Author Topic: Flare3D's Future  (Read 6447 times)

Kenny   «   on: February 24, 2016, 12:23:09 PM »
No updates in over a year?  Everyone who has supported Flare3D over the years deserves a statement from Ariel and the Flare3D team.  If support has ended, please tell us, so your supporters can make good decisions.  If there is no financial future, then open source Flare3D for those who still believe in it.

At the very least...Make a statement to the community.

Thank You

Christian   «   Reply #1 on: February 26, 2016, 07:32:38 AM »
Ariel: Make a statement to the community!!!!

Ariel Nehmad   «   Reply #2 on: February 29, 2016, 09:50:05 AM »
Hi guys, sorry for coming so late to this, it hasn't been easy lately for flare3d as you can imagine.

There is not much to say really, we're trying to stay afloat, as the large sequence of Adobe's non sense decisions and attitude really had hurt us, badly!.

At the moment, as you can see, there is not much in development, but Adobe hasn't added a feature worth investing time in the last year either.

Sadly, no big company is investing on flash these days, and it's not surprising, Adobe doesn't seem to be doing that either, so, how can anyone invest anything on a scenario like this? it does not make any sense.

You guys know all the work, excitement and love we were putting into flare, and I have personally suffered a lot from all this, but the saddest part is that I was a strong (if not huge) supporter of the platform, I loved it, and they manage to push us down time after time, and the history repeats again and again, here are two pages of just 3d dead engines! :( , and I can add even more to the list.

At the moment the most we can do is wait and hope that all this no sense will go away.
And as I said before, we're trying to stay afloat, it's just really hard to project what the future would be, but if there is any opportunity will go for it!, we would love to bring it back to life!.

We're still here to help, and if there is any critical bug you guys need to be fixed, please feel free to post it in on the bugs section we'll be taking a look at those as soon as possible.


Baris   «   Reply #3 on: March 01, 2016, 08:08:48 AM »
Hi Ariel,

I have used a significant number of 3D engines during last ten years. Flare3D was different. It was so logical, so full of sense, so easy to adopt as I did it myself.

What about porting Flare3D to Javascript (via TypeScript)?
In Javascript, we have two principal actors: three and Babylon.
I believe Flare3D could be placed in third place

Another option is Apple Swift.
- there is no well known 3D engine in Apple Swift. (is there any?)
- there is no cross-platform 3D engine in Apple Swift and Javascript.

In any case, I'm willing to pay 5x more per year for the flare3D.js licence offering a business support. Open source is cool but too big, too fuzzy, with thousands of crappy examples without the right solution. Sometimes we need business support for a simple small clean library with responsible people behind ready to answer.

Are you able to estimate how much Flare 3D needs, to be ported to these environments? Perhaps you have already enough potential client to crowdsource this job.

(We are porting our Flash software to javascript via TypeScript. And it works pretty well. Concerning Flare3D part, we had to write our zf3d loader in three.js. But unzipping in JS is not the most efficient in mobile environments, so we unzip in server side and download assets separately. The money I spend to develop this equals to 10x Flare3d licence cost, so there is a business logic in what I'm saying.)
« Last Edit: March 01, 2016, 08:59:32 AM by Baris »

Kenny   «   Reply #4 on: March 01, 2016, 09:16:24 AM »
Ariel, thank you for adding some insight into what is happening with Flare3D.  Your engine is the most advanced and complete beyond any other.  I fully agree with Baris.

AwayJS has moved on to Typescript and is a natural progression if ActionScript is truly no longer viable, although I truly don't believe that.

At one point you talked about changing your pricing strategy.  I was looking forward to paying month to month and would gladly pay for updates. 

Flare3D 2.8 is truly the most viable for 3D development and I wish you would finish and release it and perhaps charge for the update.  Flare IDE is soooo important to rapid development.

Adobe continues to listen to people using Starling, so why not Flare3D?  Coming features I'm exited about are "Asyncronous Texture Upload", "Antialiazing" and support for Apple TV (tvOS).  http://forum.starling-framework.org/topic/air-in-2016-survey-and-christmas-tale/page/6#post-89672

What can we do to get involved or help?


Baris   «   Reply #5 on: March 01, 2016, 10:23:21 AM »
... ActionScript is truly no longer viable, although I truly don't believe that ...

A project like Flare3D needs a lot of R&D which requires an investment. For investors and decision makers in B2B clients obsessed with HTML5 and Apple, you can use all the arguments about why Flash is viable but, they just don't give a shit! Flash is a big NO! No investment = no project.

Also, when we have started to port spacedesigner3d.com from flash to typescript, we have concluded:

- Browser + WebGL is slightly faster than Flash in real life projects.
- In return, the browser is severely less stable than Flash.
- And debugging and profiling in the browser is very hard (Chrome Profiler compared to Adobe Scout)

If the lifecycle of a tech investment is ten years, JS is a secure enough for investors. At least, this is what they believe until the next Steve Jobs says the contrary.

In short Flare3D needs funds. If there is no intelligent investor, let's crowdfund it!





« Last Edit: March 01, 2016, 10:25:00 AM by Baris »

solomon   «   Reply #6 on: March 01, 2016, 12:10:03 PM »
I think, yet another WebGL framework is worse move than try to stay afloat.

Flare should revisit overal product experience and definitely stay within Flash in this year.

Kenny   «   Reply #7 on: March 01, 2016, 12:17:07 PM »
I agree that Flash in the Browser is a loss on most businesses now.  However, I have spent the last 10 years working for companies who think differently.  Browser based 3D is slow to catch on.

I do believe Adobe AIR is most important for cross-platform development on Mobile and Desktop.  Adobe AIR is second only to Native for all cross-platform solutions.  Starling is huge and just released 2.0 yesterday!

I personally haven't seen a performance bump with WebGL, but as mentioned there is huge stability issues and lack of important profiling tools like Adobe Scout.

This engine has sooo much potential for rapid game development across many platforms.

Crowd funding is an awesome idea!  Go for it!  I'll rally people behind it.

Ariel Nehmad   «   Reply #8 on: March 02, 2016, 05:31:57 AM »
Hi guys!, thanks for all the interest and positive feedback! we're taking it all!
And thanks for all the nice comments :)

A few months ago I started with a c++ port, the main idea was to get JS (via asm.js) and native performance on Android an iOS. (don't get intimidated by c++, the engine API is even simpler than in as3 :P )

Now, the tricky part is the scripting. We can get native api's or wrappers for all of them, or maybe go for something like HaXe.

Other possible path would be to embed some scripting VM lang like JS, AS3 or even python and maintain all the rendering and logic at native performance...or even a mix of both.

Doing a straight port to JS it may be also possible as well, but it will limited only to JS.

What do you think?
« Last Edit: March 02, 2016, 05:37:19 AM by Ariel Nehmad »

Marcel Klammer   «   Reply #9 on: March 02, 2016, 10:14:41 AM »
I was a supporter from the first hour (Flash 10 without Stage3D, long long ago), paid every year for the Flare3D license, even in Januar 2016. I need Flare3D in current Adobe AIR apps (Kinectv2, mapping tracked body skeleton to 3d character + cloth). So I would appreciate a release of v2.8 a lot.

That put aside:

With our face tracking we went from AS3 only to a C++ port to support all possible platforms. So far the AIR ANE for iOS and Android is based on that C++ port, also the JS export is working well using emscripten.

Our face tracking is quite stand alone though. A 3D engine is much more work in terms of integrating with a  platforms GL stuff. But I think it would be worth the shot for you. As I told you already a year ago, go for JS as a platform, get your great workflow (3DS exporter, IDE) out to the thousands of JS developers out there. You have all the know how to get this working. So do it :).

And don't mind the Flash platform too much, have your C++ port work as a native extension for iOS/Android/Windows/Mac for Adobe AIR with an AS3 API.

Best Regards
Marcel

Kenny   «   Reply #10 on: March 02, 2016, 01:03:18 PM »
And don't mind the Flash platform too much, have your C++ port work as a native extension for iOS/Android/Windows/Mac for Adobe AIR with an AS3 API.

Thank you Marcel, I think this is an excellent idea!

Tommy Svensson   «   Reply #11 on: March 02, 2016, 01:31:52 PM »
Hi Ariel!
Thanks so much for your statement.. We all know the state of things regarding flash these days and its not a great situation. Even if flash/AIR really is (still) a great platfom.

I would love to see a Haxe port. Not a pure TS/JS port, but a crossplatform target version of Flare3D .. so we can use if for flash, js, mobile.. and targeting even C++ for Desktop. I think a good decision would be to disassociate a bit with the "flash" label and just focus on the gamedev side. Haxe has really matured now, and i think its a really great option. (Check out KHA or BabylonHx with OpenFl)

Because I think Flare3D is a great engine; the best one ever written in AS3. And the IDE is also just as great. I would gladly pay for a Haxe version! And marketed as a crossplatform 3D engine with a amazing editor is bound to attract even more customers!

Regards,
Tommy

« Last Edit: March 02, 2016, 01:42:42 PM by Tommy Svensson »

solomon   «   Reply #12 on: March 02, 2016, 01:53:36 PM »
I share the idea of moving toward ANE. Besides if you introduce fallback to pure AS3 then both AIR and Flash Player may be supported. This pure AS3 version would be refactoring side effect either way. Gradual move is generally better for commercial products.

Flash Player -> Flare AS3 API -> Flare AS3 Library
AIR -> Flare AS3 API -> ANE -> Flare C/C++ Library
JS -> Flare JS API -> Flare C/C++ Library

pj   «   Reply #13 on: March 02, 2016, 07:23:59 PM »
If you guys are looking at porting Flare3D to another language, I'd strongly suggest looking into Haxe. It's a great language with heaps of really awesome language features... it's everything AS4 could have been.

Additionally you can target Javascript / Flash / AIR Desktop / AIR Mobile / Windows C++ / Java, etc, This way you can keep all Flare3D's current targets and simply add more, all while maintaining a single code base.

You can also build on top of OpenFL for it's Stage3D implementation, which will make the port super fast, Away3D has already been ported: https://github.com/away3d/away3d-core-openfl and I've ported Starling: https://github.com/Haxeling/haxe-starling

Or you could build on top of Kha, which is basically a common API for various 3d rendering engines, OpenGL, WebGL, Stage3D, DirectX, etc... however this won't be nearly as straight forward because it will most likely require quite a bit of a rework of the core code that currently interfaces with Stage3D.
« Last Edit: March 03, 2016, 06:14:20 PM by pj »

Baris   «   Reply #14 on: March 03, 2016, 04:27:05 AM »
I believe from a business point view, the best way of interesting the largest number of developers into Flare3D is not having the largest spectrum of supported platforms and languages. And it's not about the powerfulness and beauty of the languages used to develop your software. If it was the case, we, the web developers, were not forced to use the stupid Javascript today.

We need Javascript (and Flare3D.js) and here are our reasons:

- The decision maker of the business client requests it.
- The investor in technology demands it
- Research of Development projects uses it
- The junior developer who just integrated the team knows it
- The most recent development tools are made for.
- And it works in the browser

Finally, we opted for Typescript, made by Anders Hejlsberg, who is the father of Turbo Pascal, Delphi and C#... Because we believe the key point is not cross-compiling to Javascript as Haxe do very well, but fixing Javascript, until it fixes itself over ECMAScript updates. It's less comfortable, perhaps less productive and less mature than Haxe but it represents the Javascript of the future.

And It looks like, based on trends, we are not alone. (see the attached  graphic)
Attachments:
« Last Edit: March 03, 2016, 04:29:50 AM by Baris »