Sign in to follow this  
P.Lumumba

From Long to Hard Mode

Recommended Posts

What are your thoughts & ideas about the difficulty settings ?

[For starters, KSP is great stuff, despite anything below.]

I just passed the early stages of my hard career game. (stock, standard hard settings)

Did one Mun landing and a grand tour of Minimus.

After that I wanted the final upgrade of R&D, so I did a ton of missions, and got there in the end.

Anything interplanetary is still to come.

I have two complaints about career mode in it's current setup :

  • When I finally had the 6.000.000 to upgrade, I also had about 8.000 science sitting in the bank.
    Plenty to complete the tech tree.
    No need to drain Kerbin, or the Mun. No need to visit other planets.
    (I did use the administrative building to transfer rep into science at 20%, but I learned my lesson now)
  • You don't set how difficult your game is, you set how long it will take to get the money.

Squad seems to be working on balancing science & research, so enough about that.

I don't remember reading anything about them planning to rework the difficulty settings.

Share this post


Link to post
Share on other sites

Rep into science was all you did? Heh. I got 5000 science from a single mission only to get a station into solar orbit using a Funds -> Science strategy at 50%. I've yet to actually unlock the last tier of R&D, though, so I can't use it yet, ahaha.

And yeah, you're right in that at the moment the difficulty setting is more of a cash controller than any actual difficulty setting. Something that could be done to remedy that might be to make contracts require more precision in their completion, or just simply biasing the contract generation towards creating more difficult contracts slightly earlier on.

(Also, since this concerns the game's development, I've moved it to the Development Discussion section.) :)

Share this post


Link to post
Share on other sites

Do the strategies only take science from contracts?

I am sure they do, but the thought crossed my mind right now.

Next time I start a new save I would have installed a mod to kill science from contracts - but that would kill some strategies too?

Anf while we are talking about it: Does the difficulty setting affect collected science or contracts only?

Edited by KerbMav

Share this post


Link to post
Share on other sites

I remember in my old games that the funds -> science! strategy only worked for returns and not advances, but in my new game it seems to be both. (WHY did I get rid of that strategy? What was I thinking? :confused: Must get it back)

Share this post


Link to post
Share on other sites

I think one way of increasing the amount of different missions required on all levels, and to have difficulty scale somewhat, would be to remove science from random contracts entirely, and do away with that broken outsourced R&D strategy. Science from any scripted, one time only contracts such as altitude records and achieving orbit would still be allowed however.

The idea would be to pushplayers to go further form home in order to advance in tech, and higher difficulties, where there is less science avaialble would require lower tech solutions. As long as the total science available in the solar system is signifcantly more than you need to complete the tech tree then it wouldn't be overly restrictive.

Share this post


Link to post
Share on other sites

I've been saying this since difficulty settings came in. It is not difficulty, just a "grind" slider. Difficulty would require changes in gameplay difficulty.

Share this post


Link to post
Share on other sites

The career mode is no longer about grinding as much science as possible. Science experiments are now pretty much useless, as you'll get enough science anyway by flying the missions that give you money. Instead of first grinding science to unlock all possible tech tree nodes, and then grinding money to upgrade the R&D, you should concentrate more on playing the game.

Early game is the missions you fly with tier-1 R&D. This is the part of the game, where you explore Kerbin, Mun, and Minmus.

Midgame starts when you upgrade to tier-2 R&D. Now it's the time to go interplanetary. In my current game on Moderate, I completed manned missions to Duna, Ike, Gilly, Pol, and Bop, and sent probes to Eve and Dres.

Endgame starts with tier-3 R&D. This is the time for ambitious missions. In my current game, I have exploration contracts for Laythe and Vall active, and I'm considering starting a full Jool-5.

After you have unlocked the tech tree and upgraded the facilities, the game is effectively over.

Share this post


Link to post
Share on other sites
I've been saying this since difficulty settings came in. It is not difficulty, just a "grind" slider. Difficulty would require changes in gameplay difficulty.
You don't actually have to grind though, you can get at least 4 million credits on hard difficulty, by simply following through the scripted altitude-orbit-exploration contract line, while not doing any random contracts at all. Played like this, hard does actually pose a few chalenges during the early game.

Grinding is a choice rather than necessity.

The problem isn't even that grinding is a possible way to get through the challenge, that seems fine for this type of game. The problem, is that grinding is probably the 'optimal' play for maxing out as soon as possible, or if it isn't, its very close, and its certainly the easiest path!

Contracts within the Kerbin SOI simply give way to much of everything relative to a more challenging contracts outside of it.

After that I wanted the final upgrade of R&D, so I did a ton of missions, and got there in the end.

Anything interplanetary is still to come.

This type of situation really ought to be the exception rather than the rule.

To avoid this the payouts in the Kerbin system (except those obviously brokenly low) need to have a massive cut in ROI so that interplantery missions offer better returns. Perhaps even apply diminishing returns down to a reasonable minimum, on all random contracts by type and situation, to force would be grinders to either swap the contracts they are grinding, or move to another the SOI for their fix.

Edited by ghpstage

Share this post


Link to post
Share on other sites

Doing something repetitive and boring IS grinding, that's what grinding is.

The KSP reward structure is unlocking stuff, so you unconsciously want to do that. The science needs to be more useful, not just as points.

Share this post


Link to post
Share on other sites

IMO a higher difficulty level is one that leaves less margin for error.

Not allowing to revert fits in well, no missing crew respawns too, but I'm not convinced about the diminishing returns of missions & inflated building prices.

My ideas so far:

  • I did very much enjoy being challenged to stay under 18tons at first, and under 30 pieces later. (You can do a lot with just 30 pieces if you have to)
    Instead of the building costing more, maybe they could make them more or less restrictive depending on the difficulty level.
  • Lower the building costs & make research more expensive, forcing you to pick to parts you really need.
  • Maybe they could change the pay up front & on completion ratio depending on difficulty, so successive failure will take you down faster on the more difficult settings.
  • Autoaccept scripted missions & limit the timeframe. (maybe after every [yourNumber] other missions)

You don't actually have to grind though, you can get at least 4 million credits on hard difficulty, by simply following through the scripted altitude-orbit-exploration contract line, while not doing any random contracts at all. Played like this, hard does actually pose a few chalenges during the early game.

Grinding is a choice rather than necessity.

The problem isn't even that grinding is a possible way to get through the challenge, that seems fine for this type of game. The problem, is that grinding is probably the 'optimal' play for maxing out as soon as possible, or if it isn't, its very close, and its certainly the easiest path!

Contracts within the Kerbin SOI simply give way to much of everything relative to a more challenging contracts outside of it.

This type of situation really ought to be the exception rather than the rule.

I agree.

I agree that grinding was my choice, but it seemed like the reasonable thing to do.

The game has two basic things to max out: tech tree & visit all planets/moons. (preferably with your orange suited kerbals for max XP)

Doing the long trips is much easier with the best gear, so I wanted the tech tree maxed first.

By far the easiest money was doing all kinds of scans & surveys of Minimus (half of them while I was there doing a science tour anyhow...)

Those missions paid almost as much as the scripted Duna, Ike & Eve missions, so yes, it seemed like the reasonable thing to do...

Diminishing returns sounds good as a counter. (but it would make random station & base building much less likely, and that is how I like spending my profit a space program...)

The science gained from missions should just be cancelled altogether.

Reputation is still a bit of a vague concept to me.

You know how much money you have, and what you can buy with it, same goes for science.

But it's not always too clear how much reputation you have, what it does and what difference it would make if you would get this much more...

Edited by P.Lumumba

Share this post


Link to post
Share on other sites
Doing the long trips is much easier with the best gear, so I wanted the tech tree maxed first.

Isn't that against the point of playing on Hard?

There are really only around two parts that you don't get with the tier-1 R&D that are useful for interplanetary trips. Docking ports make complex missions easier, as you're no longer restricted to linear staging. Nuclear engines reduce rocket size significantly, if you're going beyond Duna and Eve. Everything else just adds more variety to your missions.

Share this post


Link to post
Share on other sites
Isn't that against the point of playing on Hard?

There are really only around two parts that you don't get with the tier-1 R&D that are useful for interplanetary trips. Docking ports make complex missions easier, as you're no longer restricted to linear staging. Nuclear engines reduce rocket size significantly, if you're going beyond Duna and Eve. Everything else just adds more variety to your missions.

This is loose from the difficulty settings, it's just logic.

The point of playing on Hard is the game being more challenging. (Not making harder challenges for yourself within the game)

If all you need is an internal challenge, you don't need more that a sandbox.

I enjoy (& expect) imposed challenges in a game...

Share this post


Link to post
Share on other sites
You don't set how difficult your game is, you set how long it will take to get the money.

What is difference between laborious and difficult. In most cases it is unpredictable events. Real world is difficult because there are not predefined rules that when you make certain things you get some amount of money and "research points" and when you have some given number of them, you achieve some known destination. Real research, planning, manufacturing and flight are full of surprises. Sometimes they are good and sometimes nasty. Difficulty is to react them, learn from them and trade between risks and costs, both of which are typically only partially known. They are very difficult to simulate in computer program. You could randomize some events, but in any case there are very limited number of preprogrammed possibilities, which are known and their probabilities are known.

And it is not all. The most difficult things are human interactions. They are impossible to simulate in computer program and even humans can not predict them. They happen on all levels, workers make mechanical mistakes, managers leave and new have different opinions, public reactions to failures are totally unpredictable and politicians have always short electoral term (relative to development processes of significant space missions) after which budgets have to negotiate again etc. There are infinite amoun of effective variables and infinite things that can change.

Share this post


Link to post
Share on other sites

I've been playing on moderate difficulty cash settings, purely to reduce the building upgrade grind.

I like working within tight mission budgets: that's interesting and fun. I don't like having to launch thirty satellites or fly spaceplanes to Minmus a dozen times in order to unlock RAPIERs and action groups.

Even on moderate, it's still tedious. And I've been spending so much time raising cash that I haven't gone anywhere past Minmus except with probes that are still in transit.

Share this post


Link to post
Share on other sites
What is difference between laborious and difficult.

Let's use maths as an example (100% predictable [at least the little I know is], as I want KSP to be):

  • Easy : 2 + 3 + 8 + 12 -6 * 2 =
  • Hard : ln(e/4)^(2/3) =
  • Laborious :
    2 + 8 *4 - 6 =
    4 + 6 / 3 - 5 =
    8 - 2 * 5 - 1 =
    12 + 3 - 1 =
    ...

In most cases it is unpredictable events. Real world is difficult because there are not predefined rules that when you make certain things you get some amount of money and "research points" and when you have some given number of them, you achieve some known destination. Real research, planning, manufacturing and flight are full of surprises. Sometimes they are good and sometimes nasty. Difficulty is to react them, learn from them and trade between risks and costs, both of which are typically only partially known. They are very difficult to simulate in computer program. You could randomize some events, but in any case there are very limited number of preprogrammed possibilities, which are known and their probabilities are known.

And it is not all. The most difficult things are human interactions. They are impossible to simulate in computer program and even humans can not predict them. They happen on all levels, workers make mechanical mistakes, managers leave and new have different opinions, public reactions to failures are totally unpredictable and politicians have always short electoral term (relative to development processes of significant space missions) after which budgets have to negotiate again etc. There are infinite amoun of effective variables and infinite things that can change.

There have been a few suggestions here that would make the game more difficult without the need to randomize, or simulate human interaction.

I don't think the game needs random events to up the difficulty.

New player (easy) don't seem to be struggling that much to make it to the moon eventually, but forcing them to do it with only 30 pieces might cause some headaches.

Or forcing them to get it right on the first or second attempt.

Or [insert your idea...]

Share this post


Link to post
Share on other sites
I like working within tight mission budgets: that's interesting and fun. I don't like having to launch thirty satellites or fly spaceplanes to Minmus a dozen times in order to unlock RAPIERs and action groups.

Even on moderate, it's still tedious. And I've been spending so much time raising cash that I haven't gone anywhere past Minmus except with probes that are still in transit.

I'm also playing on moderate, and I'm currently building a Jool-5 ship, without having unlocked RAPIERs or action groups.

The current system, where facilities are expensive but missions are cheap, works quite well with many different playstyles. You still need money for something, but you can build your rockets the way you like them, and play without reverts/quickloads if you choose so. If you make mission budgets tight, you basically force everyone to play with reverts and quickloads and to build either fully recoverable rockets or rockets with cheap SRB lower stages. This is a huge improvement over 0.24 and 0.25, where money either didn't matter at all, or you were forced to abuse the silly recovery system.

Share this post


Link to post
Share on other sites
Let's use maths as an example (100% predictable [at least the little I know is], as I want KSP to be):

  • Easy : 2 + 3 + 8 + 12 -6 * 2 =
  • Hard : ln(e/4)^(2/3) =
  • Laborious :
    2 + 8 *4 - 6 =
    4 + 6 / 3 - 5 =
    8 - 2 * 5 - 1 =
    12 + 3 - 1 =
    ...

It seems that you mean learning. Hard becomes laborious when you learn to do it. Your example is perfect. I have studied math so much, that I do not feel power or logarithm functions difficult. But they can be laborious if I have to calculate their values by pencil and paper or program microcontroller without floating point arithmetics.

But it is also difference between computer simulation and real world. There are quite small number of things and interactions in the game and you can learn them by playing couple of hundreds of hours. And on the other hand, possibilities in real world are practically infinite and continuously changing and you can not learn everything during lifetime.

New player (easy) don't seem to be struggling that much to make it to the moon eventually, but forcing them to do it with only 30 pieces might cause some headaches.

Or forcing them to get it right on the first or second attempt.

Or [insert your idea...]

I suggest that difficulty have an effect to ISP values, joint strengths, electricity and resource consumptions, aerodynamic effects etc. But even then you learn how to do things after some trial, error (or even calculations) and it becomes work. And if you make game too difficult, no-one have interest to study degrees in space engineering to play (and even then you can not plan whole mission but instead small detail of some component in spacecraft). Even real research work is 99 % pure boring work and 1 % fancy innovations and learning, which converts difficult things to laborious.

Share this post


Link to post
Share on other sites

I'm agreeing with you that the builing costs are to high (at least in comparision to the other ressources) and the science return from contracts is way to high at the moment.

But to get to the discussion how to make the game harder (but longer). I do not really like the idea of random events. How would you like to couple some techs to specific missions. Like (not the best example) you only get the idea to develop ladders after the first extraplanetary landing and notice how useful they would be.

Some could be for the LV-N, you only get the idea after your first interplanetray flight

Heavier rocketry only after launching a vessel with more than x mass...

RTG only after mission to the outer system (less solar power available)...

...

Share this post


Link to post
Share on other sites

I started a fresh career (on normal as a benchmark) with 0.90 release. I've only been playing since some point in the late summer, BTW, I'm a noob. It was trivial to upgrade/unlock everything, and I used no "strategies" as one, I'm sort of confused by them, and two, they seem overpowered. I'm not yet even to a good Duna Launch window, even with liberal use of time compression. So 1955 to 2055 in ~100 days kerbal time.

Looks like the tech tree needs warp drive or something, so I can work towards something before the kerbal year is out.

Share this post


Link to post
Share on other sites

I've got the same problem as Tater. Time is a weird thing.

My launch window to Duna is ~160 days out and I'm on tier 2 of the Tech Tree. I'm pretty sure I'll complete it just motoring around killing time and doing contracts long before the window even opens up. I tend to be a bit obsessive about 'efficiency', so I see no reason not to use all that time to my advantage before we even leave for the Solar SOI, though I do need to go stick a habitat out there.

I'm somewhere between a mediocre and a decent player. I still crash space planes, I screw up my dV needs, I poorly estimate the dV needed to 'bounce' between landing sites, etc. Hard... isn't. Some of the limitations are fun, like 30 part/18t trips to Mun. I really don't have much reason to leave Kerbin's SOI though to reach 'endgame'. I've also got all the time in the world. I can launch 12 rockets a day if I want, so there's no real feeling of pressure. The only things that 'pressure' my time are my own maneuver nodes.

I'm not quite sure how to fix that though. How do you make it so that Duna, 170 days away, makes SENSE as the next exploration target, and yet you can't complete the entire tech tree and fully upgrade before that launch window even arrives? Okay, yes, you CAN launch to Duna earlier if you want. Heck, you can Jool-5 on day 1 if you launch enough dV into orbit. That's not Hard though, just time consuming.

Share this post


Link to post
Share on other sites

Yep. I listened to kerbalcast for the first time in a while, and they were talking about how "hard" 0.90 was on career. I'm baffled. I don't revert or use quick saves except if there is a bug (I had a ship stick to the launchpad, and later got it aloft and the payload docked to a minmus station then stuck there, as well (after another one had all the radial parts explode). Anyway, I only revert rarely (like I get to the pad and realize there is crew in an un-crewed launch, or I forgot the solar panels as I look it over on the pad, etc). I had a couple failed launches (no loss of life!), too. I ended up turning a few ships into probes (one a"light" impacter).

I'm in no way being careful about funds, or even thinking much about it. I'm aiming for a day 248-270 set of Duna missions (doing a sort of mars direct thing), and I have everything unlocked except a few plane parts (which I have no interest in), and all buildings fully upgraded (not strategies cause I don't use those, or the runway or hanger, cause I don't use those, either). As of last night I'm day 191.

I think I need Kerbal Construction Time next career.

The FP contracts contain some good/fun stuff, but they are as badly thought out as stock in many cases. Making contracts "hard" for no reason is sort of goofy. Why would I need to observe some random area below 5400m, and others above 7000? Why are the related surface science components not below the area I had to observe (there are like 3 EVA/surface things close to each other, then always a below XXXXm someplace utterly different).

You get the idea.

Share this post


Link to post
Share on other sites

Reducing the Kerbin-Mun-Minmus science modifier might do the trick...

Share this post


Link to post
Share on other sites
Reducing the Kerbin-Mun-Minmus science modifier might do the trick...

More science is coming from contracts than exploration, though.

Personally, I'd nerf Outsourced R&D and buff Fundraising Campaigns and Unpaid Interns all by a factor of ten, heavily reduce the payout for satellite contracts (I can do half a dozen √50,000+ sat missions off a single √15,000 launch) and substantially increase the payouts (in both √ and science) for achieving interplanetary exploration milestones. Incentivise exploration over commercial launch grinding.

I'd also add a couple more steps to each building upgrade sequence. The cost and capability jump between upgrades is huge; it needs more granularity. Give players a sense of continual progress rather than game-halting cliffs.

Share this post


Link to post
Share on other sites
More science is coming from contracts than exploration, though.

Personally, I'd nerf Outsourced R&D and buff Fundraising Campaigns and Unpaid Interns all by a factor of ten, heavily reduce the payout for satellite contracts (I can do half a dozen √50,000+ sat missions off a single √15,000 launch) and substantially increase the payouts (in both √ and science) for achieving interplanetary exploration milestones. Incentivise exploration over commercial launch grinding.

I'd also add a couple more steps to each building upgrade sequence. The cost and capability jump between upgrades is huge; it needs more granularity. Give players a sense of continual progress rather than game-halting cliffs.

I think I agree with all of these propositions.

The game should be pushing us into exploration & gathering science from exploration.

But it seems to be more about the general game than about the difference between the difficulty levels.

I strongly belief that a craft able to land on the moon in easy mode, should be just as able to land on the moon in hard mode.

And should always be able to land on the moon.

In other words, I don't like the idea of adding random failures, nor do I want to see them changing the isp / thurst values depending on the difficulty level.

Maybe an obligation to recover the remains within a certain timeframe when you kill a kerbal [normal] + a hefty fine [hard] ?

Share this post


Link to post
Share on other sites

I would change the science parts so that you gain only negligible science and lose reputation, if you use them without a contract. Science from contracts makes sense, while science from space janitors fooling around and doing unplanned experiments doesn't.

Or maybe you would need a contract or a scientist kerbal on the ship to get any benefit from science parts.

Share this post


Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this