Termination of financial support starling and feathers

[денисг22484613]

Guys, apparently the AIR is no longer an Adobe priority. Tell the developers when you are planning to end AIR support ?

[davidrenelt]

PixiJS is a bit like what Flash does. It's canvas based, but it caches sprites as WebGL textures. It doesn't have the finesse of the vector/bitmap render engine of Flash, but with a bit more time you get similar results that perform very well even on lower end hardware (like tablets/phones). The display objects share a very similar architecture to Flash (i believe it was designed so because it's done by former Flash devs). To package that application you can use Electron. It's all a bit hacky  .. but it's not like we have any choice.

Overall i see Animate's future as a asset creation tool, Actionscript doesn't deliver anything another tech couldn't, so in the end it is just a question of which rendering-engine you target. I expect Adobe to acknowledge this more and more with future releases. Maybe there is going to be even specific export mechanism for certain targets (like PixiJS or Unity)

I have yet to wrap my head around Unity. Coming from Browser-Tech/Flash this is a very strange environment for me. But clients now specifically request it.

The bottom line, i strongly believe that we all have to move on, the timing is up to each individual and this is what Adobe is doing right now, they are giving everyone the chance to jump ship at an opportune time. The "silence" is all they can do at this point.

[PedroFernandezEteria]

pixi is more a game engine, than an engine to develop interactive multimedia. It's more comparable to Starling than to Flash / AIR / Gpu mode is what I was talking about. And Starling has its drawbacks in terms of workflow.

The main problem is not only the api, but the workflow and the "runtime" that finally executes a development. Canvas, as "runtime" has many problems (more than flash runtimes).

Canvas text tends to be crap (at visualization quality and execution speed, depending on circumstances). Especially when the canvas is scaled to the size of the browser window. You can use bitmap fonts in some cases, yes, but this takes more work, and in some cases it is not possible because there may be many types of characters.

Video integrated in canvas is crap. In the end many developers choose to show the video in a lightbox with another framework based on DOM.

Many developments are based on working in parallel with DOM, which means that you have to jump from one API to another, losing consistency, and therefore development agility. That is, it is like working with patches, very "manually".

It is difficult to work with many different contents in the same canvas loaded at the same time in memory. Most times you have to create and delete assets at runtime because if not, the development will be slow in the lower hardware.

A direct practical example. We have an interactive with 80+ screens. Each screen is "different", with texts, vectors, bitmaps, etc. With everything in memory, with our own framework that works with assets in swf and AIR gpu mode, we can run it at 60fps totally smooth in different hardware (windows, Android app - even on lower end hard -, etc). If I can have this possibility in a technology, why not take advantage of it?.

On canvas (browser runtime), it is totally impossible to have so much information in memory. It does not matter that you work with bitmaps or with textures ... You have to develop an asset management system (creating / removing when needed), because if we leave everything in memory, in slower hardware everything slows down by the simple fact of "being there" (although it is not visible). Developing a load / unload system, also implies, even if you do not want it, more development time.

On the other hand is the workflow. The workflow is fundamental to have good agility in the development of interactive products. In canvas there are hardly any solutions that allow working with a decent editor in which you can place assets visually and then use programming to access them. That is, a Flash like workflow, with screens, elements of different types (vectors, texts, bitmaps), exported to swf, in which later you can work in an IDE to program all the screens and elements, using vectors, or bitmaps in gpu as needed.

It is true that in js there are some visual editors, but either they are focused on games, or they do not work with "your favorite api". That is, it is not possible to use swf assets from pixijs. Will it be possible in the future? It remains to be seen, but at the moment we are in the present. On the other hand. Getting good results with pixi like solutions, sometimes is related to assemble everything manually (perhaps in the future using a visual editor the assets will not be sufficiently optimized)

With any api based on webGL (like pixijs) you can get good results. Nobody denies that. But you get them at the cost of working more, since it's all more manual. In some cases you have to work not a little more, but much more, depending on the circumstances.

Adobe's decisions are independent of what I explain. It is irrelevant. Browsers are not the right runtime for this type of project.

Therefore, do not try to match what now exists at the technological level in this area, because what existed "before" was a superior solution. It does not matter where you look at it.

[davidrenelt]

Well, betamax was superior to VHS I don't think you'll have any choice in the matter eventually.

[Phasor1]

Obviously there are not many options. But an option for interactive multimedia that is based on the runtimes of the browsers, for me it is anything but optimal.

I think Haxe / OpenFL is a better solution for the simple reason that you do not get "stuck" in a runtime, but you can compile in "native" for each platform. And speed has plenty, knowing how to do things.

P.D.: I am the same Pedro Fernandez, but from my home account.

[zwetan_uk]

At the end of the day the choice is simple:

either you believe "web trump all" and then trying to shoe horn

Flash technology into HTML will make you very angry and desperate,

so yeah to build applications, you will naturally move to something else

because without plugins in the browser there is no sane way to replicate

the Flash experience there

or you believe "web limit all" and then trying to escape those client limitations

you will explore where else you can publish: desktop, mobile, cli, etc.

there you will compare what you can do with Flash technology outside of the browser,

and oh you realise you can actually do a hell lot, suddenly HTML is not so sexy anymore

and Flash is good enough

but that's it, you can not have it all

Things like Unity, Xamarin, Flutter, etc, are not necessarilly better at publishing for the web

and when it comes to publishing for the desktop, mobile, cli, (anything but the web)

there are certainly not far superior to AIR

it's not so much what kind of applications you want to build but WHERE you want to publish them

> TS for web and Xamarin for Apps but we are experimenting with ionic as well with the idea

> of having only one codebase (TS) for everything, results are encouraging so far.

You can perfectly use TS to publish for the web and keep using AIR to publish to everything else

still this is the same problem: what is your main target? web or all the rest?

having both from a single tech is not possible, you will have pros and cons

mostly centered around what you favour more (web target or other targets)

in my case I still build for web target but then I keep those target lights (no heavy SPA),

so either using TS or something like Apache Royale is good enough, but pure JS is good too

> Maybe we should get back to the point of the topic though.

the original topic "Termination of financial support starling and feathers" has gone off topic a long time ago

[ASWC]

That's not really the argument here. We have dropped Flash completely and are working 100% with HTML5 and (at least for now) pixijs but that's doesn't mean this is what we wanted to do. This was imposed to us and yes we need to deal with the lack of features, lack of performance, lack of this lack of that, that was never a problem with Flash. Flash was much better in many ways (except for the Sound API that was crap) but Flash is not an option anymore and only HTML5 and webgl are available (and maybe later WebAssembly). So that's the situation, that's reality, I don't think anyone here is saying Flash/AIR is not an awesome technology, it's great/best technology that is setup to fail by its owner.

[Leo Kanel]

In the meanwhile, Adobe remains silent, the only response I managed to get both from email, tracker, forum was just "We are aware of the issue" with 17days to go. Really mature

So we are still waiting for a response on when AIR will update to iOS 12.1 and when for Android 64bit.

[hferreira80]

Meanwhile, I'm looking for Xamarin and Flutter.

[Leo Kanel]

I already went with Flutter for all new projects. However, many clients with AIR Apps, still ask, when we are going to update to 64bit and I have nothing to say...

[PippoApps]

New 32 beta out today:

iOS SDK Upgrade AIR Runtime is now built with iOS 12.1 SDK, which enables AIR developers to use ANEs built with iOS 12.1 APIs without using the – platformSDK switch while packaging with ADT.

Fixed Issues

iOS SDK version warning on submitting air apps to app store (AIR-4198787)

Great job AIR team.

[hferreira80]

Are you joking, right ?

No Android 64 bits !

There priorities are definitively wrong.

I'm already moving away from AIR in respect of Mobile.

[PippoApps]

It will be added. It just takes some patience.

[hferreira80]

Patience !

Google announced a long time ago.

They are delaying to last minute and push this non sense releases that no ones care or understand.

A warning is more important than a 64 bits blocker ?

Yes, I believe that we may get this update but my patience is over and I moving right now away from AIR on mobile space.

Live with AIR on Mobile is like live with rope around the neck.

Good luck.

[rewb0rn]

Nonsense release? iOS 12.1 was required to upload updates to iTunes from March 2019. So it was a higher priority than Android 64 Bits. Get your facts straight

[hferreira80]

And when this requirement turns up ?

As I said before, live in the limit.

Well, AIR is good for cardiac

[Leo Kanel]

Yeah but there are alternatives.

If you want to fast start with Flutter there is even a good alternative to starling. Try the ActionScript to Dart guide.

[hferreira80]

YES and I started now to move away from one of the best runtimes and the worst managed one.

After the last few days on evaluations (including Flutter), I choosed Xamarin.

Anything, right now is far from best safer than AIR.

[PippoApps]

I think we all understood at this point that you are going to move away from AIR

[hferreira80]

After years with (I don't know how many) posts without a single answer I needed to said that.

Thank you for listening (reading)

I'm fill much better right now, that I decided to move from this RIP AIR.

A few ones have the privileged to get noted with their products and many others try but fail miserly, but be able to destroy the best, should be considered a crime

[PippoApps]

Just received a comment on a nasty iOS bug I filed a few weeks ago only:

"We have done a AIR beta today, you can download it from Adobe AIR 32 Beta | application development - Adobe Labs which contains the fix for this issue."

I love you AIR team.

[rewb0rn]

@Leo Kanel: In all decency, I think we understood that Flutter is your prefered SDK now, but this is not the right place to spam it with. Have some respect for the fellow Air developers. Thanks

[hferreira80]

I understand your point but I don't think that Leo is spamming anything.

It just stat a point.

Seems now a days that AIR developers have afraid of other technologies and when Adobe send a response it's like god answer.

[mrfrasier]

I don't see anyone here afraid -- but the topic is about getting answers from Adobe and sharing info on updates, when it seems you just keep harping on about how you ditched AIR and moved on. That's great, so why keep posting about it in a thread discussing updates to AIR?

[hferreira80]

Just because that !

We don't have an answer !

[mrfrasier]

OK? So you've moved on, why do I keep getting email notifications from this thread with you saying the same thing over and over? You moved on, so move on.