Anger's Alternative

ok, let me get some things set straight before i continue...

i like moon breakers and i don't want the game to disappear. that slowly seems less and less possible unless we hear from WJ or anyone from IMBA / Ubernet. so what is this about then?

since about 2005-ish ive been researching and planning a space dogfighting game. part of that has been studying game programming techniques, languages and even entire game engines. this was a personal project for my own amusement and just to see if i could come up with something better than the gaming industry had delivered thus far in this genre. when i found MB, i put all those plans on hold having found a game with most of the mechanics i would have wanted, an active community and (at first) a prominent future of features for this game. things looked good indeed...

here we are now a few months later, and im picking up where i left off but with a few differences...

my game is no longer just a personal project; in the increasingly likely event that MB has hit the end of its development i want it to not just replace MB, but become something better than MB ever could. to do that i will need the help of the people who understand what does and does not work in this type of game, ie, the current MB community. in effect i want the game to be a community-created successor that makes use of the fruits of my research and insights.

if MB picks up again? the project won't be sidelined, but run in parallel to MB. in all honesty i dont want to compete with MB - theyve put time and money into making MB and i do not wish to steal the spotlight from underneath them; so instead it will go on to be an alternative with MB credited as the inspirational source and i'll let people decide for themselves.

the game / project currently has not yet been started, nor does it have a name yet either. i cannot make any promises to the fate of this project at this stage. but. im dead set on making a solid effort to get this thing going.

the core of the game engine will be a c++ based client; the game code, menus, items, gameplay & such will all be scriptable. to that effect the client will be a separate project from the game itself, allowing the community to work on the game without worrying about specifics or how the client does its magic.

the project will have these stages:

1. set up project (currently here)
- set up SVN
- set up project website to track progress / coordinate people (bug-tracking site will do)
- organize teams

2. develop the client core
- create & test c++ integration with Lua Runtime (+STDLIB for file IO & net sockets), bass & openGL
- create & test Lua Bindings to high level bass / openGL funcs (eg, play sound, draw models).

3. develop the client toolkit for Lua Runtime
- develop/test the gfx engine to allow: UI, effects, sprites, model/texture rendering, primitive rendering, display list support.
- develop/test the sound engine - doppler, 3d occlusion (if bass supports it, i think it does), (possibly) DSP support.

4. develop the game engine core
- create game base objects and extensibility rules (will do myself for 99% of it).
- create & test network play code.

5. develop the game
- create & test menu's.
- create & test: basic ships, basic game modes.
- modify the client toolkit as required.

6. extend the game (Alpha Stage)
- add, test & balance game objects & menus as deemed required.
- modify the client toolkit as required.

7. beta and beyond
- it is impossible to predict beyond this; alpha stage will shape the game beyond any recognition.

what i am looking for as priority:
project management assistance - this thing will get big quickly if a lot of people vouch their support. probably the most urgent role.
coders for the client (ok to good c++, helps if you have some experience with opengl / bass / fmod / network / 'nix & mac dev).
coders who have worked with the Lua 5+ runtime before.
client testers.

what i am looking for once the client is in a stable state and ready for game dev:
anyone with game design / balancing experience.
anyone who likes working with scripting languages.
modelers / graphics designers / audio engineers / etc.
more testers.

rather than replying, ive added a poll (for 2 weeks) to show numbers of each role vouched so far. please be honest and realistic about this - the figures in 2 weeks time will say a lot about where this project will head. if you feel undecided about this altogether, all i can do is ask that you keep an eye on how this project is going (updates will be regular, and im not disappearing anywhere any time soon ) and offer you the chance to jump in and assist at any stage. anyone who does help in any meaningful way will not go unnoticed, but in the end this is for you to enjoy too...

again, i cannot make any promises that this will ever be completed. but if we happen to end up with enough people showing their intent to assist and support this parallel project then that's a different story; we may very well already have all the resources and people we need to create something truly fantastic... that also works...

(please, please pm me for sensitive questions, but most ill answer here anyway)

- The Anger

I'm too rusty with my C++ to be much help coding, but I can help organize and eventually alpha test. Working on my PMP, so this should be good practice.

I have no coding experience, but I would like to playtest your new game.
If you could, would you mind posting a link to download anything needed?

Nightwing wrote:If you could, would you mind posting a link to download anything needed?

if possible there will be no additional things to download besides the program to test. however, i will aim to do this via SVN (offsite version-controlled code repository), so at least you will want to become familiar with how that software works as well as perhaps installing one for yourself. you wont have to fiddle with download links or anything like that after setting up SVN.

one ive used before and use at work is Tortoise SVN (google it, download link depends on OS), but see for yourself which one you like.

re organization - in 2 weeks time we will have some figures to work with, so lets schedule in a live-chat session amongst those planning / organizing this to decide what / where / when / how to proceed, using what, and so on... probably over vent. anything discussed will go in this thread and progress from there.

timezones and work / commitments will make things difficult (i have work 9-5 weekdays for one), but im happy to organize a time any time outside of that (even in the middle of the night) so i figure i have the most flexibility in that regard... so of those who want to help manage this thing, anyone care to suggest a suitable time?

I'd be willing to help however I can. I'm not much good for programing or creating models, but I can definitely play test and help balance.

I am possibly in.
my c++ is rusty too but i will help where i can especially game concept
& modelling in Blender (DAE Export)

I've been thinking along the same lines. MB is in many ways quite a simple game both in terms of the assets and the code. I may have misunderstood your intent here though. Are you looking to create the bulk of a new engine from scratch?
Have you considered existing solutions? Perhaps you ommitted mention of them so as not to confuse the issue, but assuming there is a suitable off the shelf solution available, you'd be saving a lot of time by using it.

(I took a cursory look at various components like ogre (graphics) , bullet (physics) OpenAL Soft (audio) and also more comprehensive solutions like Panda3d or unity. Panda already features python bindings and support for bullet and OpenAL among others).

i have considered some all-in-one solutions, but so far i have found them too constricting in design for what i have in mind.

panda3d looks like it might be able to though... but then, i hate using py (i like lower level languages, easier to make them behave how you want), and the main reason i dont want to use a package like that is for optimization reasons too. it's a personal taste and coding ideology; high level language for logic, low level language for hardware / processing power. things like panda3d are the other extreme, in that it's over-structured... kind of why i stayed clear of the unreal engine though to a lesser extent here...

edit - better put: given the choice between having a good package that locks you into certain design choices, and making your own engine piecewise (from your own versions of gfx / sound / physics or otherwise), i will pick the latter every time because of the choice, flexibility and familiarity.

the-anger wrote:
re organization - in 2 weeks time we will have some figures to work with, so lets schedule in a live-chat session amongst those planning / organizing this to decide what / where / when / how to proceed, using what, and so on... probably over vent. anything discussed will go in this thread and progress from there.

timezones and work / commitments will make things difficult (i have work 9-5 weekdays for one), but im happy to organize a time any time outside of that (even in the middle of the night) so i figure i have the most flexibility in that regard... so of those who want to help manage this thing, anyone care to suggest a suitable time?

Timezones will indeed be difficult. I am based near Toronto, Canada, though Anger and I often cross paths around 6:30-7:30 EST (11:30-12:30 GMT).
I'll talk to Snubby and get his blessing to use his Vent server, or we can come up with another chat solution.

my knoledge of programming and coding is next to none but at least i can give you my time for alpha testing

Realistic physics would be nice. When I started playing Moon Breakers, the limited speed was very annoying to me. The bullets don't inherit the speed of the ship, which was also annoying.
It would also be nice to have different environments, some with gravity and some with atmosphere.
Can you give us a more detailed description of your game ?

the-anger wrote:i have considered some all-in-one solutions, but so far i have found them too constricting in design for what i have in mind.

panda3d looks like it might be able to though... but then, i hate using py (i like lower level languages, easier to make them behave how you want), and the main reason i dont want to use a package like that is for optimization reasons too. it's a personal taste and coding ideology; high level language for logic, low level language for hardware / processing power. things like panda3d are the other extreme, in that it's over-structured... kind of why i stayed clear of the unreal engine though to a lesser extent here...

edit - better put: given the choice between having a good package that locks you into certain design choices, and making your own engine piecewise (from your own versions of gfx / sound / physics or otherwise), i will pick the latter every time because of the choice, flexibility and familiarity.

moral of that bunch of nonsense - dont post when youre almost dead asleep...

ill look at panda3d some more, i only skimmed their api in 10 mins... it looks like its barebones enough to pull off the object structure im after... ill come back on that one. ditto for unity (a lot of these i judged by their api specs and feature lists instead of playing with them, time constraints)... if there are others about as barebones that use Lua, ill jump on it, but i suppose im not below using py if i must

Abstractness wrote:Realistic physics would be nice. When I started playing Moon Breakers, the limited speed was very annoying to me. The bullets don't inherit the speed of the ship, which was also annoying.
It would also be nice to have different environments, some with gravity and some with atmosphere.
Can you give us a more detailed description of your game ?

there isn't too much to say about it yet (its a bunch of ideas and a big list of no no's in my mind, being put into something practical for the first time) but here is what is cooking at the moment...

physics...
it will respect the laws of motion, for one... ive seen a lot of good things about bullet (and a lot of all-in-one's either supporting it or packaging it) so likely this is what will be used unless something better / more suitable comes along.

ships will still be constrained by speed in some sense (so the gameplay is practical) but generally speaking the flight physics will be modeled on MB for playability, keeping in mind zero-g physics. there will be little stopping you from accelerating to stupid speeds, and i dont want to inhibit that as a rule, rather, im considering mechanics to prevent people from flying away from the action too far. nothing decided yet, currently considering something like an on-board auto-pilot that regulates ship rotation and velocity which may or may not be possible to disable. mostly im holding off on that until the scale of the gameplay is more defined (MB-sized? around a planet? around the entire solar system every game? etc).

scale is up to you guys, whatever you feel would work - is something MB-scale appropriate for a space shooter, or do you want to fly about the entire solar system for ctf? i can conceive of either and i dont want to decide that on my own lol... im trying to avoid an MMO, but intend on fairly large teams (teams of 64 should be a good goal to start with) if that helps...

gameplay will be very flexible - game modes will have the power to introduce / ban item classes / specific items, as well as setting the scoring and winning conditions.

ship customization was the centerpiece of this idea - ive taken inspiration from MW series as a base, with somethings made simpler and other things expanded.
ships will be assembled by chassis + components attaching to the chassis (wings / fins / external modules).
the chassis / wings / fins can fit internal modules up to a weight limit (and TBD a 'space' criterion - MW has crits, working on something that has a physical meaning but not overkill to consider).
modules will introduce additional stats / behavior to the ship.
modules can have sub-modules to no defined limit - sub-modules augment the behavior of their ancestors.

anything independently tracked in the game (in unreal an 'actor', i just call them entities) will start with the bare minimum of properties with their entire behavior determined by modules added to it. only ships (and possibly smart projectiles) will have a weight / space constraint governed by a module that enables placement onto a ship by a player in the first place.

this module system is what i absolutely must have lol which is why im so picky about the engine... i have not seen a meta-programming technique like the one ive thought of for this purpose before, i call it 'attribution' at the moment; it can be implemented within existing frameworks but it will be a chore to do so and the result a hideous mess of unmaintainable code resembling a house of cards. im not even going to try doing it in c++ or any strict language because it will have to make use of classes (or re-implement it yourself) and using those class systems will look worse than the windows SDK in compiler abuse...

I like the ideas on large-scale maps and ship customization...
If you'd like, I could come up with ideas for base chassis types (Bomber, L/M/H fighters), shapes (Square, round, triangular, stick), components (Wings, engines), and subcomponents (Upgrades).
I wanna help!

Nightwing wrote:I like the ideas on large-scale maps and ship customization...
If you'd like, I could come up with ideas for base chassis types (Bomber, L/M/H fighters), shapes (Square, round, triangular, stick), components (Wings, engines), and subcomponents (Upgrades).
I wanna help!

sorry lol, i should have added the following... (at work atm, helpdesk stuff, hard to keep personal things in mind lol)

there are currently 3 ship classes with no plans for more yet - L/M/H.
within each class there will be a sub-class, the default being fighters. other sub-classes could include bombers / snipers / flag running ships / etc... specialties / tactical and the like... often will be mission specific. edit: furthermore, ship models should be re-usable into tactical re-fittings, eg, an L10 fighter could be re-purposed to be a light bomber L10-B. (i cringe when i see same model numbers representing different models)

wings / fins will be generic but can request specific compatibility with certain classes / subclasses / chassis.
wings can mount fins.

chassis shape is undecided, depends on tech which depends on back story / setting which is currently being worked on.
modules are TBD for the same reason, once there is a framework or some pattern / flavor going then by all means feel free to suggest / design components / modules of your own

engine will be fitted, so will be a module. engine will be an external module fitting on engine-only external ports (why stop at one engine / thruster?). engines can therefore be fitted onto some wings / fins too.
ships will have a reactor separate to an engine for other fitting reasons and additional fitting criteria directions to explore.

but yeah... try to hold off design until the setting / tech level is more defined (soon from the looks of things), though im not stopping you from being imaginative in the meanwhile - might even help shape the tech / setting, but no guarantees it will be used this early on...

Although it might make the game a bit more complex, you might want to add some visual distinction between the team/faction ships. Rather than the MB Red-pirate Blue-gov, you could have different species. I'm thinking of ship design similar to that in endless space (if you've not heard of it, you should look up some screenshots). I imagine you would have one standard-ish terran species with really utilitarian, boxy ships, and either a space-elf style species with very wavy streamlined ships, or an insectlike species with very skeletal, lashed together looking ships. Sheesh, this really makes me wish I could draw with any proficiency.

Vonstapler wrote:Although it might make the game a bit more complex, you might want to add some visual distinction between the team/faction ships. Rather than the MB Red-pirate Blue-gov, you could have different species. I'm thinking of ship design similar to that in endless space (if you've not heard of it, you should look up some screenshots). I imagine you would have one standard-ish terran species with really utilitarian, boxy ships, and either a space-elf style species with very wavy streamlined ships, or an insectlike species with very skeletal, lashed together looking ships. Sheesh, this really makes me wish I could draw with any proficiency.

already in consideration, still working out the species / tech-based-factions and other divisions. they will have their own designs, but as before, anything visually design related should best wait until the back story / setting is at a stage to give some direction on that.

That seems entirely reasonable!

I really love the idea of larger scale maps! and have it realistic to this solar system... but there would have to be some sort of reason to not land on a planet, like not having areodynamic wings for instance, i.e. said ships cannot hold themselves up to the gravity of a planet. this could also make it possible for faster travel using the planets gravity as a sling-shot for ctf or something like that

well important points here but
what about the engine and the network programming ?
This will be a difficult point - server management too.
Fin X has attributes y are just some really minor changes ingame
and different looks just a matter of design.
I would feel better discussing this stuff after there is a possibility to fly around in the engine on alpha cause the rest can be changed afterwards anyway... and one big thing that sux in MB are the crashes !
Lets start small until stuff is running fast and correctly... totally fine with one faction design for starting ... long way to go .....

Thuufir wrote:well important points here but
what about the engine and the network programming ?
This will be a difficult point - server management too.
Fin X has attributes y are just some really minor changes ingame
and different looks just a matter of design.
I would feel better discussing this stuff after there is a possibility to fly around in the engine on alpha cause the rest can be changed afterwards anyway... and one big thing that sux in MB are the crashes !
Lets start small until stuff is running fast and correctly... totally fine with one faction design for starting ... long way to go .....

nail on the head lol

the framework to make the small changes actually small changes has to be created...
engine choice and network protocol affect this too... (the module method ties in with the network protocol / state sync quite nicely by design, but needs to be implemented first, yes)

its why i havent bothered with details like what ships will be available / actual gameplay and so on until way way into alpha / pre-alpha except where the engine / framework depend on it (ie, backstory)

I wouldn't be able to help with any of the technical stuff at all, I'm afraid. I only wish that I could offer some sort of help with this project. The only possible thing that I could do is come up with suggestions for gameplay...probably more of an annoyance than any sort of help
I was thinking that as this game is labelled an alternative WW2 flying game, the the ships and their relative weapons load outs could be more closely based on WW2 aircraft. Disposing of guided missiles altogether and concentrating on machine guns,cannon and rocket weaponry. I think this would certainly increase the skill factor ingame.This might also go a long way to keeping new players interested. A bit more sustained play and not having to respawn every 10 seconds because of guided missiles, simply must be more appealing.

Drummerman wrote:I really love the idea of larger scale maps! and have it realistic to this solar system... but there would have to be some sort of reason to not land on a planet, like not having areodynamic wings for instance, i.e. said ships cannot hold themselves up to the gravity of a planet. this could also make it possible for faster travel using the planets gravity as a sling-shot for ctf or something like that

we all love our big maps now and then... the problem i have with a solar system-sized arena for MB-scale + 1-2 orders of magnitude in speeds is that its too damn big lol... people forget just how damn f***** huge space is, even with 100,000 players the odds of running into someone outside of planetary lanes (as the crow flies planet-planet + grav sling if you must) is virtually nonexistant.

i like and hate how eve solved this - only allow FTL to map-significant places (planets, the sun, asteroids, structures / stations) and gimp the hell out of sub-light speeds. to its credit though, the 3 travel modes are realistic in relative scale - sub-light averages in 100's of m/s to the extremes of (now) <10km/s (formerly <20km/s on average and <50km/s for the insane with 300km/s ive seen done on the test server back then); the FTL warping is on the order of 0.5 - 15 AU/s (at this speed it still takes on the scale of hours to travel between stars, if one could); jumping is instant but limited by distance.

dont get me wrong though, i would love to wage interplanetary war & having to warp from location to location to get to the battlefield, but that kind of scale i feel is best done as an MMO where you can take your time getting around... for shorter games i want to find something to blow up, not spending the entire time chasing after people for laps around the sun... my thoughts on it anyway... if anyone can think of a way to make it work, by all means lets hear it lol

Ruffjustis wrote:I wouldn't be able to help with any of the technical stuff at all, I'm afraid. I only wish that I could offer some sort of help with this project. The only possible thing that I could do is come up with suggestions for gameplay...probably more of an annoyance than any sort of help
I was thinking that as this game is labelled an alternative WW2 flying game, the the ships and their relative weapons load outs could be more closely based on WW2 aircraft. Disposing of guided missiles altogether and concentrating on machine guns,cannon and rocket weaponry. I think this would certainly increase the skill factor ingame.This might also go a long way to keeping new players interested. A bit more sustained play and not having to respawn every 10 seconds because of guided missiles, simply must be more appealing.

every bit helps...

i dont dislike the WW2 theme, but it is very limiting and imo wears thin in space... the skill factor won't be removed by something like tracking missiles, if it will have them they will be done 'right' (proper space physics will take care of that for 80% of the problems MB has, the rest is a lack of foresight in how they 'track' being totally unrealistic) as well as adequate counter-measures offered (more defense options, possibly have point-defenses on heavier class ships, pretty much anything limited to our collective imagination really).

but then, some people like respawning every 10 seconds, few there may be... otherwise things like the 'excessive' quake 3 mod would never have been created (shotgun fires rails, gauntlet instakills, machinegun fires like a mounted air-craft minigun, rockets & rails spammed at machinegun pace, bfg spawning grenades on impact, etc...)

There's this too: http://polycode.org/ (+lua, +network MIT licence)

You could prototype the game with an off the shelf game engine then see if you needed more. If so, you could replace components with your own as you create them and would have an existing framework to test them with. (until you conclude that a ground up rewrite is necessary anyway) Assets could be created in parallel for the prototype and should easily be transferable to a future system.

Open third party libs (licence permitting) can be thought of as malleable templates anyway since you can modify them to suit your needs as and when the need arises. Well funded commercial libs are likely (but not guaranteed) to be superior in terms of their feature set and quality though. That's the way the cookie crumbles. Games like MB aren't really pushing the envelope from a technical standpoint though.

The thing is, if you anticipate being the primary developer then you need to look long and hard at your own motivations. If you were really excited about rolling your own engine to pull in and glue together anything you wanted to adopt (sound, physics) then without that additional source of enthusiam, your motivation may ultimately evaporate and the project along with it.

If you just want to make "that game" then it's probably not such an issue. At least if you prototype it you still have the possiblity of implementing your own stuff down the line if needs be. Just be careful to do what you want to do and not what other people want you to do.