AFAIK they do
They need them for aircraft attacks also
:up:

This I call an instant reply. Thanks BBW.

Currently working on a solution without the risky Hex-Offsets.
Instead I use the Random Folder Function of Sh3-Cmdr. Works well.

Current problem is the fine-tuning. What about these values:

Probability, that NO merchant of a convoy uses searchlight (SL): 80%
Probability, that ONE merchant ship type uses his SL: 15%
Probability, that TWO merchant ship types use SL: 5%.

(Tankers and merchants without weapons will never use their SL).

Comments welcome.
h.sie

Wow, can't belive how fast your getting this to come together.
Nice work :yeah:

Oh yes.

Hi Jim,

still waiting for the Kaleun Bechers. Hope they arrive this weekend.
If not, I'll phone them again.

h.sie

No problem H....in your own good time.

Much appreciated http://www.psionguild.org/forums/ima...s/thumbsup.gif

Do the values change with the year of the war?

yes a little bit, because currently two conditions must be true to make a merchant use his light:

1) Light must be made visible by SH3-Cmdr random function.

AND

2) light must be available in the (unchanged) .eqp file

Because of 2) there is a time-dependence according to the availability times in the .eqp files.

But it's possible to make the time dependence much more obvious/dominant using the time-dependence of the random function in SH3-Cmdr.

But to use that, I need data: The propability in dependence of the time. Or in other words: At what time of the war did merchants use their lights seldom and at what time more often? Any historic data available??

h.sie

In order to make halfway precise calculations regarding probabilities, I need to know the following (maybe someone can help me):

If in Campaign_RND there is a convoy ship entry like:

Origin = British
Type = 102

And in the Roster Directory for Britain there are, lets say, 5 different ships of type 102 available at the time the convoy is spawned:

Does every of these 5 ships have the same probability (20%) to be spawned or are there some ships preferred over others?

I remember to have asked that before and I also remember the answer, but it was not precise enough.

Thank you.

EDIT: DONE. Don't need information to this topic.

The only person I know who might know the answer is BBW.

Hi irish,

thanks. last night I made some tests with random generated convoys where Class is NOT given. It seems that every ship class which matches the "Origin" and "Type" condition is spawned with the same propability. There seems to be no preference for some ships. This I needed to know in order to make some over-all estimations regadring probability.

Regarding the time - dependence of it all: I guess that the later the war, the higher the probablility a ship uses the lights???

Some historians here?

If I get no other information, I'll make a slightly increasing probability for the searchlights during the war.

Little Preview:

Since I have no other historically based information, I have to guesstimate and I'll start with the following values:

"p_all_dark" is the probability that no merchant of a convoy will use his searchlight, thus the convoy keeps dark (except escorts):

Time dependency of p_all_dark:

p_all_dark (1939) = 95%
p_all_dark (1940) = 80%
p_all_dark (1941) = 70%
p_all_dark (1942) = 60%
p_all_dark (1943) = 50%
p_all_dark (1944) = 40%
p_all_dark (1945) = 30%

Otherwise, one or two (or, very improbably: more than two) merchant ship types will use their lights.

There will be 18 armed merchant ship types involved in the randomisation:

9 Cargo / Freighters (Type 102)
6 Troop Transport (Type 103)
3 Replenishment (Type 100)

Tankers (101), Coastal vessels (104) and unarmed Cargo are not involved in the randomisation and thus will never use their lights.

Warships with anti-uboat-weapons (destroyer, escorts, frigates) are also not involved in the randomisation. they always use their searchlights.

Big warships like Battleships, Cruisers are also not involved in the randomisation, but one can chose whether they should use their lights all the time or not.

Mod will use the Random-Folder function of Sh3-Cmdr. and will NOT use the risky Hex-Offset-method.

Comments welcome (maybe the quantities have to be adjusted)
h.sie

....continuation of my monologue :):

Doing the final tests before publishing, I unfortunately found out that this mod in its current state conflicts with the external camera. When pressing the "," or the "." key on the keyboard in order to show the next / previous unit you get an "camera out of sectors" error. This surely has to do with the "searchlight-nodes-moved-out-of-visible-range" workaround.

On the other hand: This mod is for realism. And the external camera does not have anything to do with realism - so there is not really a conflict between them.

But in spite of this, I currently try a different solution: Instead of moving the searchlights nodes out of visible range, I deleted them in the .dat file. To be honest, I expected a CTD because now there are references to the searchlights nodes in the .eqp file but no such nodes exist in the .dat file. I was very surprised that nothing happened and all worked fine, even the external camera.

But I have to do intensive tests now to be sure all works.