Ultimate Smash Friends: Updates, Feature/idea requests, suggestions

[KizzoGirl]

Hey there

I just want to bring attention to a project called Ultimate Smash Friends [1], that a group coders all over the world are working on. It's an open source fighting game, made in the spirit of the smash bros game series. Believe me when I tell you this: this game will become better than even Melee, and by far (because I'm working on it, and will make it so : ).

It's a work in progress, as this is an open source project and is constantly being worked on. I encourage anyone interested in creating the next generation of smash, take a look at the website and check out the code [2]. Anyone can contribute (code, documentation, artwork, etc.).

[1] Webpage - http://usf.tuxfamily.org/wiki/Main_Page
[2] Source code repository - https://launchpad.net/ultimate-smash-friends
USF Documentation - http://usf.tuxfamily.org/wiki/Documentation
USF Developer Forum - http://forum.freegamedev.net/viewforum.php?f=33

 

[the_dannobot]

looks pretty cool! I'm already working on my own little smash clone in XNA, so I don't really have any time to contribute, but good luck!

 

[KizzoGirl]

I updated the documentation page at USF Doc [1] with future plans, to entice smash coders to contribute. Plans include:

=== Wishlist ===

==== Complete superset of previous games ====

USF aims to become a game that will necessarily be better than any of the predecessors (why waste time working on a game that's not better than the game you're currently playing?). This will be true precisely because it will accurately mimic everything that is important about the previous games. This leaves no reason to continue playing them.

==== Custom Content (Characters, Stages, Items, Physics, etc.) ====

You just know the fun will never end once you get to play smash with any character you want, designed and created by you. USF will let you change anything you want. Users will be able to share and upload characters and stages that they have made. The possibilities are endless.

==== Data ====

Entire tournaments (pools, tourney matches, friendlies, times, results, statistics, etc.) can be recorded seemlessly over the network and stored in a database.

==== Mobile Play on Android OS ====

It is possible (I think) to play USF on the Android OS with android-scripting [1] (to allow Python to run) and a Gamecube-to-USB adaptor [2]. This would allow you to play with others on the bus or waiting in line, or wherever. While I have yet to see anything in Python run on Android, let alone with pygame, this is more than likely possible, and definitely worthwhile.

==== Artificial Intelligence ====

Narrow or general AI could be implemented.

Primitive (narrow) AI could be written in a simple manner, and will allow for simple/repetitive training AI (a dogdy/escapist/campy/opportunistic AI, an aggressive/0-death AI, etc).

General AI could be implemented via [http://www.opencog.org/ OpenCog], and would allow one to create an AI that can play any character well. It is also possible to somehow have an AI read a history of gameplay footage of yourself, and then "learn" how you play, allowing you to play against yourself.

[1] - http://usf.tuxfamily.org/wiki/Documentation

 

[KizzoGirl]

Future plans regarding Android Developer Challenge 3

The following is an email I sent out recently to the developer mailing list of USF.

----

Hello

I just wanted to inform listeners about my future intentions regarding
USF. A lot of motivation for me to work on USF has been in hopes of
the project competing in Google's Android Developer Challenge 3 [1] later
on this year. I am not even sure if it is going to happen again yet
but I love being prepared. I believe that USF has a good chance at
competing in such a contest, and if it does, the project would receive
a large reward (tens of thousands of dollars) to fund further
development.

What that means then, though, is that USF has to be in a workable,
playable, and enjoyable state so that the game gets ranked high by
judges (I believe that you can only win the contest by impressing
arbitrary people, so it has to be good). To that end, the game should
look good, be quick, and be fun.

Based on my own experience with the game series, I recommend that our
first phase of development strive towards making USF accurately mimic
the gameplay physics and mechanics of the Melee series, my reason
being that Melee is by far the most popular of the Smash series, and
will be attractive to the smash community that I am a part of. It
would be good if a lot of gamers from the smash community volunteer to
test the game and let us know of any annoyances or bugs.

With a combination of the above, I think USF could stand a good chance
of competing for the ADC, and it would be worth trying.

--
Kizzo


[1] Android Developer Challenge

 

[KizzoGirl]

Please post in this thread if you have any ideas/suggestions that you would like to see incorporated into the next generation of smash.

To get your mind jogging, check out the USF wishlist.

I myself would enjoy

1.) user profiles (so you have all your data saved)
2.) network play (local LAN would be best, for latency reasons)
3.) database backend (so that all matches and statistics are automatically/seamlessly recorded and viewable later)
4.) data sharing (share created characters/stages/items/etc. over the net)

 

[KizzoGirl]

I've made a lot of changes/updates to the Documentation page in case you folks wanna have a look. Of particular significance is the Roadmap, which outlines the future of what USF will look like for us. I also added more sections to the User Manual to help layout things for everyone.

The important thing about the first release is that it should result in a game that's indistinguishable from Melee ONLY in respect to character/stage physics/mechanics. Lots of things still arn't implemented yet (shields, rolling, dodging, ...), but give it time, we'll be rolling.

 

[HentaiKing4Everr]

intrested but....busy =[

 

[KizzoGirl]

USF gets a forum for devs

USF now has a forum for use by developers, hosted by FreeGameDev.net [1][2]. Here is where the developers will discuss the goings on about the development of USF. You might be interested in seeing my latest post about the heads up display (HUD) and the future of USF regarding the combination of contents (characters, levels, stages, items, gameplay physics, optional shields, etc.) [3].

[1] FreeGameDev - http://forum.freegamedev.net/index.php
[2] USF Dev Forum - http://forum.freegamedev.net/viewforum.php?f=33
[3] Something to look at - http://forum.freegamedev.net/viewtopic.php?f=33&t=233

 

[the_dannobot]

Well, personally if I were going to work on this project, I'd like the design document to be fleshed out a little more. Like what will the menu flow be? What do the characters look like? What are their moves like? Why are both x & y mapped to "jump"? What file format are resources in?

Like just as an example, the last Smash clone I did with team of like 10 people, and we had a 200 page design document with every little detail of the gameplay, data flow charts for the programmers, and asset requirements for the artists & sound designers. Now i know 200 pages might be overkill for what you're trying to accomplish here, but by designing out every tiny detail like that we were able to finish the game in 2 months.

I've looked at OpenCOG and it seems to be overkill for a lightweight videogame AI. Honestly, if you have good online multiplayer no one is going to play against bots anyway haha

Anyway hope this helps. Cheers!

 

[KizzoGirl]

Well, personally if I were going to work on this project, I'd like the design document to be fleshed out a little more. Like what will the menu flow be? What do the characters look like? What are their moves like? Why are both x & y mapped to "jump"? What file format are resources in?

Like just as an example, the last Smash clone I did with team of like 10 people, and we had a 200 page design document with every little detail of the gameplay, data flow charts for the programmers, and asset requirements for the artists & sound designers. Now i know 200 pages might be overkill for what you're trying to accomplish here, but by designing out every tiny detail like that we were able to finish the game in 2 months.

I've looked at OpenCOG and it seems to be overkill for a lightweight videogame AI. Honestly, if you have good online multiplayer no one is going to play against bots anyway haha

Anyway hope this helps. Cheers!

Thanks for the suggestions, as I actually haven't thought about a design document yet. This is the first game programming project I've been seriously involved with, and I now remember that the design document is a useful thing in game development. I'll run this by the other devs.

And yes, I know OpenCog would be too much for USF, but I'm a programmer for OpenCog too, and I am trying to do as much with it as I can. I KNOW using a weaker AI system would be sufficient, but for my own academic purposes, as well as purposes related to OpenCog itself, it would be useful for me (personally) to pursue such a goal, and that's why I'm even considering OpenCog. I know it's overkill, but that's what I want : ). It would make me (and a bunch of other AI friends I know) very happy to see an unbeatable USF AI controlled by OpenCog.

 

[KizzoGirl]

USF and Google's Summer of Code 2010

USF is applying as a mentoring organization for the annual Google's Summer of Code event [1]. If USF gets accepted, a certain number of students will be sponsored by Google to write code for USF. This will greatly speed up progress, and many of you can help (and get paid : ). I'll update on if we get accepted or not.

[1] Google's Summer of Code - http://code.google.com/soc/

 

[HentaiKing4Everr]

becareful of double/triple posts...Just continue editing your first post =\

 

[KizzoGirl]

becareful of double/triple posts...Just continue editing your first post =\

Oh yeah? Is that what it looks like? Could you give an example of what posts you're talking about, because I genuinely don't know what you're talking about. I'll take a guess though of course: are you talking about my posts that have a bolded heading and are intended to act as announcements (Android Developer Challenge, Forum, Google's Summer of Code, etc), or are you talking about the posts where I'm just quoting someone else and responding to what they said (like this post) ?

I'm guessing that the first type of posts that I described to you is the kind of posts you're talking about. In such a case, are you suggesting that instead of the posts for the ADC, the forum announcement, and the GSoC announcement, you suggest that I just update the first post with the same such information (instead of making a post about it)?

This is just making sense to me as I type it, so I will do that.

 

[Super_ness]

I'm pretty sure that's what he meant. I don't really care about double posting, but I don't want you to get infraction points.

 

[HentaiKing4Everr]

its what the words mean "Double post". The reason Im warning you is because of infraction points. Too much of it = ban.