But I did not refer to commercial engine use, just I mentioned the Blender community as a whole. Of course commercially BGE has never been advantageous, but then depends on the point of view of who is using it or analyzing it.
What I find is that the points of those who want to use commercially BGE are being taken seriously. As for enthusiasts, unfortunately limbo.
Honestly, I am disgusted when I start reading posts about the current situation of BGE and then place the trade issue as the primary focus for a possible improvement of the same by the BF (FB on my tongue).
What is clear to me most of the complaints about the BGE is that the fps drop dramatically, and these same complaints ask only that the income is stable.
If the most viable solution is to cut the BGE as Game Engine suite, go ahead, save time users it prolonging this useless discussion, since no official dev provide any concrete information.
This is not a criticism, just an outburst on my part.
bge is a great concept, that no one ever really pushed hard for since it’s integration in blender…that is what the problem is, not that it is a “silly idea” …ton and co never attracted or put the right amount of manpower behind bge for it to be truly competitive. That is a fact…there have been developments, but no push like we’ve seen for bmesh, new particles system, deps graph rewrite or cycles…as far as I can tell a lot of things come down to ton just not being interested, as we get older we see more things as being “gimmiky” and not usefull…I think that is how ton has been about a lot of things…he said it about ngons, colored wires, and I’m sure some other stuff…meh. The future of games inside blender is sketchy, and the fact that ton is sirting the “game engine” moniker scares me more…I think it is the end, but only time will tell.
The reason is simple: There are better options, both very cheap AND open source. When you can get UE4 for $20, there’s simply no way for any open source project to compete. Anyone serious will forego a couple of cups of coffee and grab UE4 or Unity, and anyone who is diehard tied to FOSS will grab Panda3d, GDevelop, or something similar whose capabilities long ago left BGE in the dust. So that leaves people using BGE as who? People who use it only because it’s part of Blender? People who use it only because they refuse to learn something new? Stubborn people? What is the impetus to continue developing it when the problems with it are so entrenched in its core, and (no offense to diehard BGE fans) nothing of note has ever come out of it? It would make more sense on pretty much every front to put more time into consolidating to a high-quality viewport and simply easing the pain of moving back and forth between Blender and more modern game engines.
I myself moved to unity long ago, but I always liked the idea of the bge, although I never used it for anything but to tinker.
I would be more inclined to agree if the issues facing the BGE could actually be solved without a complete rewrite, but between a code-base that actively turns off potential developers and the issue of the GPL, there’s not much that can be done with the existing solution.
and anyone who is diehard tied to FOSS will grab Panda3d, GDevelop, or something similar whose capabilities long ago left BGE in the dust
Don’t forget Godot, it’s already pretty capable and they’re getting close to going ahead a releasing the first candidate for version 1.1 (with node-based shader editing, more advanced code-completion technology, and full shading/lighting capabilities for 2D games). The only standalone FOSS engine as of now that is a full creation suite including a script editor.
Couldn’t have said it any better.
The only thing holding you back currently in the game engine, is the number of polygons you can draw per frame, as well as some issues with rasterizer and render… The viewport Fx branch has code in it that can be shared with the bge,
Physics has issues with traveling at high speed on triangle meshes,
This can be fixed by rewriting the bullet wrapper, to use the current CCD system used in blender core.
Doing 2 things Ton wants anyway,
Name something you can’t do with logic in blender you can do in a modern engine?
The issue with the current systen is it not well documented in the engine,
What do I mean by this?
Some good examples of usages, included in the blender package by clicking a link in the game settings window.
Like
A first person shooter setup (Modular and simple)
Guns(2 or 3)
Player armature
Player hit box,
Camera rig
A simple Ai example Leveraging the pathfinding code in blender
And with python you can script about anything you can’t do with logic.
The reason no one has made a full game, is we have 90% of a badass engine
My own game and team are gearing up,
Also some credit needs to be given to Moguri, Kupoman, HG1 and PGI, and LordLoki
The codebase is huge and intertwined,
And trying to fix it is not easy…
If the core devs took a look at how to untangle and rewrite what was there to be more moduler, developers could replace just the render or just logic or?
Right now all the *hooks into operating things are not flexible enough to swap modules without herendus effort.
If this was done, like logic bricks? Use that module, like nodes? Use that,
And standalone capacity just means booting blender into a game.
You remove it,
Someone will just write a standalone script that morphs blender into a standalone player.
Blender full install is not as big as 1/120 of planetside 2
You keep missing the point, BPR. The BF developers are not going to look into it in the way you want because they don’t see it as a priority. They aren’t interested in, motivated by, or otherwise looking into game development because, frankly, the money & management focus is coming from Blender Institute’s open film projects. It’s hard enough getting focus for game artists, let alone game developers unless you throw them $10K and up (& even then, you get an exporter not a game engine overhaul).
The kinds of work the BGE needs to make itself competitive (in terms of capability, not commercial application) to the offerings available today in (both open-source and commercial engines you can get for less than $25) is phenomenal… and not really of much benefit to the film projects. Even someone as enamoured of the engine as you are listed several large jobs that need to be done by someone well versed in the code, which means an experienced developer not assisting the Blender Foundation/Institute in regards to Gooseberry.
Honestly, spend the $20 to see what UE4 can offer you (complete with source code, samples, tutorial videos, and the capability of commercial sales on all major distribution platforms). You too cheap/broke to splash $20 on an engine with source code? Look at Godot or (commercially without source) Unity. And that’s just the popular options. Crysis is out there too for even cheaper than UE4, Doom3’s engine is available in GPL (if you’re a “free software only” purist), Sauerbraten (Cube2) & Tesseract allow you to edit the world whilst playing & viewing the game in realtime (MIT license), etc.
The BGE is simply too far behind to catch up without a concerted & focused effort by the Blender Foundation/Institute. Unfortunately, the BFI is too focused on film projects to do that and no-one is waving the sums of money at them necessary to change that focus.
I have used all of them,
None are wysiwyg
Blender I can model, skin, rig, animate, add logic click go, and have the actor land on his feet. No break in workflow,
The viewport is the limit I face at the moment, all of my game works,fine.
If you’ve really used all of the engines I listed and still think the BGE stands up favourably to all of them, then you are looking for things in a game engine that the vast majority of game developers are not (& I honestly have no idea what that would be).
There simply isn’t enough time saved using Blender & the BGE together to make up for the lack of flexibility, functionality, and performance you get elsewhere. At least for me & everyone I’ve spoken who actually works on games for a living (indie & AAA). So when you say it is better for you than the rest, I honestly have no reference point with which to analyse your point.
That said, whether or not we agree on which engines are better, the fact remains that the BFI developers are currently (and according to their plans, for the foreseeable future) occupied by features that will assist in the Gooseberry open film project. The game engine requires a concerted effort by skilled developers, familiar with the code, and over a significant time period in order to fix the issues even you have with it. Even if you are 100% right about the BGE & I’m 100% wrong - I don’t see the resources being dedicated to the job whilst-soever the Blender Institute has an open film project running.
Viewport Fx branch OpenGL, would fix much of what is is wrong from what I understand,
The bge and viewport use the same code when the viewport is in gsl mode.
Also, no one knows how to use the bge,
What is missing that you need?
Premade pay assets?
Do you know how to use events and python together?
What can’t you do that is not graphical. OpenGL /shader speed related?
Also we have many blender game devs, and just got a New dev, PGI
They work for free.
If the core devs helped for a month after the open movie…
BTolputt; I don’t think you’ll be able to convince BPR that the BGE has major issues both development-wise and usability-wise.
Once he gets a passion about defending and trying to promote something, there is pretty much no stopping him until said product either disappears altogether or it becomes a major force in its respective industry.
Do note that he has mentioned before of his brain literally being wired differently so you’re going to get nowhere as long as the BGE is still in Blender. One of the only ways otherwise is for the devs. to simply remove the BGE before the 2.74 release so we can clear the way for bridges with other engines and ease the pain for that part of the community.
LibLoad / LibFree need work*
This is a major system that could use attention - used in large open world games
Light system needs to delete all lights at runtime, and add stand ins,
Then populate only the closest 8 lamps or 16 etc,
memory leak in video texture somewhere-
Add anything that needs fixed to this list please.
It would fix “some” of what is wrong, but it is nowhere near “most”.
No-one knew how to use Unity, UE4, Sauerbraten, Nexus, etc when they first came out either. People saw that they were worthwhile and went to the effort of learning them so they could make cool stuff with them. The thing one should ask yourself is why do those people who know the BGE (and they exist, you’re not a unique snowflake 
) decide to abandon that knowledge? I humbly suggest it has something t do with the (lack of) utility they see in the engine.
Quite a lot actually. However we’ve done this list before. You didn’t accept it then, I am not going to run around in the same circles for you again. If it makes you happy, the one you cannot get around is the requirement for GPL compliance on any & all code I develop used by &/or linked to the game. This rules out two platforms I develop for & is distasteful/painful on the rest.
I am more than skilled enough in Python to handle the BGE. For what it’s worth, I’m not a big fan of Python for games because of it’s memory footprint, speed, lack of sandbox capability, issues with multi-threading, etc. It’s not a lack of knowledge that makes me hesitant to use it, it’s the fact I know more than enough about it’s flaws.
Power of two based snapping when putting together an environment from prefabs, auto-LOD terrain, good-looking procedural content (trees, buildings, litter, etc), octree-based occlusion, GPU particles, Android/iOS platform compatibility, built-in behaviour trees, runtime GI lighting whilst editing the scene, etc, etc.
There is a veritable tonne of things that the BGE cannot do that the alternatives can. I’m not exactly sure why you think asking that question helps your case
That’s great. Have they got a commitment from the BFI that their patches will be merged into trunk? How long are they going to stick around? How much time are they able to commit per week working on the BGE? Getting the BGE up to par with the alternatives isn’t something that you can accomplish just spending twenty minutes a night on the task. It’s a huge job touching parts of the Blender internals that make the BFI developers wince at the thought of newbies messing around in.
You would still be nowhere near the alternatives available. It’s not something that is going to be fixed in a development sprint. It’s a fact that there is no official focus on the engine, hasn’t been for some time, and no chance there is going to be in the near future. The best bet for BGE improvement is a fork that allows the BGE devs control over code that the BFI developers won’t let them patch… and we’ve seen just how many of them are around :rolleyes:
Yeah, I’m beginning to see that. There is hope (you were initially unconvinced) but it would appear he is arguing from a sunken cost perspective (he’s done so much in the BGE he doesn’t want to admit that effort is/was wasted). Very hard to get through that because people that have invested a lot of time/effort into a project don’t want to accept they’re not going to get as much from it as they thought. It’s actually a deep-rooted psychological drive that affects decision making at all levels (from government through to gamblers at a poker table).
The bge is an idea, that each artist that uses it, uses it free,
and in return help it grow.
“If that works, and our animation system is smooth and fast, and physics unified, then we only have one main target to upgrade: A decent logic editor and a way to have logic and playback work smoothly inside the animation system.”
If I understood well those targets, the Interactive mode is nothing more than a mode that makes everything realtime?
Is it really intended to replace the BGE? BGE could stay as it is an keep fixing bugs with future update.
The BGE works quite fine actually. I’m using it since some months, so I knew criticisms before to start using it but its integration within Blender is so great that I give it a try. And I don’t want to go back on another engine, I really love it. I tried Godot which is great and I also tried Unity to see how behave FBX export and I found that it’s so boring to make exports, to make sure that your rig setup will work after export, etc. Stay in Blender make that a lot easier.
Blender Game Engine is not perfect of course, but when I do something in the Game mode, a rig for example, I’m 100% sure that it will behave how I want in game. Same thing for material even with a complex shading node setup: the viewport in texture mode displays what I’ll get in game. And that’s awesome, I really like it.
Does the new viewport will be available in the BGE as well? If yes, shouldn’t it increase performance? and what about using openlgl 2.1 or newer version? will it affect the game engine?
I’ll be very sad if BGE disappear, but very excited about Interactive more. If it makes “bpy” as interefctive as “bge”, it would be very very great. Imagine dynamic creation of mesh “in game”, etc.
I was starting to be convinced because of the recent nature of BGE development (it was in a state where game developers were adding workarounds in each subsequent release rather than removing them thanks to the slowly increasing regression count). Then I looked at the emerging Open Source solution known as Godot, and I figured that there’s not much use for people to use the BGE anymore because you now have access to a better solution that isn’t a cut down free version or a subscription-based product.
The BGE used to be one of the only FOSS solutions that was built in a truly generic way (no taking other game frameworks and changing the rules) and was a full creation suite, Godot upended that fact and is now the open source world’s most promising solution of that type with its superior code base and consistent development. There’s even two other independent solutions being written by Blender users that plug into Blender in a very similar way to the BGE (which further brings its future into question because they were developers who hit a brick wall with the messy BGE source).
Not refuse to learn anything new, so I’m giving a chance for Godot, which compared with Unity seemed more attractive.
Anyway, the fact Ton want to integrate the BGE in something that will perhaps be called interactive mode indicates that BGE will not be cut off from Blender. From what I understand, what worries the current BGE users is the possibility of not having to export a game executable. In this regard, I believe that will not be too difficult algume create an add-on covers to do the same run-time role.
Anyway, we have to wait to get to 2.8 or more Blender version to have a concrete idea of futro Blender suite. Until then, everything is nothing more than mere speculation, in some cases inflamed.