05-21-2011 02:18 AM
I guess It really has been a long time since I was an active developer if the current PB tools are so painful for folks to use. Clearly some of you folks never worked on anything prior to theyear 2010 or something. Anything where I am not using a hex editor, an EPROM burner, and a logic probe is a win.
Haha, what is this? Are you being condescending, or pulling the old "walked uphill both ways to school" kind of thing, or what? Whether or not you think it's a bonus to have some sort of gui or whatever is irrelevant. The fact is, I bought a Droid from someone on craigslist, and within about 90 minutes, I had the development environment all ready to go, and was deploying directly to the phone, yet it takes HOURS upon HOURS to get the PlayBook right, and that's not even counting the waiting time for keys. I just don't see why it has to be so hard and complicated with keys, and passwords, and silent failures, and doing things manually on the command line, and setting up PATH variables, and setting up custom builders, etc etc. I'm a very good BlackBerry programmer, yet here I am asking questions about installation and configuration. That means there's a serious problem. How could I possibly recommend this to a casual or novice programmer?
The point is, I should have been able to plug this thing in, install a driver, enable development mode, and deploy straight to the device, no questions asked. That's how it is with Android and iOS. That's the standard that developers expect in 2011.
05-22-2011 09:10 PM - edited 05-22-2011 09:13 PM
I was pretty down on the Playbook process as well.
Then I submitted my app through Apple App Store. Wow, what a joke. You can literally not even submit your binary file if you're not running OSX, they removed the feature completely from their developer website. You need multiple cerificates depending if you're developing or releasing, multiple provisioning profiles as well, and it also is a multi day processs just to get signed up.
So, RIM is somewhere in the middle I think, Android is just amazing and basically a dream experience compared to anyone else, Apple is just ridiculous (at least as a Flash developer, it's probably way easier as a native xCode dev) .
05-22-2011 10:20 PM
The only problem with Android is that nobody is buying anything, and it's a complete cesspool of the worst of the worst apps along with outright stolen source code. Actually, there's a couple problems, but I'm not gonna go there in this post. We can talk about that later, when we all move to Android since we can just port to PlayBook from it.
Anyway, I just wanted to give y'all helpful folk a little update. I'm not entirely sure what did it, but I think the difference was adding the author and authorId tags to my XML. It's described in this video for anyone who needs help in the future: http://deleteaso.com/playbook-debug-tokens-and-sig
Back to RIM. So, I deployed a bunch of builds to the PlayBook tonight, cutting out this collision and that collision, and making things not draw or spawn, and it seems there's no one big thing killing my game. The PB just can't handle the game, period. It's not like my collision is bad. I check each object's possible collisions once per frame. I've seen other slow PlayBook games, some of which obviously had some work put into them (though I can't say how efficient their code is). A RIM employee told me the flash implementation is slow, which I'm not sure whether to believe or not. At this point, I just feel like there's nothing I can do to bring my game up to 60FPS on the PlayBook, the same way it is on my computer. It's horrible. I mean, it's totally playable, but it doesn't look nearly as good without the fluid movement. The only thing I can think to do at this point is port it to iOS since it's written in AIR, and see what happens. Maybe it's an AIR thing. Maybe it's RIM's fault. I dunno. All I know is that I deeply regret starting this project.
Thanks again for all the help getting me running. At least I know it wasn't a quick fix now.
05-22-2011 10:43 PM
AIR 2.6 / 2.7 are supposed to have significant performance improvements, and PlayBook currently only runs AIR 2.5. AIR 2.7 is supposedly targeted for PlayBook by the end of summer (based on posts made by RIM in this forum). Perhaps your app will perform better then.
05-23-2011 12:51 AM - edited 05-23-2011 12:51 AM
The Android market is definately **bleep** in terms of sales. My free version gets 150 installs every day consistantly, my paid version sees maybe 1-3 sales a day. It's brutal, I'm anxious to see how it will do on the iTunes App Store, where I've ditched the trial version and am just doing the paid.
Btw, is your game using a blitted rendering engine? If you really want good performance in flash for games on mobile, you have to ditch the displayList. Using something like flixel is a huge win.
05-23-2011 05:20 AM
@shawnblais - think i asked you on another thread, but do you have any estimate of your piracy rate on Android? I've seen other folks writing about piracy rates of 80-90% over there.
05-23-2011 08:52 PM
I dunno, I have analytics that tells me whenever someone installs a paid version, and judging by that I have virtually no piracy. But then again, if they are blocking network access for the app, then I'd never know.
The nice thing about Android is the sheer volume. You get so much traffic, that at least some of those users are going to purchase your app. My goal right now is really to harness that traffic somehow, but I'm starting to think the best bet is maybe an ad-supported free version, with a paid ad-free version, for Android specifically.
There must be a reason Angry Birds went that route...
05-23-2011 11:36 PM
I personally think users percieve a "donation" version as better than a "paid" version, even though they are the same thing. Something to think about.
Also, about Android development, I have no idea where to go with this question, but I realized earlier today that my computer no longer recognizes my Droid properly. Ever since I installed the PlayBook drivers. HMM. If I plug it in, the Droid knows it's plugged in, and the computer sees a Moto a855, but it won't see it as a mass storage device, and I can't get Eclipse to pick it up anymore. I'm starting to think that a divine force is attempting to push me away from mobile development. Ideas?
05-24-2011 12:50 PM
Not really what I am saying. I am by no means experienced on these devices and am not a heavy duty coder -- been producer-ing rather than coding for a very long time now. But what I am saying is that with limited experience or know-how, I had pretty much no trouble getting things built, installed, tested, signed, and submitted. The turn around waiting for keys was a little iffy, but it was a one time thing. A couple of simple batch files works to drive the command line stuff more easily. All this whining ("it's hard so I'm not gonna play anymore") about the tools is ridiculous. Can they be better? Sure. Are they blocking development in a serious way? Not for me and many others. If I want to bitch about the dev system, I'll complain about something I can't easily work around like the lack of support for the sensors, the lack of a native dev environment, etc...
05-31-2011 10:16 AM - edited 05-31-2011 10:19 AM
How nice for you. Did you set it up on my system? No? Then I guess you have no idea about how difficult installation can be when it doesn't work properly. Why are you leaving such worthless comments?
Just stick to "producer-ing" and leave the dev talk to developers.