Today I have seen the light - it's called Flash Lite!.

J2ME is fast coming to a close for personally after developing around 10 games over the last year and while there is money in it still, I am looking to newer platforms now given my late entry as a developer into J2ME in 2005 - and basically I don't want to make the same mistake!

Both Flash Lite (FL) and XNA are on the development cards for me now but this day by day diary is intended to relate to FL only from here on inwards and how the transition will be made to incorporate FL into my day to day life as developer.

1.1 Step into the Lite:
-------------------------------

OK I have had a cigar, mulled it all over several times and looked at a load of FL games and demos to get a feel for the level of quality currently on offer and have decided to jump into the pool of lite.

I want to make money from this down the line as an ongoing business so it's not just for fun. As I have written games for most of my life this has to be an exercise in quick learning and fast production of titles if I am to make anything useful happen. This time next year the market (like j2me) will already be flooded with FL content so it's now or never!

1.2 The first day:
-------------------------

OK some of what I might say from here onwards you may not like to hear, but it is the ramblings of a developer starting from 100% nothing with FL from day 1 and how I went about it.

First I do my homework and spend about 4hrs looking at other peoples FL stuff and related blogs on the net. I'm like a giant sponge sucking in all the info I can get into my brain and at one point have about 15 windows of 'cack' open on my dual screen setup - that is a lot of information! Eventually I get most of it into my head and feel like the Neo in the Matrix when he states 'I know Jujitsu!'

After looking at various FL games and titles I quickly come to the conclusion that this is not a lot like j2me, in fact all the content I have seen so far is homebrew generated FL - some of it not bad and some of it graphically really bad.

The first thing I need to get is the FL SDK so I sign up with Adobe and get the required files and bit and bobs. I also open up a few torrents on the net to get what I want but it comes down the wire as German SDK so that's no good to me.

Eventually I get the entire SDK (Trial) for FL 1.1 and 2.0 and am left looking at a hefty price to pay for the Flash 8 Pro IDE when the 30 days expire – oh well I march on regardless.

I takes me about an hour to set up the FL IDE and SDK because I installed the stupid German version first which was a pain to actually get off my machine to put the trial version of the SDK on instead - jeez. After fiddling for about 10 minutes further I load in a FL .fla and see what the deal is - Christ! what a difference it looks like FL is going to be like Shoot'Em'Contstruction kit or Amos all over again and as a veteran coder of the harder variety this all looks too easy to be true.

After getting over my shock of just how a J2me title is actually written and produced compared to j2me or just about any other game I have ever written, I am left with the thought that I can probably knock out a good game every couple of weeks or less - cool!.

Then I face the nightmare again I had with Midp1 vs Midp2... not unlike this we have FL 1.1 and FL 2.0 versions?... I quickly check the list of supported devices and my own FL test devices here (Nokia N70 & N80) whilst going back through those 15 screens of info, and soon realise that we are stuck with FL 1.1 for development if I am to make any money out of it later.

Once I had the SDK up an running and had looked at a few more fla source files and final swf's I then think, ok how to I make the swf? (the end product). After fiddling with the IDE for about 2 minutes I change a few target device settings and hit the publish button - it compiles the swf from the fla - and seconds later out pops the swf amazing - is that it?... yes it is! Man is this looking good compared to all those j2me builds and almost a joke to the seasoned coder :)

I soon realise also that if it is that easy to build the target build (that would probably need to be optimised later for the target devices) then every body will want in on this if they have been a mobile games developer. Within 45 minutes I have a rotating sprite scrolling backdrop (crap bitmap effort not tiles yet) and sound - awesome! though I am still compiling for 1.1 and not 2.0 so I don't know just yet what the trade-off or major differences between the two actually are, so I don't get too much more excited just yet heh :)

I then think OK give me a week or so and I could probably knock up a game alpha build of some description so for now I leave the IDE and concentrate on the actual business end of things.

Distribution... where is it as compared to all other platforms I have developed on?... well there is some but you have to look hard for it. Some developers I have come across already seem to have their finger in the FL 1.1 pie and have distributed FL game titles and apps with larger publishers and some are left out in the cold for now. To cut a long story short here I contact Dale at Mocket after reading about his planned entry into the FL distro scene on 18th October and after a few business related questions about % rates and exclusivity etc. I eventually sign the NDA... as I have now signed the NDA of course I will not mention any further dealings with them other than if they turn out to be good or bad in the months ahead of course, as for now it all looks normal and a good place to start (or getting the door open enough for product pitching later).

Now at this point, and we are coming to the end of the first day (if your still with me then thanks! and I promise future updated to this FL dev diary will be much shorter heh) and I am left in a fairly upbeat mood regards this as a viable platform to make money over the next year alongside everything else I code on - early days for me but it does look viable as a business venture to waste my time on - time being the most valuable commodity of course to any professional developer (read... professional as in you get paid for what you develop - amateur as in you don't of course like everything else).

Mostly at this time I have my dev machine setup with demos of FL SDK's and such like and have not actually acquired the full licensed versions of nay software just yet - I'm still on trial with the whole FL experience and I will know when it is time to cross over into the lite fully from past experience mainly and shell out for the full versions etc.

I look to increase my FL toolbox to make FL games etc. further... so now at the end of the day I'm getting tired but have just enough brain energy left to  download a FL decompile after finding one on Google... I have always used decompilers since the 80's and I'll be honest if you want to learn a particular code set first, then bugger the cost of books etc. just decompile and 'learn' but never copy another fellow coders code is the rule. If you blatantly copy code a) you are breaking the law in most cases and b) you will never be a good programmer just a parrot trust me I learnt the hard way back in the late 80's ;)

I whack and swf I downloaded into the decompiler (another demo version) hit a few obvious buttons and out pops the deconstructed flash lite script and objects - Marvellous! :)

I then go for another cigar and I switch off for the day which was about (11 hours solid flash lite'ing) and overall a good day all in all - even after I have switched off for the weekend I know I will be back to take another in depth look at the decompiler and most likely acquire the full version shortly for further self training purposes of course heh :)

The Flash Lite Diary is a contribution to the blog by Adrian Cummings. Adrian spend many years in the gaming industry with experience in coding projects for Gameboy and mobile phones. Please visit his company at www.softwareamusements.com.