Why no serious game on Godot can be seen yet ? Godot is well promoted but still, i saw nothing serious coming out of it
There actually is a notable 3D title made with Godot 3.0 on Steam.
There’s also another big title slated to come out this year.
I’ll just toss in the list of Godot games currently residing on their own indieDB pages (with some of them using the 3D engine)
Heres one im working on:
Theres a few issues with the 3d side of the engine that wont be fixed till 3.2 Vulkan release. I think thats the main reason why we are still to see many 3d games…
Name some serious games.
ha ha ha
your game looks nice.
What do you define as “serious”?
Keep in mind that it’s only been maybe 3-4 years since version 1 came out. To have the titles mentioned already says a lot about its vitality as a respectable engine. It has a lot going for it such as the massive number of user-submitted patches that add features and fix bugs.
I again refer to the question above, is there is a hard metric for this or is it flexible enough so no matter what projects are posted, the goalposts are moved in an attempt to show the majesty of UPBGE?
What browser are you using?
My link shows a list of 92 games on Google Chrome, could it be some cache issue you’re having?
Bro…He is trolling you
Just look in the search bar
So why is this thread here, are people in the BGE forum really seeking an honest discussion of what has been done in Godot, or was it done to create a punching bag for UPBGE fans?
I think is only a useless flame war
just landed on this movie
I’ve now been using Godot for about a year, making 3d stuff in it in a professional capacity.
- Performance of the rendering engine is abysmal. It is unusable on integrated GPU’s.
- Screen space effects are extremely performance heavy [eg default SSAO is unusable on a GTX 1050] and you can’t write your own.
- Performance of the physics engine is abysmal. More than a handful of mesh colliders slows it to a crawl. Physics scenes that run fine in BGE really struggle in Godot. [note: I don’t think this is actually Godot’s fault. We’ve been working closely with the guy who implemented it, and I’ve looked at the integration, and nothing seems to be wrong]
- 3D asset import pipeline sucks
- 2D textures/sprites and things are all done in screen-pixel-space. Great for “pixel-perfect” 2D games, but this sucks for HUD’s, GUI’s and just about everything else.
- Using CI/CD with the engine is an exercise in frustation. Every couple weeks something will break the godot-server from exporting pck files properly
- Using a custom scripting language for “simplicity” may sound great, until you try to do things in it and discover super strange bugs, missing features and other odds and ends.
- Tool scripts are an exercise in frustration, bringing segaults every other day, and seem to be unable to clean up the scene properly.
Before anyone accuses me of being too pessimistic, it’s bad enough we:
- Are writing writing/integrating Nitrogen
- Developed the godot-blender-exporter, which now meets our needs [but isn’t fully general purpose yet]
- Wrote some tools for style checking GDScript.
- Are actively working on the physics bindings/integration.
- Are actively investigating either writing future projects in other languages [through the GDNative bindings, C# integration etc] and are investing in developing the language parting part of godot as a separate part of the engine that can be used standalone to do off-line checking etc.
- We discover and report a new segfaults every couple days, and frequently submit fixes for them, or work with others to find/solve them.
- Regularly report and attempt to solve sometimes serious usability issues.
- As far as we found, the best open source 3d game engine.
- Still a long way off actually being great.
If you can’t tell, I’m quite tired of working with Godot. We’ve hit significant issues with just about every part of the engine. My motto with BGE was that most games never hit release due to developers losing motivation or lacking skills - most of the time not to do with the engine directly. But with Godot, the closer we come to release, the more and more issues we encounter with the engine itself.
Guilty, coded myself into a corner more times than I can count. I just get depressed when a project goes over the level of complexity that I can handle – and most of the time I’d rather re-write things from scratch after I learn from whatever mistakes I made, than go back and fix the thousand lines I wrote on top of them. But it’s not so often I come to a point where I’m forced to drop an idea because of the engine itself, there’s usually ways around it’s quirks.
Then Godot, never felt comfortable enough with to keep trying and get better at it; you’re not making me think I should reconsider. But I do think the demos look pretty.
There were a few beta releases which had a massive performance regression when it came to scenes with collision-capable objects, but all of my stuff runs perfectly smooth in 3.1 official.
I’m not sure if it’s your workflow or the type of game you’re making that causes Godot to give you an endless supply of problems, but I have yet to run into a truly showstopping bug on my end.
Now a few things such as the render engine being fairly un-optimized is true, but that has not stopped my projects from running at less than 60 FPS.
We tracked master all the way up till 3.1 release, and then froze it there. The weird thing is that the physics runs way faster if you don’t launch from the editor! Like 20FPS -> 60FPS difference. Also yes, this is a very physics-heavy project, and we’re running 4x physics substeps. (which shouldn’t be a problem, in BGE I’ve run up to 12 on fairly complex scenes)
I’m not fully sure where the difference in physics engine comes from. I suspect it’s as simple as the version of Bullet. BGE is based on bullet 2.8 or so, and Godot is pretty close to current master (3.something). I’ve had a look through the binding code, and there’s nothing wrong as far as I can see.
Oh, there’s nothing show stopping, there’s just a sufficient quantity of little things to make working with it unpleasant. Minor annoyances are find when you’re working with software in your free time as a hobby, but if you’re working with it 40 hours a week, they all add up to be a real pain.
Here’s a fun one: If you change/export an asset, you have to close any scenes it’s open in before you let godot reimport it. If you don’t, the re-import will fail and the resulting scene will be corrupt (objects in the wrong place with meshes from the old scene etc. etc.). Right, now try iterate on an asset or animation. [or for even more fun, try to develop the blender-godot-exporter]
Thanks for the extensive review. As the BGE is now discontinued unfortunately, I am looking for alternatives now and then but found none of the straightforward ones fitting to my needs. Coming from an engineering background here is what I always liked about the BGE:
Logic Bricks: Yes, they can be hard to debug sometimes - but for simple things (if Keyboard pressed then Do XY) the time from idea to implementation is literally just a few seconds. No remembering/searching of API/syntax, no typos, no compiler setups.
Full Integration: It’s fully integrated in Blender, which makes onboarding of new (non-computer graphics) people easier. For UE4, Unity and Godot, you have to teach a person a modelling tool and a game engine which will end up in a huge loss of time and will not gonna work.
Python: Straight workflow, no compilation. Having integrated Python support with NumPy and all the goodies available in the Python World.
Physics: Collisions, Raycasts, Forces, Torque, constraints - all coming out of the box.
I have seen that things that are < 300 lines of code in BGE/Python (running near real-time and looking great) need 10x as much in other environments with an extensive overhead. I hope there will be a proper replacement some day of the BGE as “interactive mode”.
Good luck with your further endeavors with Godot!