More realistic patrols

[mike_espo]

I know how most of us would like to see some kind of system failures during patrols in career mode.

I expanded from the excellent work done by tedhealy for roleplaying/random breakdowns on patrol:

this is what I came up with. Feel free to comment:yep:


Random Failures


Roll Percentage die every 7th day; 5th day for S boats. First determine if a failure exists. If so, roll on table to see what type of failure it is. Check to see duration of failure and repair time. If a second roll results in the same system, there is a 25% chance the system is damaged to one more level.
Random failures occur 9% of the time. 15% on S boats. Add 5% if boat not overhauled on last patrol.

Random Failures Table

% Die Roll

Result


0-15

Radio Failure. Cannot send or receive messages.

16-29

Radar Failure. Cannot use radar. Roll to see which radar is affected if multiple radars are available.

30-38

Batteries. Batteries will not hold charge. No more than 2/3 speed available while submerged. If second damage result, than no more than 1/3 speed available while submerged.

39-46

TDC. Cannot use the TDC until repaired.

47-59

Torpedo tube leak. Randomly choose torpedo tube and that tube cannot be used until repaired.

60-69*

Pressure hull leak. Use S3D (if available) to change Crash Depth by % die result.

70-79

Diesel Engine failure. No more than Std speed available while surfaced. If second damage result, than no more than 2/3 speed available while surfaced.

80-89

Air conditioning failure. The crew must remain at battle stations for failure duration or 1xD6 hours per day if unrepairable to simulate increased fatigue due to lack of cool air.

90-100

No Failures.


1-30: No Failure. System is repairable in a few minutes
31-80: Minor failure. Failure repaired in 1XD6 hours (round up)
81-94: Major failure. Failure repaired in 1XD6/2 days. (round up)
95-100: Failure cannot be repaired at sea.

*: There is a 5% chance (15% on S boats) that the pressure hull will develop a leak whenever test depth is passed. If a leak occurs, the boat cannot go below test depth. Alter by S3D the crash depth if so desired.

Sorry about the text. I copied/pasted a word document. :oops:

[Rockin Robbins]

I'm just hoping they fix that for SH5. The die rolling technique doesn't work because you are experiencing scheduled failures, every 7th or 5th day. It only APPEARS reasonable. But reality is much more random in event as well as timing of the event. No "roll the die" system will approach a system built into the game.

For instance it's day four since I last rolled and I know I have three days of trouble-free running. Might as well attack that convoy now eh? In real life your radar could quit just as you were initiating the attack.

So nice reasonable sounding idea but......FAIL!:D

[ichso]

If you're already rolling, then roll the die beforehand to determine the span until the next failure might occur. Then if that time comes, roll the die again to see whether damage *really* occurs or otherwise it is just dropped until the next random time span has passed.
If damage occurs, use results from your table.

Another idea might be to bind those propabilities to crew efficiency factors in some way. A well worked engine compartment might be less prone to damage and wear-off then one with a lot of noobs in it ;).

I like the basic idea, maybe some SH5 dev is reading this ? :D

[Armistead]

Be nice if we had failures..like the TDC going down during an attack.
Also be nice if charges blowing above you sent you down another 100 m, under you blow you towards the top.

[mike_espo]

If you're already rolling, then roll the die beforehand to determine the span until the next failure might occur. Then if that time comes, roll the die again to see whether damage *really* occurs or otherwise it is just dropped until the next random time span has passed.
If damage occurs, use results from your table.

Another idea might be to bind those propabilities to crew efficiency factors in some way. A well worked engine compartment might be less prone to damage and wear-off then one with a lot of noobs in it ;).

I like the basic idea, maybe some SH5 dev is reading this ? :D

Great idea!:yeah:

[Rockin Robbins]

As long as you're rolling the dice, you are establishing a future time period during which you know with 100% certainty what your status will be. That can never be realistic. Rolling a second time to establish the duration of that failure state or the time before another failure state changes nothing. In fact, it is as unrealistic as no failures at all.

Unless failures can happen at any time without warning the system only introduces the tedium of old-fashioned board games to our computer game for no gain. That's called failure.

[AVGWarhawk]

Set your roll for the dice more randomly....do it when the weather changes:up: Weather changes are not as predictable. If it is really bad seas, roll the dice twice for a double whammy.....you know water and elecronics do not mix. :D

[Rockin Robbins]

Rolling dice is so......20th century!:woot:

[ichso]

Unless you don't want to simulate every breath of your men and every splash of water pushed aside by the propellors of your boat you will have to do it by simple dice rolling.
How else would you want to implement a damage system that presents you failures without you knowing when it happens if not by chance ?

You just have to set the parameters right and you have no clue if and when something is going to happen.

I think in real life there are no propability based events. Only when something happens where too many minor things come into play such that we can't predict the outcome, we just call it randomness. And if in a submarine little things in the innermost of a machine break that cause some failure and which no one can predict because while running the machines constantly you just can't check them to 100%, then such failures would appear very random also.
And such pseudo randomness we want to put into the game by using the pseudo randomness the computer gives us. And there are enough ways to do this without leaving the player with any clue whether and what is happening.

[Armistead]

I almost think it's funny when you get shelled and have a huge hole in your boat it may be only 10% damage. I would prefer anytime a shell hits a compartment, the damage should be high and forced to return home. This running around with several holes in your boat attacking TF's is silly.
One hole....sent home. It would also make people think before using the deckgun.

[Rockin Robbins]

Unless you don't want to simulate every breath of your men and every splash of water pushed aside by the propellors of your boat you will have to do it by simple dice rolling.
How else would you want to implement a damage system that presents you failures without you knowing when it happens if not by chance ?

You just have to set the parameters right and you have no clue if and when something is going to happen.

As long as you know when you are going to roll the dice, you have eliminated true randomness. How do you respond to my situation above?

You know your state. You know you don't roll the dice again for 24 hours. Therefore you attack with 100% certainty that your malfunction status will not change during your action. That totally defeats the purpose of the bad malfunction scheme. That is because a human is throwing dice on a schedule. It's the Heisenburg uncertainty principle all over again. The observer influences the event, and in this case the observer games the flawed process for very unrealistic gameplay. We did it that way in the bad old days of board games because there was little alternative. It didn't work then either.

On the other hand, if a computer game is factoring in malfunctions, it can first of all set modifiers: an engine crew with elite rating means breakdown probability of .015% per hour, or whatever. Heavy seas could modify the percentage, throttle settings, time at full throttle, depth, all could in a continuously variable way not attainable by dice influence your chances of malfunction. This malfunction possibility could be evaluated each second of gameplay if you wanted, probability modified random number generation.

You have NO IDEA when a malfunction is probable. You have no idea WHICH malfunction is possible. You have no idea when multiple failures will happen at once (something not contemplated by the 20th century dice). And when you are spinning dice only a tiny percentage of equipment failures can be covered unless you want to make a list of several hundred possible malfunctions of a hundred different pieces of equipment and spin the dice for every one of them every second. Good luck getting to the sub game.

Any way you want to cut it, a human throwing a die or dice is too predictable to introduce any real gameplay improvement. However it is certain to introduce tedium, the problem computer games were produced to eliminate. Tedium kills wargaming. It is worse than non-random events. It is worse than not having breakdowns at all. And when there is no gain from the tedium it just becomes a silly thing to contemplate.

Let computers do what computers do best. Don't try to substitute a horse and buggy for a Ferrari.

[ichso]

I think you really imagine dice being rolled here, do you ? If you talk about computers and die rolling in one sentence you don't really mean 'die rolling' you just refer to the pseuso randomness the machine gives you. But the term of rolling a die is very common in propability theory. Would you actually think we wanted to simulated or use a die roll for deciding about malfunctions ? :DL

Then, trying to make use of the system in mike_espo's idea would involve choosing a time span, randomly. This might be a few days long it might be 30 minutes long. It gets chosen internally and you absolutely don't know about it. Even if a damage occured you can't be sure whether the new time span will give you a new damage in the next few hours or whether it will last during the next 10 days without one.
Also, if this time frame expires it might still end up in no malfunction at all because whether one occurs or not is also chosen randomly.

Of course this whole system seems to be quite complicated to be implemented. It would be easier to simply just get the combined number of days all submarines of a specific types where in patrol during the war and divide the number of engine failures they had by this.
Then if you get something like ~1 for example, you would be left with a ~1/3 chance to get an engine failure during a 100 day patrol.

Both system could be factored by crew effectiveness/experience, sea state and whatever. Only the second one is easier to implement ;)

[Rockin Robbins]

I'm afraid that Mr Espo is talking about really rolling a physical six or more sided die with multitudinous spots on it. At least SOMEONE understands my problem with that.:up:

[mike_espo]

Yes. A lot of good points here. But really, all events in life are probabilities. It is possible to assign values with percentile dice to certain events: i.e. random events for breakdowns.

I do see the RRs point of how unrealistic it is to assign every 4th day or whatever to see if events happen. A method to determine when rolls would occur would be a good idea. :hmmm:

Just trying to make our wonderful game more enjoyable as 100% reliability is very unrealistic.

I wish this was already incorporated into our sim. I should think it would be easy to do so using weighted values.

[Munchausen]

It would be easier to simply just get the combined number of days all submarines of a specific types where in patrol during the war and divide the number of engine failures they had by this.
Then if you get something like ~1 for example, you would be left with a ~1/3 chance to get an engine failure during a 100 day patrol.

Both system could be factored by crew effectiveness/experience, sea state and whatever. Only the second one is easier to implement ;)

For engines, you'd also need to consider the type engine (HOR, for example). For radar, you should consider the length of time it's been running without pause (overheating) and the outside air temperature (also overheating).

:-? In other words, failure rates should be linked to various pieces of equipment and how they're used/abused ... adding another factor to the game: the skipper's option to argue against the installment of HOR engines.

[Rockin Robbins]

Uh, captains weren't consulted, they were assigned. Even if they were picking up a sub out of construction there were parameters he had no choice about. But the bottom line is the number of modifying parameters, types of equipment and characters of breakdown makes a system of throwing physical dice a travesty more unacceptable than 100% reliability. It is worse because it is entirely arbitrary and overly simplified. It would leave the vast majority of real possibilities uncovered and still at 100% reliability unless the system were so complex all gameplay would stop.

[vanjast]

It would be great to have unpredictable machine failures..

As for realsim, I'd like to give an example of SH3 (100%,RuB,Real Mechanics, RealNav + other smalls). I could still obtain a minimum of 24K (highest was 85K) per patrol through out the war period. What did become difficult was evading the escorts, but I always managed this.

I'm not sure whether this was my 'rebellious attitude' and methods used, coz comparing this to RL patrols, my patrols seem far fetched - I never died ??
Maybe a game cannot give a RL perspective of things, and the question is really in the game AI... this has always been lacking even with our so called advanced tech - not wanting to start a Cyborg war of sorts:D

[vanjast]

not wanting to start a Cyborg war of sorts:D
Maybe this is the beginning ... _ _ _ ... ... _ _ _ ...

[Rockin Robbins]

We have the advantage of hindsight. There is a critical mass of knowledge that we have that none of the real participants could have. So we game the situation, knowing that in early war we set American torpedoes on the surface, when the real captains set to the correct depth, ran under the target, but never knew it. All they knew is that they missed. Whether they went under or passed ahead or astern they didn't know. We do and we compensate for our "forbidden knowledge."

[ichso]

Actually - not necessarily. It depends on whether one plays the game just to beat it or whether one tries to put some roleplaying aspects in it. Try to gain the perspective of a sub skipper and act accordingly. Like setting the torpedoes as deep as it is supposed to be done because you act like you don't know that sometimes they will run too deep.
If such things give you more emersion and therefore more fun, why not?

[Rockin Robbins]

That's exactly true and why I don't get crazy trying to "correct" everything when I can just change the way I play to compensate. For instance I don't worry too much about the excessive accuracy of visual target placement on the nave map. I just know which is which and don't take any measurements off visual positions.