Do you fly test flights in career mode?

[zarakon]

I'm curious how many people here incorporate test flights into their space programs, whether to test a launcher before putting a real payload on it, to test that parachutes won't rip your Duna lander to pieces, or anything else.

So, which do you do?

A. Never test, just launch!

B. Test in a different save file or with reverting to avoid the costs.

C. Test in the same save and pay the funds.

If sometimes B and sometimes C, choose whichever one you do more often.

I don't do it very often, but I sometimes do for major missions, and I pay the costs.

Edited January 19, 2015 by zarakon

[Superfluous J]

I think I've tested pretty much every rocket I've flown in the past year. For me, a "test" is to put the rocket on the pad, fire it off, watch what happens, and then revert the flight to make changes. This iterates anywhere from several times to several dozens of times, depending on the complexity of the rocket and the difficulty of the target and/or mission.

I usually don't test beyond getting the ship to LKO or MAYBE doing whatever ejection burn I'm going to do, but sometimes I will also Hyperedit parts of my ship to the destination and see if they have a chance to succeed there. Eve and Tylo are biggies for that sort of thing.

For both of these, I used to use a separate sandbox save, but these days I find it easier to back up the persistent file and then do whatever I want, and then just restore it when I'm ready to actually launch.

After this, though, if I'm launching I launch and deal with the consequences. I try really hard to only restore from quicksave when the circumstances are very extenuating (game crash, bug, and before the Trajectories mod, I allowed quicksaving to get aerobraking right).

[problemecium]

Sorta mostly B. If the test is more or less successful and I can recover at KSC, I'll just do that and pay the "R&D costs". Otherwise I just revert to VAB until I have something that isn't totally derp.

[Slam_Jones]

Well, usually I'm confident enough in my designs to just launch them, but when they do fail, I go with option B. Reverts all the way.

[Kerano]

Testing isn't really optional for me, especially when designing a new super-heavy booster stage or a new SSTO spaceplane. I usually revert test flights, but only up to LKO. Any missions beyond there become the real deal.

[Itsdavyjones]

I do a lot of testing, and it costs me to do so, since I Kerbal Construction Timer installed. Loads cheaper to test then have a rocket launch and crash.

[RogueNine]

Yes, I test perform test flights. If the rocket explodes then it's a test flight.

[timsfitz]

Longest test and revert I've had was a multiple satellite launch to Duna. Got there only to realize I forgot the probe cores, so had to revert. What a waste of time.

[Frozen_Heart]

I test the craft in career but I'm madly cautious to try to avoid wasted funds. I carefully do a checklist to ensure that everything is going to work how it should.

[Liowen]

I do a combination of both, but normally I build in a sandbox save and test there just to keep my craft list on my career save cleaner. I usually have two saves running at all times; one for building and one for doing. Doing it like this just makes life a touch easier, and gives me a place to do some of my more...ahem....crazier things.

[Pecan]

Where's the option for "How the hell can I waste time in career mode, I've got all these test flights to do!" ?

[CalculusWarrior]

I use Kerbal Construction Time, so none of the above. It's simply fantastic to simulate a rocket to make sure all systems work the way you want them to, and if it does violently explode, you're only out a few hundred funds, rather than several thousand.

[merendel]

Mostly I just build and launch for simple missions. I'm confident enough in my building technique to at least get to orbit and most major issues will show themselves by then and I can find a way to remedy if needed. There are 3 situations I test in a sandbox save however. New heavy launchers, Multi launch assembled in orbit missions/craft, and anything that is soposed to fly with wings.

The heavy launchers normally take a few iterations to make sure fuel is flowing properly and staging works, after that I can subassembly it under any payload without difficulty.

Large orbital assemblies I normally hyper edit into orbit and put them together just to make sure everything fits properly and is able to fly. Once I've got all those kinks worked out I load the designs up in my non simulation save and attach them to launch platforms to send up and assemble legit. I normally don't do this for stations, just anything that will be moveing after assembly. Stations I normaly have enough docking ports on that I can slap a module in wherever and it will work so no simulation needed.

As to the airplanes, I play with FAR. While that should be enough said for the initiated for everybody else just know it takes many tries to get a plane that wont flip out badly and I'd rather do test runs in a simulation mode.

Edited January 20, 2015 by merendel

[sewerbird]

Doing a spaceplane-only career save, a successful test usually only costs me a touch of fuel. If the test plane crashes and the kerbal dies, it happens for real - thankfully all my pilots (if not planes) have come home

[Streetwind]

As others have mentioned, Kerbal Construction Time's simulation mode is fantastic for testing. Instead of gambling on whether or not you can recover a test vehicle (especially one you spent weeks waiting on to be built) and lose or regain the full price, you simply pay a flat fee (depending on the vehicle and the testing parameters chosen) to enter a simulation instance. You can simulate anywhere you have legitly been to before, in orbit or on the ground, even modify the point in time... but when you exit the simulation mode, the game reverts to where you were before and nothing you did in the simulation actually ever happened. Except that the funds are gone, of course.

This is even more useful than you think, since some of the most complex things you build are also those that take the longest ingame to finish being constructed. So if you have a lot going on at once, and you don't get much time to play, you might suddenly end up with something coming out of the build queue that you placed there an RL week ago. I can't count the number of times I've used simulation mode simply as a refresher of my memory on what the heck I built back then, and how it was supposed to work...

Edited January 20, 2015 by Streetwind

[prophet_01]

I usually don't need to do that many testflights (anymore). I don't count the ocasional launch abort due to missing solar panels, fuel lines or chutes as a test flight

I don't build that many heavy lifters. I have two lifter designs that cover the range from 15 - 50t. I simply adjust the number and kind of bosters and do small tweaks to the amount of fuel in the center stage. I usually go with aditional 200 dv to orbit and stick to rather high twr's for my upper stage designs. I usually end up with launch vehicles that are customised to lift a specific payload, but I'm quite good at it by now. However, with regular rockets I usually don't mix up crewed launches with the delivery of heavy components (stations, interplanetary stuff). I don't care that much if an unmaned vehicle crashes and my crewed launches are meant to be rather simple (nothing more fancy than apollo-style)

I do use very heavy shuttles and some small space planes in the 'late game' (tech tree mostly done). Those designs do require many very expensive test flights before I even put a pilot into the seat. I usually have to stop development halfway, do some contracts, and continue to burn money there is no way in hell that those designs will ever break even with those development costs, but they are pretty...

- - - Updated - - -

Doing a spaceplane-only career save, a successful test usually only costs me a touch of fuel. If the test plane crashes and the kerbal dies, it happens for real - thankfully all my pilots (if not planes) have come home

Not bad. I'm definetly not that good when it comes to aerodynamic designs and have crashed quite a few planes and shuttles

[Janos1986]

Always.

I also perform tests with various amounts of fuel and cargo to test the maneuverability with shifting CoM.

[eddiew]

I use Kerbal Construction Time, so basically yes because the penalty of not doing a test flight/simulation is a huge waste of time. It charges a decent % for each 'simulation' though, so you don't feel dirty about it

[cicatrix]

If reverting can be considered 'a test flight' then yes, I test everything. Still, it's more 'cheaty' than doing proper tests (and pay for them, of course). Therefore, I can't pick any poll answer. I launch, see what happens, if there is something wrong I revert, change things and launch again.

[SVlad]

Kerbal Construction Time is great for simulation. It's useful in hard mode, allowing to made preflight simulation, but forcing to complete mission without saves and reverts then.

[rynak]

I test repeatedly with revert.

And i don't consider it cheating, because there is no way to "simulate" a lot of things pre-launch. I don't want to spend the time landing over and over and over again, just to see how a minor change affects things.

[Brainlord Mesomorph]

Test and revert several times during construction.

Then I pay for a full test flight, to LKO or the Moons if its a Lander.

Edit: usually the one I pay for starts out a another test and revert, but it goes well and I decide to save that one.

The Mun is littered with Landers and Rovers that didn't work, and flags showing where the rescue ships landed. I'm able to deorbit and recover most of my failures from LKO.

Edited January 20, 2015 by Brainlord Mesomorph

[cicatrix]

I also test my landers with Hyperedit because there's no way to simulate gravity / atmosphere.

[WWEdeadman]

That's why I keep reverting on in career. If I have no Idea whether it will work, I just try it out, and revert if it doesn't. Changing to another save every time would be annoying as hell to me.

[monophonic]

I never test. I will manipulate time to undo catastrophic failures though. I find it more realistic than reverting and saying "it was just a test." I dunno... maybe I watch too much Dr Who?