Making Science More Fun

[Pthigrivi]

Why go through the trouble though? Right now people can plonk down and experiment away. Instead you'd like them to stop what they're doing and go back to KSC to sift through hundreds of potential experiments and sign agreements for everything. Isnt this just replacing science points with a bunch of extraneous inflight UI? Are we asking squad to reengineer half the game because you don't find science points believable? 

31 minutes ago, Panel said:

That seems very backwards. Apollo 11 knew that it would land on the moon and preform certain experiment months in advance. 

I tend to think this is the difference between flying an actual moon mission and playing a fun game.

 

[Wjolcz]

21 hours ago, Panel said:

That seems very backwards. Apollo 11 knew that it would land on the moon and preform certain experiment months in advance. 

True, but that way the player would have the flexibility needed in case when not all of his/her objectives went according to the plan (forgot one of the experiments/a parachute/ran out of fuel, etc.). They could pick missions before and after the launch. It would also be very handy for someone who would want to extend the mission of his/her vessel.

A good example of this is the currently ongoing Dawn mission. It is now orbiting Ceres, but the Dawn team thinks they can go to some another minor planet. If this was a case in someone's game, he/she would simply pick another set of missions connected to what they want to do next.

Edited June 8, 2016 by Veeltch

[Wjolcz]

17 minutes ago, Pthigrivi said:

Why go through the trouble though? Right now people can plonk down and experiment away. Instead you'd like them to stop what they're doing and go back to KSC to sift through hundreds of potential experiments and sign agreements for everything. Isnt this just replacing science points with a bunch of extraneous inflight UI? Are we asking squad to reengineer half the game because you don't find science points believable? 

What? This makes no sense. The player could do experiments with and without accepting the missions. But the completed missions would help you to complete the program, which would unlock more programs. Basically what we have now, except you pick the theme of missions yourself and get rewarded with cash and rep instead of science points (which are the main reason why the game still needs "tweaking").

Ok, I get it. You are trying to strawman me to death, because I'm a pest here. Fine, I will leave and never bother this thread again.

[Pthigrivi]

1 hour ago, Veeltch said:

Ok, I get it. You are trying to strawman me to death, because I'm a pest here. Fine, I will leave and never bother this thread again.

Im really not trying to do that. These aren't strawmen, they're logical implications. The science system may not be perfect, but science points do one thing pretty elegantly: give weighted rewards for conducting experiments. If you're not weighting different experiments by a points system then you need some other system to replace it. You're talking about throwing them in with missions, which may sound simple enough, but there are thousands of potential experiments in the game, hundreds on any given mission. For every experiment that has value, literally every time a player right now would right click an experiment (which is intrusive enough) you would have them sift through half a dozen drop downs to manually input that experiment into your mission summary, press accept, then go ahead and do the experiment. Maybe that happens before they launch, maybe they have to do it in the middle of a mission. Either way thats a LOT of UI fuss, and potentially more of a hassle than what we already have.

Edited June 8, 2016 by Pthigrivi

[mattinoz]

3 hours ago, Pthigrivi said:

I tend to think it best to keep things as simple as possible. 

. Start simple: Yes but to me you just can't make it fun without making in harder over time to test the player as they learn

[Wjolcz]

19 hours ago, Pthigrivi said:

You're talking about throwing them in with missions, which may sound simple enough, but there are thousands of potential experiments in the game, hundreds on any given mission. For every experiment that has value, literally every time a player right now would right click an experiment (which is intrusive enough) you would have them sift through half a dozen drop downs to manually input that experiment into your mission summary, press accept, then go ahead and do the experiment. Maybe that happens before they launch, maybe they have to do it in the middle of a mission. Either way thats a LOT of UI fuss, and potentially more of a hassle than what we already have.

The only major change to the UI would be program tabs in the Mission Control and missions in them, which wouldn't even expand when clicked on ("Perform temperature scan in space high above Jool" is IMO enough of an explanation about what the mission wants from you). You get what you have to do in the title and once that is selected you simply click "Yes" or "No" button. The current UI for missions is more complicated than that.

Do you want me to draw it for you?

Edited June 8, 2016 by Veeltch

[Pthigrivi]

On 6/7/2016 at 5:45 PM, Veeltch said:

The only major change to the UI would be program tabs in the Mission Control and missions in them, which wouldn't even expand when clicked on ("Perform temperature scan in space high above Jool" is IMO enough of an explanation about what the mission wants from you). You get what you have to do in the title and once that is selected you simply click "Yes" or "No" button. The current UI for missions is more complicated than that.

Do you want me to draw it for you?

A good mission collects dozens of readings. Do I have to do that for every one of them? And then re-do them every time I change my mind about something? Am I penalized for doing so? If not, can I collect money for a mission and then decline for no penalty? I just don't think you realize all of the implications removing science points has, they're tough to deal with. 

Edited June 8, 2016 by Pthigrivi

[Pthigrivi]

23 hours ago, mattinoz said:

. Start simple: Yes but to me you just can't make it fun without making in harder over time to test the player as they learn

Yeah... there could be a combination here. 5th Horseman's biome idea is interesting but there are places like on the Mun where the Midlands span half the body, does it make sense that a single impact in this zone would map the whole thing? You could make it so they had to be some minimal distance apart to count, like 10k or something, but I still don't know if their radii should stack. Performing this is kind of like a scan, so it doesn't seem right that scanning the same area twice would still gain the same rewards. You should really have to scan unique areas. That would also help put a bit of a cap on how much science could be gathered from any given body. I think it should work like this:

You'd get that network effect, (albeit not in that 3d sense Mattinoz mentioned) and you'd have to think somewhat carefully to get the best results without it being too easy to exploit. You could encourage this kind of thing by keeping the sensor range relatively small. I also do like the idea of different level scientists producing bigger radii. Another thing you could do is make the science pay-out linked to total ore concentration within the scanned area. Not only would this add another layer to the challenge, but players would be inclined just to scan the most heavily concentrated areas seen from their orbital scan rather than feeling like they had to do a whole planet. 

Edited June 8, 2016 by Pthigrivi

[vger]

I wonder if a robust 'programming language' could be derived from this as well, using drag & drop nodes to keep it from becoming overcomplicated.

For example, if you want to take a temperature reading at a certain altitude, you need an altimeter and a thermometer. Then in the programming interface, configure the thermometer to take a reading the specific altitude is reached. Really do think that a lot of new science toys are needed, along with a reduction in the amount of science attained per experiment.

I really love the seismometer idea. Keep in mind some worlds may still be geologically active though, so you could still bag science on some worlds without having to knock on them.

 

A couple more ideas to add to this.

Dust-Buster - Genesis probe solar-wind collector analogue, could also serve additional purpose for grabbing cometary debris if we ever get those in the game, Europa-style fissures, kuiper belt, sun's corona, etc

Radicalizer - Measures the ambient level of radiation in space or on a planet's surface. Higher tier science leads to a more advanced device...

Kamma Sensor - Scan for Van-Allen belts from high orbit and detect radiation belts - radiation belts could also be a hazard that needs to be avoided or at least, as little time spent in them as possible. After being properly-mapped, a visual representation of the belts can be toggled. Also helpful for sending the Dust-Buster-equipped probes to collect high-energy particles.

 

 

Edited June 9, 2016 by vger

[Wjolcz]

1 hour ago, Pthigrivi said:

A good mission collects dozens of readings. Do I have to do that for every one of them? And then re-do them every time I change my mind about something? Am I penalized for doing so? If not, can I collect money for a mission and then decline for no penalty? I just don't think you realize all of the implications removing science points has, they're tough to deal with. 

The points are tough to deal with because they exist. That's why we get the "tweaks" all the time and why the parts tend to move around the tree every update. The more types of points there are the longer they take to balance. Especially when they have influence on each other.

A "good" stock mission (I'm assuming you mean a single contract here) doesn't give you the final reward if you miss one of its objectives. Good luck getting all of them checked if you forgot one of the experiments specified in it. But that's what reverts are for, right?

The mechanic of multiple and independent program goals (missions) rewards you for each experiment performed independently.

Think about it:

1. Go to the Mission Control we have now -> cycle through a bunch of RANDOMLY GENERATED contracts until you find the one that interests you (has the exact combo of goals that you actually want to do). How much time did it take you?

2. Now let's go through the "science-points-free" concept of career mode: Go into the admin Building -> pick the Jool (it's only an example here) Exploration Program -> go to the Mission Control -> there's the tab called "Jool Exploration Program" -> a long list of missions (experiments that can be performed around Jool) rolls out -> you pick the ones that you want.

Now tell me which one was easier and faster to go through? Let's not assume you were super lucky and you got what you wanted right away in the contract lottery in the current game.

[Wjolcz]

Sorry for multiposting but I'm unable to edit.

IMO declining the goals/missions shouldn't be penalized. Each program would give you money to complete the said program. Each completed goal would give you rep and help you progress. A number of completed goals would make your program ready for finalisation (you get rep for it).

The experiments give you rep, the rep give you money and the money gives you tech and science equipment, which when performed give you rep.

That way the circle is closed and no resource is useless. Even if the whole tech "tree" (which isn't really the tree that we know and more of a bunch of loosely connected nodes) would be completed there would still be plenty of things to do (science experiments which fund your program).

[Pthigrivi]

Right, but you aren't just selecting one experiment. You're specifying each experiment part in each instance it will be performed. You're selecting:

Jool: Goo: High above Jool

Jool: Goo: Low over Jool

Jool: Goo: High above Vall

Jool: Goo: Low above Vall

Jool: Goo: Vall Lowlands

Jool: Goo: Vall Midlands

Jool: Crew report: High above Jool

Jool: Crew report: Low over Jool

Jool: Crew report: High above Vall

Jool: Crew report: Low above Vall

Jool: Crew report: Vall Lowlands

Jool: Crew report: Vall Midlands

Jool: EVA report: High above Jool

Jool: EVA report: Low over Jool

Jool: EVA report: High above Vall

Jool: EVA report: Low above Vall

Jool: EVA report: Vall Lowlands

Jool: EVA report: Vall Midlands

And on, and on, and on, and again if you decide on a different flight path or landing site. Or you could just go there and do the thing and automate data retrieval. This isnt about making new problems. Its about fixing the ones that currently exist.

 

Edited June 9, 2016 by Pthigrivi

[Wjolcz]

3 minutes ago, Pthigrivi said:

Right, but you aren't just selecting one experiment. You're specifying each experiment part in each instance it will be performed. You're selecting:

Jool: Goo: High above Jool

Jool: Goo: Low over Jool

Jool: Goo: High above Vall

Jool: Goo: Low above Vall

Jool: Goo: Vall Lowlands

Jool: Goo: Vall Midlands

Jool: Crew report: High above Jool

Jool: Crew report: Low over Jool

Jool: Crew report: High above Vall

Jool: Crew report: Low above Vall

Jool: Crew report: Vall Lowlands

Jool: Crew report: Vall Midlands

Jool: EVA report: High above Jool

Jool: EVA report: Low over Jool

Jool: EVA report: High above Vall

Jool: EVA report: Low above Vall

Jool: EVA report: Vall Lowlands

Jool: EVA report: Vall Midlands

And on, and on, and on, and again if you decide on a different flight path or landing site. Or you could just go there and do the thing and automate data retrieval. This isnt about making new problems. Its about fixing the ones that currently exist.

 

That could be easily dealt with by applying filters.

[Pthigrivi]

How? All of these things have to be specified. You aren't doing all the science on Vall, you're just doing some of it. In your version we have to tell the game in the form of drop downs everything we plan to do. Maybe you don't have the gravoli yet, maybe you're splitting up your mission with probes and crewed sections and they're doing different landings and orbits on different moons. These things get really complicated in practice. Again, could spend 2 years re-writing and rebalancing half the game just to add a bunch of redundant UI or you could just go there and do the thing and automate data retrieval. 

Edited June 9, 2016 by Pthigrivi

[Wjolcz]

Again, sorry if this doesn't get merged.

I have nothing against the automation of experiments. You would pick the goals, fly in there, do a bunch of manouvers, go high, go low and see how many of them you got completed. That's a really good idea.

1 minute ago, Pthigrivi said:

How? All of these things have to be specified. You aren't doing all the science on Vall, you're just doing some of it. In your version we have to tell the game in the form of drop downs everything we plan to do. 

That's exactlt why there should be no penalty for declining the previously activated goals.

 

3 minutes ago, Pthigrivi said:

Maybe you don't have the gravoli yet, maybe you're splitting up your mission with probes and crewed sections and they're doing different landings and orbits on different moons. These things get really complicated in practice. Again, could spend 2 years re-writing and rebalancing half the game just to add a bunch of redundant UI or you could just go there and do the thing and automate data retrieval. 

Since the program is all about the SOI there's no problem with that. All you have to do is to collect the experiments. Who cares if you split your probe in half if one half is going low and the other high? The experiments are gathered and that's all that really matters.

[Pthigrivi]

What Im saying is the way different experiments and their values get specified now is through science points. You go to the place you want to go and do the experiments you want to do and the game handles the rest. Making every instance of that a matter of manual input seems like a step backwards.

[Wjolcz]

3 minutes ago, Pthigrivi said:

What Im saying is the way different experiments and their values get specified now is through science points. You go to the place you want to go and do the experiments you want to do and the game handles the rest. Making every instance of that a matter of manual input seems like a step backwards.

I agree. That's why I think the automation of experiments is a great idea.

[Pthigrivi]

36 minutes ago, vger said:

I wonder if a robust 'programming language' could be derived from this as well, using drag & drop nodes to keep it from becoming overcomplicated.

For example, if you want to take a temperature reading at a certain altitude, you need an altimeter and a thermometer. Then in the programming interface, configure the thermometer to take a reading the specific altitude is reached. Really do think that a lot of new science toys are needed, along with a reduction in the amount of science attained per experiment.

I really love the seismometer idea. Keep in mind some worlds may still be geologically active though, so you could still bag science on some worlds without having to knock on them.

 

A couple more ideas to add to this.

Dust-Buster - Genesis probe solar-wind collector analogue, could also serve additional purpose for grabbing cometary debris if we ever get those in the game, Europa-style fissures, kuiper belt, sun's corona, etc

Radicalizer - Measures the ambient level of radiation in space or on a planet's surface. Higher tier science leads to a more advanced device...

Kamma Sensor - Scan for Van-Allen belts from high orbit and detect radiation belts - radiation belts could also be a hazard that needs to be avoided or at least, as little time spent in them as possible. After being properly-mapped, a visual representation of the belts can be toggled. Also helpful for sending the Dust-Buster-equipped probes to collect high-energy particles.

 

Sorry I missed this, there are some great ideas here vger. I really like the idea of combination experiments, I might spend some time imagining ways this could work. I'd also been thinking about a general radiation sensor. What would be super cool would be volumetric mapping, picking orbits that swathed out large areas rather than just scanned the surface efficiently. One thing Im trying to do is link each experiment to some useful bit of information for the player so to me this kind of thing would really shine if radiation became an actual gameplay concern. I know Nertea is working on something like this for his mods, not sure if it will be environmental at some point.

14 minutes ago, Veeltch said:

Since the program is all about the SOI there's no problem with that. All you have to do is to collect the experiments. Who cares if you split your probe in half if one half is going low and the other high? The experiments are gathered and that's all that really matters.

Right, but in your version you have to input it into your mission each time, for every instance, in order to get credit. Filters aren't going to help you that much. You're still sitting in the admin building specifying dozens of experiment instances when you don't have to. You're right, Mission Control doesn't make it clear that you can just go out to the Mun and be rewarded with a World First. You also can't get advances for any old place you'd like to go. Also right-clicking on parts gets kind of old after a while. Also there are some grind issues. The solution to those problems isn't to turn every right-click into 6 filtered drop-downs though, its to fix the specific issues people are having with the existing system.

[Wjolcz]

Or do you mean the way the player would have to pick things himself? The goals and experiments could be hidden, of course, and simply auto-collected. That gets rid of the Mission Control goal cycling and clickiness, I guess.

But I think that it is always nice to keep track of your own goals. Having it in your in-flight UI is even better. If you set yourself a goal of, for example, achieving a Jool orbit then you should be able to check the state of the goal anytime and anywhere. That way it could also serve a role of a mission planning tool.

11 minutes ago, Pthigrivi said:

Also there are some grind issues. The solution to those problems isn't to turn every right-click into 6 filtered drop-downs though, its to fix the specific issues people are having with the existing system.

Depends on how it's balanced. Right now we have too many resources (money, rep and SP). It's the main reason why it sometimes can get grindy and why the strategies were first OP and now are almost useless.

There's no hazard of grind if you can't exchange the SP for money, or rep (the closed circle mechanism posted somewhere above).

[Pthigrivi]

To a degree, yes. But why make it mandatory? This just a general sense I get from people and their comments on this board but it seems like when people are having issues with Career its because there are a few clunky, overbaring bits going on that are taking people out of the experience.  The bit for you is you don't find science points relatable, but thats not really an issue for most people and its effecting roll-play, not gameplay. Right-clicking experiments is a bit of a chore. I'd like not to make it more of a chore. Unbelievable as they may be science points are also really useful, they help players understand the scientific value of what they're collecting. They can look at the pay-out and see that a surface sample is much better than thermometer scan, and that a sample from Eve is much more valuable than a sample from Kerbin. Knowing the relative values of those experiments is really important information to the player and it doesn't really make sense to bury it in Missions, with no real way of seeing the result until your rep goes up and then calculating in your head how much rep converts to money and then how fast money becomes a new part and yada, yada. Tying those points directly to the tech tree makes it obvious that "Oh, I need x more science. I bet I can get that from a small probe to Moho." They also create trade-offs; "I can put up a satelite and get X money, or I can do a scan of Minmus and get Y money plus Z science." If all that matters is getting rep because rep=money then there's nothing really to juggle. You just do the thing that gets the most rep and everything flows from that. 

I know they annoy you, though Im still not exactly sure why, but I promise you, they matter, and they actually make things much simpler on the front end. 

Edited June 9, 2016 by Pthigrivi

[vger]

Maybe it's just that science points feel too "meta." They're being used like 'experience' would be used in a RPG, and once you have enough, you just level up. In that case, you're practicing combat to get better at combat. It's a bit more obscure with science because it's more of a stretch to think that you can build a better rocket motor by examining goo in orbit.

It would be wonderful if different worlds yielded new alloys or something. Enabling R&D to manufacture parts that are lighter without sacrificing durability. But that takes away from the sandbox element of the game.

[Wjolcz]

8 minutes ago, Pthigrivi said:

To a degree, yes. But why make it mandatory? This just a general sense I get from people and their comments on this board but it seems like when people are having issues with Career its because there are a few clunky, overbaring bits going on that are taking people out of the experience.  The bit for you is you don't find science points relatable, but thats not really an issue for most people and its effecting roll-play, not gameplay. Right-clicking experiments is a bit of a chore. I'd like not to make it more of a chore. Unbelievable as they may be science points are also really useful, they help players understand the scientific value of what they're collecting. They can look at the pay-out and see that a surface sample is much better than thermometer scan, and that a sample from Eve is much more valuable than a sample from Kerbin. Knowing the relative values of those experiments is really valuable information to the player and it doesn't really make sense to bury it in Missions, with no real way of seeing the result until your rep goes up and then calculating in your head how much rep converts to money and then how fast money becomes a new part and yada, yada. Tying those points directly to the tech tree makes it obvious that "Oh, I need x more science. I bet I can get that from a small probe to Moho." They also create trade-offs; "I can put up a satelite and get X money, or I can do a scan of Minmus and get Y money plus Z science." If all that matters is getting rep because rep=money then there's nothing really to juggle. You just do the thing that gets the most rep and everything flows from that. 

I know they annoy you, though Im still not exactly sure why, but I promise you, they matter, and they actually make things much simpler on the front end. 

 

2 minutes ago, vger said:

Maybe it's just that science points feel too "meta." They're being used like 'experience' would be used in a RPG, and once you have enough, you just level up. In that case, you're practicing combat to get better at combat. It's a bit more obscure with science because it's more of a stretch to think that you can build a better rocket motor by examining goo in orbit.

Pretty much this ^^^

They simply make things go backwards. Now it's: Gather science -> build better things.

In a proper career it should like this: Spend time and money on tech -> build a probe with sensors -> understand the world sorrounding you (and get rep for it, because kerbals like to read about new scientific facts, but mainly to close the loop).

Well, and there's a dose of roleplay from me for today.

[Pthigrivi]

Yeah, I mean its a matter of perspective. Its kind of a Realism v Kerbalism kind of thing. I feel like I mentioned (in PM maybe?) that I make table top games as a hobby so Im really focused on nice tightly functioning game mechanics. To me if the game is fun and people aren't bogging down or taking advantage of gnarly exploits and there are great tradeoffs and challenges and rewards then all is well, but perhaps I do discount the realism folks too easily. I mean there's a reason RSS and RP-0 are so popular. 

Everything aside this has been one of the better dabates I've had here man. You're a smart dude. 

[Panel]

As for how to keep missions organized:

Rather than have each mission be to do x experiment in y biome, why not make two types of mission? One type could be to do a flyby/orbit/landing at the specified body, and the other could be to collect x science points from the body.