How do you deal with all the shortcomings of the game ?

[NikkyD]

I am serious!

 

I just finished my horrible Duna Expedition. In a sort of mental debrief i am wondering what i could have done better.
The setup wasn't bad, it's just that KSP can't handle its own mechanics...

Let me start at the beginning. Mission was Duna + Ike, with a Rover that drives to as many biomes as were timely manageable. The rocket consists of 2 parts, a nuclear driven Traveler-Strage with about 5k dV and a lander with 3 Kerbal and a Rover.
1) It was a MESS to build a vehicle that can be unloaded and loaded back up with all science on board including the possibility to store experiments. I know that some players build monster trucks that are huge as a house but i have a problem with that.
I just hate inefficiency! So much that i regularly quit KSP and have to let off steam elsewhere... great. So i built a really small vehicle with the help of TweakScale. I was small enough to fit/drive inside a 2.5m service bay.ofc driving in and out
was easy on Kerbin, on the other planets the wheels suddenly could no longer manage to drive in. The service bay was my "landing gear" so it was on the floor. And ofc the wheels popped inside the bay everytime the ship moved even tho the rover was secured with
a docking port.
Building vehicles to reach as many biomes as possible is one of the BASIC and FUNDAMENTAL things in the game BECAUSE of how the career mode is made. But there are almost NO PARTS to help out. Yes there are wheels... but HOW THE HELL do you transport those vehicles ?
There are cargo bays... sure, and how do you land something with a cargo bay in a way that you can reload the rover ? Ahhh... the huge plane with ramp. Seriously, that is the ONLY option ?! Like no one would ever want to build a rover and drive around on planets.

But hey, with enough cheating and mods somehow i got the rover thing done.

2) Reentry. It's a good thing that reentry finally is more deadly, but mainly its deadly because there are no parts to build proper rockets that can survive it. I know there are heatshields, but how do you build a ship that has a somewhat aerodynamic nose, and engine pointing in the
oposite direction AND a heatshield that points in the same direction as the engine ?! Good thing for me that a service bay is almost better than a heatshield but still i have to mount so many parts radially and they will pop.
I mean how do i protect landing gears that i HAVE to attach radially because the only thing with a bay is the plane gear stuff. I can't wrap everything in a fairing that i rebuild every time. There is this awesome inflateable heatshield, but it is MAX SIZE and heavy AND does not allow
and engine other than radials. And radial engines SUCK (efficiency). I only need ONE nerva so radial placement is off the table anyway unless there would be smaller nerva so that i could split up 1 into 3 or 4.
Yes i could do all that with TweakScale BUT when i tweak a part i can no longer see its stats as i could in the part description. Makes planing harder. Plus i have the feeling that the 4x weight scaling is a little bit cheat.

So how do i design a PROPER rocket that has to do multiple aerobrakes ? I spent AN HOUR until my ship finally slowed down enough through duna that i could orbit without blowing up.

Why was that an issue ?

3) So to plan a mission i have to look at community budget maps. Why ? because the game itself provides NADA. You can't even plot a course at a future time to see how many steps and dV would be required. The game simply provides NOTHING but its silly course plotter.
How much dV can i aero off at duna with my design ? NO WAY of finding out other that cheat/beaming the ship to duna and trying it out. Oh yeah, thats practical.
How lost would i be without Alexmoon's charting tool... how will the planets be aligned in 300 days from now ? Stock game CAN'T give you the answer. You need mods mods mods. And this isn't something special, this is the VERY BASIC in a game that deals with rockets and space travel.

4) have you ever looked closely at the weight of the stock components ? Some are so rediculous. The smallest wheels weigh 50 kg, same as the bigger ones... WHY ? Both extendable ladders weigh 5 kg but one is twice as long ?!
Batteries however. An average car battery weighs a good 10 kg and in KSP the smallest disc-battery weighs the same. What a bargain! All the parts... it's like 5 years ago someone quickly scribbled some numbers together, late for dinner or whatnot and BAM they made it into release and have never
been touched again, EVER.

5) Cockpits. Now with the system of 3 professions it is somewhat obvious that a lot of players will want to bring 3 kerbals with them however there are only TWO cockpits that allow 3 or more and they are utterly bad.
The one is a plane cockpit and the other one weighs... That would all be no problem IF there were nicely stackable 1-kerb-cockpits. But the Mk1 can not be stacked properly and the lander can has horrible aerodynamics.
It would take a developer TEN MINUTES to overhaul those numbers, but no one does it.

 

So i am playing this game since... dont remember... and still so many vital parts are missing. Instead we get a career mode that feels like occupational therapy for ADS. Unity 4 to 5 and imho it still looks the same and i have everything maxed.
Every year we get like 3 new parts, YAY. Most of the time they are not balanced, Duh.
How do you deal with it ? Install 20 mods and once more unto the breach ? I just can't build those cartoon rockets with parts sticking out left and right, i really hate that. So how do you all deal with it ?
And please don't answer by suggesting even more mods, my limit for a year is reached.

 

I REALLY like this whole space stuff but KSP just gives me more and more headaches with each patch.

Edited May 6, 2016 by NikkyD

[Rabada]

Regarding your question about heatshields and engines: Look up how NASA designed the skyscane to land the Mars Rover Curiosity. I often use similiar designs to land probes on Duna and various moons. I would post pics but I'm on mobile.

Edit: Also I find that aerobreaking is not entirely necessary to get a craft into orbit of Duna or Jool. I prefer to use a small burn to get captured into an elliptical orbit and then I use reverse gravity assists using Ike with Duna and Laythe with Jool. In fact I find that 90% of the time I can plot a reverse gravity assist using the big moon around Jool, (I'm having a brain fart, the major moon that isn't Laythe or vall) where I don't need to burn at all to get captured into a good orbit.

Edited May 6, 2016 by Rabada

[Renegrade]

15 minutes ago, NikkyD said:

4) have you ever looked closely at the weight of the stock components ? Some are so rediculous. The smallest wheels weigh 50 kg, same as the bigger ones... WHY ? Both extendable ladders weigh 5 kg but one is twice as long ?!

Batteries however. An average car battery weighs a good 10 kg and in KSP the smallest disc-battery weighs the same. What a bargain! All the parts... it's like 5 years ago someone quickly scribbled some numbers together, late for dinner or whatnot and BAM they made it into release and have never been touched again, EVER.

Highlighted for truth.

MANY of the values both don't make sense AND don't fit into the progression.  They feel like placeholders.  The costs are crazy too - four tailfins is about half of a Skipper price-wise.  A plastic nosecone is the cost of a small engine.   Especially suspicious are items that cost 500 (either exactly 500 or close to 500).

Also I suspect we still have the #lolstat problem of ox-stats being the best cells - I haven't run figures for them since before the last set of solar changes but I wouldn't be surprised to find out they still have the best energy generation per mass and per cost..

[bewing]

For most of your objections, I'd say that's what sandbox mode is for. It takes five minutes to slap a megamax lifter onto whatever design you want, boost it into orbit, and timewarp it to Duna to test. Testing in advance is very practical, quick and easy. You want to know where the planets will be in 300 days? Start a new game, port in your persistent.sfs file, timewarp ahead 300 days, and look. I don't find I need mods to do any of these things.

For reentry, you need to be clever. Aerodynamics will get your stuff to the ground whole with no fuss -- and heatshields never will.

As far as efficiency goes, you can only start measuring efficiency once things actually work correctly. Until that point, every design is inefficient. If it has to be heavier in order to work, then that's what you have to do.

[regex]

Last time I went to Duna, maybe pre-release, I dropped into the atmosphere from 70km orbit on a Terrier and a 2.5m quarter-tank, no issues.  I mean, yeah, if you're coming in from interplanetary space and you want to straight land and aerocapture heatshields are the name of the game, but if you do yourself a favor and put yourself into a stable orbit instead of trying to subject your spacecraft to the rigours of an atmosphere you'll do much better.

Edited May 7, 2016 by regex

[Waz]

Why would you reload a rover back into a rocket? To explore Ike and Duna all in the same mission? This sounds like the basic cause of your problems. Yes, KSP is not an everything simulator, you can't expect every possible idea to be streamlined. My suggestion is to use KAS/KIS to build the rover from a bucket full of parts, not entirely unlike how the Apollo rover was "some assembly required" (and it was left behind, they didn't try to pack it back up and drag it all the way back to Earth).

[Enorats]

While you have some points regarding stat imbalances (which I have to point out are VERY easy to fix yourself by editing the part files, even a total beginner can do that), much of your trouble seems to be in engineering solutions to problems. Much of what you suggest can and has been done. You don't need to have a ramp to get a rover back on board a ship, especially not on a low gravity moon. Give it some RCS and visit Ike first. Your rover can then fly on and out of the bay if needed. Leave it on Duna, no need to reload it. Don't bother stowing experiments again, take the data and leave them. Use interstage fairings to shield unwieldy lander bits until you actually land. The smallest cylinder tanks used as a core with three 1.25 meter ranks placed radially (with or without decouplers) around it can fit inside one of the larger fairings nicely, even with landing legs attached to the tanks. 

 

And.. yes. Mods make or break the game in many ways. That's sort of expected. Squad isn't a large team, so a huge community of dedicated modders is understandably going to be able to do a lot more in the long run. Now that it's unity 5 and 64 bit we can use a great many, so it's up to you to decide which you want.

[[email protected]]

sub'd

[msasterisk]

A thing you can do is have heatshields attached to robotic claws and RCS. Then you can move them in place for aerobraking, and remove them for engine burns. However, if realism is your thing, that won't work.

[Renegrade]

2 hours ago, Enorats said:

While you have some points regarding stat imbalances (which I have to point out are VERY easy to fix yourself by editing the part files, even a total beginner can do that)

I'd recommend ModuleManager for that.  End users can use it too y'know, it's not just for mod authors

[Camaron]

Before I go on, let me just say: I love this game. I've put in probably 5,000 hours of playtime altogether. I'm glad to see that you put together a coherent list of concerns rather than just flaming on about various easily solved issues like some have.

However, some of your issues  come from trying to do something unrealistic, or getting upset that something that is exceptionally difficult is difficult. Taking a Rover back home with you. Pointing an engine and a heat shield in the same direction. You're trying to do a lot of things that simply shouldn't be done because they unnecessarily complicate things.

As for your complaints on lousy parts, I'm with you 100%, and will even add a few complaints of my own.

Most of the models are just plan bad and the textures are even worse. I have a considerable amount of experience in modeling and texturing for games, and it aggravates me how many models aren't even centered properly!!!

Centering a model is modelling 101, guys. I know for a fact that whoever is making KSP's models has been doing so for 5 years now and the fact that he has only recently gotten to the point where I can call his work "mediocre" is disappointing to say the least. If I didn't have better things to do than remodel all of KSP for free, I would have done it a long time ago. I understand the problems that come from faulty floating-point rounding in modelling programs, but as a seasoned modeller, these issues should be figured out by now.

Then we have the issue where a lot of the attachment points aren't quite right, leaving us to correct these novice modelling mistakes manually with the offset function. The assets from which we construct our ships are quite sub-par. Once upon a time I could excuse this, since the KSP team is a tiny Indie dev group and the people in it didn't have extensive experience. But.... it has now been around half a decade. The person building these assets should be an absolute pro by now.

[NikkyD]

15 hours ago, Waz said:

Why would you reload a rover back into a rocket? To explore Ike and Duna all in the same mission? This sounds like the basic cause of your problems. Yes, KSP is not an everything simulator, you can't expect every possible idea to be streamlined. My suggestion is to use KAS/KIS to build the rover from a bucket full of parts, not entirely unlike how the Apollo rover was "some assembly required" (and it was left behind, they didn't try to pack it back up and drag it all the way back to Earth).

Why NOT ? On a 10t lander, a 500 kg rover is nothing. The problem isn't the weight or space, its that THERE ARE NO PARTS to properly pull that off even though it would be so easy. As i have done it, 2.5 service bay and a rover that fits into it. Now make this stuff on purpose and not makeshift and it is absolutely no issue, but thats just it, there is no such part. And it's not just the reloading of the rover, its the landing itself. Putting a rover in a box is a standard requirement imho, but they dont have parts for it... there is just so much stuff missing, it's like they never even played it themselves or they would have realized within 1 hour what kind of stuff is missing.

[FullMetalMachinist]

2 hours ago, NikkyD said:

Why NOT ? On a 10t lander, a 500 kg rover is nothing. The problem isn't the weight or space, its that THERE ARE NO PARTS to properly pull that off

So your response to "why are you making things hard on yourself?" is "why would I make them easy?" 

As you say, there are no parts to do what you want, but that doesn't mean that it's a shortcoming of the game. It just means that you're making an engineering challenge that, big surprise, is challenging.  

If a shooter game doesn't give me a ski pole to stab people with, it doesn't mean that's a shortcoming. It means I need to adjust my play style to the game's tools. 

[Sandworm]

"Putting a rover in a box is a standard requirement imho"

 

Really?  I cannot think of any realworld rover that landed in a box.  They all land on either atop some platform or hanging below a skycrane, both of which aren't difficult in KSP.  My SOP is to hang the rover underneath the lander, using radial jets on a pancake-like fuel tank, then either hand on legs attached to the tank or land directly the rover wheels.

[waterlubber]

If something in game happens that should not happen under normal circumstances (i.e due to physics bugs, etc.) I'll use the debug menu and HyperEdit to fix it.

For example, if something gets blown up due to a glitch I'll HyperEdit a new one there.

[KSK]

2 hours ago, Sandworm said:

"Putting a rover in a box is a standard requirement imho"

 

Really?  I cannot think of any realworld rover that landed in a box.  They all land on either atop some platform or hanging below a skycrane, both of which aren't difficult in KSP.  My SOP is to hang the rover underneath the lander, using radial jets on a pancake-like fuel tank, then either hand on legs attached to the tank or land directly the rover wheels.

Apollo rovers. Not quite in a box but all folded up in a nice neat package on the the side of the LEM. Pull a lanyard, package folds down, wheels pop out and off you go. Good luck doing that in stock - probably not impossible but boy is that going to be a butt ugly lander.

For puttering around the Mun and Minmus, doing flags and footprint missions, maybe a medium sized space station, stock KSP is great. For the most part you have the parts you need and the  trajectory planning tool is wonderful - in fact I would strongly argue that the map screen and the way it gives you a very visual feel for something that's inherently very complex, is KSP's stand-out feature. That XKCD cartoon with the 'learning about orbital mechanics' hockey stick graph? That's the map screen right there. Most importantly of all, iteration and learning by failing is quick and relatively painless.

Go beyond Kerbin space and it's clear that the spaceflight part of the game really hasn't advanced very much since the planets were first introduced. Worse, the tools and 'learning by failing' philosophy that work really well for shorter missions really don't scale gracefully to the added complexity and timescales of interplanetary flight. It's obviously perfectly possible but without mods it's nowhere near the compelling experience that the cis-Munar game is.

The stock game isn't great for actually building and doing stuff in space or on the planets either. Again - it's possible but it's very rough and ready. Yes, there are mods, but some improvements in the stock game would go a long way too. Plus - and it's a personal thing, I get a bit tired of the 'mods will do everything' attitude. Some folks are happy with a framework game that they can flesh out with mods - that's fine and I respect it. I prefer something a bit more substantial which can then be improved or tweaked a bit further with mods rather than relying on them to build my whole game for me. Mods as the icing on the cake if you like, rather than substantial ingredients in it.

Last word - Squad are a small team and they've put a ton of work into KSP. On my side of the fence, I'm well aware that it's easier to criticize than create. I'm also one voice amongst hundreds when it comes to ideas and opinions on the game. With all that said, I still think OP has a valid point. How much of what we attempt in KSP is hard because its intrinsically difficult and how much of it is hard because of arbitrary limitations in the tools and parts available? 

[lobe]

A stiff drink and positive thoughts.

[N_Molson]

I totally agree that some parts, especially batteries, have out-of-common-sense mass values.

Also, a lot of parts have quite surprising heat-resistance values (a battery pack full of acid, as an example, should blow very early when heating too much - and that would make radiators useful -)...

[RocketBlam]

You have some valid complaints, but I can't respond to a wall of complaints, I don't suppose anyone wants to do that.

KSP has to balance being fun for beginners, and being fun for more advanced players. And also being fun for superhuman, amazing players that have returned a Kerbal from the surface of Eve. Ahem.

So, what I'm saying is, for KSP to be the game I really want it to be at this stage, it would be overwhelming for those people that have trouble landing on the Mun.

That being said, a lot of your complaints could be solved with some mods. Infernal Robotics is really good for making things like a rotating arm that can deploy a rover. I'd look into that.

Another solution, possible with stock, is to put the rover on top of the craft with a decoupler, and a small rocket to shoot it up and away from the main craft. Then use a chute to drop it back to the ground nearby, or rockets if no atmosphere. In fact I think I'll try that right now.

This may sound crazy and complicated, but you should see some of the nutty things NASA has done, that have actually worked.

Part of the fun of KSP is coming up with new, clever solutions to problems like this, without having all of the tools you think you need.

Edited May 7, 2016 by RocketBlam

[Fallarnon]

I think @RocketBlam summed things up rather nicely, I was sitting here trying to think of the best way to post my thoughts for consumption here and his last line caught my eye just above the editor.

9 hours ago, RocketBlam said:

Part of the fun of KSP is coming up with new, clever solutions to problems like this, without having all of the tools you think you need.

You're having trouble doing things they way you've decided they should be done @NikkyD but who's to blame when you try to do something knowing you haven't got the tools in place to do it? Squad can't possibly anticipate and design for every single approach players might come up with. That's where modders step in. I'd much rather have Squad focus on things to improve everybody's experience like porting to Unity 5 for 64 bit support than coming up with part packs or ensuring that the battery weights make logical sense.

 

On another note, does anyone else have trouble taking a thread seriously when the poster insists on shouting to add emphasis? You'd think we were on a plain text board or something, jeez.

Edited May 8, 2016 by Fallarnon

[TMS]

To buy and play KSP is to accept that the game is massively under-resourced, under-developed and will rely on the free efforts of a community of modders to get the game resembling a finished state.

Rule of thumb: don't play the game for at least a month after a patch.

[gchristopher]

On 5/6/2016 at 2:21 PM, NikkyD said:

2) Reentry. It's a good thing that reentry finally is more deadly, but mainly its deadly because there are no parts to build proper rockets that can survive it. I know there are heatshields, but how do you build a ship that ...

This is EXACTLY what is wrong with Reentry! You can design for reenry, but the results use a limited selection of parts and tend to be monotonously similar, which greatly detracts from the freewheeling sense of fun that makes Kerbal more appealing than a strict simulator. So many more fun designs could be made viable if you could just pick a part and add some thermal tiles at the expense of a little more mass. This is my biggest turn-off for newer versions of the game. 

I hope they can revisit that general topic someday. 

(Gripe #2 for me is the replayability of the contract system, but that's a lot harder to address than reentry flexibility shortcomings. The devs have my sympathy there.)

[John FX]

It seems there are two camps generally when it comes to this sort of question. The first wants to design craft in a limited setting, with limitations imposed by the game design, in order to successfully fulfill a desired mission. This camp generally does not want information tools.

The other camp as far as I can see wants to design craft in a limited setting, with limitations imposed by physics, in order to successfully fulfill a desired mission. This Camp generally does want information tools. I get the feeling the second Camp is more open to mods.

You can't really say one playing style is wrong and the other is correct because it is just a matter of preference.

As games designers, Squad should provide a game where both playing styles can be accommodated. This would mean providing the tools for information that are needed for one side but making them optional so the other side can turn them off.

To my mind this would mean essentially making a physics setting for difficulty and a game setting for difficulty so you could choose whether your in game difficulties are imposed by physics or the game...

[Superfluous J]

I deal with the bugs by playing all the other games that do what KSP does only without the shortcomings.

Oh wait there are none.

Edited May 8, 2016 by 5thHorseman

[Vermil]

Right now, I deal with KSP by not playing it. That's right, I'm taking a pause because it seems like crashes are regular since 1.1/1.2. It will be better eventually.

But otherwise, I've found it's all too easy to blame the game for your own inadequacies. So I deal with the game by experimenting and testing, and testing again, ...and testing...

And progressively adapt my vehicles so they do work inside KSP's universe as it is. And, yeah, I don't use any mods.

What's more, this progressive testing and development is probably what I find most fun about KSP.