Slitherine

During the last episodes of my LP series, I was loosing my astronauts on a more or less frequent basis. Sure, my playstyle could have been a little bit too aggressive, but I wasn't really able to observe a clear pattern for the catastrophic outcomes.

The general skill level and component quality determines the chance for having an issue. However, the issue itself is creating a random number and this random number is then compared to several cases (failure, glitch, catastrophic outcome). The cases seem to be influenced by the overall reliability -> better quality, lower number. Unfortunately, the roll for the random number seems to be completely free.

As an example(Episode 90), I had 92.2% chances of success, 5.89 chance for catastrophic outcome and a random number of 4.10. Later in this game(Episode 91), I had 81% chances of success, 14.27 chance for catastrophic outcome and a random number of 64.36.

This finally brings me to the conclusion, that the chance for a catastrophic outcome is basically just based on luck. Sure, when the chances for a catastrophic outcome are lower, then there're also less random numbers which can trigger it, but in general, the random number can be still lower and I will loose my men. In this case, the only chance to prevent my astronauts from dying in a catastrophic failure seems to be just to lower the chance of having an issue by having the best components, best flight controllers and best astronauts assigned to this mission. Otherwise, as soon as I enter the issue state, I can be unlucky with the random number and loose my astronauts.

Unfortunately, it is impossible to achieve a mission reliability of 100% and with each mission step the chances of having an issue will increase. In other words, I won't be able to fully protect my employees and they'll need to deal with sudden issues during their mission. Additionally, due to the fact that it gets harder and harder to reach very high percentages, launching a mission as soon as the components reach a reliability above 80% seems to be the most practicable way.

In this case, it would be really amazing if you, Ignacio, could give us some additional insights into the calculations of the issue state. How are the chances for glitch, failure and catf influenced by the overall reliability? How is the random number generated? Is it really just a luck based outcome and the only chance to get my men back alive is to lower the catf to reduce the amount of "bad" random numbers?

Hello Sebastian,

Apologies for the late reply. For every step, the chances of success are computed using the mission components involved in it, along with the astronaut skills used in that step. If that fails, Flight Controllers kick-in and a number is computed by multiplying the involvement of the mission components with the skills being used by the flight controllers. We call this number 'saving throw probability of success', which is a number between 0 and 100 and which will always be lower than the combined reliability of the mission components being used, since the skills of all your flight controllers will never be 100%.

We compute another number called 'chance to fail', which is 100 - 'saving throw probability of success' and we then proceed to compute two thresholds:

maxValueForCatastrophic = chanceToFail * 0.75f;
maxValueForFailure = chanceToFail * 0.9f;

We roll our 100-sided dice again and, based on the outcome, we determine the outcome for the mission step (i.e., failure, catastrophic failure, minor glitch or minor glitch solved). By having an overall greater reliability and more skilful flight controllers, the chances of your second die roll falling in the 'catastrophic' area get smaller.

Hope this helps!

Cheers,

Yep, this really helps. Thank you for your thorough explanation!

(Apparently, I've broken too many mirrors in the past )

I have to ask exactly when is this calculation taking place?. If I save prior to assembly or save after assembly and run with CTF result and then load the save at either step when the mission runs again I get exactly the same result which would indicate that the numbers generated are not random at all.

aisdad wrote:I have to ask exactly when is this calculation taking place?. If I save prior to assembly or save after assembly and run with CTF result and then load the save at either step when the mission runs again I get exactly the same result which would indicate that the numbers generated are not random at all.

Hello aisdad,

Welcome to the forums! The numbers generated are random. The reason you're getting the same results is because we store the initial seed used for the PRNG, along with the number of times it has been used, in the savefile. When you load an existing game, we initialize the PRNG and leave it in the exact same state, so you'll get the exact same results.

We do this so that we can easily reproduce all the problems reported in the forums by loading a savefile and get the exact same output experienced by the player.

Cheers,