Let's Rebalance the Tech Tree

[Pthigrivi]

49 minutes ago, Lord Aurelius said:

I would also like to add that part balance is also a critical part of balancing the tech tree. Just moving some of the parts around doesn't completely solve the issues since the part costs (and in some cases their specs) have already been tweaked to reflect their position in the tech tree.

Absolutely. There are a lot of parts so taking a fine-tooth comb to all this stuff isn't a trivial task either, but it would go a long way to smoothing things out. I also agree on simplifying the transmission and science payout system. I'd actually like to see different biomes produce different bonuses, so landing in a specific canyon might yield more science than landing any old place in the 'Midlands'. You should really also be able to produce a biome overlay in map mode with the survey scanner.

Then we get into the slightly deeper stuff like anomaly contracts and life support. I'd like to see all that (and a graphics and art assets pass too), but those get into some deep investments in development time. There are a couple of other threads on the anomalies right now. I'll be honest the existing anomalies are cute as easter eggs but don't do much for me as a real considered part of the game. I'd rather see them as part of a larger system of geological landscape features players could explore, and if that were the case receiving contracts to visit specific places would be totally great. You'd want to think carefully about how this played out though. You don't want them just linearly connected, nor would it mean much if they were totally random. Maybe the contract pops up after you've scouted it in Kerbnet? Something like that.

Life support too is a bit of a doozy. Im using USI-LS and its really great, Id recommend it to anyone who's interested in LS. It's also about as simple as it can get and still be a challenge, and even at that you really have to carefully think things through. Many players don't want it, which would be fine, obviously it would be on a difficulty toggle, but stockifying USI is a bit lower on my priority list. 

Edited November 16, 2017 by Pthigrivi

[tater]

My issue with scale is about the resultant player design choice options more than anything else.

3.2x with stock parts works fine, and if the Mun was larger than 3.2x, then perhaps players would have a real design challenge for the Mun. Note that they’d still have ez mode via Minmus.

[Spricigo]

2 hours ago, tater said:

Then play sandbox. 

No. Thanks for the suggestion but sandbox don't offer any significant advantage for me. OTOH career mode make more accessible the metrics that I use to evaluate how successful a given mission are.

Actually, you seem to be the one to keep distance from career mode given your perception of it as an awful system,  incorrectly designed and problematic.

2 hours ago, tater said:

This thread is connected to the career system.

Correction: This thread is about the tech tree. That is just one of several aspect of the career mode and not even a feature exclusive of that mode. 

And, even if carrer as  whole was the subject, it is not an excuse to disregard the opinions of those that like career mode as it stands.

2 hours ago, tater said:

The career system should put the player's space program into a kind of context. . 

And it does, just is one that you in particular and some other players feel is not ideal. KSP was developed to be that way, it don't ended that way accident.

As for your actual proposal of how to fix what you see as issues of the career system,  I see absolutely nothing that can't be done with mods or even with self imposed limitations.  Sorry, but I can't consider the "issue" anything but solved.

 

[Superfluous J]

On 11/8/2017 at 3:06 AM, cfds said:

Personally I believe that the whole career mode and tech tree are another victim of the whole console debacle

<snip>

This seems to be a pretty good recap of the past couple years. Though you forgot "get out before the whole thing collapses."

Edited November 16, 2017 by 5thHorseman

[Spricigo]

1 hour ago, Pthigrivi said:

The question for me is, if they are going to keep developing career mode, what are the biggest bang for the buck changes to improve things. Stuff we could reasonably see tweaked up for 1.4 that wouldn't involve totally rewriting the code. For that I'd love to see:

Regardless of my position pro/against your ideas I have to acknowledge you put a lot of though on it.

In my humble opinion each point is worth to be considered/discussed in it's own thread. And that could give better chances to gather support to push the implementation of one or another.

[Pthigrivi]

33 minutes ago, Spricigo said:

Correction: This thread is about the tech tree. That is just one of several aspect of the career mode and not even a feature exclusive of that mode. 

Haha don't worry this thread went off the rails ages ago. 

I mean, I'm sure the Devs don't ignore this section completely, but I have to have a certain humility about how seriously our scattershot suggestions can possibly be taken. Who knows. Maybe some good comes of talking about it. At any rate they know their schedule and have a much clearer understanding of the feasibility on this stuff. 

Edited November 16, 2017 by Pthigrivi

[tater]

Mods cannot solve the problems with the career system, they can only make it less awful. 

The rewards in career are not useful metrics for judging mission success, really, and the link between planetary science (all KSP “science”) and rocket parts makes no sense. A mun rock doesn’t inform rocket engine design.

[Lord Aurelius]

40 minutes ago, tater said:

...the link between planetary science (all KSP “science”) and rocket parts makes no sense. A mun rock doesn’t inform rocket engine design.

Totally agree. That said, I could potentially see certain specific experiments (not generic science points) being prerequisites for some parts. For example, getting a probe into space would provide the necessary environmental hostility information to design a crewed space capsule.

[Spricigo]

46 minutes ago, tater said:

Mods cannot solve the problems with the career system, they can only make it less awful. 

In that case what you proposed also don't solve the problems, only make the system 'less awful'. Since everything you proposed can be done with mods.

Do you see any possible real solution or just a pile of unsolvable (and serious) issues? 

12 minutes ago, Lord Aurelius said:

That said, I could potentially see certain specific experiments (not generic science points) being prerequisites for some parts. For example, getting a probe into space would provide the necessary environmental hostility information to design a crewed space capsule.

Somewhat like missions to unlock parts? That could work, but would require a complete new system. 

[tater]

11 minutes ago, Spricigo said:

In that case what you proposed also don't solve the problems, only make the system 'less awful'. Since everything you proposed can be done with mods.

Do you see any possible real solution or just a pile of unsolvable (and serious) issues? 

Somewhat like missions to unlock parts? That could work, but would require a complete new system. 

You'd have to go back YEARS to see my ideas about the career system in this game on this forum, including my first post back on 0.24 days.

I lack the will to type out the pages of suggestions I used to post before they gelled career up to the mess it is now. What I proposed is completely gutting career and starting over. Seriously, there is nothing about the current career system worth keeping if the goal was a good career system.

The scale issue is a generic game problem, and only interacts with career to the extent it doesn't provide any replay ability, since there is without question a "best" way to get most places. A crew landing and return on Eve or Laythe are certainly more variable, but according to numerous posts on this forum over the years, most players never leave the Kerbin SoI. Many go a long time before landing on the Mun (a long time for me was a few sips into my second Guinness, though to be fair, rendezvous and docking in Kerbin orbit took me rather longer).

Managing a space program is the stated objective of career. This requires that the game have a decent economic system at some level (it lacks this).

Budget matters (fixed budget if a "national" program, otherwise contractual obligations, along with perhaps some investment (pegged to "rep?")

Time matters. (time to actually build a craft for launch, do R&D, required milestones towards goals, else a rep hit which can decrease budget/investment, similar to KCT).

Failures would be good. Testing or having a failure then results in a delay time to find the cause, which reduces the chance of the same sort of failure in the future. This is important for the idea below...

A foil would be ideal, since it provides context. The current game implies competition ("firsts," and astronauts stranded by OTHER programs). Add the competition, either commercial, or national. Want to be the first to the Mun? Pay attention to what the other nation is up to, and perhaps send a risky mission before you have the parts you think you need. You either need "firsts" for nationalistic reasons, or you need certain capabilities to steal market share in commercial competition.

Another factor that would aid replay of career is "fog of war." This means that you only know what you could know from telescopic observation about the solar system, which also means a different solar system (if desired) every game. Like the fact that Mars was visualized with "canals" before Mariner 4 showed the cratered surface. 

 

 

[Lord Aurelius]

21 minutes ago, Spricigo said:

Somewhat like missions to unlock parts? That could work, but would require a complete new system. 

Yeah, that's kind of what I was getting at, and I know that it would require some new systems and a LOT of balance work to get it right.

The idea is that when you get into a specific situation where you don't have the technology to go farther, that inspires the R&D people to come up with the parts you need for the next step.

Which also gives me an idea for a semi-related thing. IRL, the biggest challenge with any kind of mission to Venus is the crushing pressure and high temperature. KSP doesn't model any of this. It would be neat to have some part upgrades that would be required for those extreme conditions. Not that Eve isn't difficult enough already (at least to return from).

3 minutes ago, tater said:

Another factor that would aid replay of career is "fog of war." This means that you only know what you could know from telescopic observation about the solar system, which also means a different solar system (if desired) every game. Like the fact that Mars was visualized with "canals" before Mariner 4 showed the cratered surface. 

A procedural seed option for KSP would be awesome. I just did a quick search and found this: 

Unfortunately it's an external application for now, but hopefully at some point it will be integrated into the game.

Also during that search I came across some old dev comments on why they didn't do it, and it had to do with being able to share the game experience. That has some merit, but at the same time if the seed is visible, people will be able to share great seeds and be able to share experiences that way.

Would definitely add a lot to the replayability of the game.

[tater]

I also forgot that I long ago suggested that the tech tree partially abandon generic "science" points in favor of funds, but also that specific missions might be generated to accelerate R&D. Instead of random part testing that makes no sense (landing gear on a suborbital flight over the Mun, or something equally stupid), you'd get testing for parts under R&D that could accelerate dev time, and increase reliability.

Science could also be broken into a few broad types. Basic Science includes planetary science, astronomy, etc. Medical Science would be kerbal factors in space. Rocket Science would be engineering. Collection of samples, etc is the first type, a crew on a station for X months is medical, etc. Some parts could have an ante of different type of science. Other parts might have very specific data required. Instead of one type of chute, have chutes specific to each world. Want to use a chute on Duna? Send a probe to sample the atmosphere. That sort of thing.

[Lord Aurelius]

2 minutes ago, tater said:

I also forgot that I long ago suggested that the tech tree partially abandon generic "science" points in favor of funds, but also that specific missions might be generated to accelerate R&D. Instead of random part testing that makes no sense (landing gear on a suborbital flight over the Mun, or something equally stupid), you'd get testing for parts under R&D that could accelerate dev time, and increase reliability.

Science could also be broken into a few broad types. Basic Science includes planetary science, astronomy, etc. Medical Science would be kerbal factors in space. Rocket Science would be engineering. Collection of samples, etc is the first type, a crew on a station for X months is medical, etc. Some parts could have an ante of different type of science. Other parts might have very specific data required. Instead of one type of chute, have chutes specific to each world. Want to use a chute on Duna? Send a probe to sample the atmosphere. That sort of thing.

Using funds primarily for the tech tree makes perfect sense. That's how research generally works IRL.

That also is a great idea for the part missions, I agree that right now they're generally stupid. Unless I get one to test a part on the launchpad (free money), they're generally not worth bothering with at all.

Multiple science categories makes a lot more sense than what we have now. The idea of sampling a planet to use parts there is another related thought (extension?) to the unlocking parts via missions thing. Additionally, it could be used to limit where Kerbals could EVA. If you haven't sent a probe there before to take measurements, the Kerbal's suits (or even their capsules) might not be appropriate for the environment. Especially on Eve and the sunny side of Moho.

[tater]

Yeah, crew parts would require a certain number of medical science points, for example. That or there would be cumulative accounts of what missions you had flown, and how them would impact your understanding of kerbal factors. With life support in the mix, this would become a rationale for extended orbital flights to improve life support. The difference between more closed loop systems and open loop systems are typical a trade off of complexity vs mass. bring everything open loop, or save the raw materials via recycling, at the expense of more complex systems (possible failures).

For me, it's about meaningful design choices. If you knew your LF might fail, would you send 2 of everything, for example, or 1 craft with backup LS systems?

[Pthigrivi]

16 hours ago, Lord Aurelius said:

Totally agree. That said, I could potentially see certain specific experiments (not generic science points) being prerequisites for some parts. For example, getting a probe into space would provide the necessary environmental hostility information to design a crewed space capsule.

While this is absolutely my favorite suggestion on this board, it does come with pitfalls. When parts are linked to specific mission types, it forces players to execute those specific mission types to proceed. Some of those, (sending a probe to space, for example) most people might like. Others, (Testing parts in different environments or rescuing a stranded kerbal for instance) might be more polarizing (see the contract system). A series of overprosciptive missions that everyone must do to progress if they want to or not is as as bad as a series of entirely random ones. Science points may be more abstract, more bland even, but they have the advantage of being interchangeable. That means if Mark just likes sending probes everywhere first and Cindy likes tooling around with high performance planes they can both--in theory--basically do their thing and can both unlock everything and progress. 

I do love the sound of this game, it may even be something you can construct with making history. It will, however, be an entirely different game with an entirely new set of limitations. I cant imagine thats within the scope of career mode's current development. 

 

 

Edited November 16, 2017 by Pthigrivi

[Lord Aurelius]

15 minutes ago, Pthigrivi said:

While this is absolutely my favorite suggestion on this board, it does come with pitfalls. When parts are linked to specific mission types, it forces players to execute those specific mission types to procede. Some of those, (sending a probe to space, for example) most people might like. Others, (Testing parts in different environments or rescuing a stranded kerbal for instance) might be more polorizing (see the contract system). A series of overprosciptive missions that everyone must do to progress if they want to or not is as as bad as a series of entirely random ones. Science points may be more abstract, more bland even, but they have the advantage of being interchangeable. That means if Mark just likes sending probes everywhere first and Cindy likes tooling around with high performance planes they can both--in theory--basically do their thing and can both unlock everything and progress. 

I do love the sound of this game, it may even be something you can construct with making history. It will, however, be an entirely different game with an entirely new set of limitations. I cant imagine thats within the scope of career mode's current development. 

I totally understand your concerns with forcing the player to do specific mission types. If the devs ever get around to trying any of this stuff out, a huge amount of playtesting will be needed to see how it actually works in practice. It's one thing for us to be armchair devs theorizing about how to make this game more awesome, and another thing entirely to actually put it into practice and see how it holds up.

One feature I really hope Making History ships with is the ability to create a series of missions, not just a one-off. That way players could construct their own careers/campaigns and use the mission triggers to simulate some of these things.

[Pthigrivi]

1 minute ago, Lord Aurelius said:

I totally understand your concerns with forcing the player to do specific mission types. If the devs ever get around to trying any of this stuff out, a huge amount of playtesting will be needed to see how it actually works in practice. It's one thing for us to be armchair devs theorizing about how to make this game more awesome, and another thing entirely to actually put it into practice and see how it holds up.

One feature I really hope Making History ships with is the ability to create a series of missions, not just a one-off. That way players could construct their own careers/campaigns and use the mission triggers to simulate some of these things.

Absolutely, and Im serious this could be awesome and Id totally play it. 

[tater]

If R&D were not bought with Mun rocks, but funds, then these missions to test specific elements could merely reduce the price, and speed R&D. Hence none are required.

Other elements might be required. You must fly kerbals in space for some time period to develop the next generation of life support in a timely way, or you need a space probe to some altitude (the kerbal version of discovering the Van Allen Belts). It's possible to create a system where specific missions are not required by fiat, but because the player would need a necessary understanding of the solar system in scientific terms to make certain sorts of leaps.

 

 

[Spricigo]

1 hour ago, tater said:

You'd have to go back YEARS to see my ideas about the career system in this game on this forum, including my first post back on 0.24 days.

I lack the will to type out the pages of suggestions I used to post before they gelled career up to the mess it is now. What I proposed is completely gutting career and starting over. Seriously, there is nothing about the current career system worth keeping if the goal was a good career system.

It was good enough to keep you hooked for years now. Perfect? No. But not that worthless mess you pretend it is. You'd be playing something else if it was.

Anyway, I think there is some interesting ideas in how you think career should be. I prefer the game as it is now but that is because I'm a toy player that want more liberty.

[Yakuzi]

19 hours ago, Spricigo said:

It was good enough to keep you hooked for years now. Perfect? No. But not that worthless mess you pretend it is. You'd be playing something else if it was.

Sandbox has been good enough to keep me hooked for years.

Career is a complete mess (it's more of a KSP beginners guide with some random side quests thrown in) and hardly challenging/interesting/rewarding. I haven't played career since it was introduced and probably would've stopped playing KSP altogether if it was the only game mode. Career could be amazing if some decent game and user experience design is implemented (as discussed above and the multiple dedicated threads that have been posted on the subject over the years). But alas, it doesn't seem Squad is inclined to give career the overhaul it needs...

[Spricigo]

8 hours ago, Yakuzi said:

Sandbox has been good enough to keep me hooked for years.

Nonetheless a lot of players play carrer, and enjoy career. No reason to give any less consideration to their opinions than to yours.

Also, KSP has been good enough to get you hooked. If you play career, sandbox or science is irrelevant. 

[eddiew]

Tbh, my issue with the tech tree is not so much that things are in a weird order, it's that it's not long enough nor expensive enough. At standard settings, it is fully unlockable with just Mun/Minmus - possibly just Minmus - with the result that many players go from entry level to end game and never do anything with the mid range parts.

For my current playthrough, I've taken the community tech tree, added Near Future and Space Y, and stretched the costs. Up to the 300 band is about the same, meaning you can get to your first orbit without massive pain, but the 1000 level now costs over 4000, and the end-game stuff is 24,000. Conquering the moon (SSRSS) got me some rover wheels, resulting in a science station around Mars and several surface expeditions to maximise each landing. I've been sending resupply ships and replacement landers to the station rather than launching repeated round trip missions, because efficiency has become important. Once all the data gets processed I'll have nervas, and a choice of pursuing 3.75m tech, or spaceplanes. I may or may not be able to do ion drives.

To me, being presented with the requirement of making decisions because I can only afford some of the next tier is interesting. Being forced to 'make do' with the mid-range tech results in a more challenging game, and means I will actually get to use things that I'd have simply skipped over before. Particularly with regards to modded parts. I have never previously used the lowest tier of Near Future drives, but now the steps from one node to the next are large enough to make it unlikely that one mission will unlock multiple tiers.

TLDR: massively inflate prices at the expensive end of the tree, while keeping the early days tech at a similar price, and your game will change drastically