Destroyers Discussion (getting rid of pin point drops)

[CB..]

just to make clear i'm not actually going to be making any mod at all so i'm in no way in competition with the concept/aims/work of the thread

i'm just hoping to get some better gameplay for my-self --

this lone DD thing- is a puzzle--happens every time i play---(again with the cfg deletions)

every thing actually is brilliant sometimes as in the one i have just tried the DD's detect me on the way into the convoy and somtimes only after i torp a vessell --- good gameplay--

but every time i get stuck with one DD that is impossible to shake even after it runs out of DC's to throw--

strangely i noticed that as each time the DD was not the same one (if i hit the same convoy as last time) so there's not much to gather from the sensors on this one--- BUT it dropped on me that the final un shakeable lone DD i end up with is in fact the first DD to detect me--all the others have a very natural pattern to their abilitys and can be shaken--

the depth edit didn't help either--

there's a kind of general flow coming about the cfg stuff

and the sensor.cfg entrys (normally associated with the u-boat crew) as G has said--and following on from waht you were saying eailier--

if we can remove entrys from the sim.cfg --can we add new ones?


how about changing the normal sim.cfg entrys for the sensors to the same as the sensors.cfg
ie so the sim.cfg hydrophone section looks like this

[Hydrophone]
Hydrophone range factor=1 ;[>=0]
Hydrophone fog factor=0 ;[>=0]
Hydrophone light factor=0 ;[>=0]
Hydrophone waves factor=0.2 ;[>=0]
Hydrophone speed factor=0.5 ;[>=0]
Hydrophone enemy speed=0 ;[>=0]
Hydrophone aspect=0 ;[>=0]
Hydrophone noise factor=0.5 ;[>=0]
Hydrophone sensor height factor=0 ;[>=0]
Hydrophone already tracking modifier=20 ;[detection probability modifier]
Hydrophone decay time=150 ;[>0] already tracking bonus decay, in seconds
Hydrophone uses crew efficiency=true ;[true or false]


same as the sensors.cfg

maybe waht is happenning with the lone DD ( and perhaps others depending on the circumstances) is that it gets a sensor boost a "tracking bonus" and it just simply never decays so it becomes uber where-as all those DD's that join in the attack don't get any "tracking bonus" so behave normally--- this would explain why things seem out of whack very often and not following the rules--

it must be follwing some sort of rules it's a computer game!

what do you say G'? as the tracking bonus would give the DD a significant boost to it's sensors it may start detecting and behaving in ways that exceed the normal sensor settings--?


i can 't think of any other logical reason why one DD should prove impossible to shake whereas all the others behave normally--

this may apply to some of the uber DD abilites aswell sense-ing your turns when it logically can't do so-

i'm going to try a run thru using the edit and see if it affects anything at all

thinking on it might need to be edited to read this if i follow the logic of the language--


[Hydrophone]
range factor=1 ;[>=0]
fog factor=0 ;[>=0]
light factor=0 ;[>=0]
waves factor=0.2 ;[>=0]
speed factor=0.5 ;[>=0]
enemy speed=0 ;[>=0]
aspect=0 ;[>=0]
noise factor=0.5 ;[>=0]
sensor height factor=0 ;[>=0]
already tracking modifier=20 ;[detection probability modifier]
decay time=150 ;[>0] already tracking bonus decay, in seconds
uses crew efficiency=true ;[true or false]


as the sensors.cfg doesnt have section headers for each sensor and the sim.cfg does--

[gouldjg]

I am stuck on that one also CB.

I think one unit is selected as the commander or is that related to soemthing else.

To be honest here, I am quite clueless at the moment,

I do say that there is noting wrong with sometimes adding a line into the cfgs. I do it in zones.cfg and have also done it in crew settings for fatigue models.

Try an obvious line and see if anything happens. It would be cool if it does work, even if only one or two things work then it is a leap forward.

What I am a bit clueless to is what speeds do the Hydrophones on DD account for.

Should I be detected at sensor max range at flank, full, standard, etc etc.

At the moment the only thing I have done is set the crew all to 4 on happy times mission and all campaign files. I then sail in between the convoy and on comming DD's so am about 7000 mtres of from any possible contact. It seems like a ideal start point for my tests.

I have set the detection time to 0.5 for hydro and left the rest of the cfg etc alone for the moment.

Hopefully this should start giving me results on the standard deviations between speeds and ranges of detections for this crewtype=4.

Once I get some solid results, I use the maths to set the other sensor ranges.

This is why I would prefer to edit each sensor to suit rather than the cfg. Obviously the cfg may need tweaking at some point but I would love it to remain as close as original at this moment in time till I get the first base line.

So lets say I have a passive sensor that has a max range of 6000 mtre,

Is that max when I am at flank or max when I am at standard. Any help from historians amongst you would be great. What would be the best gameplay choice bearing in mind we use time compression which often bites us in the ass now and again.

[Ducimus]

Originally Posted by CB..

i see we're going to go round and round untill we vanish up our own perfectly controlled ; perfectly standardised exhaust pipe here; ESPECIALLY if we all use the same mission---i'm out of here !!

I was only trying to put my finger down on how the passive sonar behaves in the DD's. Its by far not an all inclusive test because its only a single DD (not several) and the seas are NEVER as favorable in acutal play as they were in the test runs i was making.

I'm only trying to tinker my own game so i can enjoy it better. Im not looking to make some offical mod.

All that said, im going to post my last findings and then let you guys do with the info ive found (if you want to call it that) what you wil with it. Then im off to tweak my own game and start playing again :P

Anyway

Waves factor=0.5
Speed factor=15
Noise factor=0.5


Crew rating 3 escort:

1/3d speed he detects me at about 4000 - 4500 meters

Ahead slow, he detects me at about 2600-3000 meters

Slow and silent running, he doesnt see me at all, passed right under his keel.


Personnlay i think im gonna tinker with it a tinsy bit more so it forces me to drop 1or 2 knots at silent running when hes about 1000 meters or he detects me. I think running 3 knots at silent speed, right under his nose, i shouldnt be able to do that. 1 or 2 knots yes, but not 3. Just my personal settings.

Of course none of this will take into account of rough seas. I think im going to try zeroing out the wave factor and see if that just makes weather a non issiue and then just work with the noise level which for lack of a better word, might as well be labled "amplifier".

[Redwine]

@ CB :

I experienced that alone unshakeable DDs too CB, but it is gone after i touch the sensors angles.




About my experience at moment, i am sticky with to found an average setting to not have ubber or dummy DDs.

In this way i start up tp touch decoys.

As an experiment, and ashot in the darke, i rise up all 5 or 6 diferent decoys values.

I duplicated life time, and triplicated effective surface.
Saddly i can not manage noise, is the same problem than sensivity is a long secuence of numbers into the hexe file.
Any way i can copy the higger decoy noise value and apply it to all them.

the results is made a intentional provocation, against ubber Buckley and Evarts.

Then when detected, i lauched 6 decoys all at same time.

Was an spectacle, all 5 DDs was around the decoys as "flyes around honey" launching all they have onboard, helgedogs, barrel depthcharges.....

I was able to scape from that hell at silent running with any disturb.
Settings looks to be too much, these settings made the sacape so easy.


This fact can be used to manage ubber and that alone unshakeable DDs experienced by CB.

We can set files for a hard game play on crew rating 3, ( CB needs to rise up, others may be needs to drop down) and let the tweaked decoys to evade the ubber Evarts and Buckley DDs on crew rating 4.

[CB..]

Originally Posted by gouldjg

I am stuck on that one also CB.

I think one unit is selected as the commander or is that related to soemthing else.

.

was just about to say the same thing--- i just ran a patrol against aconvoy using the sim.cfg hydrophone entrys i posted in the prvious thread---

went pretty well again but couldn't shake the first DD that detected me---

but i did pay very close attention to what it was doing and how the other DD's behaved around it--- this time i was ahead of the convoy- so i could let it drift over head-- i wa then able to get the uber DD stuck behind some merchants long enough to get up and torp a tanker-- and as the DD's maneuvered around the merchants i watched the uber DD closely--
it had absoloute priority over all the others---as the DD's circling got confused by the merchants they somtimes got in the way of the uber DD

and they allways gave way to it--even if actually making a DC drop---they would go into flank reverse to get out of it's way and reverse back over their own DC's damaging themselves in the process--very strange as the uber DD would only have had to make a very minor course correction to avoid the attacking DD---it was allmost as if all the other DD's were afraid of it !!

something and nothing i don't really know if we can't edit it then there's not much we can do with that---

i'll try adding the sensitivity entry to the hydrophone stuff ( i'm not having any success at all with AI_sensors.dat edits--tho i'll keep trying)


RED can you post the way to edit the decoy file? i want to have a look at it---

i have a minor idea regarding equiping the DD's with the decoy launcher instead of one of it's K/Y guns ( what is a Y gun anyway?)

i wonder if it will automatically launch a decoy at the same time as DC's if i replace one of the K guns with a decoy launcher--some thing along those lines --other than that im stuck in a bit of a catch 22


i've replaced the Y gun in the Clemson with the boldtype1

like this

[Equipment 19]
NodeName=W03
LinkName=BoldType1
StartDate=19380101
EndDate=19420101

[Equipment 20]
NodeName=W03
LinkName=BoldType1
StartDate=19420101
EndDate=19451231

i have no idea what will happen---no idea at all- but as all my convoys have an extra squadron of 4 clemsons atacched to them what ever does happen i'll be sure to see it--if the game doesn't crash that will something--

there's isnt any mention of the bold in the subs equip files so i can't make a direct comparison of the correct way to add it (if there is a correct way to add it)

should be on a torp tube really but the DD's don't use their torp tubes so thats a dead one--

if some thing like this would work then we can edit the files to allow the DD's to use bold 1 making it les efefctive than stock and start the player sub on bold 2 instead making that less effective to match the original bold 1--

[Marhkimov]

I honestly don't know much about group commanders and setting a DD as a group commander, but I have a hunch that if the group commander is destroyed (whether it is a DD or not), then the DD (escorts) will begin to suspect that a u-boat is around. This is probably when they start actively searching for the u-boat.

Inversely, if you attack a convoy and destoy a ship that is not the group commander, maybe the DD's are clueless about the presence of a u-boat.

Does anyone have any conclusive evidence about my hunch?

[Redwine]

Of course CB ....

just use any free hexeditor, sure you have one at this time of jobs in SH3.

Open the file Sensors.sim....... do not confuse with .dat.

Look at the column at right and search for amun_Bold

There are many.

Each one as a secuence of settings as follows :

bubbles
life_time
surface
noise

After bubbles is a call for the visual effect.

Let one point free after life_time and surface

life_time.
surface.

the number start up at the second point after the words.

if i not remember bad, the first decoy has a surface of 250 m2

you will read

00 00 74 43

Into a hexeconverter you need to put it in inverse way to read the surface number :

43 74 00 00 = 250

If you want to input a new value in example 500 m2

insert 500 into a decimal to hexe converter 32bit y you will obtain :

43 fa 00 00

Then inser the number in the inverse order into the same place you obtain the original value, starting at the second point after the word surface

00 00 fa 43

Now your decoy will have 500 m2 effective surface.



Do the same with life time, starting at the second point after the sentence life_time


Repit the procedure for al 5 or 6 diferent decoys.



I do not know how to edit noise, wich is the most interesting here, but increasing surface has a good effect on active sonar.

the sentence bubble is refering to the bubble visual effect, not intersting for this topic.



Belw you can found the :

wpn_BoldsLauncher wich is the launcher with the setting :

bolds_count, wich is the quantity of bold inside the sub, but i dont know how to tweak it.

[gouldjg]

All sounds very promising really to me.

I think we are all doing a good job in our own rights here because we are all trying different things and have now come to the agreement to try and prove or dispell our results with a language we all seem to be understanding. i can see where you are comming from etc blah blah blah.

I looked at the commander settings in the campaign files and all I could find was a reference to the subs when commander = 1. All the rest =0.

Anyway, I sometimes too get the feeling that one attacking DD gets priority over the rest. I suppose there is some form of group rules that they play by.

I like both stories about the decoys and possible methods to implement a blind zone during DC attack. Cool stuff going on there.

Anyway guys, some news from my side,

Please be patient with me as am just nipping into a different direction at this moment in time so I can prove or disprove somethings that keep tinkering in my mind.

At the moment, I have all crewratings=4 on everything that moves on the sea :rotfl: :rotfl: :rotfl: a bloody iceburg will probably attack me at somepoint knowing my luc.

I have left everything in the sim.cfg alone and am doing my changes in the ai sensors.dat which may look like I am taking a step backwards.

I have run a couple of missions on the Happy times mission as i stated before.

Test 1 = detection at flank/30mtres at approx only 1800 -2000mtre max with default sensors, and default cfg apart from detection time for hydro being 0.5.


My first tweak at AI sensors is as follows due to the fact that the DD detection is on third short, I have decided to add 3000 mtres to all passives.

Test 2 = Detection at flank/30mtres = 3500-3800 mtres

results = further tests needed to be conclusive but improvements noticed.


Hopes and desires

Like you guys, My main motive is my own satisfaction with the game and a possible new look at damage should things go to plan.

Also like you guys, I am not really interested in gaining Kudos in this area, because quite frankly, I feel that whoever does the mod, will have to keep on backing himself up against critics etc a bit like Beery with the fatigue mod and deck gun.


I am not bothered about losing other crewratings for the use of just one as long as that one works to my own plaaybility.

I am not being thrown off by the figures in the ai sensors dat file. To me, if I decide that I have to set all passives to 3x as much as their historical values to get my version of sensible ranges in game then so be it. That is of course If my plan is correct .

At the moment I am working to this line of thought before looking at uber DD's, decoys etc. I am just preying that such changes do not make them start detecting me in silent mode, which is of course a risk I must make and hours I must spend proving this.


If this line of thought fails for any particular reason, I am quickly going to catch up to you guys and start to really get involved with your lines of thought i.e. cfg changes. I do think I may need to do some CFG chages at one point anyway.

Please bear with me while I just either prove or disprove something.

[Col7777]

I just ran a test using a cruiser with 4 DD's attached, I got a sighting then sailed straight for it. I was not detected at peri depth and running silent, I raised the scope fired 2 torps then lowered and continued without changing course.
I sank the cruiser and the DD's started to look for me but with no real hostility. I continued my course and left them circling where I was roughly when I fired the torpedoes, they eventually formed up again and carried on their course.

The reason... I edited out the Sensor.cfg

So does this prove the Sensor.cfg is read as a kind of back-up to the Sim.cfg?

[gouldjg]

Col7777

I cannot say for sure. Are you on about the sensor.cfg as in the supposed sub affecting sensor cfg?

On a side note, it is nice to see that hydro detection is also affected by aspect of sub but again where does this come from? or is just the fact that once directly in between oncoming escorts and convoy, I start to circle whilst waiting.

The threat indicator is a great help here, I have never had it on so it is new to me.

[CB..]

Originally Posted by Col7777

I just ran a test using a cruiser with 4 DD's attached, I got a sighting then sailed straight for it. I was not detected at peri depth and running silent, I raised the scope fired 2 torps then lowered and continued without changing course.
I sank the cruiser and the DD's started to look for me but with no real hostility. I continued my course and left them circling where I was roughly when I fired the torpedoes, they eventually formed up again and carried on their course.

The reason... I edited out the Sensor.cfg

So does this prove the Sensor.cfg is read as a kind of back-up to the Sim.cfg?

yup Col i reckon that with out any doubt at all there some sort of reversed logic cross referencing going on between the two cfg files - i have no idea what the trick is but there's something going on

tell you something even more likely to be difficult--i reckon SH3 has the same old "kick in the rubber pots" as SH2 had up it's sleeve ---just to make life really REALLY confusing (and i'm dreading saying this) the AI works differently in the campaign than it does in single missions---sigh -- i know-- i know---i'm going to have to have a look at trying to check this out---but my thought on this is that it doesn't matter how i edit the cfg's or dat now that i'm testing from the single mission launcher-----

( even tho the mission im using only uses shipping that is stored in the campaign.rnd like th scapa flow mission does --- the mission file only has the player sub written into it--every thing else in the mission comes directly out of the campaign files-- i thought this would be a easier way for me to test than running a dang full campaign patrol each time ---argggh!!!)

---i'm getting stuck with my lone uber DD unshakable without remorse or pity and often without even depth charges will circle me making runs directly over me for ever --i thought it was part of the uberisation (if there is such a word lol! ) due to the removing of the hydrophone entrys from th sim.cfg---but theyre all back in and the ai_sensors.dat is back to stock and it's still happening--

opposite problem from normal---except that i never play the single missions EVER!! i only ever play the campaign--i'll go back and run a proper campaign patrol from the career game launcher and i'll bet i'm stuck with 100% useless DD's again--- dang---

if so at least this would explain the two differing problems from the same game--but it'll be a pig to sort out if it's true--

on the plus side the Clemson with (the big boobys) bolds didn't cause any problems in the game -- i've just been DC'd for a long time by one and nothing un toward happened---tho try as i might i couldn't see any bolds in the water--but it's a start--

this is a weird one--

aha i know what i can do i can edit the flotilla cfg to place the start location for one of the career home ports to smack bang in the middle of the main convoy routes--that'll save my sanity a little--(what's left of it)

perhaps this relates a little to the cfg stuff perhaps the campaign engine and the single mission engine uses the cfg entrys in a different way no idea -some where to start--

cheers Red for the bold info will get on it- fingers crossed

[Ducimus]

Misc:

- Im positive that regardless of crew rating, there is randomization involved on detection. How MUCH that randomization is, im guessing is dependant on the crew level. The reason i say this is because of varing results under the same test conditions.

-Sim.cfg
Noisefactor seems to be arough adjustment
Wave factor seems to behave like fine tuning.

With Noise factor of 0.5, and wave factor of 0.0 i was detected about 2000 meters from a crew rating 3 DD. At slow speed, i dont remember if i had silent running on or not, but i do remember that i repeated the same test with a noise wave factor of 0.25 and wasnt detected. Curretly experimenting with
Wave factor 0.25 (default is .5)
Noise factor 0.5 (default is 1)


-Thoughts on crew rating and DC's as a whole.
I noticed crew rating 3 is nowhere near as accurate as 4, but a marked improvement over 1 and 2. After watching a crew 3 most of the DCs were not even close. Ive come to the conclusion that the devs adjust DC's with an accuracy of 5 and radius of 40 to accomidate for crew rating inaccuracies. Adusting from the stock values i think neuters their effectiveness. The side effect however is crew4's are deadly.

-Crew rating 4' seems to act the most consistant, and aggressive, and responsive, but even then behavior is not 100% predictable under same test conditions.


- Group commander theory:
I have a bad feeling your right. On the U505 mission regardless if the DD's rating is 1 or 4, il'll bet they'll all respond and migrate to your vaciniity the instant you torpedo the carrier.


- Being depth charged at 300 meters by 5 Crew 4 DD's is cool :P
BTW, did you know the AA guns get all tweaked and warped ? I took an ashcan right above the conning tower, wrecked EVERYTHING. If it had hit anywhere else id probably been sunk. At 25 accuracy and 20 blast radius, i had 3 DC's get close enough to do damage. 2 atop the conning tower, 1 near the stern, the rest were just alot of noise.EDIT: I was at 300 meters, so the falling time made for the DD inaccuracy, at 150 meters id have been pummled.

[Ducimus]

Forgot to add,

Im gonna try a war patrol with
wave factor 0.25
Noise factor 0.5

WIth crew rating 3 DD.s

I just want to see if these adjusments change their situational awarness at all.

All The tests i ran are not completely accurate because:

1. Map is in 1940, DD's are equiped differently in 1940 vs say 44. So test was only against 1 type of DD passive sonar im gussing.

2. Only 1 DD, not 2, or 3.

3. Their not escorting anything.

etc etc..

But it did give me an idea on how the sonar is working.

[Col7777]

All good stuff fellas, can I just go back to my last post where I said I edited out the Sensor.cfg, there were four DD's, three of them circled then the fouth joined in, it was the last one, the fourth that stayed that bit longer after the other three resumed on their course.
CB is right as some of you have already said, one hangs around that bit longer, even though this time it didn't hang around too long.

[gouldjg]

Ok Ok
:rotfl: :rotfl: :rotfl:

I guess we are going to need to test these things in campaign mode from now on. (what a bummer cos I am lazy.)

CB can you give any clues as to what year to start at and where best to go in both base and patrol grids.

I kind of aggree with the consensus here about random behaviour regardless of crew ratings.

Like I just noticed before, I think hydro is also being affected by aspect. Either that is just something to do with my direction as opposed to where the propeller aims at or it has something to do with something.

Man, I thought this was starting to get easier not harder .

I will leave my settings as they are for a few missions in campaign mode and then start to report back results.

I have to admit, I tried my first campaign and headed staraight to scapa and indeed, unless I did something completely stupid, the DD can be as blind as bats even though I had set sensors.dat.

This is all very frustrating but I am not for giving in yet.

Remember some posts back, I mentioned I had seen a lot of jargon about random this and that. I forgot where it was and I spent all day flicking in every file that day (Very boring).

Good work CB, Col7777 and ducimus.

It looks like we have to delve into a massive can of worms here.