Diminishing returns per body

[eddiew]

Hey all,

Shower thought for the day 
What would happen if your science return for an experiment was dictated not only by the number of times you'd run it in that biome, but also the number of times you done it on that body?
E.g. first Mun surface sample gets 100% (of current setting), second gets 50%, third gets 25%. This would stack up with same-biome diminishment.

Reason for thought
These forums are full of people who maxed the tech tree in Kerbin's SoI, and I'm not convinced that's "fun"^TM 
Sure, you can dial science gain down to 10%, but that ultimately pushes players to repeat, repeat, repeat the same thing. A diminishing return strategy would reward firsts and encourage players to get out of their comfort zone - which generally works out as a good thing.

And yes, I know we can just be disciplined and promise ourselves that they'll only do n landings on any particular body, but ultimately most of us don't have that level of willpower. Sooner or later, I'll realise I need just 50 more science points for a thing I want, and I'll crack and send a cheap expedition to Mun rather than designing a massive Jool mission for it. Just because it's easier and I can 

Just an idea for discussion, if anyone's of a similar mind 

 

Edit; alternative as raised further down the thread:
The first time a specific experiment is transmitted/recovered from a particular body would reward double, or even triple, science.

Edited February 29, 2016 by eddiew
adding a thing from deep down in the depths

[Matt77]

I like this idea because I am one of those people.

I like a slow game and I don't like that every mission unlocks at least two nodes, meaning redesigning my rockets again and again. So I may be a minority...

 

[Nich]

Not sure where I land on this.  Devs put a ton of energy into the Mun Biomes I feel bad not visiting every crater.  At the same time I am not sure I want to totally explore the Mun everytime I start a career.  Perhaps make it a slider an option.

I to am guilty of oh hell only need 25 science for my new toy better go check the MPL and heaven forbid excrements only 21 TIMEWARP

[GluttonyReaper]

I personally always felt the system was self-regulating. really - landing more than few times on the same body can be really tedious, so I assume people who get bored of that really quickly would just find another body to land on. The other option is making the science farming more interesting in itself - putting a refueling station in orbit and reusing the same lander is arguably more interesting than several missions direct from Kerbin. Either way, KSP has always been about making your own fun, rather than the game providing it for you. That said, I think it would be better to totally revamp the science system in general, rather than introducing more limitations on the current system - the way proposed in the OP, at least IMO, just makes one aspect of the game even less fun than it already is, to effectively make other aspects appear more fun, which I don't feel is something the game should rely on for encouraging progress. 

[lajoswinkler]

It's a decent idea, but I'd need to think more about it. One thing that comes to my mind is - will it decrease the progress rate a lot? I'm also not a fan of maxing out the technology tree in Kerbin's system by grinding, but the thing that bugs me the most is how slow the progress sometimes occurs. Repetition annoys me a lot.

I think simulations would have to be made to predict the outcomes of your economical model.

[Joonatan1998]

I agree the game should incentivize progression, not grinding Mun/Minmus forever.

The rates for the first landing probably would have to be buffed from the current ones to not make the game too hard.

Halving the rate for each biome may be a bit too much, maybe have each biome give 3/4 or 2/3 of the previous one.

[kujuman]

I disagree that this is something that should be a forced change in the stock science mechanics, for several reasons.

First, in response the the first rational for the change given (that maxing the tree in Kerbin's SoI isn't fun)--well, that's how other players enjoy the game, and it's unreasonable to limit what others can do to have the stock game better fit one's own play style.

Second, in response to the comment that just a quick mission to Minmus to grab a bit extra science, that's not inherently a bad thing gameplay wise. Sure, it can help unlock a node we want/need for something (say, getting better engines for an SSTO that *almost* works), but that seems like a good thing--one doesn't have to launch a massive Jool mission to make play in LKO more interesting.

Third, as weird as the science system is (finding out the temperature of a valley on Dres helps my engineering team build a bigger battery? Umm, sure...) finding out the environment from different parts of a body tells you about *the body*, and not just about the rock you landed near. If I land on an animal and it has fur on its back, I would later want to do a fly-by of its belly to see if it has four legs, and land another probe on its feet to see if it has claws or hooves, and then rove to the mouth to inspect the teeth, and the exploration of the parts tells a story about the whole. And for players who don't use biome maps or lists, maxing a body is a puzzle in itself--hmm, there's a farside crater I landed in, maybe the other craters are interesting as well! Oh, Duna should have loads of science! But there aren't distinct features...hmm...where should I try?

I think a mod like this though could be very popular (for the reasons stated), just as the mods that reward one for really in depth exploration and development of a single body are. But I think the stock game does a reasonable job of giving the player a choice of where to direct his energies, and enforcing a particular playstyle isn't what makes KSP. :-)

[Renegrade]

This idea has come up before (well, broadly at least.  The stacking rules might have been different back then), and I supported it before as well, and continue to support it now.

 

9 hours ago, kujuman said:

First, in response the the first rational for the change given (that maxing the tree in Kerbin's SoI isn't fun)--well, that's how other players enjoy the game, and it's unreasonable to limit what others can do to have the stock game better fit one's own play style.

It's unreasonable to limit what others can do to have the stock game CONTINUE to fit one's own play style. *cough*

The real problem here is that people who prefer the existing "balance" (I use the term loosely and scathingly) could easily re-establish it by hauling the science gain slider to the right when setting up a save, but people who wish to play with the diminishing returns style system are unable to do so at any slider setting (setting it to 10% or less isn't the same, for example, as you end up with a really large early grind which redefines boring to new levels of hideous).

Also we're ignoring the elephant in the room of the fact that the tree is just a random hodge-podge of parts without any real progression (I think Harv described it as an 'extended tutorial' that 'reduced newbie part confusion by restricting the available parts'), but that's a subject for another thread.  Also ignoring the MPL issue...

By the way, in your analogy, when you landed on the feet of the animal, you already knew it was furry, had fur on the top and bottom, had samples of aid fur, probably had enough data to determine if it was warm blooded or not, and most likely had a rough idea of what sort of feet it had and number of legs etc - your return from that mission would be significantly less than the first landing on the back.  You still learned something, but a lot less than the first landing.  Well, assuming the first landing worked.  If it crashed and burned, you probably only learned that it was solid and crashing into it is a bad idea, heh.

 

[eddiew]

Thanks for the thoughts guys   I appreciate that it's not an idea to suit everyone, but I do feel like the game would do well to reward players for famous firsts. I set my science right down at 10% because I want to be lured out to Duna and beyond for the multipliers, and I explicitly don't want to have maxed the tech tree until I'm way out around Jool and Eeloo - but the opening grind just to get up to 2.5m parts becomes painful.

As a different option to decay, how about:

The first time a specific experiment is transmitted/recovered from a particular body would reward double, or even triple, science? 

This would probably be much easier to implement, and it avoids telling players they're collecting 80 points when they're walking on the Mun, which then recovers for 20 due to DR. If you had two Mun missions running at the same time, then the first one back gets the bonus, and there's no complications like who collected what data first, or whether you have that same data in transit on another vessel. I believe this puts it in line with the way the admin building's thingy for increased science works?

Just an idea

Edited February 29, 2016 by eddiew

[Joonatan1998]

1 hour ago, eddiew said:

 

The first time a specific experiment is transmitted/recovered from a particular body would reward double, or even triple, science? 

Would be (much) easier to make some of the experiments biome independent.

 

I think there could even be only one experiment that cares about the biome for surface/atmosphere/orbit.

Samples from surface, atmosphere analyzer for atmosphere and gravioli detectoru for orbit.

Obiviously the science values would have to be rebalanced.

 

 

 

Edited February 29, 2016 by Joonatan1998

[Alshain]

Eh, I hate the tech tree. I'd rather max it fast so I can start playing the game.  Though I'd really rather have a Funds only game mode, so if they added that I wouldn't care.

Edited February 29, 2016 by Alshain

[Pez_]

I think this could be less artificially solved if the reset/data collection mechanics were tweaked so the number of biomes you could get data from were limited by the equipment you can take. Then the science processing lab is more useful if you can build it into the mission as well. 

[Superfluous J]

I just independently had almost this exact thought. I was looking into making a mod for it and did a search and came across this thread. Figured I'd bump this up before actually trying to mod it in.

All my mod needs to do is know how much of the experiment's science has been gotten from that world, and how much that experiment's science is worth in a single biome (that number's always the same, right?). Then it would knock down the science gained based on that. Ideally, it should also report that diminished value when you take the experiment, with the caveat that you may get less when you return it (as you do now)

So let's say temperature data is worth 12 science per biome on Mun, and you've got a probe up there hopping and transmitting for 50% return. The first transmission would get you 6 science. The next time you take a reading (from a new biome) it'd be worth 1/2^0.5 or 1/1.41 or about 71% or about 8.5 science. You transmit that for 4.25 science and now you have 10.25 in the bank. The third biome would be worth 1/2^(10.25/12) or 1/2^0.85 or about 55% or 6.6 science, transmitted for 3.3. And etc. You get the idea.

No clue how to handle capping transmission. Maybe force 100% of capped transmission return so you can only transmit data once per biome. Though ... Ugh. I hate details.

[Wjolcz]

Uuuuuuuuugh. Any thread that proposes more tweaks to the career mode will get bombarded by me. I'm very sorry.

Not really actually.

How about this: let people pick what kind of resource they want to get for performing science and completing contracts. Let people choose which currency (rep/sci/funds) they want to use for: buildings upgrades, tech tree, launching vessels.

+ keep the current career tweakables as they are of course.

The current system is bad. Look at what is happening with this subforum. Everyone is asking for something completely different when it comes to the career mode. It's because we're stuck with this crap of a system and it really, really needs to change drastically. More tweaks won't help it. Stop getting stuck in that Stockholm syndrome. "Boo-hoo! Don't ruin my saves please! I have so many vessels! And my tree is complete!". Well guess what. You are not alone. I too had a save full of on-going missions. Then a career mode "tweak" came along and I decided to kill it with fire because I couldn't play the way I wanted anymore (had an SSTO-focused space program). Thanks SQUAD. You made me really not want to restart it.

Game Mode Creator is the only way to fix all the problems with all this.

Edited October 10, 2016 by Veeltch

[Tau137]

On 2/26/2016 at 5:05 AM, eddiew said:

Shower thought for the day 

 

What would happen if your science return for an experiment was dictated not only by the number of times you'd run it in that biome, but also the number of times you done it on that body?
E.g. first Mun surface sample gets 100% (of current setting), second gets 50%, third gets 25%. This would stack up with same-biome diminishment.

Reason for thought
These forums are full of people who maxed the tech tree in Kerbin's SoI, and I'm not convinced that's "fun"^TM

Yep, I had the exact same suggestion, thread ongoing (moderators - merge the threads, perhaps?).  Using 1/2 or 3/4 multiplier for subsequent biome experiments (on the same body) will:

1. Make sense (rock from Polar Crater might be different than a rock from Midlands, but is it is not likely to bring a revolution in Munar geology);

2. Make grind inefficient and thus undesirable (for those playing Career/science; others can play Sandbox all day long if they wish without being affected by this change);

3. Gently guide players, old an new, towards exploration of new planetary bodies instead of making it easier to grind the same old Kerbin SOI, and yet will allow some grind if absolutely necessary or desired.

 

Thumbs up!

 

P.S. Considering that this thread is more than 7 months old, it does not look like developers care much about this particular aspect of their game, ether though it does not take much effort to implement (as opposed, to, for example, implementing castrated remote tech equivalent, or fixing landing gear, or implementing planet cloud cover, or implementing planet axial tilt, or {list your idea here}).  Really sad.  I really want to finally PLAY the game, but with every update I see the same story: either the game is majorly broken (<1.1 with limited RAM, or 1.1 with screed surface physics), or the mods that I consider "must have" are not updated yet... then another update comes is and the cycle continues (C) ME

Edited October 11, 2016 by Tau137

[eddiew]

18 hours ago, 5thHorseman said:

No clue how to handle capping transmission. Maybe force 100% of capped transmission return so you can only transmit data once per biome. Though ... Ugh. I hate details.

I would say 1 transmit per biome is totally fine... landing multiple goo canisters to scratch out the last few points from transmitting isn't a gameplay I'd encourage. Not that it's my call re stock, but since we're talking mods...  

8 hours ago, Tau137 said:

P.S. Considering that this thread is more than 7 months old, it does not look like developers care much about this particular aspect of their game, ether though it does not take much effort to implement (as opposed, to, for example, implementing castrated remote tech equivalent, or fixing landing gear, or implementing planet cloud cover, or implementing planet axial tilt, or {list your idea here}).  Really sad.  I really want to finally PLAY the game, but with every update I see the same story: either the game is majorly broken (<1.1 with limited RAM, or 1.1 with screed surface physics), or the mods that I consider "must have" are not updated yet... then another update comes is and the cycle continues (C) ME

Tbh, I'm not entirely sure the devs have ever used a suggestion from the suggestions forum - but @5thHorseman's thoughts about modding the feature in give me hope   That said, it does sound like 1.2 is much more stable, and with a largely new dev team, they may start looking to build on it, rather than losing all their time squashing bugs.

Edited October 11, 2016 by eddiew

[Superfluous J]

47 minutes ago, eddiew said:

I would say 1 transmit per biome is totally fine... landing multiple goo canisters to scratch out the last few points from transmitting isn't a gameplay I'd encourage. Not that it's my call re stock, but since we're talking mods...  

Don't hold your breath (you're already looking quite green) because I poked around the API docs and I can't make head nor tail of it. I'm not ruling anything out, but I honestly think it'd be easier to wait for Kopernicus to update to 1.2 and then just make single-biome maps of each world

[eddiew]

56 minutes ago, 5thHorseman said:

Don't hold your breath (you're already looking quite green) because I poked around the API docs and I can't make head nor tail of it. I'm not ruling anything out, but I honestly think it'd be easier to wait for Kopernicus to update to 1.2 and then just make single-biome maps of each world

Would make a bit of a mockery of most custom contracts though; they'd all turn into "land anywhere you like" ^^;