Questions about discretion

[Fahnenbohn]

Hello every body,

I would like to talk about a strange situation I encounter with GWX v3.0 : so, it's night, i've spotted a convoy and calculated an interception course. Now, I am well placed to attack. A destroyer is headind the convoy. So, I am thinking it won't repair me, because it's not aware of my presence... I let it pass 150 meters near me, I am absolutely silent and submerged. And it detects me every time !!! :wah:

How is it possible, as I am submerged and not detected yet ?? :hmmm: I don't understand at all. Maybe there's something I don't know about Allies' detection's strategies ! .... ?

Thanks for answering if you have any idea !

[makman94]

Hello every body,

I would like to talk about a strange situation I encounter with GWX v3.0 : so, it's night, i've spotted a convoy and calculated an interception course. Now, I am well placed to attack. A destroyer is headind the convoy. So, I am thinking it won't repair me, because it's not aware of my presence... I let it pass 150 meters near me, I am absolutely silent and submerged. And it repairs me every time !!! :wah:

How is it possible, as I am submerged and not detected yet ?? :hmmm: I don't understand at all. Maybe there's something I don't know about Allies' detection's strategies ! .... ?

Thanks for answering if you have any idea !

hello Fahnenbohn,

it is the sonar that is tracking you. the sonar is an active sensor,sending a ping which hits on your hull and returns back. it doesn't matter if you are silent or not. AI sonar is effected be your distance and depth.
http://i57.tinypic.com/33bh0s3.jpg

[Hitman]

But won't the active sonar start pinging only when a passive system has first found the contact? :hmm2:

As far as I know, SH3 only uses one sensor at a time, and it uses by default that which has the biggest range/detection ability available. It then switches to active sonar if he has found a sub and it is submerged.

[makman94]

But won't the active sonar start pinging only when a passive system has first found the contact? :hmm2:

As far as I know, SH3 only uses one sensor at a time, and it uses by default that which has the biggest range/detection ability available. It then switches to active sonar if he has found a sub and it is submerged.

hello Alberto,

if you got into the sonar's area you will get pinged .
i can tell you this from my tests where i made some fixed single missions: my boat was stationary submerged and the target was setted to pass in front of me at about 300-400m. the target was indentified (this allows you to see on navmap page the sonar's area) and as soon as i got in sonar's area i was pinged (for some delay that may be on this is responsible the settings at sim.cfg file)

i think (not sure though) that you are right that sh3 uses only one sensor at time but if sonar is the first one that tracks you ,you will get pinged

[Fahnenbohn]

Hello makman94 !

Thanks for your answer ! I already knew what you told me, but my problem is on this point : i think it's unrealistic to be detected although the ennemi doesn't know you are here... In fact, the sonar seems to work underwater as the radar does above water, even if it doesn't send a ping : you are detected as soon as you enter in the detection's area.

@ Hitman :

It would be nice if SH3 was using by default the sensor which has the biggest range/detection ability available, and then was switching to active sonar if he has found a submerged sub. But it doesn't happen like that in my game !


So, i wonder if there is a solution to avoid this problem by changing the parameters in sensors.cfg or sim.cfg or else ... ?

I give you my own parameters :


In sensors.cfg :

Sonar range factor=1 ;[>=0]
Sonar fog factor=0 ;[>=0]
Sonar light factor=0 ;[>=0]
Sonar waves factor=0.3 ;[>=0]
Sonar speed factor=0.1 ;[>=0]
Sonar aspect=1 ;[>=0]
Sonar enemy speed=0 ;[>=0]
Sonar noise factor=0 ;[>=0]
Sonar sensor height factor=0 ;[>=0]
Sonar already tracking modifier=20 ;[detection probability modifier]
Sonar decay time=5 ;[>0] already tracking bonus decay, in seconds
Sonar uses crew efficiency=true ;[true or false]


In sim.cfg :

[AI detection]
Lost contact time=15

[Sonar]
Detection time=20 ;[s]
Sensitivity=0.03 ;(0..1)
Waves factor=0.5 ;[>=0]
Speed factor=20 ;[kt]
Enemy surface factor=200 ;[m2]
Lose time=30 ;[s]


Is there anything wrong ? Do you have any suggestions ? ...

[makman94]

Hello makman94 !

Thanks for your answer ! I already knew what you told me, but my problem is on this point : i think it's unrealistic to be detected although the ennemi doesn't know you are here... In fact, the sonar seems to work underwater as the radar does above water, even if it doesn't send a ping : you are detected as soon as you enter in the detection's area.

....

i don't know if it is realistic or not as i don't know how the AI sonars worked in reality back then. i mean that if the sonar devices pinged all the time at their working area then it is well done in game too but if the sonars were just turned on manualy when something had to be searched and the crew started some specific pings then yes , you are right at calling it unrealistic.


So, i wonder if there is a solution to avoid this problem by changing the parameters in sensors.cfg or sim.cfg or else ... ?

I give you my own parameters :


In sensors.cfg :

Sonar range factor=1 ;[>=0]
Sonar fog factor=0 ;[>=0]
Sonar light factor=0 ;[>=0]
Sonar waves factor=0.3 ;[>=0]
Sonar speed factor=0.1 ;[>=0]
Sonar aspect=1 ;[>=0]
Sonar enemy speed=0 ;[>=0]
Sonar noise factor=0 ;[>=0]
Sonar sensor height factor=0 ;[>=0]
Sonar already tracking modifier=20 ;[detection probability modifier]
Sonar decay time=5 ;[>0] already tracking bonus decay, in seconds
Sonar uses crew efficiency=true ;[true or false]


In sim.cfg :

[AI detection]
Lost contact time=15

[Sonar]
Detection time=20 ;[s]
Sensitivity=0.03 ;(0..1)
Waves factor=0.5 ;[>=0]
Speed factor=20 ;[kt]
Enemy surface factor=200 ;[m2]
Lose time=30 ;[s]


Is there anything wrong ? Do you have any suggestions ? ...

i don't think that you will manage anything through the sim.cfg file . the parameters there are only for delay this detection .(you could possibly even eliminate it but then the AI will have no sonars at all).
i am afraid that the way the sonar works is hardcoded stuff and only tweaks at executable files can change the sensors's behaviour

[Fahnenbohn]

Yes, it may be hardcoded. So, it's a future improvment to do !! I will suggest it to h.sie, I believe he'll be interested in ... I've already suggest him another improvments like no seagulls at night.

Now, i would like to understand several things :

What do these parameters change in game ? :
- Sonar already tracking modifier
- Sonar decay time
- Detection time
- Speed factor
- Lose time

:salute:

[sublynx]

The way the detection works is unrealistic in a technical sense. It very likely didn't work like that during the war.

However the lack of knowledge how the detection works is very realistic in an experiential sense. Not exactly knowing what the escorts were capable of, was exactly what the U-boat commanders faced. The beauty of the game is partly in trying to figure that out in only a few convoy attacks in constantly changing weather and escort type conditions.

[Fahnenbohn]

However the lack of knowledge how the detection works is very realistic in an experiential sense. Not exactly knowing what the escorts were capable of, was exactly what the U-boat commanders faced. The beauty of the game is partly in trying to figure that out in only a few convoy attacks in constantly changing weather and escort type conditions.

Mmm.... I'm not convinced...:-?

[makman94]

Yes, it may be hardcoded. So, it's a future improvment to do !! I will suggest it to h.sie, I believe he'll be interested in ... I've already suggest him another improvments like no seagulls at night.


i think that H.Sie is not modding sh3 anymore so better contact Stiebler. he is the only one that can answer to such question


Now, i would like to understand several things :

What do these parameters change in game ? :
- Sonar already tracking modifier
- Sonar decay time
- Detection time
- Speed factor
- Lose time

:salute:

the first two are parameters for your sonar device.
the ''Sonar already tracking modifier'' is a detection probability modifier.as explained in the cfg,once a contact is detected it will lose it very hard
the ''Sonar decay time'' is the time of fading of your sonar line once the contact is lost

the rest three parameters are for AI Sonars
the ''Detection time'' is the delay at the detection once the AI has locate you
the ''Speed factor'' is a modifier that taking into account your speed in order to calculate the final probability for AI to detect you
the ''Lose time'' is the time that AI sonar will track you since its last ping.

[Fahnenbohn]

Thanks for the explanations makman ! I will try some tests... :ahoy:

[irish1958]

I seams to me the only simple solution to the problem would be to use CMDR to randomize a series of different sensor files. This would keep the element of uncertainty. I can think of a number of drawbacks to this but war, performance, and competence are all uncertain and variable.
It has always bothered me that I could be detected if completely silent, deep and unexpected.

[BigWalleye]

I seams to me the only simple solution to the problem would be to use CMDR to randomize a series of different sensor files. This would keep the element of uncertainty. I can think of a number of drawbacks to this but war, performance, and competence are all uncertain and variable.
It has always bothered me that I could be detected if completely silent, deep and unexpected.

I suspect that there were RL skippers who were also bothered that they were detected while they thought they were"completely silent, deep, and unexpected." Although they might not get a chance to complain about it. I have no personal experience, but I believe war can be like that.

[Stiebler]

I've been asked by Makman94 to look at this thread.

I wish to point out that the standard advice for Royal Navy warships was to keep active sonar (pinging) and active radar on at all times while at sea. That was the recommendation (but not an order) in 1945, I read it in a contemporary Royal Navy magazine - for sailors' eyes only. Some serving British sailors had asked whether it was wise to keep pinging and using radar, in good weather, when the U-boats could hear the pings a long way off, and the U-boats were fitted with radar detectors.

The magazine's argument was that use of eyes and hydrophones to pick up U-boats might not be effective, but the sound of pinging and radar detection would force the U-boat to run away or dive deep quickly. It is true that in 1945 most of the U-boats were deployed in shallow waters around Britain's coast - the magazine's recommendation was a scarecrow tactic.

I see no reason to make the hard-code changes outlined in previous posts, although doubtless it is possible. It would take a long time to find the necessary code, though.

Stiebler.

[makman94]

I've been asked by Makman94 to look at this thread.

I wish to point out that the standard advice for Royal Navy warships was to keep active sonar (pinging) and active radar on at all times while at sea. That was the recommendation (but not an order) in 1945, I read it in a contemporary Royal Navy magazine - for sailors' eyes only. Some serving British sailors had asked whether it was wise to keep pinging and using radar, in good weather, when the U-boats could hear the pings a long way off, and the U-boats were fitted with radar detectors.

The magazine's argument was that use of eyes and hydrophones to pick up U-boats might not be effective, but the sound of pinging and radar detection would force the U-boat to run away or dive deep quickly. It is true that in 1945 most of the U-boats were deployed in shallow waters around Britain's coast - the magazine's recommendation was a scarecrow tactic.

I see no reason to make the hard-code changes outlined in previous posts, although doubtless it is possible. It would take a long time to find the necessary code, though.

Stiebler.

hello Stiebler,

yes, there was this question floating around as how the real sonar devices worked.
according to the magazine ,we have one clue now that the sonars ,indead, were working 24/7,
so yes, there is nothing to be 'fixed' here,devs made it the right way.

thank you for the interesting input !:up:

[BigWalleye]

I've been asked by Makman94 to look at this thread.

I wish to point out that the standard advice for Royal Navy warships was to keep active sonar (pinging) and active radar on at all times while at sea. That was the recommendation (but not an order) in 1945, I read it in a contemporary Royal Navy magazine - for sailors' eyes only. Some serving British sailors had asked whether it was wise to keep pinging and using radar, in good weather, when the U-boats could hear the pings a long way off, and the U-boats were fitted with radar detectors.

The magazine's argument was that use of eyes and hydrophones to pick up U-boats might not be effective, but the sound of pinging and radar detection would force the U-boat to run away or dive deep quickly. It is true that in 1945 most of the U-boats were deployed in shallow waters around Britain's coast - the magazine's recommendation was a scarecrow tactic.

I see no reason to make the hard-code changes outlined in previous posts, although doubtless it is possible. It would take a long time to find the necessary code, though.

Stiebler.

Interesting, indeed! It is certainly different from present-day doctrine, which is to first make contact using passive sensors. From accounts I have read, it is also somewhat different from WW2 IJN tactics. IJN did sometimes use active sonar for search, but more often would search initially with passive. Continuous pinging was not unheard of, but was unusual enough to cause comment. US submariners regarded it as a mark of an inexperienced ASW team. As Stiebler said, it was considered a "scarecrow" (good word!) tactic. The purpose was not so much to engage the enemy as to warn him off.

[irish1958]

It appears we have two different questions at play here. I have no problem when an active sonar ping picks my sub up. I can always hear it at greater range than the destroyers can hear the reflection, and therefore take whatever action needed to either attack or evade. The original question was the ability of the destroyers to always find you with passive sonar no matter how silent your sub is or how deep it is. Did this really happen? And what, if any, did thermal layers have on this effect (in game, via SH3CMDR)?

[makman94]

.....

The original question was the ability of the destroyers to always find you with passive sonar no matter how silent your sub is or how deep it is. Did this really happen?....

hello Irish,

i guess that with ''passive sonar'' you mean the hydrophones ,right?
this is good question that must be checked.

i don't believe that this is what happening in game. my suspicion is that it is the sonar (by sonar i mean the active device) which is detecting you and not the hydrophones.

it is easy to create a mini single mission and test your thought. we will put in this mission a destroyer which will be equiped only with a strong hydrophone and not an active sonar device at all.the destroyer will be setted to pass in front of us at about 300-400m and see what will happen.

i will make the test and let you know

[irish1958]

hello Irish,

i guess that with ''passive sonar'' you mean the hydrophones ,right?
this is good question that must be checked.

i don't believe that this is what happening in game. my suspicion is that it is the sonar (by sonar i mean the active device) which is detecting you and not the hydrophones.

it is easy to create a mini single mission and test your thought. we will put in this mission a destroyer which will be equiped only with a strong hydrophone and not an active sonar device at all.the destroyer will be setted to pass in front of us at about 300-400m and see what will happen.

i will make the test and let you know
Yes, hydrophones only.
Thank you for your quick reply.

[BigWalleye]

hello Irish,

i guess that with ''passive sonar'' you mean the hydrophones ,right?
this is good question that must be checked.

i don't believe that this is what happening in game. my suspicion is that it is the sonar (by sonar i mean the active device) which is detecting you and not the hydrophones.

it is easy to create a mini single mission and test your thought. we will put in this mission a destroyer which will be equiped only with a strong hydrophone and not an active sonar device at all.the destroyer will be setted to pass in front of us at about 300-400m and see what will happen.

i will make the test and let you know

Good idea, Makman. May I suggest an additional test? Program the escort to pass parallel to the sub on opposite course at the same range. If the sensor is modeled properly, there should be a large difference in detection probability between head-on and broadside-on. What do you think?

[makman94]

Good idea, Makman. May I suggest an additional test? Program the escort to pass parallel to the sub on opposite course at the same range. If the sensor is modeled properly, there should be a large difference in detection probability between head-on and broadside-on. What do you think?

hello BigWalleye,
i did ,also, a mission as you suggested :up:

Yes, hydrophones only.
Thank you for your quick reply.

Irish, you were right but you were ,also, wrong too ! i will explain

get this mission:
http://speedy.sh/mMadn/TEST-SENSORS-FOR-FLETCHER-ONLY-HYDROPHONES.rar

drop it in your MODS folder and enable.
you will see two new single missions called as ''Fletcher - 400'' and ''Fletcher - 400 - 2''
at ''Fletcher - 400'' the Fletcher will pass at 400m in front of you and
at ''Fletcher - 400 - 2'' the Fletcher will pass at 400m on the starboard side. (as BigWallEye suggested)

at both missions the Fletcher is equiped ONLY with hydrophones ( specifically, with a type QClP ). don't raise your periscope as it will be spotted (Fletcher is allready too close and it is impossible ,through cfgs, to discard the ai-visuals). don't do anything, just wait for Fletcher to pass by.

now, why i said that you are right and wrong at the same time?

if you run the missions on STOCK sh3 everything works perfect on hydrophones and ,as long as you are silent , the Fletcher is not detecting you.
if you run the missions on gwx or nygm or wac (lsh3 is ok) ,the Fletcher is always detecting you ,through its hydrophones, although you are absolutely silent. so , i am guessing that you are a user of gwx or nygm or wac.

after a little search,i found that the cause of these anomalies at hydrophones are the settings of hydrophones at sim.cfg file. by setting to sim.cfg the STOCK hydrophone settings everything worked like a charm to all supermods (gwx,nygm and wac. LSH3,as i said, has not this problem).

the stock values that work ok are:
[Hydrophone]
Detection time=1 ;[s]
Sensitivity=0.03 ;(0..1)
Height factor=0 ;[m]
Waves factor=0.5 ;[>=0]
Speed factor=15 ;[kt]
Noise factor=1.0 ;[>=0]

once you use the above settings:
at some of your tests , challenge with the Fletcher by starting your engines.when you are detected ,play with the throttle and you will see that Fletcher immediately is loosing you ,especially when you stop the engines.(remember ,this Fletcher is equiped only with hydrophones).


edit:
i forgot something very critical and this is the ''Rig for silent running'' order.
with this order enabled , your boat is not detected by Fletcher's hydrophones ,as long as you keep silent, at all supermods too. so, maybe, the modders of supermods's sensors have that in mind when they were adjusting their hydrophone settings.
the boat ,even with stopped engines is a little noisy (repair works-torpedo loadings..etc) and this is what is detecting Fletcher's hydrophones.
at the above missions ,if you enable the ''Rig for silent running'' order, you will see that Fletcher's hydrophones are not detecting you, so my final conclusion is that supermods wanted it to be that way

[BigWalleye]

hello BigWalleye,
i did ,also, a mission as you suggested :up:



Irish, you were right but you were ,also, wrong too ! i will explain

get this mission:
http://speedy.sh/mMadn/TEST-SENSORS-FOR-FLETCHER-ONLY-HYDROPHONES.rar

drop it in your MODS folder and enable.
you will see two new single missions called as ''Fletcher - 400'' and ''Fletcher - 400 - 2''
at ''Fletcher - 400'' the Fletcher will pass at 400m in front of you and
at ''Fletcher - 400 - 2'' the Fletcher will pass at 400m on the starboard side. (as BigWallEye suggested)

at both missions the Fletcher is equiped ONLY with hydrophones ( specifically, with a type QClP ). don't raise your periscope as it will be spotted (Fletcher is allready too close and it is impossible ,through cfgs, to discard the ai-visuals). don't do anything, just wait for Fletcher to pass by.

now, why i said that you are right and wrong at the same time?

if you run the missions on STOCK sh3 everything works perfect on hydrophones and ,as long as you are silent , the Fletcher is not detecting you.
if you run the missions on gwx or nygm or wac (lsh3 is ok) ,the Fletcher is always detecting you ,through its hydrophones, although you are absolutely silent. so , i am guessing that you are a user of gwx or nygm or wac.

after a little search,i found that the cause of these anomalies at hydrophones are the settings of hydrophones at sim.cfg file. by setting to sim.cfg the STOCK hydrophone settings everything worked like a charm to all supermods (gwx,nygm and wac. LSH3,as i said, has not this problem).

the stock values that work ok are:
[Hydrophone]
Detection time=1 ;[s]
Sensitivity=0.03 ;(0..1)
Height factor=0 ;[m]
Waves factor=0.5 ;[>=0]
Speed factor=15 ;[kt]
Noise factor=1.0 ;[>=0]

once you use the above settings:
at some of your tests , challenge with the Fletcher by starting your engines.when you are detected ,play with the throttle and you will see that Fletcher immediately is loosing you ,especially when you stop the engines.(remember ,this Fletcher is equiped only with hydrophones).


edit:
i forgot something very critical and this is the ''Rig for silent running'' order.
with this order enabled , your boat is not detected by Fletcher's hydrophones ,as long as you keep silent, at all supermods too. so, maybe, the modders of supermods's sensors have that in mind when they were adjusting their hydrophone settings.
the boat ,even with stopped engines is a little noisy (repair works-torpedo loadings..etc) and this is what is detecting Fletcher's hydrophones.
at the above missions ,if you enable the ''Rig for silent running'' order, you will see that Fletcher's hydrophones are not detecting you, so my final conclusion is that supermods wanted it to be that way

Well-executed and useful piece of research, Makman. Answers a great many questions. Demonstrated the value of silent running at low engine speed. I was surprised, though, that there was no noticeable difference between bow-on and broadside aspects. That's useful information, also.

[irish1958]

Thank you both for your work and information. I run GWX and NYGM only.
I will download the missions and play around with them when I have time, but this info explains a lot.
I should have done the work myself. I am really not lazy, but have been preoccupied with RL stuff.
What a great forum with polite, intelligent and congenial members. Thanks Neal.

[BigWalleye]

Thank you both for your work and information. I run GWX and NYGM only.
I will download the missions and play around with them when I have time, but this info explains a lot.
I should have done the work myself. I am really not lazy, but have been preoccupied with RL stuff.
What a great forum with polite, intelligent and congenial members. Thanks Neal.

I hate when RL intrudes on my gaming time!:D

[makman94]

Well-executed and useful piece of research, Makman. Answers a great many questions. Demonstrated the value of silent running at low engine speed. I was surprised, though, that there was no noticeable difference between bow-on and broadside aspects. That's useful information, also.

hello BigWallEye,
i am glad that you found the post usefull :yep:
about your thought: i am not surprised as i see no factor for 'aspect' ratio or 'Enemy surface' at hydrophone settings. i don't think that devs have modeled this parameter for hydrophone.
on the other hand,there is the parameter ''MinSurface'' in the hydrophone sensor data (in AISensor.dat) but this is ''0'' for all types of hydrophones so it has no effect. maybe , there is the possibility ,by changing this value there, to get the desire result but not everything we see in s3d works in game. only tests will show you if this value is 'active' or not.

Thank you both for your work and information. I run GWX and NYGM only.
I will download the missions and play around with them when I have time, but this info explains a lot.
I should have done the work myself. I am really not lazy, but have been preoccupied with RL stuff.
What a great forum with polite, intelligent and congenial members. Thanks Neal.

it was a good question Irish :up:

and , as there was a way to be checked , why not ?

[Leitender]

Hi you all,

really interesting discussion! From uncountable tests with the sensors, I may tell about my experience how the sensors work in SH3.

makman

the sonar is an active sensor,sending a ping which hits on your hull and returns back. it doesn't matter if you are silent or not. AI sonar is effected be your distance and depth.
Correct, but further, the AI sonar is effected by your aspect ("bow on" or "bow off"). "Bow on" means according to NYGM 30° to each side of the bow.

Really well self explaining graphic, btw.

Hitman

But won't the active sonar start pinging only when a passive system has first found the contact?

As far as I know, SH3 only uses one sensor at a time, and it uses by default that which has the biggest range/detection ability available. It then switches to active sonar if he has found a sub and it is submerged. Both acoustic sensors work for their own AND both sound sensors can be seperately used by the exe as trigger for attacking action. Both sensors are to be used as complete attack control sensor (in contrast to radar, btw. but that´s another thing).

In fact, the hydrophone always has the bigger range, so without counter measures, the fist contact is normally made by this passiv sensor. It doesn´t matter, if the active sensor could trigger any action too, if the contact was already made by the passive means.
Hearing the pinging is only a sound function and doesn´t mean that contact is kept by the active sonar. It´s just an acoustic effect if you are within the active sonar´s detection range.

Fahnenbohm

Now, i would like to understand several things :

What do these parameters change in game ? :
- Sonar already tracking modifier
- Sonar decay time
- Detection time
- Speed factor
- Lose time You´d rather not mix values from sensors.cfg and from sim.cfg. The first both belong to the simulation of your own hydrophone operator´s simulation (sensors.cfg), whereas the latter belong to the enemy AI (sim.cfg). In your describend situation, your own hydrophone doesn´t play any part for the enemy´s detection.

Irish1958

It appears we have two different questions at play here. I have no problem when an active sonar ping picks my sub up. I can always hear it at greater range than the destroyers can hear the reflection,In theory and in real life yes, but not in the game. As already mentioned, the pinging sound can only be heard ingame, when you´re within the sonar range.

The original question was the ability of the destroyers to always find you with passive sonar no matter how silent your sub is or how deep it isThe original - real life - counter measure was: Silent running, deep diving, narrow aspect. These counter measures work perfectly ingame!

Only one thing has to be changed possibly: If you order rig for silent running, the speed will be reduced to "Ahead slow". This means in the stock configuration around 3 knots and around 150 revolutions (submerged). You may reduce the "Ahead slow" value in your NSS_Uboatxx.cfg to 0.25 from 0.40. With this, your boat will only run with about 90 revolves at 1.5 to 2 knots (like in real life) and you won´t be heard by the passive sonar. I use this value for years and if I keep also the bow to the sound source, I´m able to evade from a single enemy warship.

The reduction of the "Ahead slow" value from 0.40 to 0.25 is already included in GWX.


Big walley

Program the escort to pass parallel to the sub on opposite course at the same range. If the sensor is modeled properly, there should be a large difference in detection probability between head-on and broadside-on.The aspect doesn´t matter if there´s only a hydrophone mounted. Neither in real life nor in game. To be proofed by makman´s testing configuration.

makman

i forgot something very critical and this is the ''Rig for silent running'' order.
with this order enabled , your boat is not detected by Fletcher's hydrophones ,as long as you keep silent, at all supermods too. so, maybe, the modders of supermods's sensors have that in mind when they were adjusting their hydrophone settings.
the boat ,even with stopped engines is a little noisy (repair works-torpedo loadings..etc) and this is what is detecting Fletcher's hydrophones.
at the above missions ,if you enable the ''Rig for silent running'' order, you will see that Fletcher's hydrophones are not detecting you, so my final conclusion is that supermods wanted it to be that wayThat´s it. One small annotation: If you order "rig for silent running" the boat´s speed will be reduced to "ahead slow" from all other speeds - besides if you are already stopped. Then you will stay stopped. If you compare the behaviour in different installations, you may keep that in mind.

if you run the missions on STOCK sh3 everything works perfect on hydrophones and ,as long as you are silent , the Fletcher is not detecting you.
if you run the missions on gwx or nygm or wac (lsh3 is ok) ,the Fletcher is always detecting you ,through its hydrophones, although you are absolutely silent. so ,Once more: Agreement! This is also my testing experience. Furthermore, I can confirm your test results with the Fletcher. Be it GWX or my own optimised sensors, you could only hide if you order "rig for silent running". Different in Stock: Here, the destroyer couldn´t here me even without silent running.

i am not surprised as i see no factor for 'aspect' ratio or 'Enemy surface' at hydrophone settings. i don't think that devs have modeled this parameter for hydrophone.
on the other hand,there is the parameter ''MinSurface'' in the hydrophone sensor data (in AISensor.dat) but this is ''0'' for all types of hydrophones so it has no effect.
In my understanding, the "sensor data" functionality always contains the same bunch of values because they are needed to feed the exe, be they necessary or not.

Last annotation: The QClP sensor wasn´t touched in GWX. Thus the different results to the stock behaviour should be based on the sim.cfg settings. In NYGM, QClP has a strongly reduced range - so with this sensor you pobably won´t receive results that are comparable to other installations.

Happy testing :-)

[banryu79]

Very nice test makman94, and very nice stuff of information posted by other memebers as well.
I'm a noob here, and not a modder so I really do not know a thing about SH3 internal workings even if I always read all the info I can get about the mods I use with GWX because I like to know what I am using and how thing works and because I'm just plain curious :D

Hoping to post some useful information I include the following links:

http://www.subsim.com/radioroom/images/icons/icon1.gif GWX's "uber" AI demystified (http://www.subsim.com/radioroom/showthread.php?t=104377)
http://www.subsim.com/radioroom/images/icons/icon1.gif [TEC] Adjusting the AI - cliff notes version (http://www.subsim.com/radioroom/showthread.php?t=111395)
http://www.subsim.com/radioroom/images/icons/icon1.gif Help me making GWX a bit easier....(AI sensors) (http://www.subsim.com/radioroom/showthread.php?t=147947)

These threads helped me a lot in understanding how the sensors work in the game world.
The first one is especially useful to every user of GWX :)

[evan82]

Always the old good technique is the best. Goes ahead at silent 1- 2 kt speed when destroyer searching and is near from You. And You should turn left or right at ahed flank speed, when depth charges are in the water directly above on Your boat. Then again, is most safe to go back to 1 - 2 kt speed, and goes as deep as possible. When more than one escort hunting you, you should be more smart with your left or right turns strategy. Thats all.:ping:

EDIT. Each escort in SH3, has limited number of depth charges.

[BigWalleye]

The aspect doesn´t matter if there´s only a hydrophone mounted. Neither in real life nor in game.

Could you please provide a source for that statement? Thank you.

[makman94]

makman


the sonar is an active sensor,sending a ping which hits on your hull and returns back. it doesn't matter if you are silent or not. AI sonar is effected be your distance and depth. Correct, but further, the AI sonar is effected by your aspect...


yes ,you are right (as always ! :up:) and ,by looking now at sonar's settings ,i see that it is also effected by waves and uboat's speed too. pretty interesting modeled sensor,can we say that it is almost fully modeled ?



The aspect doesn´t matter if there´s only a hydrophone mounted. Neither in real life nor in game. To be proofed by makman´s testing configuration.


without having any idea at all, how the hydrophones worked back then , i am thinking , basis of common sense, that the sound a ship creates should be more easily heard if we are behind the propellers. i have the sense that if we are in front of ship ,the sound may should be reduced by the hull...or no?
yes, i know that in game has no difference but what about reality ?


In my understanding, the "sensor data" functionality always contains the same bunch of values because they are needed to feed the exe, be they necessary or not.


didn't know that ! and this is explaining why some parameters i saw here and there in s3d are not 'active' ! thank you Leitender :up:



Last annotation: The QClP sensor wasn´t touched in GWX. Thus the different results to the stock behaviour should be based on the sim.cfg settings. In NYGM, QClP has a strongly reduced range - so with this sensor you pobably won´t receive results that are comparable to other installations.

Happy testing :-)

yes , i had checked that but then i saw that NYGM's QClP was at 4100 (? instead of 7500m on others, if i remember correct) and i said ok ,as it will pass from 400m away it will do its job so let it be.

ps: i really wish one day for Leitender to decide to share his work on sensors ! this guy has made ,allready, an enormous work on sensors but i know that he has still a long trip to ...completion:up:

[Leitender]

Hi makman,

I see what you mean. But sound propagation in the water is non-directional. You can hear the propeller noise even if you are directly in front of a vessel. There may be a phase shift, and there may be a sound reduction if the propellers were covered by the hull, but the sound still reaches you even if you are at that position.

Compare it to sound propagation in the air: You can hear a car passing by e.g. even behind a wall. The wall cannot avoid the sound propagation. AND: It makes no difference, if the car passes from left to the right or if the car is closing or departing. Of course distance makes a difference, but it doesn´t matter if the car is heading towards you, departing from you or just passing by at a determined position (i.e. the sound propagation is independant of the AOB).

But there is a difference between air and water: Sound propagation in the water is much better than in the air because of the thickness of the medium. E.g. Propagation velocity in the air is 343 m/s in the water around 1400 m/s (4 times faster).

Historically, contacts were reported against single ships from u-boats at distances of up to 20km, against concoys up to 100km. The range of the GHG mounted in the german heavy cruiser "Prinz Eugen" was reported to be 50km. We know that whales are able to communicate with each other at distances of hundreds of kilometers (!) if they use special sound channels (and they know how to use them!). So if we talk about such dimensions, the angle on bow has hardly any influence concerning the discovering ability. So if we talk about the propeller noise of a fletcher class destroyer at a distance of 400m, there should be no difference, be AOB=0 or AOB=90, under consideration of the fact that such a sound transmitter could be heard at 20km distance.

Bigwalley

That may have answered your question? In reality there may be a gradually difference, but in the game I´ve never experienced a difference in getting listened earlier or later, depending on the destroyer´s position. makman´s tests confirm that experience. Did you made different experiences?

makman again

didn't know that ! and this is explaining why some parameters i saw here and there in s3d are not 'active' ! thank you Leitender

This not knowing either. I haven´t seen how the sh3.exe works. I only know from comparable problems, that the program needs input, be it necessary or not, else it may overflow :D. So if the exe is told to read 10 values from "sensor data", it may complain if there were only 9.

PS: Further replies soon, must urgently go to work :oops:

[BigWalleye]

Hi makman,

I see what you mean. But sound propagation in the water is non-directional. You can hear the propeller noise even if you are directly in front of a vessel. There may be a phase shift, and there may be a sound reduction if the propellers were covered by the hull, but the sound still reaches you even if you are at that position.

Compare it to sound propagation in the air: You can hear a car passing by e.g. even behind a wall. The wall cannot avoid the sound propagation. AND: It makes no difference, if the car passes from left to the right or if the car is closing or departing. Of course distance makes a difference, but it doesn´t matter if the car is heading towards you, departing from you or just passing by at a determined position (i.e. the sound propagation is independant of the AOB).

But there is a difference between air and water: Sound propagation in the water is much better than in the air because of the thickness of the medium. E.g. Propagation velocity in the air is 343 m/s in the water around 1400 m/s (4 times faster).

Historically, contacts were reported against single ships from u-boats at distances of up to 20km, against concoys up to 100km. The range of the GHG mounted in the german heavy cruiser "Prinz Eugen" was reported to be 50km. We know that whales are able to communicate with each other at distances of hundreds of kilometers (!) if they use special sound channels (and they know how to use them!). So if we talk about such dimensions, the angle on bow has hardly any influence concerning the discovering ability. So if we talk about the propeller noise of a fletcher class destroyer at a distance of 400m, there should be no difference, be AOB=0 or AOB=90, under consideration of the fact that such a sound transmitter could be heard at 20km distance.

Bigwalley

That may have answered your question? In reality there may be a gradually difference, but in the game I´ve never experienced a difference in getting listened earlier or later, depending on the destroyer´s position. makman´s tests confirm that experience. Did you made different experiences?

makman again

didn't know that ! and this is explaining why some parameters i saw here and there in s3d are not 'active' ! thank you Leitender

This not knowing either. I haven´t seen how the sh3.exe works. I only know from comparable problems, that the program needs input, be it necessary or not, else it may overflow :D. So if the exe is told to read 10 values from "sensor data", it may complain if there were only 9.

PS: Further replies soon, must urgently go to work :oops:

My "question" was not a question at all, but a request for documentation. You have given me your opinion. What I asked for was a reference to technical documentation which supports your statement. I asked because your statement conflicts with my analysis if the physics of the situation. I wanted to review the source of your statement before I presented my own opinion, which follows the same argument Makman presented. (His presentation was much less wordy!)

Sound propagation in a continuous medium is isotropic. It is the same in all directions. However, we are not talking about a continuous medium. We are talking about water in which sits a long steel tube filled with air. On one end of that tube is an acoustic generator - the screws, shafts and electric motors. If I am sitting dead ahead of that tube, and the generator is on the other end, then that cylinder of air, which is a poor conductor of sound, is between me and the radiator. There is propagation through the steel tube, but that is a very small cross-sectional area when compared to the cross-section of the air bubble. (Only the cross-section that is actual steel itself counts as steel, not the air-filler interior.)

In effect, we have a radiator on one end of a long narrow sound absorbing tube. When the sensor is directly in front of the absorber, most of the sound waves it receives must travel around the absorber. The actual area of the absorber transmits very little and so the sound level is lower. Moving the sensor to an abeam position, it becomes exposed to more of the direct transmission from the generator through the water, and the sound level rises. Directly aft, there is no occlusion by the absorber, and the sound level is highest. In fact, there may be some reflection off the steel hull in that direction, so sound level may be greater than the isotropic case.

IAC, the important point is that the sub is not an isotropic radiator and that we can expect a non-uniform, directional sound field with the strength in any direction being related to the aspect of the sub. The greatest sound level will be directly aft and the lowest level directly forward.

Your example of a car is not strictly analogous. The car has sound generators at front (engine), rear (exhaust), and all four corners (tires rolling on pavement). The sound field is not isotropic - a noisy exhaust will be more noticeable when the car is going away than when it is approaching. You can verify this for yourself. But, in general, the car produces a significant amount of noise in all directions. Moreover, the sound condictivity of the medium inside the car is about the same as the medium outside (air versus air, ignoring the metal structure and the seat cushions, etc.). The air-filled submarine is sitting in water, which has a sound conductivity (that's not the technical term, but it will do.) 3.5 times greater.

A better analogy to the sub (although in reverse) is a shotgun microphone. This is a device with a sensor at one end of a long tube of sound-absorbent material. The tube blocks sound coming from every direction except the direction the tube is pointed. This allows sounds from a long distance to be picked up, because the background is reduced. For example, you can listen to a conversation from across a crowded, noisy football stadium. The sub is the situation in reverse. The tube is sound-absorbent (relative to the surrounding water) and the sound source is at one end. When you try to listen through the tube, the sound level is significantly reduced.

A nuclear submarine has a more distributed sound source than an electric boat. The reactor and power plant are in response to occasional transient sound contacts would reach a point where they were at risk of a head-on collision! Check Peter Sasgen's Stalking the Red Bear for details of the doctrine and its use in practice. Now, a lot has changed from WW2 diesel-electrics to nuclear attack subs, but sound propagation has not. So I have to project back and say that a WW2 boat running submerged had an anisotropic sound field with the greatest sound level dead aft and the lowest dead ahead, just as modern boats do.

I can see thate your model of the sub as an isotropic radiator would be somewhat more valid in a far-field situation, where the sub is 10 km or more away from the sensor. But the anisotropy should be more significant at ranges around 1 km and under - tactical ranges for a WW2 submarine attack.

If anyone has either data or technical analysis to contradict my conclusion, I hope they will post it. I would far rather have hard technical information on this subject than duelling opinions.

As for the game, as opposed to RL, Makman has convincingly showed that its hydrophone detection model does not take target aspect into account. Whether it should is another question. I can say, without providing further detail, that more sophisticated models do.

EDIT - Some further thoughts about listening to the car behind the wall: The source is loud and the sensor (as Alexander Graham Bell discovered) is highly non-linear. If you put the wall between you and a marching band, you wouldn’t notice much of a difference! But we can refine your analogy by using an electric car instead of one with an IC engine. The electric car is quiet, but not truly silent. If we are on the sidewalk, we first hear it when it is fairly close, but we do hear it. Now go to the other side of the wall. We hardly hear the quiet electric car at all, even when it passes alongside. The wall makes much more apparent difference (to our nonlinear ears) in the case of the low-level sound. But that has much to do with the way our ears work and little to do with sound levels and attenuation.

[makman94]

seems very complicated the whole theme in real life.
if we consider that temperature of water is ,also, a great factor for sound's 'travelling' then i am thinking that creating 'sophisticated' hydrophones would have been a whole new simulator by itself.
one of the biggest lack in sh3 is ,indead, a not modeled hydrophone room.