We don' need no educashun

[Sea Scourge]

In an attempt to discover whether schooling was worth the time and cost I ran a sandbox game with the intention of developing a skilled crew through schooling and an unskilled crew through press ganging captives which the manual says results in below average pirates. What I wound up with was a crew of slightly above average pirates aside from their low courage which was the criteria I used for selecting from the captives. If I were doing it over I would save before making a selection and restore unless the pirate turned out to be really inferior. The point of the test was to see what difference skill made in the crew and I would have preferred a greater difference in the average skill level of the two crews. I also selected another captain (Bloody Mary) for her poor seamanship skills. I then ran a series of 10 cruises saving before each cruise and restoring after using different combinations of crews and cruise settings (harass, board, pound). After 20 cruises with Bloody Mary only resulted in 4 encounters I concluded the manual is correct in saying that the chance of finding victims depends entirely upon the seamanship skill of the captain and gave up on B. Mary. The below average loot from her 4 succesful cruises might be a result of her low notoriety but who knows. Another mistake I made was in only running 10 cruises regardless of whether victims were found. I should have kept cruising until there were 10 successful encounters. One thing I discovered was that my ship would be sunk once during every 10 cruises regardless of crew. I am attaching a summary showing the average loot from each type of cruise. I would say the results are inconclusive although the skilled crew shows a slight edge. I doubt it's worth the time and money required to train ordinary pirates. Obviously a lot more testing is needed to discover WHAT SHOULD BE IN THE #$%& MANUAL TO BEGIN WITH. I finally got bored sitting around waiting for ships to return and decided to try the Very High Speed option. The next series resulted in 3 sinkings so the heck with it. I quit. See the attached txt file for cruise summary. I give up on getting the numbers to line up under the headings.

[Severous]

This is not a Pink Floyd posting

Seascourge.  

Thats reinforcing the importance of Seamanship ..if only for the Captains.  There maybe more to Marys lack of success but im sticking with our theory of seamanship skill as vital for captains.

In my book saving and replaying to provide a control environment is essential and fair do.  Saving and replaying if you dont get a good pirate or good cruise outcome in any other situation is as bad as typing in a cheat code for extra dosh.

As for variation in cruise outcomes - there is a very powerful randon element at play.  Every now and then theres an encounter thats going to wipe you out. You dont know what it is and can only speculate on how many narrow escapes we have had.

I prefer to think that the random ship encounter routine is beyond our control but is influenced in part by seamanship and captain courage(flee or fight).  Once a combat encounter is underway the courage and notoriety are really really important - perhaps more so than marksmanship etc. If your high notoriety captain can syke out the enemy and high courge prevent the same to your crew then you walk it irrespective of average gunnary etc.  

Who knows (one of my two most annoying things about trop2)

So i currently thinking I go for seamanship, parrots, legs.  That said in my test island where im not playing for score im just educating when the unemployed pirates have nothing better to do and there are school slots available......hey teacher...leave those kids alone.

[Sea Scourge]

Saving and replaying if you dont get a good pirate or good cruise outcome in any other situation is as bad as typing in a cheat code for extra dosh.

I couldn't agree more and that's why I've spent so much time trying to figure out what makes a difference in combat outcome. If my ship is sunk because of poor strategy on my part I'm perfectly willing to accept the loss and learn from my mistake. If, however, my ship is sunk because the AI says my ship has a 10% chance of being sunk regardless of crew skill (which appears to be the case), then the cheating started on their part.

Once a combat encounter is underway the courage and notoriety are really really important - perhaps more so than marksmanship etc.

Maybe, but I haven't been able to verify this and my results indicate that courage and notoriety in the crew makes little or no difference. Possibly it does with the captain but I didn't have enough data to say. The problem with verifying anything is that there are an impossibly large number of variables and each variable must be tested with a large number of cruises while all other variables are held constant. So far the only thing I've established beyond reasonable doubt is that the seamanship skill of the captain is very important and worth a school. Beyond that the combat AI appears to be a random number generator. To what extent (if any) all the other skills and attributes affect the outcome remains a mystery. This allows you to believe whatever you wish about the importance of parrots and peglegs. I forgot to mention that the single most successful cruise (3000 gold) was made by the unschooled pirates. Teacher, leave them kids alone!

[Coffeebean]

Thats reinforcing the importance of Seamanship ..if only for the Captains.  There maybe more to Marys lack of success but im sticking with our theory of seamanship skill as vital for captains.

I may be misunderstanding you Severous, but I believe that the seamanship skill average for the whole ship is important. I think this because of the info shown re each ship - see pic. Leadership only counts for captains and is only shown for captains. Likewise navigation only counts for captains and officers and is only shown for those two, but the seamanship skill average is shown for the crew as well, as are the weapon skills.

It would be useful to be able to cheat to set up a nil skilled crew, a one point all skilled crew....etc. and send each variation out on 10 cruises to the same region as if for the first time each time. Of course you'd also need a lot of patience and a lot of time!
I have no knowledge of game codes type stuff but is there anything there that suggests that skill levels alter the chances of finding other ships and winning battles?

[Severous]

Coffeebean & SeaScourge (and anyone else who wants to chip in)

Without checking the manual again im recalling:

- Seamanship for captains affect encounter frequency

-Seamanship for whole crew affects encounter success once its under way...catching an fleeing enemy/fleeing yourself.

- Navigation. The whole crew is what I thought and impacts the chances of your shiploosing the wind from its sails as it moves from the dock to the map edge.  Could easily be officer/captians only but for some reason nav for all sticks in my mind.

- Keeping all variables the same still produces loads of different outcomes (AAR Weak Snow Groundhog day)

We need Charlemagne !  We need a worked example !(that doesnt need a game release or development)

[Coffeebean]

Well the manual says...  

[Coffeebean]

And yes, maybe Charlemagne could help here.

[Severous]

Doh.  I should have looked in the manual first eh.

Navigation is clear - Captain and Officers.  Me old grey matter updated.

Wonder what the Nav score is when no officer.. 0 ? or is it ignored and we get just the Captains score?  Or for that matter if only one officer in a two officer type ship? Do we half the one officers skill or take it as the average.

Seamanship...hmmm.  There is another quote from the book that i was recalling and pinning the assumption on.  

In your quote Coffeebean...theres clearly an all crew calculation referred to for duration and 'accidents'.   Now thats confusing.  I could have sworn it was navigation that would affect duration...and now we have a new term 'accidents' to puzzle over.

Seamanship of all the Crew is said to be used when determining the chance of coming alongside after a cannon round exchange.

So still plenty of         all round...ah yes please i'll have a Rum

[Sea Scourge]

Seamanship for captains affect encounter frequency

Correct and proved true.

Seamanship for whole crew affects encounter success once its under way...catching an fleeing enemy/fleeing yourself.

I seem to remember manual says seamanship of entire crew affects boarding sucess. Maybe, but unverified.

- Keeping all variables the same still produces loads of different outcomes (AAR Weak Snow Groundhog day)

Absolutely. The question is how many cruises are needed to get a meaningful average?

I'm trying a screen shot of the text file to see if it comes out more readable.

[Charlemagne]

I can see from this thread you guys want me to post something--but I am not sure what it would be?? Probably I'm not reading carefully enough.

Anyone want to enlighten me? You can post here or send me a message, either way is fine.

[Laduguer]

To what extent do Seamanship and Navigational skills affect cruise outcome (encounters), and should the players be concerned with educating pirates, officers or captians in these areas to effect the same?

[Charlemagne]

To what extent do Seamanship and Navigational skills affect cruise outcome (encounters), and should the players be concerned with educating pirates, officers or captians in these areas to effect the same?

Okay, I've got the algorithms from the engineer who coded this part of the game (his name is Andy btw). If this stuff is not clear you can post again of course, and I'll try to explain it. Anyway here it is in Andy's words:

* in making sightings:
each day each fleet checks for a new sighting from
each nationality.  We make a roll between 1 and a data-defined number
(which seems to have shipped as 350).  If that number
is less than the chance number we generate, a sighting's made.  The chance
number is determined as:
         ((s + f) * (t + r)) * ((k + m) / m)
where:
         s = fleet's seamanship value (the seamanship belonging to the
captain tying for HIGHEST LEADERSHIP)
         f = fleet size (# of boats)
         t = nation's merchant traffic in sea zone (0-9)         r = nation's known trade route traffic in sea zone (same range)
         m = max knowledge level of a sea zone (0-255)
         k = current knowledge level of sea zone (0 to max)

* in chases
the two vessels generate competing values.. the vessel's speed
times a certain data-defined weight plus its seamanship times another
coefficient, plus its cannon times another... possibly plus another factor
if the pirate king is crafty.  From this is subtracted another number from
data.  Pursuers might get a numeric boost if it's early in the game, and
fleeing schooners don't subtract that last number.  Anyway, I believe you
just roll a number in a range defined by the sum of the competing factors
and whichever vessel's value the roll indicates wins the race.

* in cannon-pounding rounds (how fights always start), seamanship is
multiplied by six and added to the ship's speed as the "speed factors."  In
other words it's about as important as ship speed in determining how many
cannon you get to fire each round: your speed factor over the sum of yours
plus theirs is the fraction of cannons firing each round.

[Sea Scourge]

I can see from this thread you guys want me to post something--but I am not sure what it would be?? Probably I'm not reading carefully enough.
Anyone want to enlighten me? You can post here or send me a message, either way is fine.

Okay, what I'd like is some solid info on how the skills of the captain, officers and crew affect the outcome of a battle. Are actual formulas used in deciding the result or is it simply decided randomly? As you may have noticed I've been trying to determine empirically exactly how important skills are and so far the tests have been inconclusive, other than proving the value of seamanship skill for the captain. One thing I noticed in running the test series was that my ship was consistently sunk once during every set of 10 cruises. This leads me to believe that there is a 10% chance the ship will be sunk regardless of crew skill. True? If not, what factors were involved that caused the fastest ship afloat with an average crew seamanship skill of 5 to be caught and sunk? The Pirate's Book o' Lore, Myths and Superstitions contains statements which are verifiably false and it would be nice to have some straight scoop for a change:) Oh, yeah. The original question was: Is educating pirates worth the time and money involved?

 

[Severous]

To Bill/Charlemagne

Thank you very much and please pass my thanks to Andy Fredricksen also.  This was exactly the sort of thing I was hoping would come along.

Im not jumping to any conclusions.  Gonna pop this lot into a spreadsheet model and see what happens.  If i may I will eventually post my observations and questions here so all can see.

Cheers !

[Charlemagne]

Okay, what I'd like is some solid info on how the skills of the captain, officers and crew affect the outcome of a battle. Are actual formulas used in deciding the result or is it simply decided randomly?

This was exactly what the previous post was supposed to do. Maybe it was too unclear. Many resolutions in the cruises including sightings and chase results DO involve seamanship as was explained above. In the case of sightings this is expressed in the long equation listed where the seamanship is represented by an "s". (see above) As far as sighting go there is no more solid information than this.

Once a sighting occurs you have to figure out what we sighted. At that point we already know the nationality since the sighting chances are rolled for each nationality seperately as explained in the previous post. There is a simple list of possible targets for each region that changes accordiong to how many points a given power has in that scenario and where they have allocated their points. Suffice it too say that the determination of the target DOES NOT depend on the skills of the pirates in any way.

Then there is a chase determination assuming both ships don't flee. Take the number 255 and split it between the ships according to: speed by ship type, and the captain's seamnship. There is an adjustment for schooners shallow water abilities and for crafty captains. Then you roll a number 0-255. If the number rolled falls within your ship's range you win the chase and either get away or catch up depending on what you were trying to do.

Then there is a battle determination unless the ship is a non pirate non naval vessel. Assuming you have a fight the first round of battle (which involves the cannons in alll cases)  does use seamanshiop as explained above. Faster  amd more seamanly shiops get a higher percentage of their shots off in cannon rounds.

In non-cannon rounds, that is boarding - seamanship does not help. I can post more battle round later stuff later but there is no formula like the above. It is respolved in rounbds in which active sailirs and officers take shots with weapons and try to score hits on enemies.  I assure you that the skills of the sailors and officers with their weapons DO strongly influence each round of battle, and that their courage determines how long they fight.

As it says in the manual education is a good thing. It increases your chances of success. But it does not guarantee success.

If you want to you can post things from the manual that you believe are wrong. I'll try to check them when I have time. Keep in mind that the manaul was written (by me) a few months before the game was done and releases. Ideally changes are explained in the readme, although I don't remember if we listed any for this game.

[Sea Scourge]

* in cannon-pounding rounds  (how fights always start), seamanship is multiplied by six and added to the ship's speed as the "speed factors."  In other words it's about as important as ship speed in determining how many
cannon you get to fire each round: your speed factor over the sum of yours plus theirs is the fraction of cannons firing each round.

Thanks. Great start and exactly what I've been looking for! I assume by seamanship you mean the average of the entire crew? Now, how does gunnery skill enter the picture?  Marksmanship? Swordsmanship? Courage? Leadership?

[Sea Scourge]

If you want to you can post things from the manual that you believe are wrong.

My postings seem to be out of phase with yours. Sorry 'bout that. I'll check before I post from now on.

Apologies also for the slurs on the veracity of the manual
One example that comes to mind though, is that the manual says crew skills increase with experience. The only "skill" that increases is notoriety. Rank also increases but that is mostly a pain in the patootie so far as I can tell.

[Charlemagne]

Battle Improvement
Pirates get a chance to improve a skill in battle each round they fight. If successful they improve 0.1 points.

Details from Andy:
Every round (of which there are several per battle), participants call the
MaybeIncreaseSkill method for whichever skill they used this round (gunnery
in pounding rounds, marksmanship or swordsmanship in boarding, gunnery or
marksmanship in harassment).  Basically that means they generate a random
value between 0 and 5 (assuming 5's still the value in data), which may be
modified per skill based on rules that elude me--maybe pirate king stuff,
maybe scenario rules, but generally speaking we can say 0-5--and if the
roll is lower than their current skill value, it goes up by some
data-defined value, which turns out to be .1... so it takes 10 successful
rolls to go up a level.  There are probably usually fewer than 10 rounds in
a battle, but it varies, and as their skill increases, successful trials
will get scarcer, so it will take a few battles--which probably means lots
of voyages--to gain skill through experience.


Education
Education offers much faster advancement.
Education improves in a bell curve within 0.2 of 2 by Andy's reading. Note that as he says this is not his code:

Skill through education ... Jon wrote this, but let's see.  Well, when
pirates graduate they do this:
         float oldValue = pirate->GetSkill(skill);
         float mean =  GameDataPath(Sac::Path('data',
'game')).GetFloat('edum');
         float sigma =  GameDataPath(Sac::Path('data',
'game')).GetFloat('edus');
         float delta = TheApp()->GetGameRandom().GaussianDev(mean, sigma);
         float newValue = min(oldValue + delta, (float)kMaxSkill);
         pirate->SetSkill(skill, newValue);

...mean in data is 2 ... sigma in data is .2.  Thus, new skill value is:
         (old skill value + delta), up to maximum skill value.
By my reading, delta is a bell-curve random number within .2 of 2.


Skills in Battle
I can remember the battle stuff pretty well myself. A pirate picks a weapon according to the type of round he is fighting. You can influence what sort of round by your orders at the start of cruise. Boarding and harassing rounds offer a weapon choice, I believe, and pirate chooses to use the best weapon for them if available. Of course if you don't have the right weapon on the ship he'll choose another weapon, excet that cannons cannot be used for boarding rounds.

Once he has his weapon he get a 'to hit' chance which is simply skill/10. If he hits it takes an enemy out of subsequent rounds. At the end of each round all unhit sailors and pirates much make a courage check. Here I don't remember the exact code, but I know that it is based only on courage altered by captain leadership. Sailors who fail the check cower for the remainder of the battle and are not killed (until their ship surrenders) and do not fight.

[Coffeebean]

Thanks for that Charlemagne.
I must admit I don't understand all of the detail. But I get the idea. There is a deal of randomness but improving skills and accoutrements (parrots for all   ) do make a difference.
I may start sacking cowardly crew, if I have more courageous landlubbers.

[Sea Scourge]

Battle Improvement
Pirates get a chance to improve a skill in battle each round they fight. If successful they improve 0.1 points.

This is very strange. I played an entire scenario and kept track of my pirates by renaming them with numbers corresponding to their original skill levels at the beginning of the game. At the end of the game none of their skills had changed. Version 1.2 is the latest update isn't it?

[Charlemagne]

This is very strange. I played an entire scenario and kept track of my pirates by renaming them with numbers corresponding to their original skill levels at the beginning of the game. At the end of the game none of their skills had changed. Version 1.2 is the latest update isn't it?

Yes, you have the latest version. But there are several things that could be going on here. First, as Andy says, they get a chance to improve after each round. A chance may not mean much success.  Also, they may have improved 0.4 or whatever over the course of the game, but rounding means that the number displayed (which does not include tenths,  would not change).

But if you had enough pirates, a large enough sample, and a long enough game this should not be the issue. Remember though that battle rounds only occur when the pirates encounter a naval or other pirate vessel. So it could be that even though you played an entire game there simply were not enough battle rounds to do the job. This is pretty good evidence that we should have made the battle improvement larger. We didn't because we wanted schools to be valuable, but we may have overdone it. Perhaps if there is another patch I'd try setting the improvement number to 0.4 or 0.3

If that still doesn't make sense (and I don't know how much fighting your guys did) there could be a bug with Andy's code, or with perhaps with the text update. So either they did actually improve, and the numbers didn't adjust on screen, or the code is called but does not work for some other reason. Clearly, both of these are possible, though I can tell you that it did work at some point in development because I remember seeing my own pirates improve. I'm not claiming that I know for sure that it worked in the final build. Something like this could have 'broken' later and not been noticed by anyone since it is such a small effect anyway.

Maybe the fact that you have not seen improvement (for whatever reason) is a good argument for pirates needing some educashun in this game.

[Laduguer]

You people over at Frog City are amazing, I've never *NEVER* seen this much interaction between a developer and their customers / fans, sans when I was admining for a MUD.

*hi5* to you all.

[Sea Scourge]

So either they did actually improve, and the numbers didn't adjust on screen,

I am running a series of tests trying to prove that skilled pirates are better than unskilled and noticed that after some battles the ship rating for gunnery improved from 2 to 2.1 or 2.2 yet when I looked at the individual pirates nothing had changed. It seems the ship average takes fractional skill levels into account while individual levels are reported only as integers? I'll try again to see if I can get on the job training to pay off.
My method of testing involves setting up a sandbox game, putting together 2 crews, training one to the max while selectively press ganging captives for poor courage and skills for the other crew. I then save, cruise, record results and restore. I now have the results of 75 cruises with the schooled crew and 50 with the unschooled. This morning something you said in a previous post set off alarms and that is that early in the game the rules are easier for combat. If so it explains why I am seeing so little difference between the performance of the two crews but also means I've blown a lot of time and effort. Can you suggest an easier method to show that a skilled crew is better than an unskilled?

Perhaps if there is another patch I'd try setting the improvement number to 0.4 or 0.3

Uh oh, here we go again, nudge,nudge,wink,wink

[Charlemagne]

Sea Scourge-

I can't suggest a better method right now. I'll think about it. The problem is that there are so many factors, including random number rolls, that are out of your control. It just may happen that you get consistently luckier with the unskilled crew. On average I am certain that the skilled crew is much better. But the sample to prove it would have to very very large to reduce the impact of the other factors.

[Sea Scourge]

the sample to prove it would have to very very large to reduce the impact of the other factors.

My goal was 300 cruises in all. The idea was to do 25 cruises resulting in contact with each of the 3 cruise modes while keeping track of the sinkings, First with my super captain (Henry Morgan) and all star crew then with Morgan and the unschooled lot. I then planned to recruit an average captain and repeat the process with the 2 sets of crews. I think if I can't see a measurable diffence in 300 cruises the difference isn't worth worrying about. I apologize for being a PITA on the subject but I want what you say to be true and I want to be able to prove it. A couple of further questions and I promise to quit bugging you.
If unskilled crews are given extra help early in the game at what point does this help end?
Does skill count only when fighting other pirate and naval ships? If so I should only be counting encounters with those ships. It would also explain why there seems to be so little difference in the gold and captives between the 2 crews so far.
Many thanks for your time and trouble.