KSP2 Tech tree and For Science! VAB (and analysis)

[Strawberry]

So for my thoughts on what of science we have seen in general, in short I think some of it is incredibly promising, some of it is more mediocre. The presentation shows that youll need to unlock the top right node to move on to the next tier, which seems interesting, I like how that changes how you move through the tech tree. Also xenon is unlocked way earlier in the tech tree which is good. The science parts seem like a mixed bag, to define what I think makes a good science part, using it must actively change your flight profile. The ones earlier on the tech tree can be simple as their point can be there just to introduce concepts to people, the latter ones should be more complex.

To start off with what I think is by far the best science part we have seen, its the diving bell. Having to send a crewed diving bell to the bottom of Eve's oceans then return it sounds like an incredibly interesting and difficult challenge, especially if we get ballasts to go alongside it. Since the diving bell probably has to be crewed, if you want to save your kerbals this leads to interesting engineering challenges (such as returning the crew in the diving bell to your rocket) that sound really fun to plan around. More science parts like this that promote you to do novel, interesting, and challenging missions would be very fun.

For what I think is the worst part, its the large purple one thats likely unlocked by orbital report (Science Sr). The craft we see the Science Sr on doesnt seem to have any power generation, likely its only thing is just it needs to be in orbit and generates over time. For the last science part we unlock, that is incredibly simple and boring. I'd be fine with a single part that is just like this that way it forces you to stay in orbit for crafts instead of just flybys, but having multiple parts whose only thing is that just seems unneeded and boring.  Science Sr. seems to need a crew requirement as well, but with no life support that isnt really all that interesting of a challenge, its only in orbit so its not particularly hard to return those crew. I think an easy way to make Science Sr. more interesting would be to give it large amounts of EC consumption, making you have to put like two nuclear reactors or lots of solar panels on your craft actively changes how you design your craft and makes putting it on more interesting. 

I think there's definitely ways to add more parts that are more similar to the diving bell, and I hope they do. I think the grand slam passive seismometer from ksp1 was honestly pretty well designed, and something like that in ksp2 would be great. If I were to suggest how to make that part better for ksp2, have it generate science passively over time to promote landers, but generate spikes of science upon impacts with the planets with diminishing returns (bonus points if crashing into different biomes would reward you differently for the seismometer).  it would have interesting synergy with the sounding rocket nose cone for planets with an atmosphere as well. Also a Solar Science part that makes you do a parker solar probe esc mission by rewarding you for getting close to the sun would be cool as well. Bonus points if you get extra science points for going over the suns poles.

[The Aziz]

5 minutes ago, Strawberry said:

Also xenon is unlocked way earlier in the tech tree which is good.

The problem I see here is very simple: it does not require you to run tests in xenon-rich environments to get access to the technology.

[Kerbart]

3 hours ago, Scarecrow71 said:

Did they remove celestial bodies from the Kerbolar system?  Because it should be absolutely required to telescope them before you go there.  You should have to image Eve or Tylo or wherever before just blindly flying there and hoping for the best.  Heck, you could make the argument that you shouldn't know they exist until you image them.  Until then, they are just lights in the sky.

That sounds great on paper but when restarting Exploration mode for the 5^th ti e it will get old and grindy. Especially if orbits are statically defined so you already know where to look. For a long time I considered this to be great and from an science/explorer view it is. But in the end this is about pushing the limit where you can get the little green guys, and not a masterclass in astronomy.

Mapping/picture taking would be a great way to unlock other missions. Long distance telescope shots of Duna lead to a request to take pictures up close. Close up pictures lead to a survey for what are the highest peaks. The result of that leads to a temperature survey from the top of one of them, and so on.

[GluttonyReaper]

3 hours ago, PDCWolf said:

From AMAs and other media, Exploration is only based on managing "resources", which'd partly or only come from what they call "Discoverables" on planets. Those "Resources" are what's used for colony and shipyard (orbital or otherwise) construction. The only apparent installment of "guided gameplay" they've talked about is the "missions" we've seen here.

Huh, sounds neat! It'll be interesting to see how that pans out, nice to hear they're going in a different direction to KSP1 at least. 

3 hours ago, PDCWolf said:

Except it didn't work like that. In KSP1 incorrectly managed heat would still build up, with proper flux mechanics and skin/core interactions so that the distance from heatsink parts to heat generators would actually matter for example. They're trying to sell those interactions as new when in reality now all parts combine into a magical heat number that gets subtracted from the heat dissipation number. Now, this new system does boil to "add heatsink parts literally anywhere".

I mean I guess? Out of atmosphere I only ever used the active radiators (which I believe were just magic heatsinks that could draw from anywhere) given they were so light, and re-entry was generally just a matter of having full heatshield coverage. The only time the heat system actually seemed to come into play was for spaceplanes, which always felt to me like more of a consequence of spaceplane parts in general just being slapped into a rocketry game without much thought and not having the logical parts you'd expect (i.e. no ablative tiling on plane parts). Whether the solution is trim the excess or to find a way to more meaningfully integrate the old complex system will depend on how the rest of the game shapes up I suppose. 

[Provisional Name 45]

I cannot wait to have an excuse for violently exploding every single Kerbal I have again. 

[PDCWolf]

10 minutes ago, GluttonyReaper said:

Huh, sounds neat! It'll be interesting to see how that pans out, nice to hear they're going in a different direction to KSP1 at least. 

I mean I guess? Out of atmosphere I only ever used the active radiators (which I believe were just magic heatsinks that could draw from anywhere) given they were so light, and re-entry was generally just a matter of having full heatshield coverage. The only time the heat system actually seemed to come into play was for spaceplanes, which always felt to me like more of a consequence of spaceplane parts in general just being slapped into a rocketry game without much thought and not having the logical parts you'd expect (i.e. no ablative tiling on plane parts). Whether the solution is trim the excess or to find a way to more meaningfully integrate the old complex system will depend on how the rest of the game shapes up I suppose. 

Say you had part A, generating heat, attached to B, which had the radiators. If A was a drill for example, it'd heat at the skin first, then convect to the core and to B's skin and core, which would then convect to the radiators. Thus, it would've been much more efficient to put the radiators on A directly, as you'd be cooling the skin and maybe have some leftover convect into the core and to other parts (thus unless you spammed radiators, heat buildup was a thing). On the other hand, if A was really a drill, you can't cool it directly and have to design with that in mind.

With the proposed heat system, you can put radiators in the cockpit to cool the heat from the engines at the other side of the ship because the whole ship is a magic mass of instantly transferred and equalized heat.

[Periple]

17 minutes ago, PDCWolf said:

With the proposed heat system, you can put radiators in the cockpit to cool the heat from the engines at the other side of the ship because the whole ship is a magic mass of instantly transferred and equalized heat.

I thought the new system was the opposite, it doesn’t conduct heat between parts at all, each part has its own equilibrium! But maybe I misunderstood

I expect there may be similar radiators as in KSP1, with some of them cooling the entire craft and others only the part to which they’re attached. So the only change to the design here is that the trick of attaching a heat-generating part to a heat-sink part and attaching the radiator that won’t work (and that was needed mostly because some parts didn’t allow surface-attaching things, notably the ISRUs and drills).

[Scarecrow71]

2 hours ago, Kerbart said:

That sounds great on paper but when restarting Exploration mode for the 5^th ti e it will get old and grindy.

The same can be said of anything when restarting a campaign for the nth time.  "Need to go here and take that temp reading again.  And then I have to launch a sounding rocket to unlock that part again."  No difference there, IMHO.

[Geredis]

1 hour ago, Periple said:

I thought the new system was the opposite, it doesn’t conduct heat between parts at all, each part has its own equilibrium! But maybe I misunderstood

It's been a while, but in KSP1 , that's how it worked, especially during reentry, with every part heating individually. It allowed for some pretty interesting ideas such as eschewing heat shields, but then deliberately rotating your craft if you desired in order to put overheating parts into the lee of your your trajectory, letting htem cool off as other parts heat up, only to rotate once again in order to cool off that new section, etc. In one of the dev commentaries in a heat discussion thread (I can't recall or find it off the top of my head) they said that they wanted to do heat on a whole-ship basis, essentially in the name of intuitiveness. At least in regards to how radiators work as I recall in dissipating heat. It doesn't matter where you place them, it'll sink from the ship's heat pool, not from an individual part or localized area.

[DancZer]

49 minutes ago, Geredis said:

It's been a while, but in KSP1 , that's how it worked, especially during reentry, with every part heating individually. It allowed for some pretty interesting ideas such as eschewing heat shields, but then deliberately rotating your craft if you desired in order to put overheating parts into the lee of your your trajectory, letting htem cool off as other parts heat up, only to rotate once again in order to cool off that new section, etc. In one of the dev commentaries in a heat discussion thread (I can't recall or find it off the top of my head) they said that they wanted to do heat on a whole-ship basis, essentially in the name of intuitiveness. At least in regards to how radiators work as I recall in dissipating heat. It doesn't matter where you place them, it'll sink from the ship's heat pool, not from an individual part or localized area.

https://forum.kerbalspaceprogram.com/topic/218512-developer-insights-21-rockets-red-glare/

[Wheehaw Kerman]

1 hour ago, Scarecrow71 said:

The same can be said of anything when restarting a campaign for the nth time.  "Need to go here and take that temp reading again.  And then I have to launch a sounding rocket to unlock that part again."  No difference there, IMHO.

I didn’t really mind that in KSP1 - sounding rockets are fun.  Heck, pretty much anything that you can launch in the game is fun.  What got to me after a while was the Contracts system being procedural and frankly a bit nonsensical.  If Exploration Mode is a bit less like a business plan for SpaceX written by a pre-alpha release of ChatGPT based on a prompt written by a second grader, hooray.

[PicoSpace]

Overall Really Excited about "For Science!" and it looks like a soild update from what we have today (and the last 9-10 months once its out). Although I'm still not a fan of how its sounds science is handled (non-repeating and push-here-for-science) take of KSP1... but I guess that's what mods are for. I'd rather of seen something more kerbalism/biome-altitude/depth dependant with repeatable science (with exponentially reducing gains, but never "0").

22 hours ago, Scarecrow71 said:

I am hoping those aren't the only 9 parts in the game.  A lot of us were really hoping for telescopes.

Agreed, I would like to see more science parts for different tasks. 9 feels "too little" and none of them are overally "big". 

20 hours ago, Sea_Kerman said:

I was hoping for some sort of scansat thing, perhaps the orbital science one will do it. It's really cool to have that sort of "minigame" where you need to get into the correct orbit and resonance and stuff, and of course it takes some time.

And you can of course do more specific maneuvers to dogleg over something you want to get more info on fast.

Telescopes, ScanNet etc. are such great mods for KSP. 

17 hours ago, Strawberry said:

Oh huh yeah good catch. Its important to note in general for this stuff is that all of this is still wip. Stuff will get changed, added, and maybe removed. Tbh main reason i posted this is because i realized one of the parts was likely a diving bell and having to make kerbals dive underwater to get science seems incredibly sick.
 

Re: telescopes, telescopes have been confirmed that they will be in the game eventually. That being said, its possible that they wont be in game until interstellar/we get asteroids as we wont exactly have things to telescope until then. The large science part with the purple thing attached to it we have seen in the past. But the old model for it had a jwst telescope while the new one has a purple greenhouse in its place. This either means that theyve decided no telescopes in general (unlikely), decided to isolate telescopes to their own parts that way it can be introduced when they fit the game, or a secret third thing that i cant think of.

All this looks amazinly cool.

10 hours ago, mcwaffles2003 said:

Was really hoping that the devs would take some inspiration from kerbalism and SCANsat here but that seems not to be that case.. I don't really want anything like mini-games for science, but I really did prefer the concept of holding a condition and passively collect while the experiments consume electricity for a duration to obtain science over simply just meeting a threshold for a moment and click a button for science.

Agreed, a lot opportunity if that ends up not being the case.

7 hours ago, Scarecrow71 said:

Did they remove celestial bodies from the Kerbolar system?  Because it should be absolutely required to telescope them before you go there.  You should have to image Eve or Tylo or wherever before just blindly flying there and hoping for the best.  Heck, you could make the argument that you shouldn't know they exist until you image them.  Until then, they are just lights in the sky.

Thinking ahead to interstellar and custom solar systems, a system like this would make sense in terms of finding & tracking new worlds / systems. Yes in Kerbol a bit "silly" since we know it so well, but once your off to Deb Deb ect, its new territory.

7 hours ago, Kerbart said:

This community is divided on many things but "Science needs a huge improvement" isn't one of them.

100%

[obney kerman]

11 hours ago, mcwaffles2003 said:

I don't really want anything like mini-games for science, but I really did prefer the concept of holding a condition and passively collect while the experiments consume electricity for a duration to obtain science over simply just meeting a threshold for a moment and click a button for science.

Maybe I'm crazy, but looking at the presentation it kind of seems like that's what it's going to be already? The pictures of the science interface did seem to have listed durations for some of the observation tools.

[mcwaffles2003]

41 minutes ago, obney kerman said:

Maybe I'm crazy, but looking at the presentation it kind of seems like that's what it's going to be already? The pictures of the science interface did seem to have listed durations for some of the observation tools.

It's confirmed that this is not the case, the times shown are the times required to transmit already collected science as confirmed by @JPLRepo, not the time a condition must be held for all the science data to be collected.

EDIT: I'll simply say that I am personally disappointed that this is the case, but I am glad that it was communicated early so we don't build up hope for a non-existent feature. Also, Im pretty surprised about the decision to keep science as is when polls I've conducted in the past show as 70% - 90% of respondents wanted something more time involved like kerbalism:

 

Edited October 23, 2023 by mcwaffles2003

[Periple]

It would be pretty easy to tinker with the mechanics of collecting science, like adding durations/science over time or what not. Doing that kind of thing based on feedback is just the thing EA is for. Maybe they’ll take the opportunity?

[The Aziz]

Just adding timers won't solve the main problem, which is basing the whole system on currency (again).

[Periple]

29 minutes ago, The Aziz said:

Just adding timers won't solve the main problem, which is basing the whole system on currency (again).

I don’t think that’s a problem. It’s a design decision. I think the really important bit is, what kinds of gameplay does science support/incentivize? Science as single currency, multiple currencies, or threshold conditions are just different ways to get there.

I will be disappointed if it ends up as “fly to biome, click button, repeat until tech tree is unlocked” again, but we’ll only see how it works in practice when we play it.

Finally, an observation: in my experience, generally speaking, the most successful and fun game designs have emerged from something that’s really simple and easy to understand but can be applied in loads of different ways, most of them completely unanticipated even by the developers. Making the mechanics richer or more complex helps sometimes but you have to be super careful because more often than not it makes things worse: harder to understand and harder to learn, and often it even does the opposite of what’s intended by creating dominant or losing strategies. Entire swathes of gameplay might just become unattractive by accident, and when you try to fix this by rebalancing, you might land somewhere everything feels samey with just cosmetic differences.

tl;dr things that sound good might play bad, and things that sound simplistic might produce emergent gameplay that’s amazing, and you really can’t tell from descriptions alone unless you’re a really good game designer and sometimes not even then.

 

[The Aziz]

1 hour ago, Periple said:

what kinds of gameplay does science support/incentivize? Science as single currency, multiple currencies, or threshold conditions are just different ways to get there.

I will be disappointed if it ends up as “fly to biome, click button, repeat until tech tree is unlocked” again

The flying gameplay will be the same, fly out there and perform science. It's the why that matters. So far it looks exactly like the bottom paragraph of the quote here, except that there's only one click and the worlds aren't made of 1500 identical biomes that somehow give new batch of science points each. And you have to wait few more seconds for the data to send. I just want some logical connection between performed experiment and tech unlocked by it.

Guess I'll just hop in and create a separate thread so it'll be more visible.

[Periple]

2 minutes ago, The Aziz said:

It's the why that matters.

In my opinion what and how are more important than why.  Breaking Ground experiments were very cool, especially the seismograph one. You had to do a sequence of things to get the desired result and precision was rewarded. You could devise similar mechanics for orbital experiments.

I think that is much more important than one currency or many, or whether you get the points immediately, over time, or on transmit. (Return missions should still be important because that’s a BIG part of the “how.”)

[Safarte]

11 hours ago, mcwaffles2003 said:

It's confirmed that this is not the case, the times shown are the times required to transmit already collected science as confirmed by @JPLRepo, not the time a condition must be held for all the science data to be collected.

This is from the IG discord, this UI indeed shows transmission times only but it seems some experiments could have a "duration" too.

[mcwaffles2003]

12 minutes ago, Safarte said:

This is from the IG discord, this UI indeed shows transmission times only but it seems some experiments could have a "duration" too.

That's a saving grace but Im not willing to put my hopes on it right now, but thanks for sharing the info <3

[Periple]

I think we should wait and see what they’ve come up with, try it, then give them feedback. This should be something they can iterate on fairly easily based on it. Getting too deep into it beforehand isn’t very helpful!

[The Aziz]

26 minutes ago, Periple said:

Breaking Ground experiments were very cool, especially the seismograph one. You had to do a sequence of things to get the desired result and precision was rewarded

And you could get ground related tech for performing those - larger or more versatile wheels, landing gear, perhaps a scaffolding for colony construction...

Using science made from a seismograph readings to buy a jet engine and a small hydrogen tank doesn't make any sense and it looks like the game is still going in this direction.

[cocoscacao]

Judging this image, a major difference is that there'll be multiple tech trees? Four separate ones in this case...

[Periple]

13 minutes ago, The Aziz said:

Using science made from a seismograph readings to buy a jet engine and a small hydrogen tank doesn't make any sense and it looks like the game is still going in this direction.

I’m not sure there’s any way to make it make sense, even with different science currencies there would be a lot of make believe. Real space science is basic research, knowing more about the atmospheric composition of Venus won’t help design better space planes.

These are game design questions and “making sense” isn’t that important, what matters is that they’re fun and support and incentivize varied, creative, self-directed gameplay without getting in the way.