Bjs is moving fast

[Nabroski]

If you a student in computer science or a branch near the game industry, you have to deal with low level stuff and an the end of your time at the university you already have 3 game engines that you call your own and probably running a small community.
If you an artist, you always mess around with people from the demoscene, html packed in png files and other crazy experimental stuff. 1Kb. 64Kb files. you really have to think about, what you are doing. 

So babylonjs aiming the 15y old who want to impress their friends with some code and the less experienced programmers, people who just wants to push their product on the market. 

A simple Raytracer is about 60Kb on Windows (Debug without optimization) 
Babylonjs(Canvas2D,Physics) downloadsize 2MB and you don't even have setup a simple scene. This is not Low -Level

I reading some dev blogs (dev is the aberration for developer (im again in low level mode, sorry)) - their really new and innovative stuff coming on the following years.

And babylonjs is moving really slow. 

I would love to see more artist coming to babylonjs (currently only Wingnut and me)I'm not against more

user-friendly coding as long it looks nice and make sense.

Game industry make the most money in the overall creative branches and i dont even know why their are less people that are aim to publish game coming to babylonjs.



 

[Wingnut]

Hi Nabs and all.  There's actually MANY artists here, and I'm more mad scientist than artist (mix lots of chemicals and see what blows up). 

I'd say a solid 60% of our regular users... qualify to be called artists, including all of the BJS founding fathers.  Many of them just don't get time to do their art.  Lots of other priorities get in the way, I think. 

Many of us are "scrambling" (survival/family/duties/whatever), and "scrambling" is not overly conducive-to artistic endeavors.  (What did he say?)

If you use a "loose definition" of artistic, then coding is artistic, too... as well as game designing, and tutorial writing "flow". 

[joshcamas]

23 hours ago, Nabroski said:

Babylonjs(Canvas2D,Physics) downloadsize 2MB and you don't even have setup a simple scene. This is not Low -Level
 

This is because Babylon isn't a low level engine. If you want a low level engine, you probably want to use raw webGL.

[GameMonetize]

Babylon.js is a ~300KB file when you serve it from a http server with gzip option turned on.

We also provide this tool to help you shape it according to your needs: http://www.babylonjs.com/versionBuilder/

[Nabroski]

first you say this was nearly 2 years ago ~100KB http://www.html5gamedevs.com/topic/10954-objects-stretched-when-canvas-resized-if-aaf-activated/#comment-64175

 

according to this at the end of 2017 babylonjs will be 400kB. geziped 300kB is probably 10 ms decompression time on a mobile. (on a sunny day)

 

i just saying that i would be kind of cool if more people with low end possibilities come to babylonjs.

 

@Wingnut yeah, everything is art, i mean visual artist, i don't know what the right term for those people is (generative art people). if i would write in german you would understand me. you have to learn german!

@joshcamas I do.

[Nabroski]

@Deltakosh

so Threejs is like Windows (NPM (3th party drivers) -just throwing things at people, ends up always in chaos)

and Babylonjs is like Appel (nice) 

this is not bad, if their is any chance in further to split this even more down. .... to get few student from Computer Science classes 
http://www.babylonjs.com/versionBuilder/

 

https://technet.microsoft.com/en-us/library/cc750820.aspx?tduid=(824661351b07b3d394245e90e45bdfbf)(256380)(2459594)(TnL5HPStwNw-UjEbcUiGE.WqYVzt_uvSdQ)()

[davrous]

No, we won't work on making Babylon.js smaller or components based. We already had this debate several times and nobody was able to convince the core team members of the interest of doing that. It would require a huge amount of work for no convincing scenarios.

Creating a simple mesh with a simple texture will often take more bandwidth than Babylon.js itself. If people are concerned about the speed of loading a page and/or bandwidth, they definitely should focus on optimizing their assets rather than hoping that shrinking Babylon.js would help them in any way.

David

[Nabroski]

Yes you are totally right, my criticism goes not in the direction of babylonjs, Its  awesome! Thank you!

it goes definitely in the direction of the users - who don't optimizing their assets. I dont know, if i scroll through the post of last two days, i would say. many of people here tending to write slow scripts. Downloading a (babylon)JS file and never triggering functions inside of an liberty or stack functions on top of an liberty provided function, is definitely going to be slow loading page. So i feel kind of sorry for them, maybe.

My though was, when babylonjs become even more "module" based. It would kind of be a better learning process. 

http://doc.babylonjs.com/tutorials

Oh, wait why my shadows are not working, - Ah, i have to download the next package. And again reading the doc. 

But whatever, its impossible to prevent the whole world to be less intelligent. (intelligence i mean: by using tools to change the environment) 
So yes, the one who are able to write better optimized apps, are also going to make a kill

[micdrop]

[Nockawa]

@Nabroski theoretically, what you said makes sense. But the reality is modularization is something that must be planned from the get go. Babylon.js is growing and we do try to make modules when it's possible. The best example being Canvas2D. I designed it from the start to have the lowest coupling possible from BJS to C2D so it could be a separate module the day we wouldn't have wanting to do so.
It was pretty easy for me too, because it's a 2D Engine, using the core engine of babylon, so very low coupling with the 3D engine.

When the size of Canvas2D got bigger, we decided to make it a separate module, for all the good reasons you know about. It's good for the main file's size, it's good for the users, for learning, even for me to keep developing C2D (I don't have to recompile all the .ts files of bjs).
But I want everybody to understand something: it was possible to do it, so we did. It's all about coupling. If you want to put physics in a separated .js, well, it's not that easy, because the coupling with the 3D engine is tight and strong. The effort it would take, the risk of major regressions that could occur afterward just make this dangerous to do.

Now you talk about 10ms of decompression, come on, that's a true valid argument you think? Do you realize how users loose time due to latency of network? 10ms to uncompress is totally acceptable.
I wish more people would contribute to see how it is on a daily basis: remember, it's not a pro project and still we're doing pretty fine!

We had this discussion internally when C2D started to be big, I mean to break down all bjs into several modules, we evaluated all the possible strategies and at the end of the day, @davrous 's point made a lot of sense to me: the assets are really most of the data that will be downloaded and refactoring bjs being really dangerous: we don't want to do that. Remember that all the time spent on this would be time that couldn't be spent elsewhere!

And most of the browser will put the file in local cache, so it's not a 300kb load every time: just once in a while!

Yes: bjs could have been design with a better separation of concern, yes it would have help to make module out of it. But that's not the reality, that's the ideal world! Reality is: like many OSS projects developed on spare time I bet than @Deltakosh and @davrous never envisioned from the start how far bjs would go! I bet they didn't start the project saying: "ok, we're going to develop the most feature complete 3D Engine for the Web in OSS!"

Criticizing is good, as long as it's going in the right direction. And honestly, my take on all of this: it's a bit of a waste of time when people criticize without helping to find the best answer and helping to put it from an idea to a concrete feature/dev/action: contributing!

One year ago @Wingnut were helping but mostly criticizing in a way that wouldn't help us and I'm glad that I saw a dramatic change in behavior from him: helping the beginners, helping us to triage the forum questions, helping us to be the link between users and the dev team and even contributing on GitHub. We need more people like that. And the more you contribute, the more you become important in the decision process. It doesn't mean we don't take the feedback of users, we really do, but my take on this: the people that will have to bear the consequences of the decisions, the actions, the code they produce are clearly more heard when a decision is to be made.

[Wingnut]

  I don't think others have as much free time... as I do.  But the subject of my year-ago rants... was specific.  I was trying to establish the league of extraordinary forum helpers.  I got piss poor response, so I decided to show fellow forum folk exactly what extraordinary forum helping IS (in my opinion).  In doing this, there was not much room left for others to participate in that.    4-8 of us helpers were kicking Q&A ass SO FAST... that nobody else had a chance.  heh

More about the specific purpose:  Those rants and blood-boils are/were attempts at setting policies for forum helping.  These won't appear as being helpful at first glance, but after the emotional reaction and pout wears off, they leave a residue.  That residual "resent" might have a Wingnut-hating after-taste, but they are still valuable for setting helper policies, and for helping other helpers... self-examine.  Essentially, I tried to make it OK to fight for altruism, and not okay to be ego-centrist (self-centered).

These things... how to care about others, no matter how annoying... is more than just a forum policy and attitude.  It is LIFE policy and attitude... and thus... it strikes at the heart.  When you question someone's heart, and put it on a pedestal for all to inspect, it's going to piss off some people.  It might not have appeared helpful, Nox... but I think in some cases... it is/was.  It is attempting to teach... Wingnut's version of...  the "spirit of teaming".   It's commie in nature, and it goes head-to-head against competing (competes with competing?  heh).   Competing, in my opinion, is an epidemic, mostly caused by money tug-o-warring in folks' lives.  Nox... it is a HUGE foe, and takes some serious heart-string pulling... to affect.

Lots of forum friends have steered and adjusted my thinking, too.  I strive to take others' wisdom and experiences into consideration, whenever possible.

Try to keep in mind what was being attempted with those rants.  They had a specific purpose against an elusive and overwhelming enemy.  It's related to my anti-capitalism ways, and my 9 years on USA Team Air Force (a commune).  Thus, it touches onto politics and morals.  Those things can "seem" off-topic to those who don't normally work-on communes and team spirit.

[ozRocker]

[getzel]

This is badass !

 

[ozRocker]

7 minutes ago, getzel said:

This is badass !

I've been putting my Blender motion-tracking skills to good use

[JCPalmer]

Amusing, but a little off topic.  Belongs in the "Announcements" section.

[Wingnut]

(Wingnut giggles and points at Ozrocker for being pulled-over by the topic police) 

(Wingnut also reports officer Palmer to authorities... for being far off-topic... with his off-topic-ness comment.)   heh

Now we're REALLY confused, huh?

Ok, but now, "BJS is moving fast"... umm... uhh... crap... I can't think of anything pertinent to say. 

Let's see... I need a universal "juicy" comment... to keep the thread alive.  USA... 1950's saying...

"So's your old man!"

There, I said it.  NOW we have some controversy and drama, again.  Maybe Ozrocker will think twice next time... before using comedy to lighten the mood of a tense thread. 

[Pryme8]

@ozRocker

Yeah we're gonna build the best 3D web gl engine ever, it's going to win biggly.... and we will make playcanvas pay for it.

[BangTao]

On 2017/1/17 at 7:32 PM, JohnK said:

@Wingnut here is a person BJS might lose, BangTao. (S)he is having difficulties with BJS and I have tried to help (her)him to the best of my ability but am failing. It might just be a language barrier but the questions keep coming and for me I do not have sufficient understanding of what (s)he wants. Perhaps you might have more success.

@JohnK
Haha,BTW,thx for your helping(and others too),and BangTao is "he",and he's not a game developer and know nothing about Graphic and image,even didn't used 3dsMax before.

Cause the work reason,he discovered this amazing.....BJS (and this moment,he can't think of a good word to express).And BJS has few "Chinese-Version-Resources" to look up.so BJS is new to him (always new),for the first time,he's  totally puzzled about the BJS's "Function"and puzzled about the "property".Although the class has descriped what these property are,but he still don't know how to use them.Then he found "PlayGround" where can "carry out experiments"....thing not happend as expect.As you can see,he had got many problems.he can't solve these problem he meet,so he looks for help in the Forum,and there r many kindly people help him.and he has make progress and still keep trying new thing so far.

His English did bad,so he attend English training class (this week)..

Have a nice weekend!