Perhaps a port of cocos2d. Yes, there's already cocos2d-x, but what about cocos2d-s? Imagine the possibilities if you combined cocos2d with sheep.
| Get the Video Plugins |
Are there frameworks/engines/libraries/... you are missing when developing?
(52 posts) (13 voices)-
@bradparks another great idea! This is something that can be easily expanded. Making a system to share levels would not take a whole year to accomplish.
I was also thinking about making a competition framework, that developers can use to create ladders and competitions. The admin can easily create competitions, and players can subscribe. Than a tree structure is created for the competition. Players have to play one game for the competition every day, and the winner goes further. After some days the finals will be played and a winner is selected.
For example:
1) admin creates competition
2) admin set rules which defines what the mission will be (get furthest in 2min, get 20000m touching minimum platforms, reach 15000 as fast as possible,...)
3) player is notified there is a new competition
4) player can join the competition
5) player is notified the competition is started, and when he has to play his game
6) player decides to play the game for the competition, and in the end the result is send to the server (admin)
7) the server uses the rules to pick the winners
8) winners compete against each other until there is only 1 leftThen it must be easy for the admin to create new maps etc for the competition (using a system @bradparks mentioned), and the players can then download the new maps and play the competition.
What do you guys think ?
Again: thank you for all the info!
Posted 6 months ago # -
bradparks
Memberhey @mrdckx ! I think the idea sounds good.... I like how it builds on the shared user contributions idea as well ;-)
IMO, I think that building the core content sharing idea first is best, as it adds real value to your product.
It's basically getting your end users to create levels of your game for you, and talk about it since they're probably proud of what they created!
You could release an app with this feature, and never do another release, come back in a year and your end users have created tons of new maps for you, and for other users of your app without you having to do a thing....
So definitely smart!
Posted 6 months ago # -
-
They declined the sharing module, because there is not enough research involved. I agree, it's a really nice project but not suited for a thesis.
Finding a good subject is more difficult then I fist thought. Thank you guys for all the help, learned a lot from this thread :) Great community!
Posted 6 months ago # -
bradparks
Membertoo bad ;-) so can you give us more details on the requirements for your project?
e.g.
- needs "research"? What's this really mean? It has to be something that hasn't been done before in some regard?
Maybe if you could post your actual requirements somewhere, or the list of what they're suggesting as topics (the ones you don't like), we could come up with something better!
Posted 6 months ago # -
Just throwing stuff out there but some stuff I'd like to see:
* real time facial gesture recognition (smile, frown surprise etc)
* a pluggable image processing framework i.e think audio but with images (edge detection, blurs, posterize, etc)
* generating vector based art with ability to output files for printing (very large)There are bits out there but a library + API would be nice :)
Posted 6 months ago # -
There are no strict guidelines or requirements. But I will try to explain what the professor told me. For example if you want do develop a multiplayer framework, you should first look at what the device is capable of, which techniques can be used and which are best applicable in that particular case. Then you can show some graphs of what you discovered and use this "research" to build the framework. Then you have successfully showed that your theoretical research is indeed correct, and the results are the results you expect.
The problem with the sharing module is that I can actually start right away with the programming. There are of course a few difficulties that have to be handled, but you could solve that with techniques that already exist. And in this case you can't show graphs of something you actually discovered.
I hope it is clear now what I meant with "research". Probably this should be mentioned in my first post, but back then I had no idea. And I wanted this brainstorm to get as many ideas as possible :)
@abitofcode Something like the real time facial gesture recognition would involve a lot of research.
Thanks for the ideas!
Posted 6 months ago # -
bradparks
Memberso it seems it either has to be:
something that hasn't been covered by the curriculum of your school
or
something that you personally don't have any experience in?
I *personally* think your professor is being a little close minded ;-) One of the biggest things programmers have zero experience with when they get out of school (and for years to come ;-) is marketing.... Which is arguably of equal importance to the actual product itself, and is a real big part of being an indie or an entrepreneur in the real world.
Most of the time it doesn't matter if you've built the best game, Os, or app in the world, if you don't know how to market it - ie get it used, played, - SOLD!
From a marketing perspective, you could release 2 games that were essentially the same game, but one with the sharing feature enabled, and one without. Make predictions (ie do research) on what you think would be the projected user base x months out, and validate it at the end of your year, using graphs before and after.
Of course "technical approaches to marketing an iOs app" may fall outside of what the professor would deem "research" ;-)
Posted 6 months ago # -
The thesis has to prove we can do scientific research, we can program (a little) and we can write a paper about it. Next year I have a lot of time for my thesis, so I would like to build a really nice product that I am proud of. The greater the enthusiasm, the better the result...
An example of a thesis that can be chosen:
Research question: can we improve the draw performance of NodeBox by porting it to OpenGL?
NodeBox is an open-source application that lets you create 2D visuals (static, animated or interactive) using Python programming code and export them as a PDF or a QuickTime movie. NodeBox is free and well-documented.
NodeBox is currently written as a Mac OS X application, using the Quartz graphics API. Multiple ports exists, for example to Qt in Linux. It is our belief that a port to OpenGL can improve performance. OpenGL provides a set of modern API's related to vertex arrays, Vertex Buffer Objects, etc. The project would study the impact of each of these features on the performance, while keeping the current NodeBox drawing API intact.
Posted 6 months ago # -
[Edit- posted this before seeing your post above]
AR or face/gesture recognition sound like good 'research' projects, both would be very interesting thats for sure. Plenty of background to look into. I expect multiple frameworks exist for both (slight guess here!), which raises the question of what you are going to add. Perhaps a new application, rather than re-inventing the wheel...
Is this a computer science dept? Should the projects be in gaming, or basically any field at all? A 1-yr project?
Posted 6 months ago # -
... handwriting recognition? There are libraries for this, of course, some v expensive. They have various limitations based on language, speed of lookup, errors etc. If you're into neural networks you could train a system to learn the users writing. I make it sound so easy ;)
Or, improved text auto-correction. God the iphone needs it.
These are quite mathematical, and ultimately could be quite difficult.
On the plus side, you can get lots of friends/colleagues involved with testing.Posted 6 months ago # -
Something with artificial intelligence or text to speech (along the lines of Festival), or sheep.
I wish there was an easy way to get something more than the basic robotic voice in Festival working on the iPhone, but it looks like a lot of work. Of course, if Apple ever ports their Mac OS X TTS, that will obliterate your project.
Posted 6 months ago # -
Yeah, was thinking about speech (Siri), but no idea where to go with it. Actually, forget that, far too difficult.
Or that AR app which translates foreign languages in real-time when you hold up text to the camera. That was so cool. but no point re-inventing it.
Posted 6 months ago # -
Ok, I'm getting away from developing 'frameworks', but I've seen a few news items lately about apps for medical use. e.g. I saw one which helped calculate the treatment for burns victims by getting the user to draw on the ipad the coverage /location of the burn:
http://medgadget.com/2011/11/iphone-ipad-app-for-better-burn-injury-management.html
Thats a comp sci phd project, but you'd need to search hard for a similar idea. Your prof might complain its not using low level opengl (just kidding ;).
Posted 6 months ago # -
Every mobile speech to text thing I've seen so far (Siri, Dragon) uploads the audio capture to a server for the heavy crunching and returns the result to the app for processing. Admittedly, Siri has an AI component, but other than that, the heavy lifting is usually done server-side. I used the Dragon solution in a valet app so attendants could speak license plates, make, model, color, etc. Kinda fun and it works well if you need a STT solution, but it's not free.
Now, an open source TTS system would be great, but that's a huge amount of work. Years ago, I helped Apple to wreck a nice beach and they had lots of people working on that.
Posted 6 months ago # -
MrMojo
MemberHere's a technical idea that I really don't know if anyone can get to work. But if you did, it'd be very interesting. And I may be totally missing something that would make this completely unworkable, but I think it can be done!
That having been said, you could try and port the Flash Player to iOs using the following opensource projects which already do alot of the work for you. Note that they're all written in C++.
** Gnash - the opensource Flash player for Linux
http://www.gnu.org/s/gnash/** GameSwf - A Swf rendering library for C++
http://tulrich.com/geekstuff/gameswf.html** MonkSWF - a Flash SWF vector renderer already running in Cocos2d
http://blog.zincroe.com/2011/11/displaying-a-swf-on-the-iphone-with-cocos2d-and-monkswf/** GameSWF port to Marmalade (Cross platform C++ game lib running on iOs/Android)
http://www.madewithmarmalade.com/devnet/forum/6024Now I haven't looked at the licenses of any of these things, but if you could get this to even reasonably work at all, then I feel confident you could find some way of turning it into a money making venture.
You could release an iOs browser that had this already embedded in it, and make it so when it saw a Flash embed tag it embedded your Flash plugin instead, even if it was only an overlay.
Food for thought! Your thesis idea could be "Can the Flash Player be ported to iOs using available opensource projects?"
Posted 6 months ago # -
Apple pretty much crushed Adobe's attempts to port Flash, and I don't think you'd ever get this through review. It would likely fall under these clauses:
2.7 Apps that download code in any way or form will be rejected
2.8 Apps that install or launch other executable code will be rejectedIf it could have been done, it would have been done by now.
In any case, Adobe Flashbuilder and Air can now export to iOS, so there's that.
Posted 6 months ago # -
MrMojo
MemberEven if Apple crushed the idea in a browser, it could still be used just as Adobe is using it - to build native binaries that don't download code at runtime - they load it from the resources of the app.
So it could still be used as a way to
** build native iOs apps using Flash without Adobe products
** package SWFs generated by 3rd party tools since there is no Flash FLA to open in the Flash IDE and publish.
Posted 6 months ago #
Reply
You must log in to post.