Tips for Hard Career?

[Ant P]

So, I was looking at some rocket designs incorporating the LES, which made me think "I've never used this," which made me think "that's because I normally just revert in a disaster," which made me think "successful launches would be much more satisfying against a backdrop of consistent failure."

So I'm going to start a career on Hard.

Other than keeping an eye on my dV/TWR during launch (I have installed Kerbal Engineer) and trying to complete multiple contracts at a time where possible, what are some techniques I can employ to optimise my performance in Hard mode?

In particular, what is a good way of escaping the habit of launching multiple suborbital test flights before committing to a specific design (do I try and launch unmanned rockets first for each new design... or do I just need to be more thorough when designing in the first place)?

(not sure if this is the best place for this thread since it isn't really an "answerable" question).

Edited October 20, 2015 by Ant P

[Jaeleth]

1. best tip of all... Even playing at hard mode, allow for quicksave. The game can and WILL turn against you even if you do everything right and you might end up loosing countless hours of gameplay. You can always decide if you accept the loss or reload at anytime and, this way, you'll be safe from bugs...

2. Periodically check if you contracts are still all there. Particularly when a recent contract is completed and always BEFORE saving game. When you notice some bug has ocurred in game, whatever type of bug, go to contracts and see if they are still all there before saving. There's a bug, in career mode, where all contracts are wiped out and a row of new ones appear. If you already spent your resources building ships for contracts that evaporate... You're done for

This is a very serious and disrupting bug! If you do not allow for quicksave in mid game this bug WILL get you and WILL kill you, sooner or later.

3. In the beginning, balancing the budget is top priority. Limit you designs to "crude and workable" forget "fancyfulness" factor.

4. You will be following certain tech tree choices. Before accepting any contract make sure that you have or are in the verge of having the required techs, otherwise you might get into a dead end... Ignore risky contracts, later on you'll have the chance to take them all if you wish.

5. Try to complete several contracts with a single launch, but you already said that, this apllies to all stages in career mode, even when you got endless tech and huge resources.

6. If you are unsure with your designs you can always try them in sandbox, consider that simulating, NASA trained astronauts in simulators time and time again before declaring them "mission-ready" and, nowdays, all spacecraft and aircraft are simulated before manufacture. But if you want to do it the really hard way avoid using kerbals in testing the vehicles, use parachutes to recover as most as you can from the vehicles and use a tested kerbal recovery system before you send a kerbal in it. Make sure you trigger the rescue sequence with the abort button and not through the staging sequence. Take time to see if there's something wrong with the design before launch, use the CoM, CoL and CoT tools frequently.

7. For aircraft use this method (used in real life): First check stability, apply throttle, let it gain some speed, then thr off, brakes apply, see if it runs smoothly on runway. Next see if it has a good CoL, TWR and landing gear placement. Full throttle, see if it accelerates fast. It should be reach rotate speed before 1/2 runway, so pull up gently on stick until nose comes up, if it does so before 1/2 runway is ok, thr off, brakes aplly. Next take it off ground... From now on there is no option but a smooth landing or loosing the vehicle. Do a circuit around the runway only. Test for optimal TWR, if nose is not too far up (high AoA) in order to achieve level flight. From there on, just do longer flights, aerobatics and or SSTO if it is the case. Remember to have an ejectable control module, with parachutes if you take a kerbal. if landing gear is ok, emergency landing almost anywhere on kerbin is feasible, if landing gear is lost (usually meaning plane suffered strutural failure) but, somehow, it is still controllable, ditching is better than crash landing in ground, unlike real world.

Edited October 20, 2015 by Jaeleth

[Jimbimbibble]

Also get the stage recovery mod. I think it makes a pretty good attempt at realism and feels non-cheaty. This will make a huge difference in your budget and enable you to design non-ssto rockets that can be fully recovered.

[Empiro]

Hard isn't super hard if your designs are pretty good. There's been threads that have analyzed things like cost per ton to LKO and basically the conclusion there is that your TWR at launch should be relatively low (1.3 or so), because fuel is relatively cheap while engines are expensive.

I didn't use the Stage Recovery mod in my game -- you don't really need it, and I'd say that it definitely shifts "Hard" career down to "Normal".

[Mattasmack]

So, I was looking at some rocket designs incorporating the LES, which made me think "I've never used this," which made me think "that's because I normally just revert in a disaster," which made me think "successful launches would be much more satisfying against a backdrop of consistent failure."

So I'm going to start a career on Hard....

I don't think a stock hard mode career will give you the experience you're looking for. The relatively low funds you get from contracts in hard mode really punish failure; instead, I found that hard mode pushed me to be more conservative in my rocket designs, to make sure each launch was successful and completed the contract(s) it was designed for. A slightly too large, more expensive rocket is much cheaper than a slightly too small rocket that has to be launched twice because it didn't work the first time.

I don't know what would be better. In terms of stock KSP, maybe a normal or moderate difficulty career with quicksave turned off would do it. (And I generally play stock, so I don't have any mods to suggest.) That would mean you have to put up with the random failures that come from game or game engine bugs, though. It is possible to rationalize them as the sort of random failures that occur in real life, but they can be very frustrating. But if you expect to sometimes fail missions or have failed launches, don't play stock hard mode!

[Venusgate]

Might I suggest just turning revert flight option off on moderate difficulty?

There are also mods out there that add a "cost" to reverts. I think construction time is one of them? This way you can simulate "testing" costs. Realistically speaking, rockets are simulated for YEARS before they actually leave the launchpad.

The issue I have with hardmode is the building grind. It's enjoyably challenging once you get your first few teir 2 structures, but the grind up to that point is a little disenchanting.

[Two]

Unless you enjoy the grind I wouldn't turn down the money rewards, it just forces you to do the same mission over and over again (90x VIP orbit just isn't fun).

On the other hand you could add a rule to never decline contracts, in which case you would be forced to do the hard ones as well.

[Jimbimbibble]

But some of the contracts would force you to do things that are impossible with your current tech, or have very little reward (test a mainsail on the surface of Duna for instance).

[Empiro]

But some of the contracts would force you to do things that are impossible with your current tech, or have very little reward (test a mainsail on the surface of Duna for instance).

Some of the contracts are impossible or completely not worthwhile, but most of the testing contracts pay a ridiculous amount if they're large parts in a different SOI.

At the end, I agree that the current career / funds system simply doesn't work very well. Either your designs are good enough, and you slowly grind your way to upgrade your buildings, or your designs aren't good enough, and you go bankrupt after a time.

[StrandedonEarth]

For my current attempt at hard mode I'm combining Kerbal Construction Time and TacLS, as well as dialing down the fund and science rewards a little to 80%. KCT allows you to simulate flights for a cost. Not sure if it's really making it that much harder though, I'm still fairly early in the career

[Alshain]

Build your lifters in advance for specific mass loads and save them as sub-assemblies. I can't stress this enough. You'd be surprised how much money you can save doing this, which is crucial in hard mode, playing around with several different build types.

For example, in 1.0 I've found Asparagus/Onion staging is almost always more expensive than using a 2 stage core with SRB's. I didn't realize this till I started pre-building lifters for this version, in 0.90 that wasn't the case and asparagus was the way to go. So far I've built lifters from 5 to 75 tons and not a single one of them has a fuel line on it because liquid fuel stages are so expensive (especially those with low atmosphere engines like the LV-T, the Mainsail, the Twin Boar, and the Mammoth).

Another tip for hard mode cost cutting on lifters, I've recently discovered, thanks to a thread in General, the Thud is a wonderful augmenting engine. It costs less than radial decouplers + SRBs in some cases so It's often better to use Thuds and some extra fuel for more TWR than SRBs. Of course it's thrust is a little lower so the Thumper still has it's uses, but even 4 Thuds + Fuel can often be a better choice than 2 thumpers + radial decouplers and nose cones sometimes. It just depends on how much more thrust you really need to get your TWR right where you like it.

Edited October 23, 2015 by Alshain