Problem with super melee!

No, they just mean you won't have AI wingmen, which is a good thing.  Because, exactly like they say, it just doesn't work.  An AI playing Chess is like a 5-year-old playing Candyland compared to an AI playing Star Fleet Battles.  It simply can't be done, not because it can't be done, but because nobody has an interest in doing it.  Chess was a science contest kind of thing or you would't have the AI that exists for chess today.  But only the US Navy might want a "chess quality AI" for what is essentially Star Fleet Battles (or 2D ACM without gravity).  And they have no use for it because the US Navy has no interest in AI, they play their simulators against each other.  I actually know someone who helped make that for them... us SFB Staff guy's got all the connections;-)

I'll even tell you the best way of doing wingmen in Star Control... since it isn't exactly giving away any kind of big secret or anything.  Lock the wingmen to the player, targeting what he targets... cheating and "sideslipping" a little when needed (or allowing the gun to swivel in the front arc... or both).  The wingmen fire when the player does.  Have them artificially "bob and weave" a little so it looks more realistic.  Tada!

Like I said, not much.  Not even good enough to even want to do, actually.  Maybe someone else might think of something better, along these kinds of lines and not any kind of actual "AI"... but this is what you are limited too because any kind of AI at all simply isn't possible.  The ships will just be a confused mess of rapidly changing "AI profiles" based on the rapidly changing conditions that will, in the end, just shake, jitter, and wiggle until it is dead.  So some simple trick of actually turning them all into one ship, the player and the wingman, is the only way of doing it without it just being a complete mess.

Oh, and my way, but only I can do that... and nobody believes that I can, haha!  I didn't mean to wind up here, but since I did, I actually can do this with Rube.  I can make a ship fight you very well... at least convince you that is what is happening anyway... but it is very limited and couldn't work in a top down game like Star Control where you have situational awareness and know what is ACTUALLY happening around you. <evil grin> 

Hmm. Thank you for clearing my fears! Are you completely sure that the ability to play as another races ship has been preserved? It was the overall tone of the entire pdf section on super melee that scared me. And the fact that this ability has been neither confirmed nor denied. I do, however, deeply wish for you to be right.  

Yes, players can control other race's ships both in the single player campaign when they become part of your fleet and multiplayer super melee. 

Thank you, Master. *cringe*

That's the way game dev works, you know. It's a set of developers who establish rules for the game that players play. So yeah, they're ALLOWING us to do that

I like board games and computer games equally.  That is one of the things about board games that are better than computer games.  Any time you don't like the way the designer made a board game, you can just make a "house rule" to play it however you want.  A lot of games develop "famous" house rules.  Like pretty much everyone knew the house rule for US Lasers in Fortress America.  

I don't like house rules in many cases, because in many cases it's just an excuse for someone who's bad at the game to ignore a rule that impedes their progress.

Doesn't "house rules" mean that the game isn't balanced well enough?

In the case of the US Lasers in Fortress America is was exactly because the rule that came with the game was completely broken.  So everyone knew the house rule to fix it, because then it was better than Axis & Allies:-)

In other cases it can be a crutch to learn something complicated.  For example, almost nobody (except the Air Force and Navy guy's, of course) learns to play SFB with the optional electronic warfare rules.  And many people who learn them once they have the basic game down decently use a house rule to make EW cost 1:1 energy instead of the square root system where you generally need 1 or 4 energy advantage for a 1 or 2 shift.  The biggest thing about using EW is that energy is very limited and once you are accustomed to the balance without EW, EW seems very expensive too you.  Playing 1:1 at first makes it easier to understand at first, and quickly teaches you why a +2 shift should be so costly, haha.

This is actually a great example of just one of the reasons why board game players are more knowledgeable when it comes to game design.  We have much more actual, practical experience experimenting with game/simulation rules because we have so much freedom to do it.  It's a part of the whole idea that to even play a table top simulation you have to actually understand everything about how it works, and then make it all function yourself.  Someone who only knows games from observing them as they control them, only knowing the interface and what they THINK they understand about about how it works, is really just "Voodoo Knowledge".  They are probably wrong about half of what they think about how it works, and how they think it works might actually be better than how it actually does work.  But there is a really big difference between table top simulations and computer simulations when it comes to learning how they function and how to actually make them.  And that Voodoo Knowledge leads to a whole lot of wasted time arguing about things that would be obvious had you actually done them before, and weren't just imagining how it had worked in whatever game you are remembering.

Okay, if you could just stop assuming that we don't have any knowledge of basic gaming outside of video games, that would be great.

Do YOU know the house rules for US Lasers in Fortress America?

I don't. I wonder what they are.

It was so long ago I don't remember exactly what it was, just that it made them fire less often.  That was the original release version, the version of Fortress America they make today would certainly have that problem fixed.  It was unplayable by the original rule, the US couldn't possibly lose.  As soon as the lasers came online the game was over.

Naw, the US was entirely capable of losing. You just had to coordinate combined arms attacks, and in particular, using leap-frog tactis with the Western forces' MASSIVE arsenal of helicopters. Like, that's more or less the core mechanic of the game: the invaders have to coordinate, or the US wins in a war of attrition. And a few well-placed blood sacrifices to RNGesus.

Personally I want the multi-ship combat. Both in campaign and super-melee. But I'm okay with it not being available on launch. Expanding the game to multi-ship is exactly the kind of major game-changer that I'd want in an expansion. And it would completely justify replaying the main game again.

That said... there are additional possibilities that could be considered now already:

• (human) super melee could already be 3v3 - no AI to worry about
• the campaign could now already have 1v3 - one human against three enemies. Not fair, but that is fine too
• similar to WarCraft 3 the game can release with a way for custom AI to be written; there will almost definitely be intrepid modders that will craft a sufficiently compelling AI

I disagree with Kavik_Kang that the AI problem is intrinsically not solvable. After all, we do not need a military grade AI. We only need a compelling AI. When StarDock says that they cannot do it - I'm fairly sure the underlying meaning is: "we cannot do it now, it requires too much time to get right and we need that time for other systems for the game". John Carmack (id Software, Quake, Doom, etc) more than a decade ago explained that it isn't hard to write perfect AI for his games - the challenge for him is to write imperfect, human-like bots.

But... have you paid attention to the history of AI at StarDock? It is something they take quite a bit of pride in. And also something that, in the long term, they spend a big amount of time on.

I'm fairly happy to wait for the third expansion if necessary.

That is the problem, there is no "compelling AI" for it, that is why you have never seen it before.  It is too complex, free, and fast moving and almost nobody outside of real world air force's has the knowledge to even begin to think about how to do it.  I can do something pretty good, after a lifetime of contemplating it, through a concept from SFB that we call "Option Points".  I could do something "OK" with my various Option Points for doing this in a "100% situational awareness" game like Star Control.  Something far beyond anything anyone has ever done before for this type of game, but it is still not "impressive", just pretty good.  In 3D where I can take advantage of the fact that you are blind to reality I can trick you into believing that you are sitting in the captain's chair and fighting a wizard of an enemy captain.  For two or three games, that is, until you become acclimated enough to it to start sensing certain patterns.

It is an exceptionally difficult thing to do, which is why you never see anyone even try.  With a lifetime of serious dedication to this particular subject, I can barely pull off a few cute tricks to fake it, and that's about it.  Making an AI defeat the best chess player in the world is child's play compared to making an AI for this type of combat that is anything other than laughably bad.

I don't think it's that they can't do it, it's that they would rather focus on making the core fun before adding that to it, and I'm confident that they'll find a happy middle ground in terms of AI difficulty. 

No, they can't do it.  That's why nobody ever does, and games that try just wind up with confused wiggling ships.  If it could be done you would have played lots of games with "smart space ships" flying around them.  But you haven't, because it really does make an AI that can beat a world champion chess player look like Candyland.  I know a lot more about this specific subject than, quite possibly, anyone.  I've just wound up having a lifetime of experience with this obscure, mostly completely useless subject, haha.  "AI" is not as sophisticated as people think it is when they hear that term.  It is mostly very simple rules used in clever ways, and there is no way to do that with this type of "naval combat".  I know exactly how to do something decent, and it really does take a lot of background knowledge to even think in the right terms about it... and almost nobody has a reason to know so much about it, so almost nobody can even begin to imagine how to do it to begin with.  I am not trying to brag or anything, it's just that an "AI" to make a ship fight you has got to be one of the most difficult things that you could attempt to create an AI to do.  

"No they can't do it." We're going to have to agree to disagree. Maybe YOU can't do it, but someone can. Stop that.

They can't do it.  That's why they said they aren't going to do it.  They could do it, really really badly.  But they have made a wise choice not to try.

With this particular thing, if I can't do it then it is unlikely that anyone can.  I really do have a tremendous amount of experience with this particular thing.  It is very unlikely that there is anyone else who understands this as well as I do.  My situation is quite unique.  You've never seen it in a game before... because they can't do it.  If they could do it, they would have a very long time ago.  It isn't just a little bit beyond their abilities, it is in a different universe.  Or to put in in the words of my own universe... "they are in the wrong place to understand."

I can do something workable in top down, but it is a magic trick.  An actual "AI" as you are thinking of it is not possible short of a two-decade-long "science competition" like happened with chess.  And since chess is brain dead simple compared too this, you should probably expect it to take a lot longer than two decades.

No, they didn't say they aren't going to do it or that they can't. They said they aren't going to do it in the initial release. That word choice is very important ant deliberate, because it doesn't completely rule out a later release with a patch, dlc, or an expansion. If you can really do it, put your money where your mouth is and show us a demo you've made that uses it.

They can have AI wingmen, but they will work very badly.  I would do exactly what they are doing, only use what you can make work well.  You can't make AI wingmen work well, so it is best not to try and have them wiggling and doing nothing almost all of the time.  I have been focused on this very subject for 40-years and am pretty well known among my generation for my knowledge of it.  It is a far more complex subject than most imagine, and the computer game industry has so far taken 0 steps towards it... meaning they are still decades away from an AI for space ships in any environment.  It's actually EASIEST to do in 3D, where the player has very little situational awareness.

As for making my games, I've been trying to do that my entire life.  The only way a designer gets to make their own games is to fund them yourself, and I've never had millions of dollars to do that.  But this Star Control thing has brought me out of my 8-year retirement and I am making one more try at the Pirate Dawn Universe.  I will finish a prototype this weekend that can be made as a board game and I will publish that as a board game while also using it as a prototype to see if I can find anyone interested in making the PDU.  So, if I can actually manage to do this, which I've never been able to before, then you will get to see what I am talking about.  But I wouldn't count on it, the odds of an actual game designer getting to work in the computer game industry have always been near 0, so I don't actually expect to succeed:-)

Stop speaking in absolutes, you sith.

Look, that's not the case. If you want something made, you don't need millions of dollars, that's bullshit. There's lots of avenues for you to make something, and they're not difficult, they just take a little bit of time. Learn how to develop in Unity, or Lumberyard, or Unreal 4, or the countless other easy to use, free engines out there. Learn C#, learn Javascript. If you have ANY inkling of programming knowledge, it's not hard to figure out where the variables are, what the constructor is, what namespace to use, how these logical indicators work, the basics of ANY function. I'm not calling anyone stupid, in fact I'm saying the opposite. If you want to make games, just do it, stop letting excuses pop up like "I don't have the money." "I don't have the time." "I don't want to make bad games." "I don't want to fail." If you want to make a future making and selling games, there's Indiegogo, Kickstarter, Gofundme. 

JUST. DO. IT. And for god's sakes, do NOT say "I'm above that." or "I'm better than that." Clearly if nothing's been released or even published, no, you're not. I've got a pc demo for a game I made coming out in the next few weeks. WEEKS. It's not a great game, but it's a game.  You have to fail a LOT before you succeed, so just do it. F!#@ all of the bullshit and just MAKE YOUR GAME. 

What's PDU?

https://www.youtube.com/watch?v=5-sfG8BV8wU

EDIT: Pirate Dawn Universe = PDU 

Question to ponder: Can PDU be potentially made out of the heavily modded SCO? 

I think he's referring to the GDD.

It's my 12 game sci-fi universe.  It is 12 game design documents inside of a single timeline, well, "broken time loop".  But what we mean by "game design document" and what the modern gaming world means by that are two entirely different things.  Three of them are playable games, most that aren't are only incomplete because they have been intentionally left that way so they can continue to evolve.

I actually have a very simple, and yet at the same time very complex playing, prototype game that I have been working on for the last month.  It should be finished this weekend.  It is the new first game of the PDU, and I threw out the old "optional game #3" to keep it at 12 games.  I am showing it to board game companies first to try and publish it as a board game first.  But eventually if that doesn't happen I will be using it as a complete game that I am willing to show people, it works well for that.  So in a month or two here I might actually have a playable PDU game I would be willing to make available to anyone that wants to see it.  With the whole beginning of the story and opening timeline.

It actually has a tremendous amount of content for a prototype made in about exactly one month... there are over 20 different "factions" that you can use in it:-)