Port Flash to android native app (no AIR)

You will need to provide a great deal more information than that to get an answer to "how much it's likely to cost".

On the low end, it can cost $0. AIR is not the only Android AS3 compiler out there. Beyond the Flex and Burrito combination, you also have several open source projects that can handle it. For simple AS3 code (not so much the rest of Flash) you can always run it through the GCC backend which can generate native binaries. If the only work involved is getting a different compiler, the cost is minimal.



On the high end, it can cost as much as a complete rewrite on the target system. Exactly how much that would be is anyone's guess with the information you provided. For a major game that could easily reach into the millions of dollars. If you are contracting out the port of your "simple games" that could be either a very high or very low six-digit number depending on exactly how simple it is, the quality of the product, and the experience of those rewriting it.


With no other guide other than "simple game", we're left to guess.

As a rule of thumb it is $10K per person-month of development time. After that it is a simple matter of multiplication.

Perhaps your simple game could be contracted out with 2 programmers + 3 artists + 1 audio + 5 QA + 2 other staff @ 1 month (quick) project = $130K.
Or perhaps your simple game would require 5 programmers + 4 modelers + 3 animators + 2 FX + 1 audio + 5 QA + 3 other staff @ 4 month project = $920K.
Or perhaps most of your resources can be reused, 2 programmers + 1 artist + 0.5 audio + 4 QA + 1 other staff @ 1 month (quick) project = $85K

All of those numbers assume there is an existing engine and they are simply hanging your stuff on top of it.

Thanks for the reply

By simple game, I mean like the millions of simple flash games out there, but to be more specific, I'm talking about games that take maybe a few days to code from the ground up with a single dev. As it will be a port of a flash game, all assets will be included. Here's one of the games I'm thinking of porting - http://www.addictinggames.com/puzzle-games/orboid-game.jsp

Thanks

Interesting game.

I haven't done a careful study of the game, only looking at it for a few minutes online.

Based on that quick study, I'd estimate nothing less than 2 months and roughly $250K to fully develop and QA that size of game at a professional quality level. That would still leave publishing and marketing and support and the remaining post-production costs on your shoulders.

Thanks.

I made that game in about 2 days. I'm struggling to see how it could cost so much and take so long to port. I've found someone to port another game, which is similar in complexity, to html5 for a considerable amount less (less than 1%).

I made that game in about 2 days. I'm struggling to see how it could cost so much and take so long to port. I've found someone to port another game, which is similar in complexity, to html5 for a considerable amount less (less than 1%).

Ah, that is a different perspective.

As I said I only looked at the game for a few minutes. I didn't do a full study, and only played about three levels. I didn't see how deep the game went, or what else it offered. I was assuming the game was much more extensive, not that I had experienced the full depth of play.



You are speaking as hiring an individual to clone your tiny project. You are not expecting any QA work, no broad testing across platforms and technologies, nor any iterations to ensure things are working as expected. You are also expecting that the first iteration is the only iteration. This is rarely the case in practice.

I am speaking of working professionally, hiring a group to port a full system. Two days of development would be pushing it for getting together a prototype; if that is all you wanted, then sure it would be enough. Professional groups will allow time for additional iterations, never expecting that the first iteration is also the final iteration. Even then, few port-centric groups would release such a thing without their own serious QA effort, typically lasting several days for QA time alone as they run the full QA checklist on roughly 40 different browsers on five or more operating systems.


Contracting work out is very much a buyer-beware situation, and you do tend to get what you pay for. If you are looking for the first case, an untested quick first-pass through the game, then yes you can find somebody willing to do it on the cheap. I would personally be very reluctant to sign such a contract for a work-for-hire project.

If the main objection to AIR is the requirement for a separate downloadable runtime, check out the new "captive runtime" feature in the latest AIR SDK. It basically packs the AIR runtime into the APK itself, thus making an app that needs no separate runtime. It's quite a bit larger than its non-AIR counterpart (8.5 meg), but it works.

Also note that a couple of tablets, specifically Nook and Kindle Fire, have AIR installed by default so nobody's the wiser.

But yeah, as it stands, there's no AS3-Java convertor in existence or in the works. I'd like a real AOT compiler like the iPhone's compiler, but I haven't heard of one in the works.

Gonna throw this in here, not really an answer to you questions (at least not directly) but might be food for thought.

Have you looked at HaXe and NME?

HaXe is a language very similar to AS3 which can compile to both Flash and CPP. NME is a library which emulates the AS3 libraries and functions, allowing you to write applications in a 'flashy' way and then deploy to Flash Player, iOS, Android, Windows, OSX and Linux. Your code runs in native c on all devices (except flash player of course) and so is nice and fast

It's all open-source and free too.

Have a look at http://haxenme.org if you're interested.