graphics, etc.
  2. Objects — Covers animation, constraints, tools, sensors/motors, etc.
  3. Physics — Covers physics.
  4. Code — Covers computation, its speed/parallelism, code language calling/hearing to and from Blender.
  5. Real-Time Interactivity — Covers lag workarounds that’re available in all Engines, ability to call to Blender by user input.

Now ask yourselves what do we need for video games or advanced AI simulations like I will be doing in Blender? We already have a UE4. I’ll prove it to yous. Look above as you read the following. Do we have sensors/motors etc for an AI robot baby? Yep. Photorealistic render after BGE runs by importing animation into Cycles? Yep. PhysX comparable physics? Yep. Code ability? Yep. GPU in BGE externally? Yep. We have it all. What am I missing??? And what should we work on first? For video games, the only thing you’re missing overlooking law and networking is nice graphics *IN BGE/etc. Armory/etc covers that soon. Only thing missing is a stable fps function.

If armatures/etc have problems in UPBGE like the website says, that’s worse than BF’s BGE.

<insert obligatory “Yet another cringelord speculation thread”>
Also the obligatory “BGE has its niche please don’t remove so long as someone is willing to maintain it”

@youle that’s perdy :]

Why cut it? Does a few extra megabyte in the installation hurt that bad? I’ve seen huge programs that take a long time to install. Keeping BGE will not only give me for one a powerful simulator, but also allow me to drive my project, and not have to redo things in other Engines, etc. Don’t cut it. Add it as a addon if you haaave to. There’s absolutely no reason it can’t exist as that little option at the top of the application window to choose your Engine from. It’s like saying we should cut back how much land we own. You can have 8 million acres on a farm.