.22 Feedback/Suggestions

[Caelib]

I think v.22 is a marvelous update, but after having unlocked the entire tech tree and spending quite a bit of time playing, I have a few suggestions:

1) Obviously, the game is still in development, but there should be some "guidance" when starting a new career mode game or new players are going to be very lost on how to advance. I feel like the game needs some form of "here are the things you can do" for the planets/moons you have visited/unlocked. For example, a new player would automatically see Kerbin in this list with of each biome and their "accomplishments" in each. For other planets/moons, this could be as simple as "High Orbit", "Low Orbit" and "Surface".

2) Building upon my previous point, it would be really nice if we had a way to view our progress of discoveries for each planet/moon. When unlocking new science equipment, it's difficult to remember where I have used what!

3) Antennae should remain in the open position if they have been manually extended; currently, every time you send a transmission, it is automatically retracted.

4) I love the concept of returning ships to Kerbin for maximum science gain, but it seems possible to achieve similar results by making multiple radio transmissions. A player can repeatedly send these tranmissions and get a lot of science data, but the process seems clunky. I get the overall idea with making sending data back less valuable then bringing it back, but my suggestion would be to streamline this where a player can gather ALL of the data at once, then the transmission of said data be sent in multiple batches depending on available power.

5) Now that we have science equipment that appears to gather meaningful data, it would be awesome if we had metrics. Sure, there are add-ons that do this, but I feel pretty strongly that the core game should at least have some more native functionality for stage-based Delta-V and TWR.

6) It would be awesome if we could setup maneuver nodes from the launch pad so we can intelligently launch into the plane of our target and AT THE RIGHT TIME -- right now it's an incredible waste of energy to launch 90 degrees and then change planes ... and THEN time accelerate and create your maneuver nodes in orbit.

7) Finally, can we please get some stock payload fairings?

Thanks for making such a great game!

[EasyAce]

Nice thread!

I would like to see the developers to change the tech three to a more logical one.

There is no reason to not being able to put together a simple plane from the start. Also there is no reason why the thermometer and barometer are so high up the tech three!

The way it is now forces me to travel around on Kerbin with rockets, either several times OR after I've been to the Mün... just to gain science to "explore" the thermometer.

I love the way science work - it should be "a must" to explore the deserts on Kerbin to make prepare for the dusty Mün travel (getting science to improve rockets),

but now I'm sending Jeb (with rockets - planes haven't been invented of course) to the desert/ice caps to dig som samples and recover... only to have to do it again with the thermometer and barometer - instruments that are essential and most probably invented earlier than the rocket engine.

Also I love that you gather science as you do a Mün flyby. The science can be used to improve/prepare the Mün lander... but the ladders are SOOOOO far down the tech tree.

FIXED ladders... so hard to invent... it has to be invented after, among others, the parachute.

I usualy land on the Mün before I've invented ladders... so poor Jeb falls of and need to use his backpack to get back into the pod... and when he goes back to the Mün again - the Kerbal engineers have still not invented the ladders... :-/

I propose that more basic items area available earlier in the game, and maybe they add more advance science equipment like drill, sample storage, particlegatherer, radiationinstruments and so on.

[Caelib]

I had similar feelings about the ladders until I realized they weren't necessary for the Mun or Minmus as you can just use your jetpack to re-enter the craft. But to your point, the tech tree seems to be organized in such a way to limit a player's ability to progress rather than scientific ability. I believe this is to combat the "paralysis of choice" problem with giving the player too many parts at the start. But I agree, the ladders should be available sooner.

[artao]

I haven't played career much yet, but I already agree with all your points.

One thing I'd like to add: PLEASE give us unmanned probes from the very beginning. BEFORE manned capsules even. That just makes sense.

[Godot]

Unmanned probes would require electric power (and Batteries and ways to generate power (aside from the rocket engine) are only unlocked deeper into the tech tree ...

also you would need some scientific instruments from the beginning on if you want to have unmanned probes make sense

(after all, in its current state (aside from recovery of vehicles that survived the flight) the only way to generate science in the beginning are crew reports)

[vetrox]

2) Building upon my previous point, it would be really nice if we had a way to view our progress of discoveries for each planet/moon. When unlocking new science equipment, it's difficult to remember where I have used what!

This is my only gripe with 0.22

Just a simple UI that shows the experiments you have done on what body and how many times you have done it. Maybe with greyed out ones to show you havent done a certain experiment but not telling you what it is!

I'm on my 3rd career start now just so I can keep track of where i've been.

[Godot]

In fact (as long as there is nothing in the stock game) there is a mod that does exactly this:

http://forum.kerbalspaceprogram.com/threads/53952-Research-Development-Overview-*Permanent-Update-Thread

[MR4Y]

I agree with all the points in the thread, also.

1) It's been a few versions since science parts where introduced, and so far they only serve asthethics. You can't read them in real time and the same painstaking method also applies to the new science parts. How long before we can access their functionalities in a menu or GUI window without having to look for the part in the ship and hope to not misclick it, to hopefully have it's menu shown to, perhaps click on the option we want? It's funny that you can have functional NERVA engines and somewhat working ion drives, but a real time readout is somewhat an unreachable ladder.

2) I won't comment about the science tree for a simple reason. I don't know how to use and achieve science efficiently. The game assumes you know how to use a feature you've been just presented. There are no tutorials anywhere. Apparently, only a few select people know how the feature work and, apparently, are not willing to share it.

3) Also, when certain parts will have functions or IVA's or finally be replaced with parts that have IVA and/or function? I know it's something small, but they have been neglect since they appeared as more parts appear and get implementation.

4) Adapters that carry no fuel would be nice.

[artao]

Well, manned capsules also require electric power. The lack of batteries in the very beginning really needs to be addressed IMHO. At least some other way to generate electricity without firing the rocket engines -- which shouldn't really generate electrical power anyhow.

And, yes .. more scientific instruments from the beginning too please. Thermometer, barometer. Simple stuff that logically would exist if rocket engines exist. We know how hot the engines are, but DON'T yet have thermometers? uhhhh LOL

[John FX]

I`d like the nodes to have a sample return requirement. For example, reports from Duna say it is a bit radioactive so a soil sample should be required to unlock the NERVA. Biomes on Kerbal should unlock parts required for space flight initially to encourage plane/suborbital development. Maybe soil samples from around KSC could unlock the plane parts...

[Caelib]

In fact (as long as there is nothing in the stock game) there is a mod that does exactly this:

http://forum.kerbalspaceprogram.com/threads/53952-Research-Development-Overview-*Permanent-Update-Thread

I haven't seen this yet, thanks for posting the link!

[MR4Y]

Well, manned capsules also require electric power. The lack of batteries in the very beginning really needs to be addressed IMHO. At least some other way to generate electricity without firing the rocket engines -- which shouldn't really generate electrical power anyhow.

And, yes .. more scientific instruments from the beginning too please. Thermometer, barometer. Simple stuff that logically would exist if rocket engines exist. We know how hot the engines are, but DON'T yet have thermometers? uhhhh LOL

That's the sad part. We have those parts since 0.18, but they require you to right-click them for a one time reading, which is unpractical and time consuming.

Also, will be too much to ask for a radar altimeter? Some low game settings makes ships not cast shadows, which makes almost impossible to land if you're running the game on the lowest settings.

[EasyAce]

I`d like the nodes to have a sample return requirement. For example, reports from Duna say it is a bit radioactive so a soil sample should be required to unlock the NERVA. Biomes on Kerbal should unlock parts required for space flight initially to encourage plane/suborbital development. Maybe soil samples from around KSC could unlock the plane parts...

Exactly! Fixed ladders, thermometer, barometer, basic battery, awefull engine and planeparts should be available at the start. If that make it to easy they could fex make it impossible to do an iva untill you've used the thermo- and barometer in space so thye "have to invent the space suit. And they should give each tech three point some text explaining what to explore for it to open. F.ex will the small upper stage engines be unlock only when you've launched a successfull part into orbit and the kerbal understand they need a more improved engine in space.

[kerm]

First off, let me say how much I enjoyed my first pass through career mode. The new animations and experiments are wonderful. The addition of subassembly dramatically increases the "fun quotient" of stock gameplay. And the introduction of biomes turns planetary exploration from a chore into a delight... I can't wait to see these rolled out to every celestial body.

Now for the bad stuff: like many other players, I have some problems with the way science is implemented. I've seen many other threads expressing similar concerns. Here are the most frequent criticisms, all of which I agree with:

1. It's too easy to spam-transmit science.

2. All experiments happen instantaneously.

3. It's too easy to max out the tech tree (took me about eight hours of gameplay, and I'm not an expert by any means)

4. There is not enough visibility on data collection.

5. The tech tree somewhat unbalanced and illogical.

Here are my suggestions for solutions. I've ordered these by my estimate of how much work they would mean for the devs:

1. Move fuel lines further down the tech tree

Once you understand how to use it, asparagus staging turns the game into a cakewalk. Fuel lines should be a tier 4 tech. Arguably they should even be in a newly-created tier 5. (This would correspond well to the real world, where asparagus staging is still in the experimental phase.)

2. Make the back end of the tech tree far more expensive

Most of the criticism I've seen focuses on the early part of the tree. These critics fail to understand an important real-world consideration: the early game MUST be kept simple, so new players are not overwhelmed. This is very important to maintaining SQUAD as an economically viable developer, and to ensuring a continuing stream of new players into the KSP ecosystem. (Anyhow, the fact that Kerbals invented rockets before ladders makes a perverse sort of sense to me.) On the flip side, though, the back end of the tech tree does not have to be friendly at all. In fact, it can be downright fiendish! Tier 4 techs currently cost 550 science a pop. I'd increase that by a factor of ten. Maxing out the tech tree should feel like an accomplishment, not an inevitability.

3. Disallow repeat transmissions

This is the most commonly proposed idea, and I totally agree. Even if you bind your instruments to an action group, pressing the send data button 144 times in a row is neither realistic nor fun. Instead, each experiment should have one maximum transmission per zone. Data-based experiments like the thermometer, grav detector, barometer and crew report should each immediately give 100%. Materials-based experiments like Surface Samples, Mystery Gooâ„¢ and Science Jr. should give 20% once, with full points only available on recovery. This is probably the most important fix and also one of the easiest.

4. Require a data-collection period for certain experiments

Things like Seismic data collection and Science Jr. should take days, weeks, or even months to complete. This encourages the establishment of bases and other permanent presences in space and on planets. Returning live Kerbals back home after long journeys should give us science too – i.e., the physiological effects of long-duration exposure to deep space and other planets should count as an experiment in itself.

5. Nerf the antenna

The basic Communotron 16 unit is totally overpowered. It can transmit flawlessly to Kerbin from the far side of Eeloo at maximum distance with the Sun in the way. The more advanced antennae are actually inferior, offering only increased transmission speed at double the electricity cost. (Transmission speed is almost never an issue.) This problem can be fixed by introducing a maximum transmission distance for each antennae, with exponentially increasing power requirements and line-of-sight considerations taken into effect.

6. Knowledge base for collected data

Everybody wants this. We should be able to know what experiments we've run where, and be able to see the collected data. This could maybe be sorted by celestial body in a handsome report? And while we're at it, why not have a...

7. Trophy case at the R&D center for returned surface samples

Sorted by celestial body and biome, with accompanying chemical analysis. This wouldn't directly affect gameplay, but it would be totally awesome. And lastly, of course...

8. More SCIENCE!

Geiger counters, magnometers, RADAR mapping, core sample drilling, methane probes, neutrino collectors, etc. etc. Also, there should be at least one multi-part experiment (Science Sr.?) so heavy that it requires orbital assembly. I feel like there really is no limit to the potential fun here.

So those are my long-winded thoughts. Thanks again to the devs for such a great game.

[Caelib]

kerm ... great response! I agree with pretty much everything you've written!

[EasyAce]

kerm ... great response! I agree with pretty much everything you've written!

Yes, so do I!

Great reply!

[Eric S]

1. Move fuel lines further down the tech tree

If asparagus staging were the only thing that used fuel lines, I'd agree, but all my decent non-low-gravity landers are dependent on them as well, mostly because fuel doesn't flow between radially attached tanks, and for mid to low gravity landers, I always have more tanks than engines, and I really don't want to try landing a tall lander. So I go wide.

2. Make the back end of the tech tree far more expensive

Probably, but the other issues need to be sorted out first. 5500 research for something as important as Mainsails would be a real bottleneck if we put an end to 4000K Mun missions. Sort out the other stuff, and then see if the tech tree costs need to be rebalanced. There's also the point that the tech tree isn't the end all be all of career mode, it may be less important once we've got a budget and other career mode features.

3. Disallow repeat transmissions

Or have diminishing returns kick in harder on transmitted results so that you can only get 50% of the max research from transmitted data, the last 50% has to be with recovered experiments, for those experiments where it makes sense that a recovered experiment would yield much more data. Sample returns (including atmo samples), Goo and Material Science would definitely fall into this category. Maybe make the diminishing returns cap for transmitted results be the percentage loss, so materials science could get the first 20% with transmitted results, past that you'd need returns. EVA reports would get 50% transmitted results, and crew reports could do all 100% with transmitted results.

Another option I've seen proposed would be a limit on the number of times certain experiments can be run with a single part. For example, a materials science bay could have enough materials to run the experiment at most 5 times. If you wanted to run a materials science test on each of the 14 Munar biomes, you'd need to take three Science Jrs, or run multiple missions.

4. Require a data-collection period for certain experiments

Disagree. Timewarp makes any time-dependency pointless unless we get life support, and even then, it wouldn't be hard to get around.

5. Nerf the antenna

I suspect that this is coming. The devs have already talked of having satellite relays, and that wouldn't make any sense with this antenna.

6. Knowledge base for collected data

7. Trophy case at the R&D center for returned surface samples

Nice, but not necessary. Some of this other stuff is necessary.

8. More SCIENCE!

It would be nice, but it could exacerbate the problems with science points being too available. Especially since most science would just be sensors where it would be hard to justify a big hit on transmitted results.

[Sathurn]

I would just Like to say +1

And while it is possible to to "game" a longer duration experiment with time warp, I would still like to see them included. When you add crew resources to the game, time warp becomes less of an option. The one thing that does need to be address is to allow data collection while on rails. Because if I have to stay with one experiment for a game "month" then I do prefer instant experiments. But if I could land a "base" then go fly another mission the longer time-frame experiments would not be unbearable. In-fact I think they would be preferable from a role-playing prospective.

[kerm]

Good! I was hoping to engineer this sort of debate. I am also willing to admit I'm overdoing it with some of these requests.

If asparagus staging were the only thing that used fuel lines, I'd agree, but all my decent non-low-gravity landers are dependent on them as well, mostly because fuel doesn't flow between radially attached tanks, and for mid to low gravity landers, I always have more tanks than engines, and I really don't want to try landing a tall lander. So I go wide.

There is absolutely no question that fuel lines (and, I should add, LV-N engines) unbalance the game in favor of the advanced player. Regardless of your personal usage, they are widely understood to be two of the game's key techs... in fact, this is mathematically provable! They should both be tier 4 at a minimum. "Gameplay value" is subjective, and difficult to support with hard data, but I hypothesize that there is much to be gained from being stuck in the "stone age" of Saturn V tech of career mode for as long as possible. Many of the better player challenges impose exactly this restriction.

Probably, but the other issues need to be sorted out first. 5500 research for something as important as Mainsails would be a real bottleneck if we put an end to 4000K Mun missions. Sort out the other stuff, and then see if the tech tree costs need to be rebalanced. There's also the point that the tech tree isn't the end all be all of career mode, it may be less important once we've got a budget and other career mode features.

Okay, excellent point, and if money budgeting is introduced this will be less of a problem. SQUAD (and indeed every developer) has to strike a difficult balance between casual "Angry Birds" gamers and hardcore "Dwarf Fortress" gamers. The issue of balance affects any modern developer, and there are no easy answers here, I know.

Or have diminishing returns kick in harder on transmitted results so that you can only get 50% of the max research from transmitted data, the last 50% has to be with recovered experiments, for those experiments where it makes sense that a recovered experiment would yield much more data. Sample returns (including atmo samples), Goo and Material Science would definitely fall into this category. Maybe make the diminishing returns cap for transmitted results be the percentage loss, so materials science could get the first 20% with transmitted results, past that you'd need returns. EVA reports would get 50% transmitted results, and crew reports could do all 100% with transmitted results.

Another option I've seen proposed would be a limit on the number of times certain experiments can be run with a single part. For example, a materials science bay could have enough materials to run the experiment at most 5 times. If you wanted to run a materials science test on each of the 14 Munar biomes, you'd need to take three Science Jrs, or run multiple missions.

Well, we agree in theory, if not implementation. Everyone understands some kind of fix is needed here. This is a critical point.

Disagree. Timewarp makes any time-dependency pointless unless we get life support, and even then, it wouldn't be hard to get around.

OK, you're right. What I actually want is a high-value science experiment that requires yearly or even monthly manual servicing. I'm not sure what this would look like (space telescope with a busted lens?), but the key point is that you must master orbital docking or precision landing for it to have value. The problem now is that you can blast through the tech tree without ever using a docking port... overriding the "need" for many of the game's key techs. But contract missions are coming soon (in 0.23, maybe?) and I suspect that will lay many of these concerns to rest.

I suspect that this is coming. The devs have already talked of having satellite relays, and that wouldn't make any sense with this antenna.

I am sure you are correct. You can tell from the in-game descriptions of the various comms devices that range is ultimately intended to be a limiting factor, and I suspect the devs have already had this exact discussion. Nevertheless, descriptions of other spaceship parts refer to both meaningful aerodynamics and significant re-entry damage, and we are still waiting on realistic implementations of both. The fact remains that a tier 0 tech is manifestly superior to a tier 3 tech.

This is a serious error! Indeed, it is the only aspect of 0.22 that is irrevocably broken. The advanced satellite antennae are as useless as the nose cones, and, even worse, they cost significant science to obtain. The presence of useless parts like nose cones and advanced antennae is the game's biggest design flaw, and these errors should not be compounded. Every part should have an immediate purpose, even in beta. Antennae need to be fixed as soon as possible. A hotfix might even be required.

Nice, but not necessary. Some of this other stuff is necessary.

If you'll permit me a rebuttal: TROPHY CASE !!!!!!!1!1!!

It would be nice, but it could exacerbate the problems with science points being too available. Especially since most science would just be sensors where it would be hard to justify a big hit on transmitted results.

This would be balanced against an overall increase of the required science points for advancement. There is a tremendous amount of player excitement about science, and indeed this is a general human quality. In the real world, actual people have volunteered their lives to increase this knowledge! (http://www.space.com/22758-mars-colony-volunteers-mars-one.html) If the devs are interested in gameplay value, this is a gold mine. We want to know the facts about the universe we inhabit, virtual or real. Give us the tools.

[Caelib]

I love the idea of a trophy case for soil samples from other planets/moons -- it's the ultimate achievement IMO.

[Lohan2008]

7. Trophy case at the R&D center for returned surface samples Sorted by celestial body and biome, with accompanying chemical analysis. This wouldn't directly affect gameplay, but it would be totally awesome.

This ^

[Pulstar]

Many of the points I wanted to make were already made in this topic or elsewhere but I'll try to be a bit original:

1) Tree being too easy to fill out with prior meta-gaming knowledge. That is you can min-max mission design for maximum yield with early and later parts, the most extreme of this being cases of Tier 0 Minmus and Mun flybys in one mission. Min-maxing and any other optimization will always happen and in every game, no matter how good the balance is (also it falls into the area of being knowledge-based challenge, while knowledge-based challenge in games is not really challenging as it doesn't test skill, but I digress), however I think some science components and the materials bay in particular are too light. Doing these large research vessels with multiple materials bays or goo pods is simply too easy by the time you reach the middle of the tree. Also the tree should possibly change with difficulty settings, with parts swapped around a bit.

Of course science parts being expensive or other limitations added in the future may kill the science combines that can easily be sent to Duna or Eve. However what should be kept in mind is that the tree should not be a grind, it should simply nudge the player towards going into new places and trying new things, without railroading him into specific missions at specific locations and stages of the game. The openness of the system in how you want to get science (and what to research) are what makes it great (more superlatives and what is good stuff saved for the end of this post).

2) Data transmission is good in principle but needs new mechanics to better flesh out its advantages and disadvantages. Antenna strengths, range and other factors being non-issues for now is the obvious one, as is the fact that the only penalty for not opting to return data is the tediousness of having to click more assuming you have enough power (bar time constraints like transmitting while falling through the atmosphere) to drain the science pool. The most commonly suggested solution is doing a separate total science available pool for returning samples where it makes sense (for goo canisters, soil samples and materials bays) which I support myself.

3) Probe missions versus manned missions. This bit is quite schizophrenic now (no offense to actual schizophrenics meant). On one hand manned missions can do more experiments, although in some cases those could be done with probes with new kinds of parts (cameras for reports, sample scoops for soil sample analysis). On the other there is the issue with transmitting benefiting probes more, as indeed does experiment return in the cases where it can be done.

With Kerbal skills added, like obvious geology, physics or other science skills, you could have a bonus to the science gotten from experiments made by kerbals themselves, derived from that skill, in addition to base science value. Of course this bonus would not drain the available science pool from an experiment case, only the base value would so that overall doing science with Kerbals would yield more science as it requires more effort. A geologist will pick the best rock sample, he can after all study the rock in more detail before picking it up than a scoop with robotic cameras can. Being on site has advantages to looking at limited resolution still photos that took minutes to reach Kerbin. The same could apply to having a physics PhD Kerbal or an engineer mess around with a Materials or Goo pod in orbit. With a skill system you can have those kerbals improve in their skills over time, this doesn't need to even be more than one variable increasing with use to make people not only want to use Kerbals but also to get them back home.

4) Long versus short term study. All experiments are instant now, but it doesn't make sense in the case of some. Of course the thing is you can time warp over any waiting element. I mentioned it in another topic, you could have two experiments on a sensor, the instant one and long-term study. Both would have different science pools and yields, but the idea is to have the latter be a smaller but still noticeable amount given after a long period of time since it began. So that most players don't feel terribly forced into time warping for it (some always will). I already said this would need a system for tracking events, kind of like the Alarm Clock mod works, to alert you when you got results you can send back.

5) Spaceplane (parts) feel awkward in Career mode's current form. This can be fixed by adding electric propellers (possibly with balloons) and biomes to other planets, the latter will of course happen but the former would really give incentive to try them out when the biomes get added. Without biomes their only use is on Laythe. This issue also applies to rovers to an extent. Now the other issue would be aerial probes being better than rovers, but airless moons and planets would still give rovers a niche. Also propellers or balloons could be further up the tech tree. Both are IMO very fun concepts of exploration. I think it would be very educational to have them and show that yes, even a balloon and an old tech like a propeller can have a use in space exploration.

6) Space Stations also feel awkward, there is another topic about that so I'll just say that while they should have a science function it should be a secondary reason to get them, with using them for resupplying or refueling the primary one. Here's the post in the other topic where I explain in detail what I mean.

The good things about career mode which should be preserved and thus kept in mind when considering making any changes:

- Not being depended on clicking once and time warping "to victory".

- Requiring effort in getting to places.

- Open structure allowing players to pick their own mission profiles, rather than railroading them into set paths with excessive science requirements for unlocking nodes.

- Encouraging players to visit new places rather than grind in the same location, giving incentive to land in those unique regions of planets Squad crafted like Dunar poles (when Duna gets biomes).

- discovering experiment results and reading them is fun, which ties into the point above. Also while some results are silly in typical kerbal fashion some are quite educational in part.

I am really, really pleased with all those good things I mentioned above. It was a positive surprise to see a system like this, rather than one of spend money, have guys research tech, get it. This is overall a very solid piece of game design that of course needs fleshing out, polish and tweaking but overall really great in how it interacts with the player and what it makes it do, most importantly taking the best part of the game which is allowing the player to do things his own way, with what he wants, where he wants and how he wants within the limits of available parts and (in the future) budget and other available resources.

Edited October 22, 2013 by Pulstar

[iLlama]

also planes are a bit useless at the beginning maybe that will need some fixing?

[Caelib]

also planes are a bit useless at the beginning maybe that will need some fixing?

I think Spaceplanes probably need their own dedicated update ... IMO, they are too difficult to build/balance/fly for anyone but the truly elite/dedicated.

[Eric S]

There is absolutely no question that fuel lines (and, I should add, LV-N engines) unbalance the game in favor of the advanced player.

Getting slightly off topic here, I feel that fuel lines are a sticky problem if you want to discourage asparagus staging without discouraging legitimate uses. Fuel lines mostly have three uses.

1) Asparagus staging or crossfeeding (directly towards the center). I'm including drop tanks without engines in this category as well, though those are more practical and barely post-Saturn-V tech.

2) Controlling shift of CoM on spaceplanes.

3) Connecting radially mounted tanks that aren't separately staged. This isn't an issue you encounter in real life often, because in real life, they'd just make a fuel tank with a custom shape, so we're not talking post-Saturn-V tech here.

I don't think anyone has problems with fuel lines for cases 2 or 3, and most people don't have a problem with basic crossfeeding, it's just asparagus staging that they have a problem with. Too much of a good thing, I guess.

Case 2 probably needs a non-fuel line oriented fix anyway, and assuming we get a fix like that, I've got a solution for case 3. Two levels of fuel lines. One that can only connect tanks within the same stage (not separated by decouplers/separators/etc) which would be useful for case 3 but not case 1, and full on as they are now fuel lines that could be used for case 1.

Case 3 could probably also be handled by making liquid fuel and oxidizer flow between stacks like all the rest of the resources but not across any of the decoupler-type items. This way you wouldn't need fuel lines to deal with this, but you're still not draining fuel from one stage to feed another except by deliberate action. I actually like this idea better than my first suggestion, but I'm not as sure that it might not have unseen consequences.

Many of the better player challenges impose exactly this restriction.

Agreed. Personally I think that asparagus staging is the natural adaptation of 100% reliable infinite flow fuel lines combined with lower than realistic TWR so I don't have a problem with that, but I still like knowing that I can get away without it (or even knowing that I can't for my really huge payloads).

The issue of balance affects any modern developer, and there are no easy answers here, I know.

Especially when you're dealing with a sandbox game where there is no difficulty setting and the only way to make something harder is with artificial restrictions or doing more ambitious missions.

Well, we agree in theory, if not implementation. Everyone understands some kind of fix is needed here. This is a critical point.

This is definitely what I see as the biggest issue in the current science implementation, and I think most people are of the same opinion. Personally, I think any of the three changes here would be an improvement. Which would be the best improvement would depend on just what effect you're trying to have. I obviously have my favorite because it pushes the player towards return missions without punishing early one way trips too much, but it does little to curb multi-SoI missions that get huge returns by doing a little science all over the place. The "charges" on the science equipment is the only proposal short of eliminating transmitted returns on at least some of the experiments that I've seen that would curb that. Then again, not sure that it would feel that abusive if we have decent limits in place that cover the single SoI or single biome missions.

OK, you're right. What I actually want is a high-value science experiment that requires yearly or even monthly manual servicing. I'm not sure what this would look like (space telescope with a busted lens?), but the key point is that you must master orbital docking or precision landing for it to have value.

Agreed, I think it would be an interesting mechanic, I just don't see how to achieve the desired result. Obviously you don't want to just let them ship all the needed resupply parts/whatever along in the original mission, because then it does become "time warp towards science."

This would be balanced against an overall increase of the required science points for advancement. There is a tremendous amount of player excitement about science, and indeed this is a general human quality. In the real world, actual people have volunteered their lives to increase this knowledge! (http://www.space.com/22758-mars-colony-volunteers-mars-one.html) If the devs are interested in gameplay value, this is a gold mine. We want to know the facts about the universe we inhabit, virtual or real. Give us the tools.

Yeah, what I said wasn't really a reason not to, more just a matter of "this might complicate balancing the costs of the tech tree."