Jump to content

The difficulty settings seem to work well Squad - Nice!


RocketBlam

Recommended Posts

I'm an experienced Kerbonaut. I have played since .23, have finished at least seven career science trees (yes, it's my favorite mode), and have landed a manned module on every planetary body except Jool and returned. Yes, that includes Eve.

So I started out in hard mode, with no adjustments, and I'm glad to say... it's pretty hard!

In this new career game (the eighth?), I have launched probes to Mun and Minmus, and have landed a Kerbal on the Mun and returned him. My current mission is a manned mission to Minmus, and I actually had to strip my vehicle down to be able to afford it. I have about 56,000 credits, and it cost, after stripping, 55,000.

God I hope this mission succeeds.

Anyway, I think you've nailed the difficulty on the Hard level. It's not any harder as a game, I mean, the rockets are not more difficult to design or anything, but you really have to pay attention to the money.

Link to comment
Share on other sites

Agreed. In 0.24 I bull-rushed through the tech tree with little difficulty other than my own pickiness, but in my new save where I have Entry Purchases enabled, it's taking ages to get through it and actually buy all the parts. I started with a strategy that trades funds for Science, and it's working rather well Science-wise, but for the first time ever I keep running into "Can't afford to launch" popups. And I'm not even doing Hardcore Mode.

Link to comment
Share on other sites

I've just orbited the Mun in a Hard career. I've only taken test contracts that are part of actual missions for science (a couple suborbitals (grabbed a soil sample upon landing, what the heck), a couple orbital flights, then my munar attempt (new rocket a little shy on dv due to a takeoff issue (forgot to hit "t"), and not enough to land and return, so just orbited, about to head home)). I have over 160k, plus 2 of the 90 tech tree nodes, and everything left of that (all parts unlocked by entry purchase).

This is no different at all from 0.24.2 career as far as I can tell (yeah, I'd be one step farther right on the tech tree, perhaps, but it feels no different at all). I've not been grinding science. It's still just as easy on hard as 0.24.2 was IMHO, even if it is certainly XX% slower that it was in that.

Link to comment
Share on other sites

I've been doing pretty well in hard mode, had lots of funds and such.. Until I decided to purchase every part I could in the tech tree. The next mission after that, I managed to blow up my launchpad, and not enough funds to repair it :P Going to be launching from the runway for a while it seems.

Link to comment
Share on other sites

I've found that 10% science is actually too low, so congrats right there Squad for finally giving me too *little* science for a change. I'm playing with 20% right now and think that may be the sweet spot. We'll see how things go and I might - just might - deign to raise it all the way up to 30%.

Link to comment
Share on other sites

Wow, I didn't realize that Hard was 60% (just looked at the settings). So hard needs to be harder, clearly. That said, all having less science would do is make me grind more for tech so that I could actually do science. PArt of the problem is still the nonsensical nature of the tech tree. Mission drives tech, tech is not driven by previous study of geology, lol.

Link to comment
Share on other sites

I do love the Custom option, verily. Though the Administration Facility can and does completely undercut any science or reputation limits you set if you set up campaigns to convert Funds to rep/science. The scale conversion is just too massively in favor of science and reputation in those cases and it undoes the difficulty quite succinctly.

Link to comment
Share on other sites

I do love the Custom option, verily. Though the Administration Facility can and does completely undercut any science or reputation limits you set if you set up campaigns to convert Funds to rep/science. The scale conversion is just too massively in favor of science and reputation in those cases and it undoes the difficulty quite succinctly.

Quite right.

The standard "landmark" contracts (reach altitude x, get to space, etc.) also pay to well, I think - but lowering the funds income via the difficulty settings makes lots of contracts mere sub-par reputation sources - although it prevents the game becoming a "launch x and test at y and z" grind.

Balancing the numbers across the game after all features are in will be fun for the devs ... :D

Link to comment
Share on other sites

Wow, I didn't realize that Hard was 60% (just looked at the settings). So hard needs to be harder, clearly.

Disagree. At 60%, many of those missions to try out prototype components won't break even, and you'll have to combine three or four on a single launcher to make a noticeable profit. It creates an environment where you MUST do many of these before you can take a trip to Mun. It also creates a system where you should go to fully reusable designs as quickly as possible, or at least try to minimize damage; I started putting parachutes and probe cores on all my launchers, just so that SOMETHING would survive to be recovered.

There are two separate issues that make even Hard trivial to beat once you get past the very start, though:

1> The "report science" missions. Once you put a tiny unmanned probe with a temperature sensor in low orbit around a body, you can complete these missions whenever they pop up, for large gains at no additional cost. For a measly 20k you can easily make an ion probe capable of reaching anywhere in the system, but each report mission from a Joolian moon (or Dres/Eeloo) will give you ~250k credits plus the usual science and reputation. (In 0.24, I made probes costing 50k that also carried a Kethane scanner, ScanSat arrays, etc. for even more utility, and they could be launched from an 8-ton spaceplane that cost barely 30k and returned 29k if you landed on the runway. VERY economical.) The flag-planting missions are similar, although there you at least have to have a manned rover/lander on the surface, which takes a lot more of an effort to set up for the more distant bodies.

SOLUTION: Transmitted data that results in 0.0 science simply cannot complete a contract. In effect this'd put a cap on how many times those contracts could be completed at all, and would really kill those 1-sensor sorts of unmanned designs. I'd prefer a sort of diminishing returns system, so that the immediate response is for players to just not abuse those contract types, but a straight cutoff seems simple to implement.

2> The admin strategies are just not well-balanced at present. In my first 0.25 career save, I took the two +research strategies at 20%. With a few orbital trips and one Mun trip, I had enough science to unlock nearly the entire tech tree, while only sacrificing 20% of my monetary income. The problem was that many of the "test prototype" contracts have huge cash rewards and tiny science rewards; in one case, I did a contract that normally awarded 1 science, but I instead gained 118. (The general useless of reputation once you hit a certain amount makes this even more pronounced, but that's a different discussion.)

SOLUTION: Cap each strategy's benefit at, say, 100% of the "native" reward in the destination resource. That is, if a contract pays 10 science before strategies are considered, then each of the two +research strategies can at most convert enough money or reputation to add another +10 science, regardless of what percentage you've set the strategy to. The percentages would still matter for those contracts that add good amounts of all three resources, but it'd keep these other types from getting out of control. Again, a more complex mathematical relationship might work better, but this way would be really easy to explain to users.

Link to comment
Share on other sites

I wasn't willing to give up my .24 save that I was working on, esp since I was alreayd using SP+

But I took a look at .25... and I quickly ran into money troubles as I paid to unlock all I could indiscriminately after my first mission (SRB+ pod to space)

Then, without enough funds for an orbital mission, I took some test part landed at kerbin contracts...

Mistake... one was a test an 800 fund decoupler... but with the setting I chose (custom, 50% cash rewards), it was a loosing proposition... but I realized that too late...

Even worse, there was a bug, and the contract didn't complete.

I decided to give myself negative starting rep... its a new program, I wanted "less than zero" since a lot of funds would have already been invested to setup the complex, with no results yet...

Result... really bad contracts.

All the other contracts seemed to be "flying at kerbin" ones that would be hard to afford just launching (given I had about 2,500 funds), and the flying part would mean poor recovery value.

Having only done 1 real mission, I started over

Link to comment
Share on other sites

Disagree. At 60%, many of those missions to try out prototype components won't break even, and you'll have to combine three or four on a single launcher to make a noticeable profit

Correct.

but each report mission from a Joolian moon (or Dres/Eeloo) will give you ~250k credits plus the usual science and reputation.

There are factors for every planet/moon, these should be lowered drastically, especially with the upcoming biomes for all remaining bodies in .26 the science points influx will become a deluge otherwise.

At least I am almost certain that these also govern the contract payouts.

The flag-planting missions are similar, although there you at least have to have a manned rover/lander on the surface, which takes a lot more of an effort to set up for the more distant bodies.

SOLUTION: Transmitted data that results in 0.0 science simply cannot complete a contract. In effect this'd put a cap on how many times those contracts could be completed at all, and would really kill those 1-sensor sorts of unmanned designs.

A two-edged solution, but I agree in so far, that the Plant Flag on X contract should be a one shot mission like Explore X.

Science contracts could be narrowed down, like "take a something-device reading from biome X on Y", and only work once/as long as it has not been done (transmitted/returned) before.

All physical experiment contracts (goo, materials, samples) could have a "return to Kerbin" requirement.

Fiddling in the requirement for processing the data in a field lab would be nice too.

But.

The downside of this is, that all that is left to do (after every contract possible under these rules has been fulfilled), will be "test part under circumstances" contracts, at least until new contracts are added.

The admin strategies are just not well-balanced at present.

Lowering the maximum percentage and lowering the exchange ratio might be an easier step first.

Although science MUST be completely sellable for funds at all times, because otherwise it will just sit there after the tech tree is complete.

Mistake... one was a test an 800 fund decoupler... but with the setting I chose (custom, 50% cash rewards), it was a loosing proposition... but I realized that too late...

What became of the idea to give the player a limited amount of test parts for free, letting him bleed funds should he not return the part (OK, not possible for most decouplers, at least not easily done) in one piece?

At least the payout on success should pay for the part, not adjusted by any other settings, plus the usual scalable benefit.

I decided to give myself negative starting rep... its a new program, I wanted "less than zero" since a lot of funds would have already been invested to setup the complex, with no results yet...

Result... really bad contracts.

All the other contracts seemed to be "flying at kerbin" ones that would be hard to afford just launching (given I had about 2,500 funds), and the flying part would mean poor recovery value.

The custom settings seem to be designed for players to torture themselves ... :wink:

Edited by KerbMav
Link to comment
Share on other sites

Mistake... one was a test an 800 fund decoupler... but with the setting I chose (custom, 50% cash rewards), it was a loosing proposition... but I realized that too late...

I've found if you shoot it up at an angle with a couple of the starting truss (that the game calls a strut) pieces attached to it, it doesn't explode on impact. Then you can recover it for money.

The game doesn't tell you about the money, but if you check your cash before and after you'll see you got it all back.

Link to comment
Share on other sites

After a few careers with grinding the mun-science in the past, with v0.25 i decided to start with 200% science to go faster to other places. Too much, i know ;). I used the "25% Funds for science"-strategy to push a little.

It was still nice until one very big contract came up. Testing the second most powerfull engine in orbit gave me over 100k Funds in advance and over 600k Funds on completion. And bam, over 8000 Science from ONE contract. Even with difficulty at "100% Science" it would be around 4000 Science, right?

No complain here, i was just a bit shocked how fast the techtree was done. I will start again with lower difficulties.

Squad has a huge project coming up. With more biomes, the contracts and all the strategies they have to ballance the science-gain A LOT. Best wishes for that. ;)

Link to comment
Share on other sites

The downside of this is, that all that is left to do (after every contract possible under these rules has been fulfilled), will be "test part under circumstances" contracts, at least until new contracts are added.

Unfortunately, yes. This is one reason I was reluctant to make that suggestion; its simplicity might make it easy to explain, but it has definite downsides. The other reason was that due to the way the tech tree is laid out you'll often send manned, all-in-one missions to the nearer bodies (Mun, Minmus, Duna, Ike, Gilly) before you unlock the techs to make unmanned probes practical. This'd make those science contracts even less likely to come up, as you'd basically be locked out of them for those bodies as soon as your first manned mission returns with the full array of data.

I suppose you could do a diminishing returns setup instead, similar to how many science amounts already depreciate when completed multiple times. For instance, instead of a set reward, have the payoff for those types of contracts be dependent on the actual amount of science gained in the process; instead of a set 250k for a Laythe science contract, have it be 10k + some fixed amount per science point awarded, with the 250k amount corresponding to the best possible case (best instrument, first time, with a return trip to Kerbin instead of transmission). A probe transmitting basic temperature data might only give 20-30k even on the first try, while a manned round-trip with a surface sample would give the full 250k.

Not only would this still give a token amount for multiple completions (the 10k), but it'd encourage you to use multiple science instruments on your probes instead of only placing a single thermometer, and encourage manned missions over just an array of unmanned probes. There'd still be a limit on the number of times you could complete it, though. The same would go for the flag plantings. Planting one in a biome you've never placed a flag in before, with a Kerbonaut you've never used for flags before, could give the best reward, while planting another one outside your persistent base would only award a token amount.

I'd really prefer to keep the award simple, but have added complexity under the hood. For instance, you could allow people to complete the science contracts the way they have been (although possibly with the instrument-dependent scaling I mentioned a minute ago), but have the RESPAWN TIMER of the contract be dependent on how it was completed. Do it the same way every time with an unmanned satellite and you might only get a Mun science mission once per year or more, but do it with a variety of different sensors on manned round-trip missions and you might see the same contract pop up almost as often as you can launch Mun missions.

But there's another solution I just thought of. This might be even simpler; just make the contracts more specific. Instead of saying to get science from a body, have the contract specify which instrument or data type is required. Maybe it'd ask for temperature, maybe it'd ask for goo, or maybe it'd want an EVA report or surface sample, and the payout would depend on the difficulty involved (mass and cost of instrument, whether it requires a person, whether it can be easily recharged). While you could still send probes with the majority of the sensor types to cover most of the possibilities, this'd greatly increase the cost of the probes (making it harder to blanket the system with cheap probes in the early game) and you still would get some contracts that just couldn't be done that way.

The same would go for flags; if the contract specified a specific biome, then you wouldn't simply be able to have the kerbonaut step outside your base every time the contract came up. Sure, occasionally he'd already be in the right biome, but most of the time he'd either have to travel, or you'd have to send another mission. The smaller (specific craters) or more remote (the poles) the biome in question, the better the reward; a basic Midlands/Highlands contract would pay only a token amount. Of course, this only works for bodies with multiple biomes, but that's the way it's all heading anyway and we ARE still in beta.

Regardless, I hope we can all agree that the current contract payoffs are unbalanced even before discussing the repeatability. If that 250k Laythe payoff required the sort of vessel that cost at least that amount then it'd be one thing, but being able to complete it even once with a dirt-cheap probe is just a problem. Heck, I've got an SSTO spaceplane design that can fly itself to Laythe and return, losing only the cost of its fuel, so even requiring a manned vessel wouldn't be enough.

Link to comment
Share on other sites

@Spatzimaus:

Great ideas. I would love to see specialiesd contracts, to go the one crater on minus for example.

And this would also lead to a stock-method of scanning for biomes. The mod ScanSat shows it, but without such a tool, specific contracts would be a bit tricky.

Link to comment
Share on other sites

Really enjoying this update, having to purchase in addition tolaying out science is great, and makes you think before you buy.

I've set myself on hard, no adjustments, have a stranded kerbal in mun orbit, whom I'll recover later, and am gearing up for a mun landing now that I just acquired batt's to give me enough elec power to last through.

Lot's to look at in the update though with management, etc. Awesome!

Link to comment
Share on other sites

The mod ScanSat shows it, but without such a tool, specific contracts would be a bit tricky.

True, but I consider it to be just a matter of time before that mod (or one that fills the same niche) gets absorbed by the development team, just like so many others have in the past. Not much point in having a biome system without some way to see them, after all; sure, you can just guess where the various biomes are, but that only really works if the bodies in question have distinctive terrain features in the first place. Plus, it gives you something to do with satellites.

And that's the major downside to this sort of proposal; it's not just that most of the bodies in the system lack biomes, it's that they lack the variety of terrains that would even form the biomes in the first place. For this sort of thing to work the biomes have to be visible from orbit, really, and that means either distinctive terrain types or major geological features. Kerbin's fine for that, and Mun has the various craters, but Minmus only has a handful of biomes (basically, the various altitudes) and most of the other moons and planets are too plain to even get that much without a major redesign.

But the instrument-based science contracts, at least, should be implementable in the current system.

Link to comment
Share on other sites

Ah, good. I was waiting for a thread like this to appear. I'll repeat what I posted in the grand discussion thread:

So far I'm having a LOT of fun in Career Hard mode. By the time I got to the mid tech tree levels, I was unlocking nodes with science but couldn't pay to unlock all the parts. This is excellent.

However, Outsourced Technology really does need to be re-tweaked. It upsets game balance significantly even at a low 10% commitment which I currently have it set to. I think the conversion rate should be reduced by an order of magnitude or more. Perhaps make conversions even worse as difficulty level scales up. I suggest 1 science for every 350 funds for normal mode, and 1 science for every 1000 funds on hard. I'm still in early game (just Mun and Minmus so far) and the contracts I'm being offered, combined with a 10% commitment to Outsourced Technology, gives me science gains that I would normally have to go to Jool for.

*snip for brevity*

Since posting that and playing a little more, I've actually started to change my stance on this a little. Sure, I can unlock the whole tech tree, but I don't yet have enough money to buy all the parts! What has effectively happened is that I can see what potential technologies are on the table (the equivalent of dreaming up ideas) but I can't use them until I invest money in developing that technology. I can see a real-life parallel to this -- anyone can submit an idea for, say, a bigger more powerful engine, but if you want to actually DESIGN one you're going to have to pay engineers and scientists to work out every last detail.

To Spatzimaus, I like those ideas. My personal take is that "transmit science from space around X" contracts should require Science Jr. or Goo, i.e. non-reusable science instruments unless you have a mobile lab. I think that solves the game balance issue right there, but so do your other ideas.

Link to comment
Share on other sites

Disagree. At 60%, many of those missions to try out prototype components won't break even, and you'll have to combine three or four on a single launcher to make a noticeable profit. It creates an environment where you MUST do many of these before you can take a trip to Mun. It also creates a system where you should go to fully reusable designs as quickly as possible, or at least try to minimize damage; I started putting parachutes and probe cores on all my launchers, just so that SOMETHING would survive to be recovered.

There are two separate issues that make even Hard trivial to beat once you get past the very start, though:

1> The "report science" missions. Once you put a tiny unmanned probe with a temperature sensor in low orbit around a body, you can complete these missions whenever they pop up, for large gains at no additional cost. For a measly 20k you can easily make an ion probe capable of reaching anywhere in the system, but each report mission from a Joolian moon (or Dres/Eeloo) will give you ~250k credits plus the usual science and reputation. (In 0.24, I made probes costing 50k that also carried a Kethane scanner, ScanSat arrays, etc. for even more utility, and they could be launched from an 8-ton spaceplane that cost barely 30k and returned 29k if you landed on the runway. VERY economical.) The flag-planting missions are similar, although there you at least have to have a manned rover/lander on the surface, which takes a lot more of an effort to set up for the more distant bodies.

SOLUTION: Transmitted data that results in 0.0 science simply cannot complete a contract. In effect this'd put a cap on how many times those contracts could be completed at all, and would really kill those 1-sensor sorts of unmanned designs. I'd prefer a sort of diminishing returns system, so that the immediate response is for players to just not abuse those contract types, but a straight cutoff seems simple to implement.

2> The admin strategies are just not well-balanced at present. In my first 0.25 career save, I took the two +research strategies at 20%. With a few orbital trips and one Mun trip, I had enough science to unlock nearly the entire tech tree, while only sacrificing 20% of my monetary income. The problem was that many of the "test prototype" contracts have huge cash rewards and tiny science rewards; in one case, I did a contract that normally awarded 1 science, but I instead gained 118. (The general useless of reputation once you hit a certain amount makes this even more pronounced, but that's a different discussion.)

SOLUTION: Cap each strategy's benefit at, say, 100% of the "native" reward in the destination resource. That is, if a contract pays 10 science before strategies are considered, then each of the two +research strategies can at most convert enough money or reputation to add another +10 science, regardless of what percentage you've set the strategy to. The percentages would still matter for those contracts that add good amounts of all three resources, but it'd keep these other types from getting out of control. Again, a more complex mathematical relationship might work better, but this way would be really easy to explain to users.

You can usually recover parts at least for in flight or suborbital, just launch straight up.

The problem with demanding science return for contracts is that its easy to milk the 1 biome planets dry, at lest so dry it demand return of an goo container or material lab.

Link to comment
Share on other sites

Disagree. At 60%, many of those missions to try out prototype components won't break even, and you'll have to combine three or four on a single launcher to make a noticeable profit. It creates an environment where you MUST do many of these before you can take a trip to Mun. It also creates a system where you should go to fully reusable designs as quickly as possible, or at least try to minimize damage; I started putting parachutes and probe cores on all my launchers, just so that SOMETHING would survive to be recovered.

Funny, I built my first Mun rocket and still have 160k in the bank, and I did few parts contracts (combined them with some other missions). My comments were not theoretical, I am playing a hard campaign, and have not reverted, quick saved, etc. I've unlocked a couple 90 science things, and everything left of that, and paid for all the parts, too. If you are low on money, and you have ever played before (this is only my 3d career, the date under my name is when I started playing) you are doing something very wrong. Money is not a problem, nor is science. On hard. I have yet to experience a "trade off."

Link to comment
Share on other sites

Funny, I built my first Mun rocket and still have 160k in the bank, and I did few parts contracts (combined them with some other missions). My comments were not theoretical, I am playing a hard campaign, and have not reverted, quick saved, etc. I've unlocked a couple 90 science things, and everything left of that, and paid for all the parts, too. If you are low on money, and you have ever played before (this is only my 3d career, the date under my name is when I started playing) you are doing something very wrong. Money is not a problem, nor is science. On hard. I have yet to experience a "trade off."

It depends on how you play. People have done Mun landings on the first launch (no science unlocks) using solid rocket boosters that "stage" by blowing up the tank/engine underneath. Could I do that? Yeah, sure, but I want some semblance of a sane design, so I try to get pretty far into the tech tree before I attempt a landing. I wouldn't say that qualifies me as playing "wrong" or being a bad player (btw, I've done a stock Eve return, so I know what I'm doing)

Link to comment
Share on other sites

I did a suborbital, then 2 orbitals, then munar orbit (I threw a few parts contracts onto rockets that were actually being used---test engine on the ground, then launch for orbit). My point is that on "hard" there is not really a chance for any trade off considerations. I wasn't building "can I land on the jun with the stuff given at career start?" rockets, just normal looking manned rockets. I was playing in a very vanilla way, and the cost of, well, anything has not even occurred to me, I could not tell you what any rockets cost because it's simply not an issue.

So if "difficulty' is to just be a balance of funds/rep/science, hard is not hard, not even a little hard. That said, I don't think that these things actually adjust difficulty. Maybe contracts could have a difficulty rating, and the career could pick based upon difficulty setting, reputation, etc. More difficult contracts might be "put a satellite with an antenna, solar panels, and this probe core into geostationary orbit" (parts not in your game yet will be provided). If you pick hard, that's the kind of thing you get instead of "get science from the space above kerbin."

Link to comment
Share on other sites

Even on hard mode I'm swimming in funds and I started with a negative rep starting point and I'm already well respected. Science is a slow trickle, but I think that funds-for-science strategy will balance me out. I haven't even landed on the Mun yet.

I mean it's alpha so balancing is shrugs right now, so whatever.

I'd recommend anyone wanting a hardmode challenge to cut the funds and rep gain by half again.

Link to comment
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.

×
×
  • Create New...