Are "science points" too generic? Does the game become a grind?

[allmhuran]

From the flurry of 0.22 information over the past couple of days I think I've gathered how the tech tree works: You explore something to earn generic science points, and then you spend those science points to unlock items on the tech tree. You can see what you haven't unlocked and can thereby choose where to spend your acquired points.

While this is a fairly obvious way to implement such a feature, I worry that it turns "science" into "economics". Players will be motivated to find the most efficient way to earn science points quickly, regardless of what that way entails. If anyone has ever played an MMO (I confess I played WoW for a few months) you'll know the drill: find some mechanism that quickly accrues XP, and then farm the ever-loving !%&@ out of that. It's not much fun, but it gets you "maxxed out" rapidly.

I wonder if, perhaps, a more directed approach would be more interesting and rewarding for players.

Consider:

- If I launch a jet and manage to fly it all the way around kerbin, I should learn something about making jet engines efficient, and unlock technology that reflects this.

- If I manage to take a really heavy plane up to 10,000m, I should learn something about making better wings by doing so. Perhaps this makes all wings slightly stronger, or have slightly more lift, or unlocks a new wing entirely, thus making it easier for me to do the same thing next time, or letting me build an *even bigger* plane next time if I want to.

- If I get a rocket into orbit, I should learn something about how rockets work in a vacuum (and perhaps boost vacuum ISP for all rockets)

As it stands it appears as though I could, say, launch a mission to Duna, earn some generic science currency, then come back and spend it on air intakes - which makes very little sense. Exploitation of the currency mechanics is also a larger concern when trying to use generic "points".

So I guess this is both a suggestion and discussion. My suggestion being: make science more directed. The game already tracks certain "achievements" in the log, and I'm sure it wouldn't be too hard to come up with others, like I have above, that give a sensible, logical reward for achieving certain milestones, which would also encourage players to explore the game and reward them for doing so.

Edited October 13, 2013 by allmhuran

[Whirligig Girl]

I completely agree. HarvesteR said he wanted players to do whatever they want, but it does not make sense. If you want to do whatever you want, play the sandbox.

[lajoswinkler]

This reminds me of monstrous abominations in Minecraft, with huuuuge farms that make bread and all that stuff people eat in multiplayer. Yeah, stuff quickly turns dull when your only goal is "experience points".

I might be wrong, but the only way to mitigate the danger you're talking about is including intelligence into the feature. AI or real intelligence by some dude who will have the job to periodically invent a new task. I don't think AI is suitable enough for this, because this needs unpredictability and cunningness.

I really wouldn't like KSP to turn into a new game ridden with massive, uninventive projects where players behave like drones. I hate that.

[allmhuran]

The neat thing is, players can still do whatever they want - and in fact, by doing what they want, they will be rewarded with *relevant* technology that makes it easier to do what they want, or allows them to do what they want in a more advanced way (bigger, faster, etc).

If I want to build planes, then I will spend time building and flying planes. And with a more directed system, that would result in me doing things that would be rewarded with plane building technology, which is great if I'm someone who wants to build planes - which, according to the first premise - I am!

And of course, there's no reason why you couldn't combine the two methods. Have some "generic" science currency that you can spend wherever you want, but make things "expensive" in terms of generic currency. At the same time, make the directed breakthroughs "free". To take an example from my OP, if I get a plane greater than X mass to Y altitude (X and Y arbitrarily determined by the devs), then I just automatically learn about better wings and get an immediate, "free", unlock or improvement in wing tech.

[jwenting]

and if you don't show people the path(s) to take in advance, they'll complain about that and someone will set up a website with a fully exposed tree for others to "cheat".

It's a classic no-win situation for the devs, whichever option they choose, someone will complain.

[boomerdog2000]

You can't really grind points since every time you do the same science in the same place you get less. It encourages you to go and explore other planets and biomes.

[Psycix]

You can't really grind points since every time you do the same science in the same place you get less. It encourages you to go and explore other planets and biomes.

So then we will grind by doing the same science, with the same vehicle in different places.

1 lander

1 interplanetary transfer stage

1 launcher for both

Launch a bunch of them, go everywhere, do all the science.

Of course there are exceptions, like Tylo and Eve, but a good lander could land on pretty much any moon or planet, albeit being overkill for a lot of bodies.

[Captain Sierra]

Even still, it's effectively a different mission each time since each planet poses a unique challenge. With a few exceptions, no two worlds are the same drill for a landing mission to. More importantly, I'd like to see you design something that can get to Jool easily on the first handful of tech levels. NERVAs look to be way up there.

[JakeGrey]

I wouldn't worry too much. This is only the first release of the tech tree, SQUAD are probably going to fine-tune it in future releases.

I do like the idea of multiple flavours of science points that unlock several items along several different tech trees. Anyone remember a really old game called Alien Legacy? That would be a pretty good example of what I have in mind.

[NeilC]

So then we will grind by doing the same science, with the same vehicle in different places.

1 lander

1 interplanetary transfer stage

1 launcher for both

Launch a bunch of them, go everywhere, do all the science.

<...>

You know, that sounds to me like a very efficient and effective space program. I would enjoy designing that lander, that IPT, and that lifter. And I would enjoy "grinding" them all over the Kerbol system.

And then I would realize that the science I had just "grinded" unlocked bigger and better lifters, IPTs, and launchers. So I would enjoy building and "grinding" with those too.

If you want to define "grinding" as "flying spacecraft to new celestial bodies", then I think you'll find KSP is chock full of grinding. I think you'll also find you're playing the wrong game.

[NeilC]

I agree that generic "Science" points might be too limiting, and that unlocking all categories of tech with the same points is not very realistic.

I recently proposed an engineering subsystem that would allow players to improve parts and reduce their cost based on how much they use those parts. In that proposal, parts are still unlocked with Science, and are just improved with Engineering points.

There's another way to approach an Engineering system, though: What if all parts are unlocked and improved by Engineering Experience Points (EEP) which can only be gained by launching craft? EEP aren't applied to a generic Engineering slush fund, but to specific parts. The tech tree is basically the parts list: every node unlocks 1 or more parts, and that node can only be unlocked by using the parts in the node(s) that precede it.

Build a lot of jets? You get a lot of EEP on your jet engine and at a certain threshold the Turbojet tech tree node unlocks.

Use a lot of LV-T30s? Unlock the LV-T45, and then gaining a certain number of combined EEP on either of those could unlock the Skipper.

The details of the tree would have to be planned and balanced of course, but I'm sure it would look a lot like what's coming out in 0.22.

But we still want to do Science, of course! It's a great concept, and it should be the reason why we're building rockets, while parts are unlocked according to how we build rockets. So I think Science should be directly tied to the finances of your space program: maybe Science is money, directly, one-to-one. You don't keep "Science points" in a generic slush fund to unlock chunky nodes, you are paid for science as it comes in. The diminishing returns would still be there - you have to keep going further afield to generate enough science to keep your space program funded.

In addition, Kerbal Congress could give you missions and funding in advance based on doing science in specific places - maybe some Kerbal gives a speech about going to the moon in this decade, and if you accept that mission you immediately get a million KerBucks and a mandate to do at least 1,000 points worth of Mun-based science before a certain date. Maybe until that date, Mun-based science is worth double the Kerbucks. This would give the game some direction and momentum, without fully putting it "on rails" with a fixed set of missions.

I think this approach would keep Science as the primary driver and focus of the game, which is clearly what the devs intend. But I think tying Science to money instead of technology makes more sense, and a complementary Engineering system that unlocks new parts would provide added depth to the game.

TL;DR:

I think Science points should be necessary to get funding, and Science should remain the reason you need to fly all over the Kerbol system. Engineering points should be used to unlock parts, and should be awarded to specific part trees instead of a general slush fund.

[PDCWolf]

The entire update, the entire dev cycle and the entire dev focus is plain retarded.

SQUAD now presents a tech tree arranging the already present parts "in a balanced way to provide new players a stepping introduction to the game". Said tech tree is unmoddable because you can't rearrange it, move the nodes, change the parts, etc. You can only add parts and change the position of the existing ones, but not the nodes themselves.

Now every time there's an update, the tech tree needs to be rebalanced and revised to match the new parts that are being added, unless they just do it half-assedly.

On a hopefully not-so-far future, we'll get resources. Now the entire tech tree is out of place, and you can unlock new engines by mining dunan soil, or create new fuels by sending a kerbal to swim, which makes no sense at all. The entire game now needs to be reworked from the ground up.

They didn't make a tech tree, they made their own grave. That is unless the tech tree turns out to be useless and/or removed in the next update when they realize how hard it is to maintain it and keep it up with the updates and new content.

All of this now combined with a retarded dev cycle with excessive bug testing for an alpha and errors being carried since the first release (wobble for example) that never got fixed.

[NeilC]

The entire update, the entire dev cycle and the entire dev focus is plain retarded.<...> they just do it half-assedly. <...> they made their own grave. <...> retarded dev cycle

I'm sorry, I can't hear any of your ideas for improvement over the high-pitched whine of your tone. Do you have any?

The devs listen to constructive criticism, and they adopt good ideas when they come up. If you're just going to vent spleen, please take it to some other thread you don't want the devs to read. I'm hoping they will read this one.

[Superfluous J]

Said tech tree is unmoddable because you can't rearrange it, move the nodes, change the parts, etc. You can only add parts and change the position of the existing ones, but not the nodes themselves.

So you can move everything anywhere in a node, and change the names of those nodes and their icons, but you're worried that you won't be able to change the nodes themselves?

Do you have a citation for not being able to move the physical nodes, even if that is needed which I don't see why it would be? Or are you too busy calling everything you don't like "retarded" to find it?

[slin3]

I like OP's idea. If I reach a new milestone/achievment/goal in KSP I should be rewarded with progress in a certain direction. Not generic science points. This would tie well into career mode because if you only do certain mission types then you will only discover technology that is relevant to these missions. At some point you would get stuck and you would have to explore different parts of the game. And you would have to choose wisely what technology path you choose because if a private investor is looking for someone to do something for him and you don't have the technology needed to do that you will loose precious funding, and you will be limited to other mission types. This allows for interesting gameplay and gives your space agency a distinct character.

Regarding the things PDCWold said: the best way to resolve the problems would be to make the techtree moddable as soon as possible. SQUAD should then put all of the old and new stuff in Tier 0 and let it stay there! SQUAD should switch their focus to the more important things like the wobble mentioned above and the development of the foundations of the game. That way everything will stay the same regarding science, career and all that stuff, it will be the same sandbox because everything is in Tier 0. The only things changing? SQUAD will (hopefully) add new features and improve/fix existing features and modders will have the ability to experiment with the techtree. When everything else is done and works SQUAD can start to mess with techtrees, career, missions etc. Everything else is a useless waste of time (IMO) because (as PDCWolf said) a techtree during this phase of development would require constant reworking and adjustments, which is a waste of time as long as we have more important problems. If SQUAD doesn't keep up with the techtree tweaking we might as well leave everything in Tier 0, hence my suggestion. If modders start playing with the techtree (and I bet they will) then players can get experience with such a system and provide valuable feedback which comes in handy when SQUAD implements the official techtree. The whole process would be much smoother and faster.

[NeilC]

the best way to resolve the problems would be to make the techtree moddable as soon as possible. SQUAD should then put all of the old and new stuff in Tier 0 and let it stay there! <...> The whole process would be much smoother and faster.

Two points:

1) Implementing the tech tree system is hard. It takes man-months of dev work and debugging time. Changing the order and content of tech tree nodes is easy. It takes man-hours using a text editor. There is no appreciable difference in terms of dev time required to produce a tech tree like we will see in 0.22, and a "flat" tech tree like you propose. Updating the tech tree is the same: hours of work, not days or weeks.

2) Squad is producing a game, not a framework. Their intent is for the stock game to have a tech tree, not a set of building blocks so mods can make one. Why would they spend months developing the tech tree code, and then neglect to actually use it? Why would they leave a core aspect of their game to be implemented in mods?

[TouhouTorpedo]

I wouldn't worry too much. This is only the first release of the tech tree, SQUAD are probably going to fine-tune it in future releases.

Honestly, I doubt that. When there is a first release of a feature, you can more or less take it as the final release. (infact it even says that in the T&C's of the game) Remember when those placeholder IVA's were updated?

they weren't.

[PDCWolf]

I'm sorry, I can't hear any of your ideas for improvement over the high-pitched whine of your tone. Do you have any?

The devs listen to constructive criticism, and they adopt good ideas when they come up. If you're just going to vent spleen, please take it to some other thread you don't want the devs to read. I'm hoping they will read this one.

I like how you removed the entirety of my post and said "I CAN'T HEAR YOU" and yeah, devs listen to criticism and accept ideas, that's why the suggestions are plagued with devs' answers (or even the media group at least) instead of mods. You need a fecal matter storm to make any of the important names post in a thread.

I admit my post was strong worded, but the point is still there and no amount of niceness or rudeness can hide it away. What you are doing is entirely skipping someone's argument because of words.

So you can move everything anywhere in a node, and change the names of those nodes and their icons, but you're worried that you won't be able to change the nodes themselves?

Do you have a citation for not being able to move the physical nodes, even if that is needed which I don't see why it would be? Or are you too busy calling everything you don't like "retarded" to find it?

As far as I know you can only change the "price" and "node number" of a part, but that's about all you can do. That's what I meant.

Anyways, there's no possible way to deny what I said. The only solution to the balancing and updating job that a tech tree at this point of development requires is either doing it half-whatever-it-is or removing it and leaving it for later.

[EDF]

-snip-

...Why would they spend months developing the tech tree code, and then neglect to actually use it? Why would they leave a core aspect of their game to be implemented in mods?

you mean like scenarios?

[Alguien]

So you can move everything anywhere in a node, and change the names of those nodes and their icons, but you're worried that you won't be able to change the nodes themselves?

Actually, as far as i have been told, all what modders can do is link mods to a node and thats it. They can't do anything about the nodes or the techtree.

IMO, career and the tech tree are a bad decision in general at this point of time/developement and the effort will be better used, as PCDWolf said, in bugs and incomplete features like wobble, aerodynamics or even the plane part rework which will force a rework of the techtree. There is a reason single player and campaings are usually done late in developement.

[slin3]

Two points:

1) Implementing the tech tree system is hard. It takes man-months of dev work and debugging time. Changing the order and content of tech tree nodes is easy. It takes man-hours using a text editor. There is no appreciable difference in terms of dev time required to produce a tech tree like we will see in 0.22, and a "flat" tech tree like you propose. Updating the tech tree is the same: hours of work, not days or weeks.

In theory? Yes. I could probably overhaul the techtree on my own in my free time without many problems. In practice? I've seen SQUAD take ages for updates that should have taken half the time, so excuse me for keeping my expectations low.

2) Squad is producing a game, not a framework. Their intent is for the stock game to have a tech tree, not a set of building blocks so mods can make one. Why would they spend months developing the tech tree code, and then neglect to actually use it? Why would they leave a core aspect of their game to be implemented in mods?

I'm not saying that at all. Re-read my post. I said they should leave it to the modders until they are further in the development of the game. Once they have everything else more or less ready they should implement the official techtree! I'm talking about releasing the techtree closer to Version 1.0 not about releasing it after 1.0 or never!

[Nobody]

While I do agree, that the methods of unlocking the tech tree needs some work, it isn't bad for a start. This will help people who want objectives placed on them enjoy the game as they flesh it out some more. Admittedly I am one of those who without a clear set objective tend to get overly ambitious and that leads to frustration and failure.

Science alone could be used to start new tech trees but add another XP based system so you need both XP in the tree and science to expand. So for example you could science the basic battery but then you would need to send out probes and rovers equipped with batteries to earn XP in battery tech before you could science the next tier of battery tech.

As to modding it would be nice for things to be a little more open so mods can be more easily fit into the tech trees as new branches.

[Rocketman1994]

Im curious, is science just points? Or is it actual data too.

Call me a nerd, but I'd love to have atmospheric data, rad info, soil analysis, geological info, temperature, how this information changes through out a year and so on.

[Rampage9112]

I completely agree, I would have liked to see variations of all of the engines. (like, instead of starting with the LV-909, you start with an earlier model, and improve it over time). The current way everything is setup is pretty dull and pointless. Also, I don't think that surface samples and crew reports on the ground should give many science points if any at all.

[boomerdog2000]

I think they did a good job with the current tech tree. It may not be very realistic but they have stated that KSP is a game first then a simulator. It seems like a good way of guiding new players, while restricting experienced ones to try new methods for their rockets.