[headdie]

Yeah, but actually, most of the FPS controls will work with simulators.  WASD (thrust & sliding) + mouse (pitch & yaw & click/right-click for primary/secondary fire) will cover most controls, and you can use Q and E for rolling (banking).  Anything else can be readily mapped to keys around WASD for ease of access.

A controller can do this too... left analog for WASD, right analog for mouse, upper left & right buttons above the triggers for roll, and ofc the trigger & secondary trigger are the primary left & right triggers.

the problem with simulators and the current console orientated generation is not the positioning of controls but the number of controls and the complexity that requires.  your average xbox pad is 12 buttons and 4 axis and there are many threads on here discussing how to try and map the main FS controls to such devices, often failing because you cant do it there are not enough buttons on a joystick or joypad and an xbox pad is about as much as the average person can handle these days because there just is not the desire to sit down and learn that number of buttons, I tried to get my 11 yr old stepson to play FS2 and he couldnt process the basic keys let alone "advanced" stuff like coms menu and switching weapons.

[jr2]

Hmm, well, FS1 & 2 didn't start you off with the advanced controls at first... they gave you 3 training missions to start with, walking you through the basics and intermediate controls.  Then, you get another training mission to learn interceptors, then another for shields... etc.

You could start out with the first mission as training, and spread the rest out as the missions get more complex, to maintain player interest.  You'd have to lure the user in.  Even FPSes do that to a certain extent, no?  People that really want to master them learn the more advanced controls (such as they are).

The problem with FS (and probably most simulators) is it tries to cram most of the training in at the beginning, so if the player isn't hooked already, it's a turn off.

[rhettro]

Without a proven market, the game companies just aren't willing to lay down the cash for a AAA space simulation.  I think space sims will be in the domain of the indy builders for sometime.  That said, I also think that as computer technology advances, the ability for novices to build their own games will increase.  Consider how far FSOpen has driven the number of mods for FreeSpace.  Consider how many more there would be if the tools for creating them were more simple to use.  There will be a point where AI assisted modding will let us create AAA titles of our own design.

[The E]

That said, I also think that as computer technology advances, the ability for novices to build their own games will increase.

Actually, the reverse is true. Creating mods for games that have modability built in is getting harder and harder, requiring more and more diverse skillsets. And that's just the games that get toolsets; the many many more that are locked and encrypted shall not be mentioned.

  Consider how far FSOpen has driven the number of mods for FreeSpace.  Consider how many more there would be if the tools for creating them were more simple to use.  There will be a point where AI assisted modding will let us create AAA titles of our own design.

Uhhmmm, no. Not really. Procedurally generated content is a pipedream at the moment; while there are parts of the asset creation process that can be automated, in the end a human needs to turn it into a game.

Oh, and if it works, whose work is it? The Modders', or the people who created the algorithm?

[rhettro]

Actually, the reverse is true. Creating mods for games that have modability built in is getting harder and harder.

At the moment that is true, more complex interactions require more complex input by the designer. But computer AI will advance to a point were the designer can tell the computer the general requirements of his game and the computer could supply the framework. It's a ways off, but definately in the realm of possibility.

Uhhmmm, no. Not really. Procedurally generated content is a pipedream at the moment; while there are parts of the asset creation process that can be automated, in the end a human needs to turn it into a game.

We don't have AI assisted programming at the moment.  There's no reason to think we won't in the future. There will be a human required to design the game, but that human's knowledge of everything going on in the background will be quite less.

Oh, and if it works, whose work is it? The Modders', or the people who created the algorithm?

Seems argumentative.  If Windows was written in C, do the programmers of C own Windows? They have a hand in it's success, but they do not own the content.

[The E]

At the moment that is true, more complex interactions require more complex input by the designer. But computer AI will advance to a point were the designer can tell the computer the general requirements of his game and the computer could supply the framework. It's a ways off, but definately in the realm of possibility.

I'm sorry, but that is still wishful thinking. "Computer AI will advance..."? No, it probably won't. First of all, what you are describing isn't AI, it's just an incredibly fancy algorithm that produces "fun". Which is really really hard to write, so hard in fact that it doesn't make sense to do it when you can just get a few creative people to do the job instead.

We don't have AI assisted programming at the moment.  There's no reason to think we won't in the future. There will be a human required to design the game, but that human's knowledge of everything going on in the background will be quite less.

I am assuming that you are not actually a programmer, or have researched the matter of what AI actually means, or what the prerequisites for creating something AI-like are. Also, you are remarkably optimistic about the willingness of the developers of said formula to actually make it public. Again, you are proposing a machine that basically takes a few random input parameters and creates a fun experience; as any game designer will tell you, that's a rather involved, highly iterative process that takes months and years of refinement until you get to a product that you can show the public.

In addition, game design is a discipline not unlike novel writing. Like a novelist, a game designer has to manipulate the emotional state of his audience if he wants his work to be remembered; Given how many do's and don'ts, written and unwritten rules there are that cover game design theory, and the knowledge of when to throw said rules out the window, the chances of someone being able to codify that into an algorithm are slim to none.

Seems argumentative.  If Windows was written in C, do the programmers of C own Windows? They have a hand in it's success, but they do not own the content.

You misunderstand the issue. Let me give you an example. Assume I have a program that, given a set of parameters specified by the user, creates a game experience. If I use that to create a game, who has done more work on it? Me, who just came up with some arbitrary inputs for that program, or the designers of that program? In this case, it would have to be the latter party. After all, the finished product, while superficially my creation, would ultimately be the creation of the algorithm.

[Sushi]

That control system is horrific for simulators.

Can you name a good space flight sim that used it?

That's actually exactly the control scheme I use for FS2, and what I used for Allegiance when I was playing it. I've never really done multi FS2, but (if I do say so myself) I was a pretty competent dogfighter in Allegiance. I wouldn't call it horrific.

[Dragon]

Actually, the reverse is true. Creating mods for games that have modability built in is getting harder and harder, requiring more and more diverse skillsets. And that's just the games that get toolsets; the many many more that are locked and encrypted shall not be mentioned.

Actually, I think this has more to do with game complexity than being adapted for modding. FS2 FRED did well with a heavily stripped down LISP scripting, ArmA II has a full HTML built into it's mission editor. LUA is practical for animations, so all animations are being defined in LUA, since this allows for making them very complex. Retail FS2 didn't had an animation system at all (other than simple rotation and purely cosmetic turret traverse). AI is now much more advanced, so modifying it is much more complicated than setting a few values. Models and textures get more detailed and more complicated with time, so an FS2 Retail quality model (which a lot of people could make) doesn't have place in modern games. I've looked into modding games like HW2 or ArmA, and I found that the most complicated part (aside from modeling/texturing, my humble skills are nothing for these games) is learning the script system they use (and they use it for nearly everything).

[rhettro]

I'm sorry, but that is still wishful thinking. "Computer AI will advance..."? No, it probably won't.

Here's a good artical on the subject. http://www.transhumanist.com/volume1/moravec.htm
Basically, it is predicted that during the 2020's, cheap laptops will match the processing power of the human brain. It's not me just being optimistic, there is a whole course of study that predicts this likely outcome.

I am assuming that you are not actually a programmer, or have researched the matter of what AI actually means, or what the prerequisites for creating something AI-like are. .

No, I'm not a programmer by trade, but I have written computer programs.

Also, you are remarkably optimistic about the willingness of the developers of said formula to actually make it public. Again, you are proposing a machine that basically takes a few random input parameters and creates a fun experience; as any game designer will tell you, that's a rather involved, highly iterative process that takes months and years of refinement until you get to a product that you can show the public.

Look at it this way, the programming required for simulation is already fleshed out in terms of physics, computer graphics etc.  Game engines like FreeSpace will be like downloading Flash games by the 2020's.  An AI that can pull dimentional data, weight, inertial loads, moments etc. out of the internet and host in the framework of a game engine is simply a matter of scale of processing power and the robustness of the data available via the internet.  The AI function helps govern the intent of the designer.

In addition, game design is a discipline not unlike novel writing.

I think that is an excellent example, but the novelist doesn't need to know how to build a typewritter in order to write a good novel.

You misunderstand the issue. Let me give you an example. Assume I have a program that, given a set of parameters specified by the user, creates a game experience. If I use that to create a game, who has done more work on it? Me, who just came up with some arbitrary inputs for that program, or the designers of that program? In this case, it would have to be the latter party. After all, the finished product, while superficially my creation, would ultimately be the creation of the algorithm.

I don't really see it being an issue.  I'm sure that the amount of effort that went into programming Microsoft Word would be enormous put against the poet that wrote his single page poem using it.

What I am proposing is an AI assisted program like http://www.yoyogames.com/gamemaker.

[Mongoose]

That control system is horrific for simulators.

Can you name a good space flight sim that used it?

That's actually exactly the control scheme I use for FS2, and what I used for Allegiance when I was playing it. I've never really done multi FS2, but (if I do say so myself) I was a pretty competent dogfighter in Allegiance. I wouldn't call it horrific.

I knew someone in college who had tracked down FS and played with a PS1 controller  hooked up to his PC.  I tried it out once for the hell of it and managed something workable with the left stick for pitch/yaw and the right for roll/throttle.  Sure, you can't cram every single obscure targeting command or comm menu option onto it, but there's a lot that can be done there in terms of shift options or condensing redundancy, and the comm menu could be handled via voice recognition, just as we currently can with voice recognition builds.

[jr2]

As far as WASD + mouse is concerned, a big plus is most FPS'ers can just switch over... not much more training necessary.

[TwentyPercentCooler]

We might get to the point where indie studios can produce better quality games if developers would ditch their obsession with having their own goddamn game engine for every release. The Source engine is a great exposition of how an engine can have a massive and useful lifetime if the initial developers make it flexible, easy to work with, and provide support. Cryengine 3 could probably see a lot more use than it currently does; it's nicely scalable and who knows what kind of capabilities it has beyond the FPS genre. Space sim in Cryengine 3 would be like an eyegasm if it was possible.

[yuezhi]

how uncivilized. Knife ftw!

[deathfun]

Okay, two things that I will address while motioning through this thread
Firstly, the colour blue for font is a terrible colour to use for font
Secondly, I disagree that the controls wouldn't work for consoles. Furthermore, having the option to bind buttons would further aid this ability.

I'll start off with a PS3 controller since, it's right in front of me.
Movement is relatively simple. You've got a single joystick that gives you directional. The other joystick can give roll. Countermeasures? Circle. Firing? R1. Secondary? R2. Afterburners can be binded to L1,  and targeting to square (the default being what's in your sights - this makes it easier to target bombs and gets rid of having to hit the corresponding command thus removing an additional buttom). Triangle can cover subsystems. If autotargeting is true, that essentially covers closest target. Double clicking square can go to closest attacking (I never really used it anyway). L2 can bring up the command prompt, and the up down directional cover, well, up and down, while X covers selection. Then we have firerate. Left and right d pad controls cover primary and secondary respectively.

If you have a headset, they could adopt a voice command system like they had in Endwar. This provides you with two additional buttons.

So tell me what I'm missing (mostly since I probably am missing something). These cover basically all that I've ever needed. Remember, you can always bind what to what later if need be. Not only that, but if they have joystick support, even better

Furthermore, I am also missing three buttons. Select, R3 and L3. They were never mentioned and can be used for whatever purposes. Additionally, you can hook up a USB keyboard provided it's compatible with the system, thus giving even more buttons.

New subject: Mod support. Console games haven't really been known for mods, but that isn't to say it isn't possible. If you have a computer, you can design stuff and submit it to a community. This community will then be available to console players via a menu within the game. This provides installable data giving you more to your game. FRED? Simple. There's been many examples of editors on console ranging from Halo, to Farcry 2. One would just have to give the same menu options and the likes, but I suppose it'd have to be either a simpler version, or the L1 brings up every possible output window. Or X double click, whatever. Hell, it could be provided on the community site and you can just do it from your computer like normal people.

Discuss

[QuantumDelta]

Lets see... what I use regularly;

On the keyboard;

Up, down, left, right, roll left, roll right, brake, countermeasure, afterburner, messaging (11 buttons at least) (no, voice comms is NOT quick enough for intense FS missions, though admittedly, intense FS missions where you need to micro AI a lot are more or less restricted to multi and a few of the better fredded mods), subsystem target, target turret, target unscanned, target new target, target subsystem in reticle, target targets target, target target attacking target, target ship on escort list, set escort target, target in reticle, target bomb/bomber, target friendly, target hostile, max top speed, all stop, equalise shields, shield management for each quadrant (4 buttons) (if you don't do it you're ****), energy management (7 buttons) (if you don't do it you're **** ), special energy management (default =scroll lock / shift scroll-lock), jump engines, change primary weapon bank, change secondary weapon bank, dual fire missile mode, hot keys(F5-12) (I must get back into the habit of using these more).

So uhh, yea. That's...... 60?
And I'm not using some that other people use all the time.
Including setting auto-targeting, match speed, and auto-match speed, different throttle settings too. Radar range (occasionally useful).

So uhm.
Yea............ No.

[The E]

I've recently tried to create a mapping for my gamepad for FS2. Despite it being a carbon copy of the Dualshock 2 layout, I couldn't get it to a level where I felt comfortable with it, and had to go back to KB+Mouse, because as QD said, there are tons of controls that are used during gameplay, and taking my hand off the pad to work them just slowed me down a lot.

[Scotty]

Lets see... what I use regularly;

On the keyboard;

Up, down, left, right, roll left, roll right, brake, countermeasure, afterburner, messaging (11 buttons at least) (no, voice comms is NOT quick enough for intense FS missions, though admittedly, intense FS missions where you need to micro AI a lot are more or less restricted to multi and a few of the better fredded mods), subsystem target, target turret, target unscanned, target new target, target subsystem in reticle, target targets target, target target attacking target, target ship on escort list, set escort target, target in reticle, target bomb/bomber, target friendly, target hostile, max top speed, all stop, equalise shields, shield management for each quadrant (4 buttons) (if you don't do it you're ****), energy management (7 buttons) (if you don't do it you're **** ), special energy management (default =scroll lock / shift scroll-lock), jump engines, change primary weapon bank, change secondary weapon bank, dual fire missile mode, hot keys(F5-12) (I must get back into the habit of using these more).

So uhh, yea. That's...... 60?
And I'm not using some that other people use all the time.
Including setting auto-targeting, match speed, and auto-match speed, different throttle settings too. Radar range (occasionally useful).

So uhm.
Yea............ No.

*Sigh*

We've gone over this before.  You can condense a fair portion of this, and easily.  I'm also sorry to say that you are, unfortunately or otherwise, not the average gamer.

Also,

Sure, you can't cram every single obscure targeting command or comm menu option onto it, but there's a lot that can be done there in terms of shift options or condensing redundancy, and the comm menu could be handled via voice recognition, just as we currently can with voice recognition builds.

[QuantumDelta]

But not enough to fit on a controller.
Even given a dozen concessions.

[FlamingCobra]

You guys see how Bungie snuck some Space Combat Sim into Halo Reach? That's how it's done. You sneak stuff in from other game genres and you gradually expose the public to that genre. Once they become comfortable with it, then the genre becomes profitable.

Well, that's my go at it anyway. I could just be talking out of my ass.

[Cyborg17]

If you used an xbox controller, you might be able to get something workable. (I don't actually own an Xbox, btw.)

Turn axes: Left stick
Countermeasure: Left stick press down
Roll: Right stick left-right
Throttle: Right stick up down.  Permanent increases like the "+" and "-" keys.  For full throttle, keep it pushed forward until you reach it.
Afterburner: Right stick press down.

Autotargeting enemy: Automatic. No key binding.  Selected on or off from start menu (usually if you have these on, you want to keep them on.  I've never had to use auto-match speed, but you could treat it the same way.)

Primary and secondary fire: Triggers
Weapons mode adjust: bumpers (cycle through secondaries once to trigger double fire mode, cycle through them again to trigger single fire, again)

Comms: adjust the comms menu to be usable by the D-pad. We're pretty used to the system we have, but you could have

1. Orders (Left)
2. Rearm (Up, also abort rearm)
3. Reinforcements (Right)
4. Cancel (Down)

1,3 go to a list of wings, where down would let you look at the next group of them.  Giving attack orders to other ships usually just messes them up anyway.
It would look something like:
1. All fighters
2. Alpha
3. Beta
4. Gamma (or more options)

For orders, the list after a wings list would be
1. Attack/Defend target (based on friendly/hostile, of course)
2. Defend me
3. Ignore Target
4. More options

1. Engage Enemy
2. Depart
3. Form on my Wing
4. Cancel

Attack target would bring up just one more submenu:

1. Attack
2. Destroy Targeted Subsytem
3. Disarm
4. Cancel

Hmm.... 5 Buttons left.....

Target Nearest Hostile (including bombs): A
Target Friendly/neutral: B
Target Subsystem/turet: X (Hold to switch between modes?)
Change Power settings: Y  (Cylces through the power settings with each press, Single quick press puts selected power setting at 2/3 of pool, held maxes power setting, tapping twice quickly resets the system)

Equalize Shields: Select


That's the best I can come up with, and it probably wouldn't work on Insane.  If it had a way to reinforce shields, maybe we could get somewhere.