[Career Mode] Contract Proposals

[tetryds]

To complement the Analogy with Race-Game you might want to learn about "Rubber band AI"

http://tvtropes.org/pmwiki/pmwiki.php/Main/RubberBandAI

Thanks a lot for that

[regex]

/!\ For clarity what I meant by Fixed-Budget, it should answer most of your question :

- Every month (example) you are given a fixed amount of money to spend for ANYTHING. Mission or not.

Noted. I assume that career mode will never have a fixed-budget and that the player is solely responsible for generating their income through direct actions within the game.

I interpreted "Do missions" as : mission-based progression. But alone it's not enough to make sense of it. As the current BETA-Career demonstrated you can have a technological progression without any mission, but right now it's not interesting enough.

What I mean by "do missions" is the classic paradigm of missions being presented to the player and them choosing one to do. The mission consists of restrictions, objectives, and rewards. By doing missions the player progresses within the game, whether through the accumulation of points (money, reputation) or the story. Since KSP is more of a sandbox, I don't expect story elements to be present (unless we're talking a "Dwarf Fortress" "you create the story" kind of thing). Missions would be presented based on some metric of the player's progression.

In that sense, this system doesn't differ all that much, it just puts the defining of objectives (and, to some extent, restrictions and rewards) in the player's hands.

* Reputation-based progression is an interesting mechanism depending on how it's done.

That's why I think SQUAD is taking that route. I simply tried to use planned "features" (well, more just a word I heard... ) in my suggestion here.

Put aside that it mean you could pretend to fail mission repeatedly to grind money, I assumed that you necessarily got 100% of the Money required to accomplish the task and kept the change in you general fund. It doesn't change my criticism.

Certainly, and it is a balancing concern. This is why I suggest no rewards or even negative rewards for mission failure, or perhaps a cascading failure mechanic where if you fail too often you are denied any contracts until your general fund is below a certain level, or something. You are correct in that the front-loading of rewards is the biggest balancing concern of this system.

The problem is that you get a recursive loop where a 100% success rate get you exponential amount of money, and suggested restriction leading to absurd situation like : "You can't refuel/save the Kerbun from the earlier attempt or you don't get as much money. (Ps: let those looser die !)"

There's no need for that sort of arbitrary restriction, even with a conventional mission system. If the player is that good, why not provide them with additional gameplay challenges like life support, reentry mechanics, or even random part failure. There are lots of additional things to spice up the game and make it challenging besides the monetary system. You could even have "hard economy" where the player received reduced rewards, or maybe a "no immediate repeats", or something.

If you balance against the most skilled player you end up leaving the newbie far behind.

the tech-tree is I hope a place-holder.

I have a feeling your hopes will be dashed, but there are mods to "correct" it.

To get sample from several part biome/planet of the Joolian system (and further Gaz Giant) efficiently you better need an infrastructure

You don't need an infrastructure to do that now, and it will still be possible afterwards I'm sure. Infrastructure will just take away some of the tedium.

The Negative Feedback I talked about is more to keep budget from escalating out of control.

I'm not really worried about that. There are already a few suggestions in this thread as to how to keep rewards under control and it is a problem that will be faced by the KSP devs at some point as well. Money and reputation sinks could be implemented to help wit the issue but I feel like that is a higher-level thing than we are discussing here.. Certainly it should be acceptable for the player to fail on occasion, but repeated failures should have some sort of penalty to make the game actually have some risk and meaning.

[regex]

To track the player experience you need very long data collecting.

There's no need for metric gathering to determine how "skilled" the player is and then adjust AI accordingly within KSP, you are assuming a much more complex system than I envision. This is simply a "build your own mission" system. You can control what objectives are available to the player using a single number, reputation (already mentioned as a future metric by SQUAD), and reward the player with additional reputation as they complete objectives. I assume that SQUAD's upcoming "Contract" system will work in a similar manner, although I could be quite wrong.

Under a normal mission system you would restrict what missions were offered to the player based on some metric and then unlock future ones based on their progression through that metric. Under this system you are doing exactly the same thing, you are just letting the player define the objectives and front-loading some rewards in order to "fake" a governmental proposal system. It's pretty simple at its heart.

[tetryds]

I just had the idea where you can pick more than one mission at once (like 3 or 4), but then have to complete ALL of them to receive your reward

They would be generic small missions, like "Land on Mün", which would disable or drop the reward for any other landing missions.

Then, landing at Mün or Duna at the begin would give about the same. (hold on before shouting at the monitor)

When you progress in the game, the missions you've remade so many times would drop it's reward, on a damped exponential way, like science.

You would also receive less money for different missions regarding the same planet or moon, to encourage and reward complex missions.

The further you progress, the further you would have to go, but if you settle up with making mediocre missions, that is okay.

But your ships will get more parts, and get more complex, thus more expensive.

You will need to go after more money, which will mean go after another planets and moons, with more missions to complete at once.

If you want to "Orbit Vall", "Land on Laythe", "Plant a flag on Bop" and "Return Kerbin" your estimated payment will go high.

And you not forced to use a single ship to do that, a relay of space stations to refuel, sending refuel for a burn back home or even the mission to send a fuel tank to your space station would count.

You would, of course, have to manage what to do, and how to do, to not overcome your current cash ammount and go broke without finishing the mission.

And it would be a good idea to link those objectives to a kerbonault or command module yet to be launched from Kerbin, to avoid grinding.

Any system we idealize will, of course, require balancing, but this way would make things a lot easier.

Out of topic, i can clearly see this working together with the "ships take time to build" idea too.

[regex]

tetryds that's pretty much how I think about this. You could have pre-generated missions, a "random mission" button that used your current reputation to construct a mission, and you can allow the player to define what they want to do. Having a general fund allows the player to build and fly craft outside of the mission system, perhaps for infrastructure or just because they want to land a ROCKETSHIP on Duna while they're drunk. You would also need to link a craft built using the contract's budget to that contract until it was completed (as in, it couldn't be used for a different contract). There some other interesting mechanics in the meat of this thread to help balance and make it interesting.

Out of topic, i can clearly see this working together with the "ships take time to build" idea too.

lol, it would work with any mission system, but that's neither here nor there.

[LethalDose]

Nothing else to say?

Just what's below:

I've sufficiently assuaged your concerns about player agency?

Not even remotely, nor did I ever state anything even close to this. This is putting words in someones mouth, which is something you seem to take offense to, but don't seem to mind doing to others.

For clarity, not once have you come close addressing the following problems:

1. Justification of increased complexity
   
2. Lack of a real problem to be addressed by the system
   
3. Explanation of how this system increases agency
   

The idea is crap for multiple reasons, regardless of what you think. If you were able to succinctly address any of the points above to justify its inclusion, you'd had done it by now.

Well, since there doesn't seem to be "any debate about this at all" I ask that you take your insults and run along. Please leave this thread to those who want to discuss the suggestion.

I've tried to discuss the suggestion, but you're not interested. You accuse others of misquoting you, then misquote them. You accuse others of insulting you, when you insulted them. You rely on misdirecting rhetoric instead of simply addressing the issues brought up.

So yes, I'm leaving, because attempting to discuss with you is a waste of time.

[regex]

Justification of increased complexity

The complexity is imagined. This system is only marginally more complex than a procedurally-generated mission system and only then because of the additional GUI elements needed, which are pretty minor. In fact, because of the need to not have a random generator, I would argue it is as complex as a procedurally-generated mission system, and brings along all the balancing headaches thereof.

Lack of a real problem to be addressed by the system

The problem could be imagined, but I hate having to do randomly generated "cookie-cutter" missions where you "take x to y, receive z". This takes the random out and puts the player in.

Explanation of how this system increases agency

Right above this quote. A random, procedural, or story/progression mission system does not have the freedom that this one does, and this has the side benefit of being able to produce two of those three with a minimum amount of additional of code.

The idea is crap for multiple reasons, regardless of what you think. If you were able to succinctly address any of the points above to justify its inclusion, you'd had done it by now.

I just did, and I have done so multiple times within this thread. I believe you are needlessly inflating the complexity of this suggestion and have not sufficiently shown me how it is any more complicated than stated in any of your replies.

[tetryds]

This is my last reply to this thread.

The problem could be imagined, but I hate having to do randomly generated "cookie-cutter" missions where you "take x to y, receive z". This takes the random out and puts the player in.

The system i presented clearly puts the player completelly under the control of the missions, and there is no complex structure to prevent them from assigning insane missions

If does, might be based on the science level or already made missions, which is directly proportional to how skilled the player is.

And

There is absolutelly no issue in "take x to y, receive z", that is a game structure which fits inside this game and its purpose, assuming the freedom is kept.

So, basically "take x to y, receive z/take a to y, receive d/take f to t, receive r or take x to t, receive w? the best option is up to you" is just what KSP is about.

[regex]

The system i presented clearly puts the player completelly under the control of the missions, and there is no complex structure to prevent them from assigning insane missions

The system you suggested is almost exactly like the one in this thread, but without the front-loading of budget to make the appearance of a governmental proposal or the grouping of objectives under one umbrella.

There is absolutelly no issue in "take x to y, receive z", that is a game structure which fits inside this game and its purpose, assuming the freedom is kept.

I don't have a problem with it either, I just want to define x, y, and z for myself, not have them randomly generated by the game.

[Kegereneku]

Well that's also going to be my last reply because you don't seem to get my/our concern about your proposal.

KSP is not a "simple game" that can work with a conventional mission system, I though you knew considering how afraid you were afraid of a linear progression. But you now deny the huge complications inherent to your proposal. Again : "You are making a problem out of perfectly normal way of playing" ...and now hide being the fact that SQUAD may work on a "mechanic" of the same name (which may have nothing to do as all with how you see it).

Plus your answers are contradictory and incoherent, ONE example amongst others : You suggested keeping the player from using a ship other the one contracted to accomplish an objective, then you say there's no need for that sort of restriction. Then you went further talking of refused-feature like Life-Support, Deadly-reenty...etc as balances mechanisms. What is that supposed to mean ? That if the player's budget skyrocket too high for its reputation Life-Support will spontaneously activate ?

FPS can generate "Easy mode" and "Hard mode" easily because they are intrinsically simple at their bases (damage vs hit-point vs time), KSP have too many variable for that

This is not a question of "balancing against the most skilled player", it's a question of keeping the progression system coherent regardless of the skills of the players. Plus, from what I could judge you are balancing against unskilled player under the reasoning that they would be I quote "refusing to advance".

In the same vein and even if you'll deny wanting that, you advocated that there's a "right way" to play the game that should be encouraged : Linearly, Mission after Mission, Keeping infrastructure as a minimal because "You don't need one anyway" or "Not that important".

Don't tell me your system "let player play how they want" since the beginning I (we) kept telling you how it don't.

In any case, I like the idea of player interacting with the mission-generator to create/influence game-generated missions. But I disagree with most of what you suggest as a "Contract Proposal".

[KASASpace]

I would think that it should really be like this:

1. You go to mission control to look for contracts

2. You find a contract to land on the Mun

3. You give an estimated budget with a maximum number (to lessen "cheating")

4. You now have three sub-budgets, development, actual mission, and the misc. budget, if going overbudget for other two, misc helps out

5. Now, you have to have at least ONE successful test flight (things work, doesn't have to get to said destination)

6. Now, if flight was successful, and you can do good stuff with the money you have to handle (good rep) you get a budget to engage the mission

7. Now, you do the mission, land on the Mun. But there are other goals you can add too

8. Now, if you get back, and have a complete success, your rep is increased, and more contract "slots" are available

9. I mean that you should be able to have more than one contract at a time, but a limited number, and you use separate funds depending on the craft

10. Now, once you have enough rep, you can start a research budget, this will allow for part upgrades later on, giving better stats to rockets

11. When you start a "program" to accomplish multiple things, you have to set a schedule and a time period that it will last(this is not a mission, it is a long list of missions)

12. these programs require "proof of concepts" which are paid for with the research budget. once go ahead is given, you can begin building a large spacecraft in orbit to go to duna or something

Of course, that's only SOME of the stuff I have up my sleeve!

[mcbmaestro]

My personal thoughts are very much the same as yours, KASASpace.

First, I'd like them to keep things simple, there shouldn't be facility upgrades or monthly operation costs, who wants to deal with that tedious stuff.

So basically, I would think generated missions could be cool if they worked right. The game could analyze what science you already have, then say you missed a crater on the Mun, the game could offer it up as a contract mission to get the missing science. When all the science on a planetary body is finished or negligable, the mission generator would see that and pick a fresh one.

Mission specific parts category. This would be parts that could be unlocked in career mode by accepting missions. They would be things that would be a burden, like a heavy satellite that does nothing but sucks up power, or maybe your required to deliver rover. I think these parts could be interesting for forcing you to be creative with how to get mission parts into your designs, but the reward would be money and science.

[KASASpace]

My personal thoughts are very much the same as yours, KASASpace.

First, I'd like them to keep things simple, there shouldn't be facility upgrades or monthly operation costs, who wants to deal with that tedious stuff.

So basically, I would think generated missions could be cool if they worked right. The game could analyze what science you already have, then say you missed a crater on the Mun, the game could offer it up as a contract mission to get the missing science. When all the science on a planetary body is finished or negligable, the mission generator would see that and pick a fresh one.

Mission specific parts category. This would be parts that could be unlocked in career mode by accepting missions. They would be things that would be a burden, like a heavy satellite that does nothing but sucks up power, or maybe your required to deliver rover. I think these parts could be interesting for forcing you to be creative with how to get mission parts into your designs, but the reward would be money and science.

That's a good idea, the mission-specific parts. Like only a certain type of engine for getting a rover to Duna or something, and only certain cores and whatnot.

[Capt Snuggler]

That's a good idea, the mission-specific parts. Like only a certain type of engine for getting a rover to Duna or something, and only certain cores and whatnot.

i agree with this but it would have to be for loosely technical reasons. i think it would be frustrating to have arbitrary illogical restrictions pushed upon you.

"use a certain type of engine to get to duna, because its harder" this is not a fun mission parameter. something like "get our crazy fragile rover to duna in one piece" would be more fun.

[tetryds]

I remember that at some dev weekly they mentioned it will be like small tasks.

Then you would pick these tasks at will, and make your mission.

But if you get too many and dont accomplish you lose reputation.

That way seems fair for me (it's what i suggested lol).

[Capt Snuggler]

I remember that at some dev weekly they mentioned it will be like small tasks.

Then you would pick these tasks at will, and make your mission.

But if you get too many and dont accomplish you lose reputation.

That way seems fair for me (it's what i suggested lol).

i guess that seems ok. its hard to tell really but i would like to see some hand written contract type jobs whit a bit of character. stuff like "take VIP to here", "deliver our sub-assembly there", "take photo of these" etc etc. each with mission specific rewards like money, parts, rep, science maybe even spacial kerbals to have on the crew?