Update, RE: a clear example.

The most appropriate place for Questions, Queries, and Quandaries regarding the nature of the Vega Strike universe and its past, present, or future history. Home to the occasional unfortunate RetCon.
snow_Cat
Confed Special Operative
Confed Special Operative
Posts: 349
Joined: Thu Jan 05, 2006 12:43 am
Location: /stray/
Contact:

Re: Update, RE: a clear example.

Post by snow_Cat »

Busy with work this week.
I would like someone else to poke more holes into what I've presented in the mean time.
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Re: Update, RE: a clear example.

Post by chuck_starchaser »

safemode wrote:Not that we need to say that the mining base directly attacks the colony ship. Perhaps operations by the mining base produce unforseen debris that becomes unavoidable and takes out the colony ship's communications and power. Afterwards, the mining ships detect the colony ship and with it having no power or communications, sees it as a mineral rich asteroid and begins mining it.
ROFL
That would be a pretty poor AI, but it's not inconceivable.
Basically leading to an increasing connection with the machines around them and a resentment of those who have left them out in space. It my mind, the entire colony would have a deeply routed hatred of the rest of humanity.
It wouldn't be too farfetched to suppose that the mining corporation that invested on this venture wanted to ensure its financial success, and was more afraid of human nature than of machines in this regard. Clark's 2001/2010-type scenario.
The mining ships AI's and Head Of Operations AI have, at the last minute, had their standard human overrides commented-out. Whenever in doubt, the machines regard the human suggestions as suspect. Thus, when one of the mining ships homes onto the colony ship, there's nothing the human operators can do about it.
And so perhaps the human miners and colonists find that there are far more valuable resources here than what the corporation expected to find, but their equipment won't cooperate.
That should piss them off, alright; --being treated like a useless cargo of dumb cows on a free ride... Why send a human crew if they don't trust their judgment at all?
So here the player can take on the challenge of defeating the machines, and play Descent for a couple of weeks, then return to Vegastrike :D
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Re: Update, RE: a clear example.

Post by chuck_starchaser »

snow_Cat wrote:...An ambitions project to begin with, however the technical difficulties involved were staggering.
The challenge wrote: "Wohld thou lahyk freez?"
At the constant acceleration of the mining ship the doppler/redshift effect approached 30%; consider that a satelite on a communications-friendly 5° Molniya orbit around the earth at 6 km/s is enough to shift a 1 GHz carrier ±20 kHz relative to a fixed station planetside. That is out, then in then back out of the human hearing range. This is also the reason that the astronauts' voices are at times indistinguishable from that which takes your order at the drive-thru.

"wŏŏ-d· yhu· līk· frīz?"
At a strait bearing from a fixed base 0.1 c becomes a shift of 111.1¯ MHz, and at 0.2 c 250 MHz. Throw in a variance of ±10 kHz from the planet rotating, another ±100 kHz from its orbit, ±670 kHz from the star floating around the center of the universe, and tack onto that frequency instabilities within {the equipment itself, external interference, reflection and incident harmonics, background noise, refraction and the MIMO effect,} and you have yourself one difficult engineering problem from a signal processing perspective.

"wĭn'dō thrē."
More difficult than predicting and then correcting for the frequency distortion whilst maintaining a frequency/carrier lock is maintaining a physical signal lock. Anyone who has ever tried to get TV-Satelite reception to work with a 'hands on' assistant knows that the minute deflection caused by a heartbeat is enough to break reception with a 'fixed' target drifting about 36 Mm (±200 km) above the equator.

[edit3: restarted my computer (for the first time in weeks) lets see what happens]

This is using reciever listening for µW, and a transmitter broadcasting MW over a distance of 36±1 Mm, to the edge of the solarsystem is ten petameters (10^16 m), and without radiooptics to focus that signal into a extremly tight beam we would need a signal measured in TW at TV satelite efficiencies to even recieve a signal. And over a distance of 4 ¼ light years (from the Sun to Proxima Centauri) that signal would need to be in the exawatt range, and that is not considering the scattering effect of thruster gasses, and the oort cloud.
Frankly, I don't see why doppler ought to be such a big technical challenge. It's true that a lot of equipment today makes a half-assed attempt at dealing with doppler. This is more a testimonial about the failure of Capitalism than anything else. You would think that market forces and competition would result in better and better products and technologies; but instead we have most capital nowadays directed at marketing and speculative investments, and less and less capital going to even the most basic R&D. To deal with Doppler effects, all that is needed is the digital equivalent of a tape recorder that records at one speed and plays back at another. The reason doppler-ed voices sound "aspirated" in most radio equipment is that, instead of the recording/playback trick, what they tend to do, for 'simplicity', is add/subtract a fixed frequency (by changing the tuning), which distorts the natural harmonics contained in speech (from integer, harmonic ratios to the fundamental, to Bessel series, if I'm not mistaken). This only happens with AM modulation and the likes (single lateral band, carrier-less single lateral band...), though; not with FM modulation or digital signals; --the latter are either received and demodulated correctly or not at all.
Without sufficent qualified staff to handle the assignment correctly it fell to the work-experience who kid figured out a way to fake a demonstration, expecting that the problem would be solved later. Using a MIMO array the image of an antenna would be projected into open space and the C/C module would follow it blindly. The 'locking' effect would be triggered by shifting the focus of the antenna creating a squacking effect at the C/C module's reciever that would be virtually undetectable by any observer not within the illusion's effect.

The following protocol was defined:
  • Carrier detect, maintaing bearing
  • Three squacks for "pay attention"
  • Five for "come here (on top of the antenna)"
  • Seven for "hold attitude and shut down for fifteen minutes"
  • Other "Ignore radio for one minute"
The initial tests / Govt. inspection apeared to go smoothly. A barge was sellected because it was 'cheap' and the poor maneuverabilty meant that the contractors had time to regain control of it should it go stray. And because completion of the first ship was years away the Govt. didn't look too closely.

When the C/C module passed, the work-experience kid left for better work, and the sub-contractor (satisfied by the pass) never actually fixed the problem, instead turning over a govt-passed black-box to RMC who never inquired further.


At the time of the launch the Nemisis (named for the type of star it was sent to) was the largest robotic ship ever comissioned, measuring 10x8x6 km at launch, and designed to expande up an additional 3 km in length during refinement.

The launch (with the endorcement of futures options by Lloyd's) spurred a series of nine waves, in total 57 ships, until obsoleted by FTL travel. Retaing the same disasterously incomplete C/C module until during the construction of the third wave the first of many inevitable collistions happened.

The Reputable Mining Corporation was bankrupted by an insufficent number of mining ships returning within the first century and more significantly a number of which that profided their own pyrothechnics when they did.

RMC was parceled and sold to the Yjere Consortium among other more successful competitors, and 'decoy beacons' sent out to reduce the number of Nemisis ships with their cargos of minerals and apocolyptic-DOOM that would find their way home.



Nemisis, Reputable and 57 are references to a long running Science-Fiction series.

[GRR2: My browser went to the "submitted sucessfully page" without actually submitting the text here, either there's something very wrong with my computer or my ISP is being [villians] with a web-cache of some sort.]

[GRR1: Stupid browser didn't actually save the text that was here, it was a lot of text.]
... Redshift and signal discrimination.
... "you thought that getting satelite TV was bad."
... Government inspection/trials of systems
... "A temporary solution"

[done: posting now to keep thread alive. plz comment.]
I think you're shooting your own foot by getting too technical. Read my proposal thread's first post again. Story writing is something that has to be done with the heart; not with the head. I see nothing of interest to humans in all this; where are the people, their hopes and aspirations (specially personal ones, rather than just desire to make money)?, where are the villains and the heroes (besides your ISP)?, where is love?
As much as most of us detest the materialism of today, it seems we're all so thoroughly brainwashed that we can't think of a story where people have motivations other than money and power. It seems our creative juices always end up funnelled into the antidisestablishmentarinist bottle no matter how anti-antidisestablishmentarian our philosophies may be. :(
safemode
Developer
Developer
Posts: 2150
Joined: Mon Apr 23, 2007 1:17 am
Location: Pennsylvania
Contact:

Re: Update, RE: a clear example.

Post by safemode »

The problem about trying to make a story into a video game like you want chuck is that we dont simulate anything but making money and shooting things.

The story has to revolve around what the game can do. It has to require the game in order to tell the story.

This means, the story has to revolve around politics, economy, combat, exploration, and reputation/hero/career building.

The backstory is much less important than the story involved with the resolution of the backstory. If the resolution isn't a logical conclusion to the backstory, then it just looks like you shoe-horned a backstory into a game that really had nothing to do with it. You could substitute pretty much any game in that situation and the effect would be the same.


I think what snow cat's coming up with can work and probably be more complex to the point where you'd like it, but it would require Much more immersive characters to interact with in the game. Well beyond the 2d sprites that spout some text or vocals. They need to convey emotion, they need to react to the player. If we had that, we could do what you want to see done in a story. If you dont have that though, then it's all just cutscene nonsense or text to read and forget about.
Ed Sweetman endorses this message.
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Re: Update, RE: a clear example.

Post by chuck_starchaser »

Not really; I don't easily forget about the text I read, if it's engaging and well written.
Take the case of Privateer, which has no emotional expressiveness for characters, and
yet it makes it very clear that while characters like Lynch, Tayla and Murphy are about
making money, and Goodin may be all about career advancement, other characters have
NOT put self-interest above all else. Monkhouse cares nothing about money; only about
the advancement of Science. Masterson puts the interests of Oxford ahead of his own;
--and in fact spends money out of his own pocket to pay Burrows for the missions. And
there's also Monte, whose aims are never made clear, but certainly are far from money
or fame. What I'm trying to suggest is precisely that we humans are not all about money,
fame, career advancement, etceteras. This is how simpleton right wing fanatics like the
Austrian School of Economics would like us to be, --to fit their economics theories.
We've no obligation to oblige them, whether in RL or in fiction.
Snow_Cat has some really good characters he presented at the beginning of this thread;
and I wouldn't suggest any changes to them; but perhaps some extra character(s) is/are
needed, whose aims are other than ownership or control of the mining operation. And
while perhaps explaining how a collision or other disaster happened at some point, this
should perhaps not be the focus of the back-story. Typical of BAD sci-fi is when a story
revolves around a new technology, and the solution to it is another new technology; or
in general fiction when a story is all about a physical situation, and the solution is just
a physical act that neutralizes the threat. Cheap Hollywood garbage.
Similarly, we have BAD AI in Vegastrike, and much of it revolves around the faction
relations and all that. You accidentally crash into a friendly ship and it becomes enemy
--as if people were that simple; or the "enemy of my enemy is friend" paradigm
just hard-coded into the engine like it's a natural law... Kill luddites to be friends with
everybody... This may be amusing at first, but such simplistic AI becomes boring to
the extreme after a while. But this is a problem that needs to be fixed at the engine
level, whereas excessive simplicity in human interactions and motives can be easily
avoided in storywriting without any engine changes.
If every quest you embark on, and every mission you take, presents characters that
all they care about is self-interest, selfish bastards whose only differences relate to
how they see their self interest being achieved; none showing any ethics or morals,
none evidencing any social concerns, not only are you helping the Austrian School of
Economics brainwash us all into believing that self-interest is all we're all about, and
therefore we should trust authorities, as somehow magically they are supposed to be
an exception to the rule, but you're also making for a hell of a boring game.
safemode
Developer
Developer
Posts: 2150
Joined: Mon Apr 23, 2007 1:17 am
Location: Pennsylvania
Contact:

Re: Update, RE: a clear example.

Post by safemode »

I think logical improvement of the AI in the game should head to it's own thread (a new one anyway) as it's definitely something that could use some brainstorming completely apart from campaign work.
Ed Sweetman endorses this message.
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Re: Update, RE: a clear example.

Post by chuck_starchaser »

Agreed. Not sure if to start a thread about it now; just that there's so many things to do, and so few people to do them. My intent wasn't to derail the thread; just to say that Vegastrike is too simplistic a game in general, right now, where the future is simply a linear scaling of the present in every respect; where accelerations are absurd; where news are repetitive ad nauseum; where missions are repetitive ad nauseum; where AI is AS (artificial stupidity); where all systems have 2 or 3 inhabited planets, plus a gas giant, and a background with nebulas; where most music has current fashion new age sounds on current fashion drum and bass, and use the current fashion 12-tone, atempered musical scale; where all weapons cause roughly the same type of damage, where shields are no different from armor in the type of damage they block, where all communications are canned; where the sound of metal being hit sounds the same no matter what ship you fly; etceteras. All this requires tons of work to fix. But in the case of stories and plots, we are not limited by the engine, and we can have more sophistication there, for starters. And more importantly, we can have the kind of sophistication that goes beyond simple mechanical complexities of multi-dimensional plots... true sophistication, philosophical sophistication, that refreshingly contradicts our present day myths and make-believes about human nature, rather than reinforces them. I'm told every day that I'm a simpleton consumer. Every junk mail in my mailbox tells me that. One has to fight back all the time... I don't want a game to also tell me that every human being in the universe 3000 years from now is a simpleton, selfish SOB. That's like a diet where you can eat is french fries, and you can only vary the spices you put on them. I want a diet that includes variety. The characters in a game should not all be selfish SOB's that only vary in the types of methods. There should be unselfish people, idealists, heroes, maniacs... whatever; just like there are in the real world. What's wrong with something as simple as someone hiring you to find a missing loved one, as opposed to kill a pirate? It may boil down to killing a pirate to rescue the loved one, but at least you've met someone with a HUMAN motivation.
safemode
Developer
Developer
Posts: 2150
Joined: Mon Apr 23, 2007 1:17 am
Location: Pennsylvania
Contact:

Re: Update, RE: a clear example.

Post by safemode »

if you are simulating thousands of units, which most have one or more "characters" associated with them, you can't hope to give them all diverse personalities, even when they are closely interacting with eachother, and if they only do it when they interact with the player it wouldn't make any sense. 99% of the interactions going on in the game are not with the player, and these other interactions are happening a lot every second.

A campaign should pick and choose characters to give them life, with the vast majority following a more regular model. we lack these special characters in this game, which makes the whole seem flat. With these characters, the player would put a face on the group, giving the group a personality it really doesn't have, because their VIP(s) has that personality.

For instance, the cylon raiders in the new BSG were basically animalistic robots, not much smarter than a dog. That is how they were perceived until Scar was introduced. Now they had a menacing intelligence that became attributed to the entire group, rather than just to Scar. Wing commander had Ace's ... We have nothing of the sort.
Ed Sweetman endorses this message.
chuck_starchaser
Elite
Elite
Posts: 8014
Joined: Fri Sep 05, 2003 4:03 am
Location: Montreal
Contact:

Re: Update, RE: a clear example.

Post by chuck_starchaser »

safemode wrote:if you are simulating thousands of units, which most have one or more "characters" associated with them, you can't hope to give them all diverse personalities, even when they are closely interacting with each other, and if they only do it when they interact with the player it wouldn't make any sense. 99% of the interactions going on in the game are not with the player, and these other interactions are happening a lot every second.
True, I wasn't suggesting modeling their brains neuron by neuron and frame by frame, though, or even different personalities; I was merely suggesting more realistic modeling of factual hostility (versus likes or dislikes) at a faction ai level, the kind of variables you would update once every 24 hours on the outside. Then, the quicker ai decisions would use those faction level variables. Whether a faction decides to go to war with another is something that can be decided once in many hours, and be more sophisticated than it currently is. And it should not be based on "like" or "dislike", but on many factors, such as,
  • do we have a chance to win the war?
  • do we stand to benefit from a war, even assuming we win it?
  • do we currently source any strategic resources from them?
  • are we currently at war?/can we afford a war on two fronts?
  • could a war with X precipitate a war with Y?
  • could we afford to be at war with X AND Y?
  • would a war with X be popular or would anti-war sentiment kill us?
Any rational faction would consider these questions regardless of like or dislike for X. The only point where like or dislike figures in is on the question of the war's popularity. So it should be possible for faction A to go to -100 with faction X, yet not go to war. And this would be passed to the quicker AI routines as "hate them; but don't shoot them", and the individual units would behave as such, in proportion to faction obedience/cohesiveness.
A campaign should pick and choose characters to give them life, with the vast majority following a more regular model. we lack these special characters in this game, which makes the whole seem flat. With these characters, the player would put a face on the group, giving the group a personality it really doesn't have, because their VIP(s) has that personality.

For instance, the cylon raiders in the new BSG were basically animalistic robots, not much smarter than a dog. That is how they were perceived until Scar was introduced. Now they had a menacing intelligence that became attributed to the entire group, rather than just to Scar. Wing commander had Ace's ... We have nothing of the sort.
Gottcha. Anyways, I was talking about AI at a different level. Faction AI? I described it in more detail in your other thread.
Post Reply