Suggestion: Expanded R&D: Time and Setbacks

[Wjolcz]

13 hours ago, Kosmognome said:

And everyone would just warp over it.

 

Unless the much bigger problem of time warping without drawbacks isn#t solved, this will achieve nothing at all.

Warping is a good thing. Doing missions in the meantime is even better.

Rep loss over time could be a really nice solution IMO. Losing rep is not as fatal as losing money. Even if it would be bound directly to annual funds you could always do an extra launch or two to get the rep back on a good level before the end of the year.

[Kosmognome]

I am not saying warping is a bad thing. Warping without drawbacks is.

 

Life Support mods are one way to add drawbacks to warping. I would love to seem some ver light life-support mods in stock (like kerbalism or Snacks). But there are other ways to give time-warp some consequences. For example having to pay salaries for astronauts.

I agree with your general idea that research should take time, and that some more interesting unlocking mechanisms are required. But as it currently stands, mechanics that just take time are useless. because warping is so extremely powerful.

Take mining for example. In theory, it should be desirable to find a good ore pot and mine there. In practice you just drop a drill and warp long enough. if you happen to be in a bad spot, warp some more on higher speed, doesn't matter at all. The lab is another good example, without drawbacks to warping it essentially gets you free, unlimited science. It would be far better if warping had some form of drawback.

 

 

[Sereneti]

i have research the woole tech-tree, but if i want to send a Kerbal to the moon, i take the moon-lander that i had created the first time...

why should i change what works?

i had build a vehicle that can carry X tons of mass to the moon, and land it safely.
maybee i can build it a little cheaper with the new tech, but why should i?
 

[Mr. Scruffy]

@Kosmognome: Yeah, the science lab... it´s just puzzling from a design POV. I mean, first they decide to bend everything over to make it work in an environment, in which time is pretty much meaningless (e.g. to name but one: 1 increasing lump sum for hiring kerbonauts, instead of wages) and then they put it in a part, that basically generates ´points´ (the in-game currency most directly tied to the sense of progression and reward in the entire game) over time.

If the scientists running that station would have to be paid in intervals, and the payment per interval would rise over the timespan of a single deployment (according to a formula, that must be carefully balanced as to not prohibit long-term kerballed missions), there would be:

a) a payoff to the science generation in the lab (first cheap, but getting ever more expensive, if you dont swap crew),

b) a reason to swap the staff out, periodically, and

c) one reason less to have a life-support system with cruel consequences in case of failure and added part-count tedium.

[tater]

Warping has no drawback, period. It needs none. TIME PASSES. That is all the drawback you need. Why does anyone think that there needs to be a drawback to warping? If missions/contracts had reasonable time limits, the warping thing would be self-regulating. That rescue mission you took? That should have a time limit of a few DAYS, not 10 years. That sat launch contract? Maybe a couple months. The Mun science? A few months. Explore Jool? A few years.

The devs have already implicitly told everyone to warp all the time, anyway, as the contract decline involves a rep hit, and you can warp ahead about a week to get new contracts. That means they want people to warp ahead a week constantly (that or they are not good at career game design and did not realize they just created an incentive to warp ahead a week all the time).

Edited May 14, 2016 by tater

[Mr. Scruffy]

Well, contracts expiring would be a drawback to time-warping just as well, just different in kind. Point is: If warping can give you something, it should also cost you something. If there is such a thing as the science lab, which pays out sci-points over time, at no cost (you dont need missions going on for that), there is a balancing issue: Once you have the nodes unlocked, launch a station with lab, feed it with data and hit warp: it will take you a long way through the rest of the tech-tree - and as long as you dont mind a high year-count in your save, it wont make the slightest difference, how long that takes. It almost turns into a tech-singularity kind of thing.

Now, i actually like the way the lab works - dont get me wrong on that, please (it pretty much mimics how the actual process of R&D should happen, as per this thread, if i am not mistaken). I like it not despite but because of timewarp. It just sort of falls flat on its face, because it is pretty much the only time-relevant thing in the game and thus generates a lot (compared to the data required) of science at no cost (past the initial set-up).

I think it´s wierd from a design POV, not because it was a bad idea in itself. It´s more like planting a tree after making sure, the place you will plant it on, is a desert. Tree good, desert bad. In this analogy, tighter expiration dates for contracts would fertilize the ground just as periodic wages for the staff would.

Edited May 14, 2016 by Mr. Scruffy

[tater]

The lab is the only issue, but it's only because science is generic. ISS only generates science that impacts manned spaceflight over time. If KSP science had astronautics, planetary, and medical science points, then new crew parts would use medical science, and the science lab would only generate medical science points, so having a million of them does little good.

[Mr. Scruffy]

Well, it would still make ´medical points´ rather moot, once you have it all set-up, if there is no cost in maintaining these labs (plus now, it seems the kerbals would be an infinite source of data, even aggrevating the problem on that side - while lessening it due to making the currency it produces less applicable, at the same time). I dont mind splitting up science, maybe, but that idea is besides the point. Which is: If time is meant to play a major role in the management game, without timewarp becoming the by-many-dreaded i-win-button, there needs to be some sort of trade-off(s) for having it elapse.

The game should be designed in a way, that makes you make it elapse a lot more than you now do (esp. in the early game), sure. Only then, it becomes meaningful trying to save it - if the difference between ´time-effecient´ play and not caring much about it at all, amounts to landing on the Mun for the first time an hour sooner or later, it simply doesnt matter. In an hour ingame time, you can collect a lot of science at the start of the current game and use that instantly to improve your techs and go to the mun much more comfortably than if you hadnt. With expanded R&D-time, the time-difference between the two would be much greater, as it would take time to develope those techs. By this, the amount of time-difference becomes potentially meaningful. That potential has to be filled with actual costs (in whatever form), though, or else, it matters nill and we might as well leave it with instantanious techs.

Edited May 14, 2016 by Mr. Scruffy

[tater]

A real time mechanic would have KSC with a base budget that is reevaluated every year (or annualized versions of contract payouts) based upon some criteria, so warping without doing anything would not be a path to victory.

Edited May 14, 2016 by tater
typo

[Pthigrivi]

So, can a player fulfill one contract and then time-warp a full year's salary out of that? What's the advantage to just getting the money up front? Why add the step of having to time warp for it?

Like, it seems to me there are some players who like to engage in a particular style of role playing and want the rules of the game to require everyone else to role-play in the same way. Except the solution seems to be to hide rewards behind a time-warp. What will happen instead is everyone will just time-warp for those rewards and won't do anything else in the meantime.

Edited May 14, 2016 by Pthigrivi

[Wjolcz]

2 hours ago, Pthigrivi said:

So, can a player fulfill one contract and then time-warp a full year's salary out of that? What's the advantage to just getting the money up front? Why add the step of having to time warp for it?

Like, it seems to me there are some players who like to engage in a particular style of role playing and want the rules of the game to require everyone else to role-play in the same way. Except the solution seems to be to hide rewards behind a time-warp. What will happen instead is everyone will just time-warp for those rewards and won't do anything else in the meantime.

Minumum amount of completed missions per year maybe? The player would have to show that he/she is actually doing something (researching the tech, missions completed) to get the funding. If they didn't the programs with min repututation required would reset, or sth (that's not really a good way of solving this, but I have no better ideas at the moment).

I mean, you can abuse the current system too. Just accept all the contracts -> build a super expensive rocket -> cancel all -> recover the rocket. Unless you're playing hard/custom you are able to cheat the system.

It's all about balancing the costs and rep hits.

Edited May 14, 2016 by Veeltch

[tater]

"Career."

It is career mode, not long-weekend mode. You can unlock the entire tech tree in less time than it should take to construct a single real rocket. It's not so much about role-playing as it is having to intentionally ignore all the terrible stuff that would otherwise make any sort of role-playing even possible. "Wow,we discovered rocketry 2 months ago, awesome that we're building near permanent bases on the Mun already!"

Balance. A new career mode would not be the time mechanics on top of the lousy career mode that exists now. Everything would be rebalanced. The payouts, the rewards, everything. So any comparison, like the one contract, then warp straw man has exactly zero bearing on how it would actually play.

If rolling out a design from the VAB to launch took a few weeks, for example (including construction time), then the Rep check might require a reasonable number of launches worth of results a year.It would take something squad would be unlikely to do, actual play testing. 

You can also "cheat" the current system by time warping to get new contracts instead of "decline."

Edited May 14, 2016 by tater

[Pthigrivi]

So, lets separate one thing out so we can focus on the issue. Let's just say instead of choosing exclusively from procedurally generated contracts there were a set of missions you could always specifically select like Mun Program or Duna Program or whatever. These programs will necessarily come with advances (some basic amount money to build your rocket) and rewards (what you get for completing the mission). Essentially what we're deciding is how should those payments be structured? Should they come in lump-sums (as they are now) or should they be paid out over time? If they're balanced presumably each program will give you something like 1/12th the amount of money it should take to complete it per month so you have to time-warp before you build. Wont players always just time-warp till the end of the year to get the maximum amount of money? Even if you're requiring them to run 3 missions per year they're still sitting around warping for hundreds of days at KSC between missions. They're also time-warping through tech-nodes, building upgrades, maybe even rocket construction? Maybe this appeals to you for role-playing reasons, it sounds like thats what you're doing already, but from a pure gameplay standpoint it is time consuming. Its a lot of extra player time spent managing schedules, entering and exiting buildings, watching the screen flash from day to night for months at a time instead of flying rockets. Maybe its more realistic, but is it really making the game more fun? Does all that time-warping (and totally re-writing the game) really address the central issues plaguing career as it now stands? 

Though I think the idea could be tidied up by just giving missions funding cut-offs instead of year-end reviews I do think we've boiled things down to the point that we've avoided obvious exploits, which is great! So now it comes down to the basic question Squad was asking back when people wanted KCT to be made stock or when Roverdude was designing the resource scanning system: does it make sense to require players to time warp in principle? Is that really feeding into what makes KSP fun? At what point does the fuss of managing schedules become more distracting and onerous than the lack of realism? Or is that kind of management and planning an important part of understanding space travel? I could see solid arguments either way.

Edited May 14, 2016 by Pthigrivi

[tater]

People time warp all the time. Either time warp is bad, or it is not bad. If warping to Eeloo is OK, then warping the same number of days for any other reason is OK.

So which is it?

I think time warp is a tool, nothing more, and it should be used if it is the tool that is needed. 

Again, the missions need not be structured as they are now, either, so comparing to any contract in the game (99% of which are simply awful) does;t help much. Say any program is given a budget divided into Minmus Month (50 days) chunks. You are right, it could be like a super "Explore the Mun" contract, or instead, it might be procedural/contextual (since the latter is now a thing). So you start a Mun Program, and then the contract office decided contextually that you need Mun missions to flesh it out. It might include precursor mission if you have not done them (say you skipped to the Mun without doing Kerbin Orbit stuff). So it might ask to do a rendezvous in Kerbin Orbit, or a docking (depending on what parts you have), and it might ask you do do that within the next couple months. It might ask for a landing site survey from munar orbit, and that might be a few more months away. The "Program" stuff might have the broad "explore" stuff included, but perhaps it could require doing X of the presented sub-contracts for completion... players choice. So pick some LKO rendezvous stuff, plus a survey of landing sites, plus one to get data from a particular landing site (which might be generated contextually from the orbital survey sites (picking one).

[Pthigrivi]

We're you the one who suggested adding a 'simulation' cost (toggleable in settings of course) to reverts and reloads? That was a great idea.  

[Wjolcz]

What tater said, but also something else is very important: The player would have freedom. They pick the types of missions they are interested in (programs), procedural or scripted, they progress either way. The major difference is the fact that right now in the career you spend a lot of time picking the right contracts instead of flying/warping. In the (our) proposed version of career you would pick a few related contracts and fly and warp instead of wasting your time in the Mission Control, which IMO would be great. The time based budget and tech research wouldn't matter that much since you are warping places anyway.

And again, there would be long term missions and short term missions, so you would have to keep an eye on those simply by visiting the Mission Control every now and then (both types would be available all the time) whenever you need a new job and to show the HQ you aren't just sitting by your desk for weeks.

[Mr. Scruffy]

@tater : I get your point, that timewarp is not an inherently bad thing and thus there is no need to dis-incentivize its use, really. But, imho, that would only give time its ´role-playing´ value, while adding incentives to ´save time´ sort of promotes it to a proper part of game-mechanics. That´s what i´d like it to be, in the end. BUT, i will admit, that baby-steps are probably the proper way of locomotion, here: If we had time elapse semi-realistically modeled for role-playing purposes only first, it would be a stepping-stone to implementing mechanics that carry it beyond that purpose.

The problem with this approach though seems to be, that for players who do not mind that role-playing aspect, really, this stepping-stone amounts to little more than added tedium and higher numbers on the ingame-clock. I can understand, how people with less spare (real) time on their hands, will find that rather annoying than beneficial: "Before, i could build a rocket and just hit launch - now, i have to do a timewarp in-between these two... why?!" (assuming implementation of something like construction time).

And as long as there is no cost to warping, this problem will remain, cause it wont be something that people can strategize and play around with. It wouldnt add anything meaningful to the game - and as such, it might (and even should) be left out altogether in the first place. So this ´implentation for role-playing purposes´ should (nay: must!) be followed up by some meat added to that bone asap, or else, this will make the game worse for many players, simply because it would force them to spend (real) time on something they do not (have to) care about, really. Being left inconsequential, time would be little more than the science-clickfest of now.

I am only okay with this RP-type implementation, because balancing periodic costs is a toughy, in my expectation, be it qualitatively (WHAT should it cost) or quantitively (HOW MUCH of it, should it cost), and should be approached from the bottom (rather set the costs too low, than too high and encroach on the ´correct´ values from below). Also, first all the time-consuming events need to be identified and quantified (e.g.: construction time y/n? how much - depending on what?...), before any costs can be attached to them. By way of having an RP-type implementation first, you´d get a more or less fixed frame of reference to attach the costs to and which they can be balanced around: It´d be like taking one ball out of the juggling act (though, if needed, it can be still tossed back into it occassionaly).

 

EDIT: Like, take construction time. If it doesnt really matter, how long it takes to build a given rocket, than things like say an expandable construction facility would be pointless. What´s the point of being able to construct faster, when time does not matter? With time having costs attached, suddenly ´sub-optimal´ transfers might make sense, too: Say, Kerbals would have a periodic cost (aka wages) that increases while they are on the same mission - now it might make sense to send them to Duna on a non-Hohmann transfer, even without life-support implemented. Also: Leaving that Kerbal stranded forever? Not a good idea, since you´d get taxed for that, periodically, and that tax would be quite high after a while (though should still be capped somewhere). Things like R&D-time (to get back to OP) would be pretty much pointless without cost-to-time (in whatever significant form) and you would very rarely try to accelerate reserach by any means, if you can just accelerate time by the factor of one-hundred-thousand (unless you plan the game to span thousands of years and make research realistically slow by default with huge, like trump-huuuge, modifiers for acceleration methods). We could have staff on the ground and/or maintiance cost, instead of these huge (dito) pay walls for buildings....

Edited May 15, 2016 by Mr. Scruffy

[tater]

What are the limits on time warp time compression mechanically? There are mods that substantially increase time warp (i.e.: for RSS). Why would the player even notice? Hit "build and launch" and, bam! the clock moves forward X weeks. All you'd notice was a different name on the button. I'm assuming a much faster warp, not the watching night to day cycles of the warp to next morning button.

Again, career is supposed to be a roleplaying sort of thing, anyone not wanting to worry plays another mode. What we have now is "career" mode minus anything that might confuse the player into thinking there is any roleplaying or suspension of disbelief at all, lol.

With the idea in this thread (which I like), time certainly needs to matter in general, otherwise it doesn't really matter though, hence the apparent digression into time mechanics. This mechanic alone (time required to research) increases the passage of in game time.

Minus some real time progression, KSP should include more futuristic stuff, since if you can go from no space travel to past what we can do on Earth now (SSTO spaceplanes, etc) in a few months, then after a few years you should have any conceivable tech anyone here can think about.

 

[Pthigrivi]

Yeah that would help as well. Another thing you'd need is a somewhat sophisticated alarm clock to keep track of all this stuff. I don't think it would be smart to make the "Build and launch" time advancement automatic as you might blow through intercepts and other things. You'd want to be able to go to a dedicated schedule window and "warp to next alarm" kind of thing.

 

Does it make more sense to just have the reputation hit and funding penalty happen on the contract deadline rather than yearly? It might be easier to scale between long and short missions that way. Jool missions for instance can take multiple years and yearly assessments might get weird with that. So instead maybe when you select the program it might say "Funding: X/month for 6 months. Deadline: 8 months, Y reputation penalty for failure."

Also: can we come up with fun names for Kerbal months? Jooltober? Krakenbruary?

Edited May 15, 2016 by Pthigrivi

[The_Rocketeer]

Perhaps what KSP needs is a manpower system.

It stands to reason that scientists working full-time in Mission Control can't be simultaneously developing new parts in R&D or planning next month's mission in the Tracking Centre. Equally engineers on the ground can't both be building a new rocket, repairing a destroyed Launch Pad and upgrading the Astronaut Complex.

So, to represent this in an abstract system, the player:

has x points of manpower;
can spend kerbucks to hire more staff, which increases x but decreases kerbuck rewards;
can fire/lay off staff, decreasing x but increasing kerbuck rewards;
must designate manpower points towards departments around KSC.

Departments:

require a minimum manpower designation to become usable;
can be optimally manned, providing dependable standard of usefulness/productivity;
can be undermanned, providing diminishing standard of usefulness/productivity over time (staff become fatigued);
can be overmanned, providing a temporary boost to usefulness/productivity that diminishes over time (too many cooks/bodies underfoot).

Science:

has 3 stages: gathered, analysed, applied;
Gathered science is produced by experiments/mission reports etc, e.g. from kerbonauts.
Analysed science is produced from Gathered Science at a given rate by the R&D centre depending on it's manpower status
Applied science is produced from Analysed Science by the VAB/SPH/test contracts at a given rate depending R&D manpower + Assembly manpower / 2.
(Assembly manpower is a single stat used for VAB and SPH)

Gathered science can only be converted to Analysed science.
Analysed Science can be converted to Applied Science, or sold for kerbucks/published for Reputation (destroyed)
Applied Science is the same as Science Points used against the current tech tree.

Edited May 15, 2016 by The_Rocketeer

[tater]

1 hour ago, Pthigrivi said:

Yeah that would help as well. Another thing you'd need is a somewhat sophisticated alarm clock to keep track of all this stuff. I don't think it would be smart to make the "Build and launch" time advancement automatic as you might blow through intercepts and other things. You'd want to be able to go to a dedicated schedule window and "warp to next alarm" kind of thing.

 

Does it make more sense to just have the reputation hit and funding penalty happen on the contract deadline rather than yearly? It might be easier to scale between long and short missions that way. Jool missions for instance can take multiple years and yearly assessments might get weird with that. So instead maybe when you select the program it might say "Funding: X/month for 6 months. Deadline: 8 months, Y reputation penalty for failure."

Also: can we come up with fun names for Kerbal months? Jooltober? Krakenbruary?

The annual bit is really an abstraction that the player might never see explicitly. You have time limits to do things, and failure to get things done on time negatively impacts what they offer you in budget for a given project. Given the idea i this thread of R&D taking time, with possible setbacks, I think folding in some sort of R&D milestones---in the form of test missions---makes sense, as the point of the career game is really to create novel mission requirements within a context. So you are working on new crew pods, and after some elapsed time (barring one of OP's mishaps) a part test mission is created that is a condition of R&D progressing. Say it's the mk1-2 pod. The mission might be to multipart, like "explore" contracts. Put craft in a suborbital trajectory over Kerbin at an altitude of X (a very high orbit, approaching munar distance). Part 2 is to achieve a velocity in the atmosphere of at least 2200 or something. Part 3 is to recover the pod. (this is sorta like the real Orion test flight). Or it might have a splashdown test (unlike many in KSP, this one would actually be sensible).

[Kosmognome]

On 14.5.2016 at 3:44 PM, tater said:

That is all the drawback you need. Why does anyone think that there needs to be a drawback to warping? If missions/contracts had reasonable time limits, the warping thing would be self-regulating.

What you are suggesting -- having contracts expire -- would be a drawback to time warping. What I mean by a drawback is simply give time warping some consequences. If you do it, something happens - e.g. contracts expire. That would add good strategic reasons to be time-efficient.

Currently, time-warping is a no-brainer. There is no downside to it, no consequence at all. Every mechanic that involves time is made impossible through that, since time itself isn't relevant to anything.Time should be relevant. It should be a resource you manage.

[Ser]

On 11.05.2016 at 7:09 PM, Brainlord Mesomorph said:

At first I thought the answer was something complex involving converting science points into “tech points.” And blah blah blah.  But it’s far simpler than that, you should direct Research and Development.  But it should take time. You should direct R&D the way you do now (in the tech tree), but the results should take time.  They will get back to you.  In weeks or months.  This will lead to pleasant, surprising pop ups “Congratulations, we’ve completed research and development in supersonic atmospheric flight.  These parts are now available in the VAB/SPH.”

Sometimes R&D does not go as planned.  You should have “setbacks”  researching any individual thing should randomly cost more money and take more time.  It would be a lot of fun, hear me out:

Even given an infinite amount of science points, it should take years to clear the Tech Tree. Just in R&D time.

I think it would make the game more realistic AND more fun.

Oh, man. That's almost exactly about the mod I try to develop time to time. Delayed research, breakdowns. Also R&D building damage and random events would cause the loss of some science results forcing you to launch the mission once again. You will be able to start research not having the complete amount of science and "feeding" it with science acquired by subsequent missions. Also the developer team, staff and workload management are among the planned features. The idea is almost a year old, then the development was suspended because the 1.1 release was announced with major GUI changes. Now it's about 10% ready for the beta.

Edited May 17, 2016 by Ser

[Sereneti]

The big problem is:
a system without time-warp ned an FTL engine to go interplanetary, even to the moon...

 

[Wjolcz]

I mostly agree with @tater but I am concerned with the concept of building and rolling out the vessels/rockets. It's realistic (and I'm all for as much realism as possible), but it's too limiting IMO. For example: I love to just build things and test them out, then recover (no reverts). The problem is if I build a rover and want to launch it, it will take (more or less) a week to 'roll' it out. Then if I wanted to tweak it, it would take another week, or so to build it again. Unless the game had some sort of rcognizing method of what kind of tweaks were made it could get really complicated and going interplanetary could be a bit problematic if someone missed the window because he was driving rovers around the KSC. IMO because of that building and rolling the rockets out should be instantaneous just to keep things simple.

Aaaaaand I completely forgot what else I wanted to say so expect a shameful act of multiposting performed by me in a few minutes!