SPEC

Development directions, tasks, and features being actively implemented or pursued by the development team.

SPEC

Postby log0 » Sat Jun 29, 2013 6:44 am

Inspired by the changes discussed in this thread viewtopic.php?f=27&t=18412&start=300 and some previous discussions (before the big crash) I'd like to make a somewhat radical proposal to the SPEC mechanics, if you don't mind.

Current state:
There is this artificial construct called SPEC ,that is required to make the game playable due to vast distances. It is constrained by some artificial interdiction rules, which are supposed to integrate it into gameplay (and make it fun?) I guess. In reality it doesn't work that well, can be just annyoing.

Proposal:
1. Remove interdiction rules completely.
2. Let the autopilot deal with approach speed/SPEC factor to avoid overshooting, drop out at safe distance.
3. To avoid units SPECing around like crazy implement a SPEC startup period/delay (4-8 sec).
4. Implement a device/method that allows to pull ships out of SPECS to be able to intercept them.

More to point 4:
Think of something akin to the EMP switch in the Matrix movies. You hit the button, your SPEC overloads and everything within a certain radius is pulled out of SPEC. As an additional effect SPEC drives of the affected units (including yours) are unusable for a certain period of time, about 10-30 sec maybe (due to the drive having to be recalibrated, capacitors reloaded whatever...).
To avoid people spamming their SPEC kill switch it should come with some costs, have adverse effects on the drive (cause damage, burn a few fuses). So after a few times (3-4) it would have to be repaired.

As a bonus this would allow for some interesting tactics:
Units sacrificing themselves to let others flee by intercepting pursuers. Pirates having specialised SPEC intercept units. Very fast, light ships without any meaningful armor or weapons, only meant to chase and pull victims out of SPEC.

The main idea is to make SPEC interactions an active gameplay element. You will be pulled out of SPEC because someone wanted it to happen, and not because you are too close to some other object intentionally or by accident.
log0
 

Share On:

Share on Facebook Facebook Share on Twitter Twitter Share on Digg Digg

Re: SPEC

Postby IansterGuy » Sat Jun 29, 2013 6:00 pm

log0 wrote:I'd like to make a somewhat radical proposal to the SPEC mechanics, if you don't mind.
After reading, this seems more like enhancements to the SPEC systems once it is fixed, more than an independent solution.

log0 wrote:Current state:
There is this artificial construct called SPEC ,that is required to make the game playable due to vast distances. It is constrained by some artificial interdiction rules, which are supposed to integrate it into gameplay (and make it fun?) I guess. In reality it doesn't work that well, can be just annyoing.
I think it is only annoying because the system is imcomplete.

log0 wrote:Proposal:
1. Remove interdiction rules completely.
2. Let the autopilot deal with approach speed/SPEC factor to avoid overshooting, drop out at safe distance.
3. To avoid units SPECing around like crazy implement a SPEC startup period/delay (4-8 sec).
4. Implement a device/method that allows to pull ships out of SPECS to be able to intercept them.
1. When you say remove interdiction completely I don't see removal as necessary to add these elements of game play, active and passive interdiction would work better in combination as I see it. Some strategies like sacrificing a ship to escape would to a lesser degree already be viable, but just not be due to a dramatic EMP pulse. Though if this became standard game play it would be not be such a surprising event.

2. Dropping out at a safe distance may not be enough when ships are still in SPEC, the ships may be moving and changing their direction too fast to maintain precision. I support slowing SPEC when approaching other warp bubbles with a partial exception for ships moving in the same direction.

3. Yea, I think when deactivated completely it should take a while to be activated again.

4. I think having EMP as a purchasable weapon would work in addition to almost any base SPEC system.

log0 wrote:The main idea is to make SPEC interactions an active gameplay element. You will be pulled out of SPEC because someone wanted it to happen, and not because you are too close to some other object intentionally or by accident.
I would like to see normal passive interdiction mostly for collision avoidance and only useful offensively if the pursuers ship is faster in SPEC. Then activated interdiction would be from purchasable upgrades and be useful for slowing surrounding ships while maintaining ones own SPEC capabilities in certain circumstances with certain technologies like SPEC busters and EMP equipment disruption. This is opposed to active and passive interdiction.
Last edited by IansterGuy on Sat Jul 06, 2013 2:06 pm, edited 1 time in total.
If you were interested I would have told you, but if I told you, you would not be interested. I can tell you are not interested because you don't already know and even if you did, it would be pointless to tell you again. Therefore in no situation will I ever tell you anything.
User avatar
IansterGuy
Bounty Hunter
Bounty Hunter
 
Posts: 174
Topics: 10
Joined: Mon Aug 13, 2012 1:49 am
Location: Milky Way, Orion arm, Earth, Canada Saskatchewan, Saskatoon

Re: SPEC

Postby TBeholder » Sat Jun 29, 2013 6:41 pm

log0 wrote:Proposal:
1. Remove interdiction rules completely.

3. To avoid units SPECing around like crazy implement a SPEC startup period/delay (4-8 sec).
4. Implement a device/method that allows to pull ships out of SPECS to be able to intercept them.
I think (3) and (4) are easily solvable by not iplementing (1). :D Because that's how it came to be in the first place (back in Elite), obviously.

1. Without interdiction and with high warp rate, approaching to use this device would be harder.
3. Startup delay would just make all the zipping and zapping around slightly less frequent. Now, decreasing SPEC ramp-up rate would make ships noticeably accelerate instead of suddenly buzzing away.
"Two Eyes Good, Eleven Eyes Better." -Michele Carter
User avatar
TBeholder
Elite Venturer
Elite Venturer
 
Posts: 732
Topics: 38
Joined: Fri Apr 14, 2006 7:40 pm
Location: chthonic safety

Re: SPEC

Postby log0 » Sun Jun 30, 2013 3:14 pm

TBeholder wrote:
log0 wrote:Proposal:
1. Remove interdiction rules completely.

3. To avoid units SPECing around like crazy implement a SPEC startup period/delay (4-8 sec).
4. Implement a device/method that allows to pull ships out of SPECS to be able to intercept them.
I think (3) and (4) are easily solvable by not iplementing (1). :D Because that's how it came to be in the first place (back in Elite), obviously.

1. Without interdiction and with high warp rate, approaching to use this device would be harder.
3. Startup delay would just make all the zipping and zapping around slightly less frequent. Now, decreasing SPEC ramp-up rate would make ships noticeably accelerate instead of suddenly buzzing away.


Then there shall also be a low SPEC ramp-up rate. I totally fine with this. ;)

Still, SPEC interdiction as is doesn't differentiate between intentional and accidental effects, is thus broken. So why not making it an active element and actually fun?

To 1:
The device would have an action radius. Think of replacing passive interdiction radius with active SPEC overload radius.

Edit:
So the basic difference would be, that the player/ai would have to trigger interdiction, and that it would have costs/penalties to do so.

Btw the ramp-up rate can be fitted nicely into this overload idea, as max safe SPEC ramp-up rate. So the overloading would happen by ramping-up far beyond limits, which would cause a destructive SPEC pulse, lol...
The same would apply to ramp-down rate. Do it too fast (by hitting the kill switch for example) and you're out of SPEC, have got a few fried fuses, and have pissed off everyone in your proximity by pulling them out of SPEC too. Now that sounds like fun imho.
log0
 

Re: SPEC

Postby TBeholder » Mon Jul 01, 2013 7:58 am

Also, currently the object's mass isn't even counted, only distance to it and Warp Interdiction value.

log0 wrote:Then there shall also be a low SPEC ramp-up rate. I totally fine with this. ;)
There's no rate as such, but physics/warprampuptime and physics/warprampdowntime in config, so it takes 5s to get the maximum warp power, 10 for NPCs, and dependency is square - which is why vessels visibly accelerate instead of just vanishing when they turn SPEC on. Set it to 30 or so, and it will be near-impossible to escape combat like this.

log0 wrote:Still, SPEC interdiction as is doesn't differentiate between intentional and accidental effects, is thus broken. So why not making it an active element and actually fun?
I don't see "thus broken" here.
Also, "accidentally approached within 5 km" should not happen in space. Ever. :twisted:
And if you chase / try to escape, here's the "active element". Especially seeing as currently it's hard to intercept a light ship even when it doesn't try to evade.
Conversely, "press the button just in right time" doesn't add much fun (especially compared to a more complex version of it during fighting itself), while letting the computer do it eliminates the "active element".
log0 wrote:To 1:
The device would have an action radius. Think of replacing passive interdiction radius with active SPEC overload radius.
When you put it like this, it sound about as different as a cut-off head from a sawed-off one. ;)

log0 wrote:Edit:
So the basic difference would be, that the player/ai would have to trigger interdiction, and that it would have costs/penalties to do so.
More of pointless "click-or-you-miss-it" fuss, when there's already something to do.
If you want it to be energy-costly, may as well be done via extra power consumption during ramping up (and thus the cost of restarting) SPEC. Kind of like charging vs. upkeeping shields. Which makes sense. This also would limit warp-escapes from combat situations even more.
"Two Eyes Good, Eleven Eyes Better." -Michele Carter
User avatar
TBeholder
Elite Venturer
Elite Venturer
 
Posts: 732
Topics: 38
Joined: Fri Apr 14, 2006 7:40 pm
Location: chthonic safety

Re: SPEC

Postby klauss » Mon Jul 01, 2013 8:57 am

TBeholder wrote:3. Startup delay would just make all the zipping and zapping around slightly less frequent. Now, decreasing SPEC ramp-up rate would make ships noticeably accelerate instead of suddenly buzzing away.


I'd fully support that move.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
User avatar
klauss
Elite
Elite
 
Posts: 7243
Topics: 55
Joined: Mon Apr 18, 2005 7:40 am
Location: LS87, Buenos Aires, República Argentina

Re: SPEC

Postby log0 » Mon Jul 01, 2013 10:54 am

"puch the button just in the right time" ... "click-or-you-miss-it"

I think you are confusing something here. It is not about time. It is about distance. Get into range (there might be even a hud indicator for this), then feel free to push the button at any time you want.
log0
 

Re: SPEC

Postby klauss » Mon Jul 01, 2013 10:59 am

And how do you get into range? Does the autopilot match velocities? Is there any difference here from interdiction, other than the mumbo-jumbo justification?

Lets focus on gameplay and not mumbo-jumbo... what changes in game mechanics do people propose? I've only seen the ramp-up time increase pop up in this cateogry, with which I agree. Everything else is just different mumbo-jumbo justifying pretty much the same mechanics we don't like.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
User avatar
klauss
Elite
Elite
 
Posts: 7243
Topics: 55
Joined: Mon Apr 18, 2005 7:40 am
Location: LS87, Buenos Aires, República Argentina

Re: SPEC

Postby log0 » Mon Jul 01, 2013 11:13 am

klauss wrote:And how do you get into range? Does the autopilot match velocities? Is there any difference here from interdiction, other than the mumbo-jumbo justification?

Lets focus on gameplay and not mumbo-jumbo... what changes in game mechanics do people propose? I've only seen the ramp-up time increase pop up in this cateogry, with which I agree. Everything else is just different mumbo-jumbo justifying pretty much the same mechanics we don't like.

Wow, my english must suck seriously, because I've posted it like three times already in this thread. So let me try again.

The gameplay change is to make interdiction event a user action, let him push a friggin button. Get rid of this, oh I am too fsckin close to something, will spend the next minutes trying to get away from it before I can use SPEC, bullshit.

And my apologies for the mumbo-jumbo stuff, it has been unnecessary indeed.

And yes, it should be autopilot/flight governors job to get you in range. Some simple hold distance controller should do it.
log0
 

Re: SPEC

Postby log0 » Mon Jul 01, 2013 11:45 am

Or shorter: Current SPEC rules don't differentiate between attack and simple flyby or approach. This is what I mean with "don't work".
Solution: Let user decide...
log0
 

Re: SPEC

Postby Deus Siddis » Mon Jul 01, 2013 12:20 pm

log0 wrote:"puch the button just in the right time" ... "click-or-you-miss-it"
I think you are confusing something here. It is not about time. It is about distance. Get into range (there might be even a hud indicator for this), then feel free to push the button at any time you want.


You might be missing Turbo Beholder's point about the issue of "insufficient player reaction time" that replacing passive interdiction with active interdiction would create.

When you are in interplanetary space, you are moving 100+ times the speed of light and so are other ships. So without an enormous effective radar range a ship will appear and disappear off your scopes in a matter of milliseconds. That leaves no time to identify the craft and decide to interdict it for the purpose of attack, docking or communication. That's why passive interdiction is necessary, it slows down encounters so that you can react.

The problem is there are too many encounters that are not interesting; solar systems are loaded with boring interplanetary haulers that offer nearly no interaction potential. What would help is if the average civilian traffic was reduced by one or two orders of magnitude and then maybe 20% of the remaining traffic was low end Pirates and 5% were dangerous enemies like Aera. Then getting interdicted wouldn't be a tedious hassle, it would be a big part of the game play (again).
Deus Siddis
Elite
Elite
 
Posts: 1362
Topics: 13
Joined: Sat Aug 04, 2007 8:42 am

Re: SPEC

Postby klauss » Mon Jul 01, 2013 12:48 pm

Deus Siddis wrote:
log0 wrote:"puch the button just in the right time" ... "click-or-you-miss-it"
I think you are confusing something here. It is not about time. It is about distance. Get into range (there might be even a hud indicator for this), then feel free to push the button at any time you want.


You might be missing Turbo Beholder's point about the issue of "insufficient player reaction time" that replacing passive interdiction with active interdiction would create.

When you are in interplanetary space, you are moving 100+ times the speed of light and so are other ships. So without an enormous effective radar range a ship will appear and disappear off your scopes in a matter of milliseconds. That leaves no time to identify the craft and decide to interdict it for the purpose of attack, docking or communication. That's why passive interdiction is necessary, it slows down encounters so that you can react.

The problem is there are too many encounters that are not interesting; solar systems are loaded with boring interplanetary haulers that offer nearly no interaction potential. What would help is if the average civilian traffic was reduced by one or two orders of magnitude and then maybe 20% of the remaining traffic was low end Pirates and 5% were dangerous enemies like Aera. Then getting interdicted wouldn't be a tedious hassle, it would be a big part of the game play (again).


@log0: exactly my point. You cannot make it a button, too many holes, and if you make it automatic, then you have nothing significantly different from what we have.

Hence my old proposal: make SPEC interdiction the realm of mines (or similarly special-purpose stationary devices). It adds the attack distinction you want, and in conjunction with Deus's refactoring removing the effect of civilian bystanders, you have no interference from background traffic while just traveling around.

Mines do not suffer from that reaction time issue (they intercept everything in their area of influence), and laying out mine fields (deciding where, deploying them, watching them in action) can be an engaging gameplay element. Not to mention being caught yourself in pirate mine fields.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
User avatar
klauss
Elite
Elite
 
Posts: 7243
Topics: 55
Joined: Mon Apr 18, 2005 7:40 am
Location: LS87, Buenos Aires, República Argentina

Re: SPEC

Postby log0 » Mon Jul 01, 2013 2:27 pm

Proposal binned. Will look for something else to waste my time on. Thanks for input.
log0
 

Re: SPEC

Postby Deus Siddis » Mon Jul 01, 2013 2:36 pm

klauss wrote:Hence my old proposal: make SPEC interdiction the realm of mines (or similarly special-purpose stationary devices).


Interdiction mines are an interesting idea in their own right but you don't want to paint over a bad feature with a new (good) one. And that's exactly what excessive interplanetary civilian traffic is, a bad feature, because it contributes nothing to the game:

Gameplay: Offers no special player interaction. Player ignores them and they ignore player.
Aesthetics: They are never close enough to be visible unless you intercept one and then only one is visible at once. Meanwhile your radar interface is overloaded with targets.
Engine: Simulating more civilian traffic in one system means less system resources are devoted to simulation of other systems or the granularity of local physics and AI.
Realism: There's nothing inherently more realistic or believable about having thousands of transport ships in a system rather than dozens.

We need far fewer and far more interesting deep space encounters. Because it is not that getting interdicted is boring but that getting interdicted all the time by witless Ox transports is boring. Imagine if on an average cargo run you encounter ships in deep space only a few time and 50% of those encounters offer special business opportunities or information, 20% are a pirate ambush, 5% are an aera raiding force, 2% are salvage opportunities and 0.5% are rarely witnessed alien creatures or phenomena. Then interdiction is exciting and surprising and so is travel in general because you are always on the edge of your seat anticipating the next encounter.

P.S. This is a consistent issue for us as a project in general. When there is a part of the game that is boring, we find a way for the player to skip past it, rather than fix it or replace it with something that is worth the player not skipping. Autopilot, auto-dock, auto-aim, consenting interdiction, etc. are all 'skip past it' solutions. When you allow every aspect of the game to be so boring that the player has to be able to skip over it, what you see is the player skipping over the entire game and moving on to Star Citizen.
Deus Siddis
Elite
Elite
 
Posts: 1362
Topics: 13
Joined: Sat Aug 04, 2007 8:42 am

Re: SPEC

Postby klauss » Mon Jul 01, 2013 3:06 pm

log0 wrote:Proposal binned. Will look for something else to waste my time on. Thanks for input.


Thanks to you too. We may not agree every time, especially on this SPEC issue ;) , but I do appreciate your input.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
User avatar
klauss
Elite
Elite
 
Posts: 7243
Topics: 55
Joined: Mon Apr 18, 2005 7:40 am
Location: LS87, Buenos Aires, República Argentina

Re: SPEC

Postby klauss » Mon Jul 01, 2013 3:19 pm

Deus Siddis wrote:
klauss wrote:Hence my old proposal: make SPEC interdiction the realm of mines (or similarly special-purpose stationary devices).


Interdiction mines are an interesting idea in their own right but you don't want to paint over a bad feature with a new (good) one. And that's exactly what excessive interplanetary civilian traffic is, a bad feature, because it contributes nothing to the game:

I think safemode was getting rid of the need to have such excessive traffic anyway. Not sure what happened with that.

Deus Siddis wrote:Aesthetics: They are never close enough to be visible unless you intercept one and then only one is visible at once. Meanwhile your radar interface is overloaded with targets.

Well, that's not really true, I do see a lot of them. In any case, as I said, safemode was fixing those issues, remember? by bringing stations closer together and forming hubs, so whatever little civvy traffic there is, it would be noticeable, visible, and engaging (since you wouldn't spend half an hour chasing one).

Deus Siddis wrote:Engine: Simulating more civilian traffic in one system means less system resources are devoted to simulation of other systems or the granularity of local physics and AI.

That's not bad. What you don't see can certainly be simulated to a lesser precision, and that's what VS does.

Deus Siddis wrote:Realism: There's nothing inherently more realistic or believable about having thousands of transport ships in a system rather than dozens.

Yeah, maybe not. I would only believe small shuttles going to and from planets would be as numerous, but not the bigger ships players interact with.

Deus Siddis wrote:We need far fewer and far more interesting deep space encounters. Because it is not that getting interdicted is boring but that getting interdicted all the time by witless Ox transports is boring. Imagine if on an average cargo run you encounter ships in deep space only a few time and 50% of those encounters offer special business opportunities or information, 20% are a pirate ambush, 5% are an aera raiding force, 2% are salvage opportunities and 0.5% are rarely witnessed alien creatures or phenomena. Then interdiction is exciting and surprising and so is travel in general because you are always on the edge of your seat anticipating the next encounter.


Indeed, we need better control over the nature of interdiction events.

I think explicit trade and patrol lanes would provide that. Ie: more explicitly scripted stuff. I know we like VS-the-huge-sandbox as a principle, but it certainly hasn't paid off as a game. Scripted content does help better manage the experience.

I wouldn't mind ideas about "randomizable scripting". Ie: a way to have the cake and eat it too. We need scripting to control just what kind of events happen to the player, because that way we can make sure they're a fun mix. But we don't want to have to script it all, it should be generic scripts that can be randomized to create a varied experience. I'm blank on that matter currently, aside from the obvious randomization points (location, number and type of flightgroups involved, etc).

Deus Siddis wrote:Autopilot, auto-dock, auto-aim, consenting interdiction, etc.

Wait wait wait... I know what you mean, and I agree in principle, but all those "auto" things are necessary regardless. Lots of people want point-and-click navigation, and it works as game mechanics really well, proved by good examples like Freelancer. There's nothing bad in recognizing their merit.

What we need is fun stuff that happens to force you to turn those "auto" things off. And we need to think about those fun stuff, not just throw features and expect people to come up with the fun stuff by themselves.

This is why I mention mines as well: mines find you. Pirates would use them. Pirates would get you very well acquainted with mines. Scripted missions too. And then you start using them on your own, but because the game showed you how.

Don't throw the player in a huge galaxy-sized sandbox with no guidance, that's overwhelming.

Ok, rant mode off.

So... fun stuff that must happen. It must all start with new, scripted, engaging missions I say.

Lets do this: if someone comes up with a realizable mission script (within our current ability), I'll vouch to code that as a random new mission type that can appear on the mission computer. It should "look" routine on the mission computer, but lots of fun stuff should happen when you take it up.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
User avatar
klauss
Elite
Elite
 
Posts: 7243
Topics: 55
Joined: Mon Apr 18, 2005 7:40 am
Location: LS87, Buenos Aires, República Argentina

Re: SPEC

Postby TBeholder » Mon Jul 01, 2013 3:38 pm

log0 wrote:Or shorter: Current SPEC rules don't differentiate between attack and simple flyby or approach. This is what I mean with "don't work".
They don't need to. "If you aren't trying to attack or dock it, don't approach it close enough for interdiction" would do. :)

Deus Siddis wrote: And that's exactly what excessive interplanetary civilian traffic is, a bad feature, because it contributes nothing to the game:
Gameplay: Offers no special player interaction. Player ignores them and they ignore player.
Aesthetics: They are never close enough to be visible unless you intercept one and then only one is visible at once. Meanwhile your radar interface is overloaded with targets.
Engine: Simulating more civilian traffic in one system means less system resources are devoted to simulation of other systems or the granularity of local physics and AI.
Realism: There's nothing inherently more realistic or believable about having thousands of transport ships in a system rather than dozens.
We need far fewer and far more interesting deep space encounters. Because it is not that getting interdicted is boring but that getting interdicted all the time by witless Ox transports is boring.
Such random encounters outside of choke points happen only because all objects are nailed to their places, so it's the same line between 2 points as 30 seconds ago and in both directions. If they moved with proper orbital velocities - even without proper prediction, just "nose to the bright dot" autopilot tracking, a ship flying from A to B and the ship flying from B to A at the same time would miss each other by many megameters (unless both are on whatever-stationary orbits :D of course), thus accidental intercept would be extremely improbable.
Though yes, only so much of bulk transport makes sense. Now, if it was actually a part of the economy model rather than jus spam of random events, the engine would know exactly how much, and interfering with the traffic would matter...
Deus Siddis wrote: P.S. This is a consistent issue for us as a project in general. When there is a part of the game that is boring, we find a way for the player to skip past it, rather than fix it or replace it with something that is worth the player not skipping. Autopilot, auto-dock, auto-aim, consenting interdiction, etc. are all 'skip past it' solutions. When you allow every aspect of the game to be so boring that the player has to be able to skip over it, what you see is the player skipping over the entire game and moving on to Star Citizen.
Aye, aye. Also, looking at the underlying problem may help (see above).
"Two Eyes Good, Eleven Eyes Better." -Michele Carter
User avatar
TBeholder
Elite Venturer
Elite Venturer
 
Posts: 732
Topics: 38
Joined: Fri Apr 14, 2006 7:40 pm
Location: chthonic safety

Re: SPEC

Postby Deus Siddis » Mon Jul 01, 2013 7:01 pm

klauss wrote:I think safemode was getting rid of the need to have such excessive traffic anyway. Not sure what happened with that.


That is scheduled to happen over many minor versions which may take years, assuming it tests out well.

Decreasing the spawning of civilian craft (and perhaps increasing the loiter time the civilian AI spends at destinations) could be an immediate fix.

That's not bad. What you don't see can certainly be simulated to a lesser precision, and that's what VS does.


But precision aside, instead of simulating excessive ships in the current system, you could be simulating a reasonable number in the current system plus every adjacent one. Or every system up to two jumps away from the current system.

I wouldn't mind ideas about "randomizable scripting". Ie: a way to have the cake and eat it too. We need scripting to control just what kind of events happen to the player, because that way we can make sure they're a fun mix. But we don't want to have to script it all, it should be generic scripts that can be randomized to create a varied experience. I'm blank on that matter currently, aside from the obvious randomization points (location, number and type of flightgroups involved, etc).


I think the problem is more that:

1) There are not enough types of things that happen in space (basically just fighting and grabbing cargo).

A low hanging fruit here would be asteroid mining, since it seems the engine already supports this, only some content is missing. A more in depth example is disabled/derelict ships that you could find and bring into your fleet and refurbish. Another example would be meaningful communication with NPCs, like the captain of a Mule asking you to meet him on board his ship, where you find him on the bridge in the form of a fixer and he offers you an especially high risk/reward mission. Or similarly, attacked ships sending distress calls that pay out to the people who assisted them against their attackers (if they survive). Or marooned pilots floating in space, that pay you to bring them back. Things like this can happen only as a result of accepting a mission at a base but they should happen dynamically in deep space based on what is happening in the simulation.

2) The dynamic universe is calibrated to spawn non-interactive blandness.

Far too many civilians, too many police, far too few dangerous criminals, too few factional task forces pushing deep into enemy territory. Random planets and station types, no uninhabited systems. No economic effects from destroyed infrastructure or delayed deliveries. We aren't using the dynamic universe to anywhere near its full potential.

Lets do this: if someone comes up with a realizable mission script (within our current ability), I'll vouch to code that as a random new mission type that can appear on the mission computer. It should "look" routine on the mission computer, but lots of fun stuff should happen when you take it up.


I would much prefer the development time be put into documenting the scripting API. And then once that is done, documenting and refactoring the existing python scripts. Then we less experienced developers could take on these scripting tasks ourselves, which should ultimately prove quite a lot more productive.
Deus Siddis
Elite
Elite
 
Posts: 1362
Topics: 13
Joined: Sat Aug 04, 2007 8:42 am

Re: SPEC

Postby Deus Siddis » Mon Jul 01, 2013 7:22 pm

TBeholder wrote:If they moved with proper orbital velocities - even without proper prediction, just "nose to the bright dot" autopilot tracking, a ship flying from A to B and the ship flying from B to A at the same time would miss each other by many megameters (unless both are on whatever-stationary orbits :D of course), thus accidental intercept would be extremely improbable.


Interesting.

Accidental intercepts are actually kind of a desirable thing though, we just have far too many overall and far too many of the boring kind proportionally.

Though yes, only so much of bulk transport makes sense. Now, if it was actually a part of the economy model rather than jus spam of random events, the engine would know exactly how much, and interfering with the traffic would matter...


True, but we can set the average amount of bulk interplanetary commerce to any reasonable amount. An Ox might make the circuit between Atlantis and Serenity once a month or once an hour depending on what is desirable for game play.
Deus Siddis
Elite
Elite
 
Posts: 1362
Topics: 13
Joined: Sat Aug 04, 2007 8:42 am

Re: SPEC

Postby klauss » Tue Jul 02, 2013 12:16 pm

Deus Siddis wrote:1) There are not enough types of things that happen in space (basically just fighting and grabbing cargo).

A low hanging fruit here would be asteroid mining, since it seems the engine already supports this, only some content is missing. A more in depth example is disabled/derelict ships that you could find and bring into your fleet and refurbish...

2) The dynamic universe is calibrated to spawn non-interactive blandness.

Far too many civilians, too many police, far too few dangerous criminals, too few factional task forces pushing deep into enemy territory. Random planets and station types, no uninhabited systems. No economic effects from destroyed infrastructure or delayed deliveries. We aren't using the dynamic universe to anywhere near its full potential...


Well, exactly! So lets implement those, instead of hardcoded engine stuff, as randomized missions. It's the lowest hanging fruit. It's something doable and beneficial. Don't you think?

Deus Siddis wrote:
Lets do this: if someone comes up with a realizable mission script (within our current ability), I'll vouch to code that as a random new mission type that can appear on the mission computer. It should "look" routine on the mission computer, but lots of fun stuff should happen when you take it up.


I would much prefer the development time be put into documenting the scripting API. And then once that is done, documenting and refactoring the existing python scripts. Then we less experienced developers could take on these scripting tasks ourselves, which should ultimately prove quite a lot more productive.


That won't happen. Because I even don't know the API that well. And what I know, is cumbersome to a shameless point.

No, newbies can't script VS ATM. But I probably could, if I were given enough incentive (ie: content - I'm no writer). Aaand... while I'm coding it, I can make it friendlier. Classes can be reused, and, most importantly, classes can be made data-driven.

I can either code a generic scripting engine on top of our python scripts, where whole missions can be made plugging classes together in newbie-friendly ways, or I can make one that just takes a text file as input, and newbies can write new text files.

Whatever I do, I can document.

So, no, the current scripting API is way too low level for newbies to manage, even with full documentation, and I wouldn't either be able to document it (don't know it that well), nor would have the motivation to do so (wouldn't immediately gain much).

The way to go about it, is having some veteran (me) do one mission, and in the process document, refactor, and improve for modders. The immediate gain (the mission) encourages the veteran (me), and the improved code base (because a veteran made it), in the long-term, pays off additionally in further contributions.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
User avatar
klauss
Elite
Elite
 
Posts: 7243
Topics: 55
Joined: Mon Apr 18, 2005 7:40 am
Location: LS87, Buenos Aires, República Argentina

Re: SPEC

Postby Deus Siddis » Wed Jul 03, 2013 10:43 am

klauss wrote:Well, exactly! So lets implement those, instead of hardcoded engine stuff, as randomized missions. It's the lowest hanging fruit. It's something doable and beneficial. Don't you think?


I think randomized missions are too cut and dry and we already have a half dozen of them. It always seems to go something like this:

You land at a base, fish the mission computer and the bar for the best mission. You select an escort mission. You leave the base and suddenly a Mule with a proper name you don't remember seeing before seemingly appears in space. It is you escortee, so you meet up with it. And no sooner than you have but a wing of pirates has appeared out of nowhere to attack your new charge! It is funny, you think to yourself, that you don't remember ever seeing pirates in this well policed system before. And then the local factional forces kills the pirates before you can get a single shot off on them. You get paid anyway! On to the next randomized mission and repeat...

And so missions have the following flaws:

  • They can't be encountered in space, only bases.
  • They aren't derived from the local game state. (A pirate entering the system should spawn a bounty mission; instead it is the bounty mission that spawns the pirate.)
  • Consequently, the local game state imbalances mission difficulty. (As in the example, the local militia swarm and kill your bounty for you.)

And because of those flaws missions don't really make interdiction any more exciting or relevant. Instead encounters with other vessels, as well as anything else interesting that happens in space, becomes an obstacle that slows down the process of finding or completing a paying mission.

The way to go about it, is having some veteran (me) do one mission, and in the process document, refactor, and improve for modders. The immediate gain (the mission) encourages the veteran (me), and the improved code base (because a veteran made it), in the long-term, pays off additionally in further contributions.


What about the dynamic universe, isn't that mostly python scripts? Perhaps this same strategy of simultaneously enhancing, refactoring and documenting could be applied to parts of that?

Because the only way to make space exploration and random interdiction relevant to the game, is to tweak the dynamic universe so that it spawns the right mix of threats, opportunities and curiosities. We need far less of do-nothing civilians. Less spawning of militia forces that remove all challenges for you. Increased spawning of hostile faction forces, even in interior systems. Increased spawning of phenomena like hoffman's blobs and beholders (I have never sighted a single one of either those things in the game ever).

And then add opportunities which occur in space due to in game events, not mission acceptance. Like have the AI comply with pirates and dump its cargo into space, so that when the militia chase away the pirates, you can swoop in to take some of the "abandoned" cargo. And instead of ships always exploding when they take critical damage, maybe they could be just as likely to become disabled, allowing you to tow such derelict vessels back to a nearby base as salvage. And have standing bounties certain systems put out on all ships of a certain faction. So if Cephid 17 gets a wave of pirate activity, local authorities will pay you X amount for every light pirate craft you destroy in their system and Y for every heavy vessel. So whenever you encounter a pirate while passing through there, it is not just a danger but an opportunity as well.
Deus Siddis
Elite
Elite
 
Posts: 1362
Topics: 13
Joined: Sat Aug 04, 2007 8:42 am

Re: SPEC

Postby klauss » Wed Jul 03, 2013 6:04 pm

Deus Siddis wrote:
  • They can't be encountered in space, only bases.
  • They aren't derived from the local game state. (A pirate entering the system should spawn a bounty mission; instead it is the bounty mission that spawns the pirate.)
  • Consequently, the local game state imbalances mission difficulty. (As in the example, the local militia swarm and kill your bounty for you.)


Well, lets fix that.

At least two and three should quite straightforwardly fixable in mission computer code, we'd add mission guards, that prevent missions from being generated in systems that don't support the right conditions.

Remember: I want tickets ;)

The first one is quite possible as well, as demonstrated by campaign missions in the privateer mods. It just requires a bit more of coding and art (as in scripting art, knowing when to trigger them and how to code that trigger).

The way to go about it, is having some veteran (me) do one mission, and in the process document, refactor, and improve for modders. The immediate gain (the mission) encourages the veteran (me), and the improved code base (because a veteran made it), in the long-term, pays off additionally in further contributions.


Deus Siddis wrote:What about the dynamic universe, isn't that mostly python scripts? Perhaps this same strategy of simultaneously enhancing, refactoring and documenting could be applied to parts of that?


Indeed quite possible. Except I'm just one man.

Deus Siddis wrote:Like have the AI comply with pirates and dump its cargo into space, so that when the militia chase away the pirates, you can swoop in to take some of the "abandoned" cargo.


I think that's already the case. Except pirates rarely stop shooting before they blow someone. AI needs lots of tweaks.

Deus Siddis wrote:And instead of ships always exploding when they take critical damage, maybe they could be just as likely to become disabled,


Indeed I was already thinking about that.

Deus Siddis wrote:So if Cephid 17 gets a wave of pirate activity, local authorities will pay you X amount for every light pirate craft you destroy in their system and Y for every heavy vessel. So whenever you encounter a pirate while passing through there, it is not just a danger but an opportunity as well.


Mission guards. That's a simple way to get that.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
User avatar
klauss
Elite
Elite
 
Posts: 7243
Topics: 55
Joined: Mon Apr 18, 2005 7:40 am
Location: LS87, Buenos Aires, República Argentina

Re: SPEC

Postby TBeholder » Wed Jul 03, 2013 7:53 pm

Deus Siddis wrote:Decreasing the spawning of civilian craft (and perhaps increasing the loiter time the civilian AI spends at destinations) could be an immediate fix.
Also true.
Deus Siddis wrote:I think the problem is more that:
1) There are not enough types of things that happen in space (basically just fighting and grabbing cargo).

A low hanging fruit here would be asteroid mining, since it seems the engine already supports this, only some content is missing. A more in depth example is disabled/derelict ships that you could find and bring into your fleet and refurbish. Another example would be meaningful communication with NPCs, like the captain of a Mule asking you to meet him on board his ship, where you find him on the bridge in the form of a fixer and he offers you an especially high risk/reward mission. Or similarly, attacked ships sending distress calls that pay out to the people who assisted them against their attackers (if they survive). Or marooned pilots floating in space, that pay you to bring them back. Things like this can happen only as a result of accepting a mission at a base but they should happen dynamically in deep space based on what is happening in the simulation.
Well, yes. Missions should be generated for events, rather than the other way around. If we "simulate when docked", let's simulate. With open missions being mostly system-wide and changing gradually as they get picked by NPC or become obsolete, rather than being randomized a landing (which is also necessary for MMO). Besides, this would reduce "I can do this, but would need to pick some hardware on another station, so the mision will disappear" to sensible risk.
Deus Siddis wrote:2) The dynamic universe is calibrated to spawn non-interactive blandness.
Far too many civilians, too many police, far too few dangerous criminals, too few factional task forces pushing deep into enemy territory. Random planets and station types, no uninhabited systems. No economic effects from destroyed infrastructure or delayed deliveries. We aren't using the dynamic universe to anywhere near its full potential.
This must depend on the system's political and economical status, right? At least as in Elite - Anarchy/Martial Law/whatever.
If tied to an economical model, however primitie, this would make some (observable!) sense.
Random basic planets and stations aren't bad anyway, but what the system does for the rest of the universe? Major military base, trade hub, out-of-the-way mining&refining area... Frontlines move, trade routes move, luxury resorts go in and out of fashion... especially when some frontlines move. :)
Deus Siddis wrote: So if Cephid 17 gets a wave of pirate activity, local authorities will pay you X amount for every light pirate craft you destroy in their system and Y for every heavy vessel. So whenever you encounter a pirate while passing through there, it is not just a danger but an opportunity as well.
[/quote] Single-ticket should not be the only mission type. There can be automatic reward missions to whoever fulfilled the condition (Elite-style bounty) and non-limited missions "for all comers" (e.g. Frontier style shortages: "Due to a flood, we need 25 units of Landscaping Equipment - ASAP!" or, say, "We need 300 tons of Emendolya for a half-continent-wide celebration - the offer stays UNTIL TOMORROW!" ;) ) are the most obvious sort.
Last edited by TBeholder on Wed Jul 03, 2013 7:59 pm, edited 2 times in total.
"Two Eyes Good, Eleven Eyes Better." -Michele Carter
User avatar
TBeholder
Elite Venturer
Elite Venturer
 
Posts: 732
Topics: 38
Joined: Fri Apr 14, 2006 7:40 pm
Location: chthonic safety

Re: SPEC

Postby Deus Siddis » Wed Jul 03, 2013 7:55 pm

klauss wrote:At least two and three should be quite straightforwardly fixable in mission computer code, we'd add mission guards, that prevent missions from being generated in systems that don't support the right conditions.


What about actually deriving the mission from the conditions. So the mission computer code looks up if there are ships belonging to factions considered hostile by the faction that controls the local system and then spawns bounty missions for those actual ships.

And so the mission would no longer spawn any ships itself, it only takes advantage of an already existing ship.

Remember: I want tickets ;)


At some point I am going to need whatever permission is required to at least edit the tickets I have created, because as these pile up, it will become necessary to re-prioritize the tickets and reset their target versions accordingly. The order tickets are created isn't necessarily the best order to implement them.

The first one is quite possible as well, as demonstrated by campaign missions in the privateer mods. It just requires a bit more of coding and art (as in scripting art, knowing when to trigger them and how to code that trigger).


It would probably have to be triggered by unit level AI for the most part.

Deus Siddis wrote:What about the dynamic universe, isn't that mostly python scripts? Perhaps this same strategy of simultaneously enhancing, refactoring and documenting could be applied to parts of that?


Indeed quite possible. Except I'm just one man.


That's why we would need to break it up into small approachable tasks to be completed over many future versions.

Like faction ship spawning density rebalance as one task (which I should first take a swing at myself since last I looked it seemed simple).
And then derive faction ship spawning rates from the amount of planets and stations it presently controls.
Another might be making station spawning derived from the presence of certain other planet and station types in the local and neighboring systems.
A follow up task would be giving stations a dynamic output value based on the above criteria and an inventory value increased by output * time and decreased by product sales.
A further follow up would be making stations' product sale prices derived from their inventory value.
And then at some point make planets use the same economic system as used by stations above.
And then derive faction ship spawn rates from relevant inventories.

...or maybe those particular example tasks are too big and would need to be broken into even smaller pieces...

I think that's already the case. Except pirates rarely stop shooting before they blow someone. AI needs lots of tweaks.


At some point we need to start a topic devoted to specific AI improvement tasks...

Deus Siddis wrote:And instead of ships always exploding when they take critical damage, maybe they could be just as likely to become disabled,


Indeed I was already thinking about that.


I guess we should make a task for that too then after some discussion of the specifics...

Deus Siddis wrote:So if Cephid 17 gets a wave of pirate activity, local authorities will pay you X amount for every light pirate craft you destroy in their system and Y for every heavy vessel. So whenever you encounter a pirate while passing through there, it is not just a danger but an opportunity as well.


Mission guards. That's a simple way to get that.


This would be unlike any ordinary mission. It would have no player action defined beginning or end; you couldn't accept it or complete it. The occupancy rate of a local system by a hostile faction would be assessed, if too high a standing bounty would be issued for ships of that entire faction (and publicly advertised somehow) and only once the occupancy rate dropped below a certain level would the bounty be withdrawn.
Deus Siddis
Elite
Elite
 
Posts: 1362
Topics: 13
Joined: Sat Aug 04, 2007 8:42 am

Re: SPEC

Postby TBeholder » Wed Jul 03, 2013 8:03 pm

Deus Siddis wrote:This would be unlike any ordinary mission. It would have no player action defined beginning or end; you couldn't accept it or complete it.
Of course you can "accept" or "complete" it. It just doesn't disappear on either event and is "accepted" by being in a system while belonging to a non-hostile faction. :)
"Two Eyes Good, Eleven Eyes Better." -Michele Carter
User avatar
TBeholder
Elite Venturer
Elite Venturer
 
Posts: 732
Topics: 38
Joined: Fri Apr 14, 2006 7:40 pm
Location: chthonic safety


Next

Return to Engine Development

Who is online

Users browsing this forum: No registered users and 1 guest