FreeOrion Roadmap

This is for directed discussions on immediate questions of game design. Only moderators can create new threads.
Message
Author
charlieg
Space Floater
Posts: 38
Joined: Tue May 25, 2004 1:59 pm

#31 Post by charlieg »

Well, jacks provides this meta-link for VS story/background info.

He also says (based on a quick look through freeorion.org) that, while the VS backstory might not be the best fit, there is a lot of species and backstory work that you might find of interest. He'd be happy if anything comes of it.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#32 Post by Geoff the Medio »

Another potential source of inspiration...

http://www.orionsarm.com/main.html

freereign
Space Squid
Posts: 53
Joined: Wed Mar 19, 2008 4:40 pm

Re: FreeOrion Roadmap

#33 Post by freereign »

The Roadmap looks good. Personally, I would like espionage sooner, but that isn't a good idea. It should stay where it is (check my sig to see why i want it sooner). Great job guys! By the time this gets finished (hopefully not too long) it will beat SotS, GalCiv, and MoO by far!
MoO: Darloks Rule!
MoO2: Custom Darloks Rule!
MoO3: Dunno cuz I'm a cheapskate

RogueWolf
Space Krill
Posts: 1
Joined: Sat May 24, 2008 5:39 pm

Re: FreeOrion Roadmap

#34 Post by RogueWolf »

I think the Moderator's, Team Leads and programmers have enough on their plate as it stands.
I'm an Operations Manager and I love this concept. Loved Master of Orion back in the day, and #2 was well thought out and had enough to make you want even more... #3 would have been good, loved the design of the star-map, and space travel, and I could ALMOST deal with the RTS style combat... if it didn't take FOREVER... But they, like this project here... are trying to do way too much way too early.

Everyone and their dog has played the Space X style games, loved them and wanted just that much more... This project can do it. I honestly believe it has all the potential to get to the point where NO ONE is asking... why can't I do this... or geez, I wish they had done that... But right now, all I read on the posts are more and more of the same ideas. At ExxonMobil where I used to work, we called meetings like that Echo's from the Boss... just going over the same stuff, over and over.

I like the fact that this group has set out a plan of attack, with different stuff needing to be completed before the next 'release' or version. But I think my point here is, that the programmers, designers and strategy consultants should step back a second. There should be some things that ARE a priority, and somethings that are not.

Space Monsters... not a priority... an afterthought that can be thrown in later in the game.
Leaders... not a priority at all... even if they have 'incredible affect' on the game play, in the end they are just number modifiers, and should not be considered a priority.
Espionage... depending on how it's implemented, also not a priority, as it too can be introduced after more essentials are up and running.

What do I think are going to be the BIG hurtles? And what issues do I currently see with the RoadMap?
1.) v0.5 - I think it is far too early to work on the AI. My reasoning?
i) This is not an easy thing to do, it'll push back (timeframe) the other essentials, which are needed to enhance the game.
ii) Later additions, changes and modifications to the game code might render all the previous AI work either useless or inept, such as not being able to take advantage of some subtle adjustments to game balancing.

2.) Space Combat should be a priority, although I personally am one to 'skip' the combat, and let the dice roll, by themselves as it were, the majority of the hard programming work is in this area. It's so big, and complex that it is like a game within a game. If done properly (and not like that HORRIBLE game I just bought Starforce Captains... don't buy it BTW.) then it will be a 'selling feature' as who doesn't like to see cool ships fly into combat guns blazing? (I know it's free, but popularity counts too.)

3.) I think that v0.7 and v0.8 should be switched around. As stated before the 'random events' space monsters and espionage can be added in once the political, leaders and races are decided.... which brings me to my final point.

4.) Moderators / Designers / Community:
Take control of the suggested races. You can't build 'SPORE' like creatures yet, and although we all wish it will eventually be able to do that, I think settling on 4 MAJOR races would be a FINE starting point. At least for the v1.0 release.
Set up a forum to Vote on people's suggested races and storylines. Take the top 4 most popular... something like that... and when glitches and bugs and other features have been added look to expand.

I love the fact everyone here is aiming for the sky. But I've been in the business for a long time... small steps forward will make a huge impact.

Good luck everyone involved.
I'll keep watching, and putting in my 2¢ whenever I can.

-Ryan Mahood
B.Tech - McMaster U.

M4lV
Space Squid
Posts: 89
Joined: Sat May 10, 2008 10:51 am

Re: FreeOrion Roadmap

#35 Post by M4lV »

I agree there. Tackling the AI was something we at our bote project did virtually last, although espionage and its AI handling was in fact the last thing, directly added afterwards. It proved to be very efficient, since all in-game variables, handlers and functions were already properly defined and so the AI could be coded one-piece balance-wise where the most important thing was the ability to cross-check and -test all routines for errors and weird behavior when interacting with each other (Ship moving AI vs. Diplomacy AI for example; it makes absolutely no sense to have your routines make the decision to declare war diplomatically and on the same time still colonizing insecure planetary systems near that enemy's border; though this may seem rather obvious and easy to handle, you'd wonder how often these conflicting states internally occur; then it helps to be able to extensively testing it and get down to the problems and creating new algorithms or adaptions to the old ones that are bullet-proof in that area).
I could only think of it being - despite the best documentation efforts - utmost complicated to achieve that with AI-coding-wise empty time in between, especially when possible former AI code contributors are leaving all the while and you cannot just call him for help to conduct necessary adaptions arising from newly-coded features.

3D space combat can be coded anytime I think. It's maybe even better to do that rather soon in such a project in order to be able to provide some additional nice screenshots and possibly draw in new attention thus new talents for the other parts of the game.

To a certain extent this is also true for the AI but since that one is extremely more difficult than whipping up a nice 3D scenery to play with, I'd rather say, do it at last. Players can always play in multiplayer (and real players will mostly prefer that anyway over single-player gameplay; also you need to consider that GalCiv II already has the crown for SP-experience in the 4X sector).

Thus I'd say do the AI at v0.8, better even v0.9 along with the general beta-testing cause you need a lot of players to test different implementations while coding the AI close-by and ready on the spot. That's what I can say as moderator of a somewhat similar project that has already undergone large parts of an alike road map.

User avatar
eleazar
Design & Graphics Lead Emeritus
Posts: 3858
Joined: Sat Sep 23, 2006 7:09 pm
Location: USA — midwest

Re: FreeOrion Roadmap

#36 Post by eleazar »

Welcome: :)
RogueWolf wrote:There should be some things that ARE a priority, and somethings that are not.

Space Monsters... not a priority... an afterthought that can be thrown in later in the game.
Leaders... not a priority at all... even if they have 'incredible affect' on the game play, in the end they are just number modifiers, and should not be considered a priority.
Espionage... depending on how it's implemented, also not a priority, as it too can be introduced after more essentials are up and running.
The fact that's something is on the roadmap doesn't absolutely mean that the discussion of the topic will lead to it's inclusion. But there's not much point in skipping space monsters... it's mostly content, with a little AI. In all it's a tiny speck compared to some of the other chunks of work.
RogueWolf wrote:What do I think are going to be the BIG hurtles? And what issues do I currently see with the RoadMap?
1.) v0.5 - I think it is far too early to work on the AI. My reasoning?
i) This is not an easy thing to do, it'll push back (timeframe) the other essentials, which are needed to enhance the game.
ii) Later additions, changes and modifications to the game code might render all the previous AI work either useless or inept, such as not being able to take advantage of some subtle adjustments to game balancing.
This is certainly not the complete and final AI at v.0.5. But it's not too early to put the hooks in the code for AI, and to do the basics of diplomacy, declare war, peace, etc. Even a really stupid AI will make it vastly easier to test the game.[/quote]
RogueWolf wrote:3.) I think that v0.7 and v0.8 should be switched around.
Not a bad suggestion.

User avatar
Bigjoe5
Designer and Programmer
Posts: 2058
Joined: Tue Aug 14, 2007 6:33 pm
Location: Orion

Re: FreeOrion Roadmap

#37 Post by Bigjoe5 »

So, nowhere in the roadmap are campaigns mentioned. I've pretty much been assuming that they'll get some attention when we have some extra manpower for coding, there's enough of campaign mode actually planned, and we actually have enough of FO done to make some initial campaign programming meaningful. It looks as though v.6 will be far from labour-intensive and we'll have the diplomacy model in place by then. Might that be a reasonable time to add some campaign stuff?
Warning: Antarans in dimensional portal are closer than they appear.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: FreeOrion Roadmap

#38 Post by Geoff the Medio »

Bigjoe5 wrote:...there's enough of campaign mode actually planned, and we actually have enough of FO done to make some initial campaign programming meaningful.
What sort of "campaign programming" did you have in mind?

User avatar
Bigjoe5
Designer and Programmer
Posts: 2058
Joined: Tue Aug 14, 2007 6:33 pm
Location: Orion

Re: FreeOrion Roadmap

#39 Post by Bigjoe5 »

Geoff the Medio wrote: What sort of "campaign programming" did you have in mind?
For starters, the galaxy on which the campaign will take place needs to be designed and programmed for each campaign, since there will no doubt be minor adjustments to the galaxy map between campaigns (exploding suns, destroyed planets, new specials, and other Precursor interference such as the creation of starlanes or new planets/systems). I'm guessing this will be the equivalent of having multiple galaxies hardcoded into the game for use in specific campaigns, unless there is a way to code modifications to a single campaign galaxy that will take effect as part of the initial conditions of a given campaign.

Once this is accomplished, the initial conditions of each campaign (colonies & ships owned by relevant empires, current diplomatic statuses and multi-governmental bodies, galactic events in progress, etc.) can be coded; even if none of the race attributes have been coded, different races themselves can still exist and be placed in initial conditions.

Only after v.8 can the final additions to galaxy creation be made (for the sake of space monsters, stranded leaders, whatever), initial conditions dependent on certain race attributes be coded (although no races will function in a fundamentally different way, so this may not be an issue at all) and all the scripted events be programmed, but the above would certainly make for a nice start.
Warning: Antarans in dimensional portal are closer than they appear.

User avatar
strooka
Space Kraken
Posts: 165
Joined: Tue Jun 23, 2009 5:34 pm
Location: Bielefeld, Germany

Re: FreeOrion Roadmap

#40 Post by strooka »

Hi there i have recently downloaded the svn Tree on my mac and i needed for à Compile very, very Long. Why don't migrate the project to Java ? I have started a project with netbeans i Hope i can Mix Java and c++ like i already did with Java and javaFX. Compile Times om Even Large projects are there just seconds, this Wohle Be a Great advantage. In Addition i have à Running javaFX java3D project which simulates the Sun System . You can Fly through the System and even 3D Models are loaded in from lightwave.
Joystick Input is Supported. Also we would have only One project file Causen netbeans exists under Win and mac , Linux i'm Not shure. What about it?

User avatar
OndrejR
Space Dragon
Posts: 339
Joined: Thu Oct 02, 2008 11:00 pm
Location: Slovakia

Re: FreeOrion Roadmap

#41 Post by OndrejR »

strooka wrote:Hi there i have recently downloaded the svn Tree on my mac and i needed for à Compile very, very Long. Why don't migrate the project to Java ? I have started a project with netbeans i Hope i can Mix Java and c++ like i already did with Java and javaFX. Compile Times om Even Large projects are there just seconds, this Wohle Be a Great advantage. In Addition i have à Running javaFX java3D project which simulates the Sun System . You can Fly through the System and even 3D Models are loaded in from lightwave.
Joystick Input is Supported. Also we would have only One project file Causen netbeans exists under Win and mac , Linux i'm Not shure. What about it?
http://forums.sun.com/thread.jspa?threadID=5373702 wrote:JavaFX 1.2 doesn't provide a script-language-based and hardware-accelerated 3D API. Neither in the FAQ nor at JavaOne 2009 any announcements were made. Concerning 'JavaFX Prism' no details are available yet. Early in 2008 Sun interrupted their contributions to the Java 3D API for the JavaFX 3D Scene Graph (SG3D at JavaOne 2008). There are rumors that it will be released this year, or later.

If 3D support is now needed or desired in JavaFX, an alternative might be integrating one of the lower or higher level 3D Java APIs. Lower level Java APIs are JOGL (Java binding for OpenGL) and LWJGL (Lightweight Java Game Library). Higher level scene-graph-based Java APIs are: Ardor3D (fork of jMonkeyEngine), Aviatrix3D, Java 3D, jMonkeyEngine, JUniversal3D (fork of Java 3D), and Xith3D. All of them do 3D rendering at least per OpenGL. Java 3D and JUniversal3D render per DirectX/Direct3D, too.

The most common 3D rendering target is a heavyweight component derived from 'java.awt.Canvas'. This is the fasted Java GUI component available. Unfortunately, JavaFX doesn't support heavyweight components in its scene. So, 3D Java APIs have to provide a lightweight component for 3D rendering. This can be added via a SwingComponent to the JavaFX scene.

As InteractiveMesh is also committed to Java 3D and its fork JUniversal3D, it was and is a challenge to make current and future Java 3D applications runnable under JavaFX as they are runnable under Java/Swing and Eclipse/SWT. But, it looks as if no other 3D community has published a JavaFX 3D application powered by their 3D API so far. Am I right?

Lightweight 3D rendering is not as fast as 3D on-screen rendering by a heavyweight canvas, because in lighweight 3D rendering all pixel data have to do a round-trip GPU -> CPU -> GPU instead of to be drawn directly into the display.

JOGL's lighweight component 'javax.media.opengl.GLJPanel' provides an actually faster alternative: direct rendering into a Java 2D back buffer (Swing back buffer ?) to avoid the round-trip. But this is declared highly experimental and isn't available as public java- or javax-classes.

Therefore, I'm curious if JavaFX will provide high speed back buffer 3D rendering and if this will be available for Java applications as well.
Also there is multiplatform open source IDE for C++ - Eclipse and therefore is no need for Netbeans. Also standard for 3D applications is Ogre3D.

planetbaker
Space Floater
Posts: 29
Joined: Sun Jan 24, 2010 11:42 am
Location: Sol d

Re: FreeOrion Roadmap

#42 Post by planetbaker »

I'd like to propose to consider an additional stage / slight design change: AI

It might be an idea to offer an API for AI so that 3rd party AI could be written. That'd have the big advantage that the AI developement needs not necessarily be done by the lead devs of this game and spread a bit the work load.
An example I can point at where such solution was implemented is OpenTTD which offers an API for AI written within squirrel, where each AI runs in its own VM. Find the implementation e.g. here: http://hg.openttd.org/openttd/trunk.hg/ ... 3e/src/ai/

The idea is such that each action which a human can do has an equivalent AI API call which the script can use to execute it the same way a human does.

But maybe this idea is also completely out-of-scope and irrational :-)
Unless stated explicitly and differently, any patches or artwork I should post in this forum are licensed under GPL v2 or - at your discretion - any later version of the GPL.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: FreeOrion Roadmap

#43 Post by Geoff the Medio »

There already is an AI API, and the current AIs are written as Python scripts, which are located in FreeOrion/default/AI

planetbaker
Space Floater
Posts: 29
Joined: Sun Jan 24, 2010 11:42 am
Location: Sol d

Re: FreeOrion Roadmap

#44 Post by planetbaker »

Oh, sorry for having wasted your time then with a pointless proposal :-S
Unless stated explicitly and differently, any patches or artwork I should post in this forum are licensed under GPL v2 or - at your discretion - any later version of the GPL.

User avatar
OndrejR
Space Dragon
Posts: 339
Joined: Thu Oct 02, 2008 11:00 pm
Location: Slovakia

Re: FreeOrion Roadmap

#45 Post by OndrejR »

FreeOrion AI is in early stage. It doesn't mean it does nothing or there is no work on it in progress.

Post Reply