Hello everyone,
it seems to me that despite the efforts of many of the forum, we still see posts that question the future of the BGE and comparisons to commercial engines. Whilst I am interested in seeing progress and it is healthy to be interested in such things, we are revising the same questions and answers with these threads. More importantly, other posts are being buried under swathes of satellite threads. If we are unable to impose more active controls which may correspond to preemptive prevention (which may be construed to be unfair) then it seems the best solution would be to devote a subsection of the forum to this topic. BGE development is a very specific part of the code base; there are few developers familiar with its style convention, functionality and development criteria, which gives some merit to isolating it within its own section of blenderartists.org
I’d be interested in other opinions. And please, let’s not fall off topic as seems to happen quite frequently.
I’m really sorry , I seem to be the one that always gets off topic. I didn’t realize that my last thread gets interpreted as some kind of war. Please feel free to delete all my posts. Is there a way to delete my last thread ?
I think my question was supposed to be more of a debugging question , or performance analysis related , but it came out wrong. I wanted to find out how much performance should somebody expect from 1 cpu core. One cpu core means , how much performance ? I guess it’s hard to say exactly because it depends on too many variables.
I support the idea completely. Not only interesting for the devs, but it keeps the forum ‘clean’. The reason that there is a forum is for a community of Blender Game Engine ‘users’ to support each other. I felt that such discussions about whether the BGE is a descent game engine or not, are not supporting anything. In contrary, these are more or less encouraging users to stop using. Of course a lot is worth mentioning and such discussions can be good for both sides to see things differently, so a separate thread for this seems perfect.
And you’re talking about a new section of the forums, right? Not just a single (perhaps stickied) thread? I’m all for another section that focuses on mainly GE development (not just feature requests).
So if I see it right it is meant to be similar to the “Beginning Blender Code and Development” thread but towards the BGE …“Game Engine Code and Development”.
I’m pretty positive towards that. I suggest to collect some more opinions here. With enough interest we can convince our admins to give us such a sub-form.
we should start rewriting the whole game engine from scratch and in a matter of 10-20 years we will have a game engine that can be compared with the commercial game engines of those time,with basic tutorials we can have noob programmers getting into the pro zones and help BGE development. There will be more features then feature request then…just my thought also hold a weekly google hangout and IRC meeting
Raco , there is a page on Python’s website that says something like "As its name implies, cPickle is written in C, so it can be up to 1000 times faster than pickle ". Don’t you feel that “users” and “non users” should know about that fact ? I mean , give or take 1000 times is not exactly the same thing. Are you saying that people should not conduct any kind of tests or debug things in any way because chances are they are going to find some bugs and that is going to make BGE look bad again ?
My test only showed that for games that are required to run on a single cpu core , BGE is going to do much much better than CryEngine and I was just trying to find out if anybody else can confirm this or my test is completely wrong. And you guys have a problem with my debug session based on what ? Semantics ? Cause I said “better” instead of “worse” ?
Can you guys confirm my findings or not ?
By the way , I think Raco and Monster should not reply to this question because they never conducted that test.
I only want to hear from people that actually conducted that simple test. Take both game engines , force them to use only one cpu core each , and see the problems from each one. If any …
You see , this is how it turns into a 300 miles long conversation. Because some people , no need to point fingers , jump into the conversation when in fact they never conducted that simple test. They just want to jump in and say “HI” and to also point it out that the test proves nothing useful. But what if it turns out that it shows something important ? We don’t know yet.
What we know is that , we’ve typed so much text already , enough to write a new game engine from scratch. Right now we are supposed to be finished and have a completely new game engine.
All of you guys think that the test is completely unfair and pointless but in fact it’s extremely fair because this is not a gpu test. We are not testing any graphics. One cpu core for each game engine and watch what happens. BPR , I really don’t know how CE functions and same goes for BGE. Now , maybe CryEngine is generating shadows on the cpu or generating the terrain mesh , I really don’t know why it needs so many cores from the cpu , but I still think the test is not useless or pointless at all.
Forget it … If you guys want to help , you should try that simple test and post your results.
A test cannot prove anything unless it is conduct in a valid manner, otherwise it isn’t a test. It would be more of a bingo game. Furthermore it would take far more to write a game engine than a few paragraphs
This small test proves that none of the game engines are as perfect or as scalable as people assume. Very simple test. We take two game engines , one BGE one CE , and we force them to adapt and survive on one cpu core each. All the game engines should be smart enough to know what they have to do automatically. And … Please report back your impressive findings.