What areas of the game engine would you like improved the most.
I would pick mainly speed of the engine especially when it comes to scene handling and handling of large worlds, when a world gets large it starts slowing down and the game engine simply needs to improve on that yet.
Heh, I think you would need to improve speed before you could improve anythig else %|
I hope all of those thingy will be improved, but I guess I will vote speed.
let’s concentrate on keeping the game engine in blender, then on what needs improved.
The graphics, right now it looks like crapy Nintendo 64 graphics, I would expect to see at least a Nintendo game cube graphic.
what about the bugs?
logic, graphics, and speed are also important, but the graphics will require updates in blender, logic will need some good time spent designing properly, and speed will need someone who knows what they’re doing [it is easy to make blender faster at the expense of conversion time to the game engine, this short time is one of the coolest things about blender].
the graphics improvements will require improvements in blender [like support of multiple uv channels]
and we shouldn’t go head first into logic options, adding new types of actuators which sense things isn’t going to make blender stay as simple. The logic proposal on blender.org was nice, but some of the actuators proposed [path finding for example] don’t make sense in the current logic system [but this isn’t a proposal of small changes]
http://www.blender.org/modules/documentation/logic_editing_proposal.pdf [1.5Mb]
oh, and new physics features [for rendering and the game engine] such as connected objects [ragdolls, veichles, uhh] would be nice too
I don’t think blender game engine is going to succeed the rate it’s going now, I would think it would be better for all of us if it use external game engine and physics to do it’s job. Why is the game engine update going so slow ? Shouldn’t this be one of it’s next priority ?
besides the fact that i haven’t seen any of your games so i can’t see why you are so concerned…
speed isn’t the neccessity right now as with 256MB GPU framerate is absolutely fine and since most are going that way anyway it’s fine although low end (32MB) version need serious speed improvements - then again you wouldn’t tell HL2 developers to increase speed so you can play it on your integrated graphics laptop!
graphics - bump mapping…a guy can dream can’t he?
logic and GUI are my biggest thing - fix those and you’ve got it made! logic will increase control and better logic increases speed, and with more speed comes more graphics! 
Why is the game engine update going so slow ? Shouldn’t this be one of it’s next priority ?
This has been over quite a lot of times. There’s simply not enough coders for the Blender Game Engine. Currently the main coder for it is Kester. Also as word has it, there’s also 1 or 2 people helping a bit. The non-GE part of blender if you check has a greater amount of coders that’s why they have a good amount of updates for each release. Game Engine update is slow because of the lack of people coding for it.
%| I would love to see a speed increase, but hey you take what you can get right? Would love to see a slew of coders that’ll slowly emerge when some of those guys pave the way towards coding for the Blender Game Engine.
Hehe, then Kester will be free to work on the mind racking things for the GE, with us other guys just hunting down and fixing bugs or implementing things wanted in the GE as well.
Jason Lin
Maybe blender.org should post a “Coders wanted” ad for the Blender Game engine. I would like to but as you can tell I’m a slow learner and reader.
what about the bugs?
logic, graphics, and speed are also important, but the graphics will require updates in blender, logic will need some good time spent designing properly, and speed will need someone who knows what they’re doing [it is easy to make blender faster at the expense of conversion time to the game engine, this short time is one of the coolest things about blender].
the graphics improvements will require improvements in blender [like support of multiple uv channels]
and we shouldn’t go head first into logic options, adding new types of actuators which sense things isn’t going to make blender stay as simple. The logic proposal on blender.org was nice, but some of the actuators proposed [path finding for example] don’t make sense in the current logic system [but this isn’t a proposal of small changes]
http://www.blender.org/modules/documentation/logic_editing_proposal.pdf [1.5Mb]
oh, and new physics features [for rendering and the game engine] such as connected objects [ragdolls, veichles, uhh] would be nice too[/quote]
The logic bricks is what make GE, its not too much of the game engine you need to modify to get a bettter control, but ratehr the GUI. Basic what I am saying is the Game Logic GUI needs to be rewritten, to accomidate a Global Hiarchy system, rather then individals objects. Blender has had gui revision with compressed menu with compression and sorts.
Its just a matter of mock ups, i.e when Logic is clicked you can get more submenus like:
Global manager
- Scene manager
+ Physics
+ Fog
+ Scripts
- Audio manager
+ Sounds here
- Object manager
+ Physics
+ Properties
+ Logics
+ sensor
+ connt.
+ act.
- Network manager
+ Server
+ Client
They way I see graphic is, yes, advance features like vertex/pixel shaders and other stuff need to visual see the model before run time needs blender’s DNA changed. But things like a Particle/Shadow/Outline actuator really dont need it…
The speed of the engine is fine. On a modern gaming platform you can get some quite good detail levels. I don’t think anyone has ever pushed the graphics to the limits.
Things like normal mapping would be nice, but who could use them in a real game? It takes huge amounts of time to create normal maps from high-poly meshes. We’d see some nice simple demos of it I’m sure, but I doubt they’d be used in a full scale implementation of a game made in the engine. Heck, has a full game even been made before in Blender, ever?
Personally I want to see better physics. No falling through the floor at random points, ragdoll implementation, effective collision detection at all speeds. This is what the engine needs.
For things falling, just up the number of physical rounds in a second. Forgot the function in the GE but its in there, ti willl help stop falling though.
I think that the problem with the Speed right now is the Graphics capabilitys are not good enogh to warrant the sluggesh Speed.
It could also use the ability to control maximum Frame Rate as well, since the Logic is affected so much from it.
And the Bugs are also a huge hole. Why, oh why does the Character fall through the floor randomly…
Where can I find that? I could really use it 
Speed is THE MOST important thing , because everything depends on it.If you have really good graffics but speed sux , then there is no point to adding those graffics because the game engine cant handle it.Everything depends on speed in ge - ge needs power to handle everything.
It’s so hard to choose!
“Speed” is obviously popular because there are many different areas that benefit from speed. Faster animation and rendering lets you create more interesting characters. Faster game logic and physics lets you create more interesting situations.
There are also different KINDS of speed. Better visibility systems could improve speed for indoor scenes. Dynamic level-of-detail could improve speed for outdoor scenes. Old computers need lower system requirements for more speed, New computers need more speed for more detailed graphics.
P.S. - I voted for Physics 
that’s exactly what i said - it isn’t the speed at all - any one who has made a blender game knows that! Speed is crucial but as said - any good gaming platform (you wouldn’t expect to play HL2 on an old 300MHz pc!) can control and perform just fine at the current speed - and what is speed? Is speed the frame rate or the proccessing time for actions and logic? I think some of you are missing the point completely - speed is essentially the core of the GE - the speed that it proccesses information - faster than you can think so i’m pretty sure that unless you plan on making a game that calculates atomic particles at 1000Fps then the speed is adequate for now…
the proccessing speed does however affect the Framerate - but like i’ve proposed in the “What would you like…” threads - a framerate limiter is essential to making standardized games - that way your IPOs don’t play too fast for faster computers and it frees up GPU time to proccess the better graphics.
Graphics - right now i’ve not seen a single game that used blender to it’s extent with graphics. As far as functionality of including more graphic capabilities (ramps, materials, textures, shadows, nor, the works) those things are still being developed for even more powerful engines - those things have to be developed when they are needed - right now they aren’t - we haven’t even come to the current limit!
As far as normal mappings - yes it isn’t something that would make a full fledged game for but it could make very nice demos and if not everything (maybe just the map not the character - or vice versa) it could work nicely - but it’s still a dream.
Yes the big problem is also with coders - how many coders are there - Kester! that’s about it - i know others have helped but he is the big one and that isn’t nearly enough. If i had time, energy, and his brains then i would help out - i also think that they should continue to develop up everything to Blender 2.4 and then get a completely new global GUI and then everything in 2.4 be mainly focused on getting the GE working even better - that would be a wonderful benefit! Atleast every other version!
Bugs - don’t get me started - although i haven’t run into too many bugs with my game project - must be doing something wrong! lol
Logic - i said it once and i’ll say it again - logic is the most crucial thing - and not so much the logic but the speed and productivity of assembling it. A new revised GUI with better layout and functions is needed and fast! This the hardest thing for would be designers as they got lost in the strings and links of the GUI to even understand the logic involved. I more hierarchy system with “smart” (like it guesses what you want to do based on other objects scripts) would be very nice. More logic is also needed - more features for controlling, more prefabs, more everything - sensors, properties, and actuators and lot’s of premade python scripts for controlling cursors, menus, credits, in game logic, motion - every other engine out there has these and it’s something blender could really use…
well, those are just my thoughts - take them with a grain of salt…
That would be nice, then we could have realistic vehicle movement like with suspension and such as that.
Two things I would like to see in Blender:
- The ability to load objects and levels while running the game. Ok let’s say that you have the following files:
- mainmenu.blend
- thegame.blend Loads third party levels, and seperate model files. So invase if anyone wants to add their own model files.
- Speed of course.
i’m not sure i understand completely what you are trying to say but i’m pretty sure that you can do exactly that by packing then unpacking data and putting them in a subfolder for 3rd party editing. You can load a game file just like you said with the game actuator. You can also have libraries which if in the same directory can actually work to be included. As far as loading a blend while remaining in the current game - just to add new objects isn’t possible however you could use the Edit Object -> Add new to do that and have them in overlay scenes on hidden layers!
I think he means the same idea I had, the abilities to load models on the fly, that way blender wouldn’t have to compute the model until you reach a certain area. Hopefully it would be done with radar or near logic.
By the way should coders redo there approach to developing Blender ? I’m saying instead assigning one area to a group of coders have all the coders work on it then move to a different area after that.