Realism- and gameplay-related hardcode fixes for SH3.EXE

[Flyingsub]

Originally Posted by h.sie

FlyingSub: MEP3 works for Makman, reaper, h.sie

Thats good for you and your friends. Wish I could do it. Thanks for the very fine mod. Regards

[h.sie]

@FlyingSub: I only wanted to say that MEPv3 cannot be the (only) issue, since it works for other people.

@Roland: Changing the number of AI-Subs makes no sense. What is your intention behind that? All is possible. But we must first find someone with assembler skills who is willing to program that. the following link would be a good starting point in case you are interested:

http://www.arl.wustl.edu/~lockwood/class/cs306/books/artofasm/toc.html

@fitzcarraldo: You have bad luck. If you are once ordered to shadow, the AI-Subs are installed properly. Congratulation!

The state of the BDU-AI cannot go directly from SHADOWING-state into ATTACK_ALONE-state without a transition state inbetween and informing you about the reason. This state inbetween could be: 1) a wolfpack attack, 2) wolfpack cannot assemble for some reason, 3) wolfpack sunk / damaged by enemy. AFTER that transition state, you will be ordered to attack alone for a couple of days, because no wolfpack is available.

BUT SOMETIMES, this important message, informing you about 1), 2) or 3) is lost. This is a stock sh3 issue, which I didn't change, since I think it is realistic that some messages get lost and the kaleun is confused. Attacks, coordination and communication were chaotic, if I am informed correctly. Please be patient. Try again.

[h.sie]

Customizing the strength of the wolfpack attack.

If you want to reduce the power of impact of the wolfpack attack, because you think the amount of ships sunk is unrealistic and too high, you can try to do the following (not tested yet):

• Open the .sim file of the Type VIIF AI-sub and reduce the amount of AP ammo.
• Use a lower crew competence (veteran instead of elite) for the VIIF

DO NOT:

• Change the number of scripted AI-subs
• Change the sensor settings of the AI-Subs
• Change the range setting in the .sim file of the AI-Subs
• Change any other setting in the .sim file except for the AP ammo.

[urfisch]

Originally Posted by h.sie

BUT SOMETIMES, this important message, informing you about 1), 2) or 3) is lost. This is a stock sh3 issue, which I didn't change, since I think it is realistic that some messages get lost and the kaleun is confused. Attacks, coordination and communication were chaotic, if I am informed correctly. Please be patient. Try again.

later in the war, you are right. as to the high amount of air attacks all boats stayed a lot of time under water, without beeing able to send a radiosignal/homesignal. so coordinating a wolfpack attack was not really possible later in the war. after 1943, when the allies used better technologies, like radar or hf/df to detect german radiosignals, the boats tried not to send more, than neccessary and combined attacks got extremly hard to set up.

but normally communication was fine. every bdu message for the boats was send in a routine, which repeated the message all the day through. so the boats could catch their signals, even they might have been a lot of time under water.

a lot of info is from the book "feindfahrten - erinnerungen eines ubootfunkers, written by hirschfeld".

[h.sie]

@THOSE WHO MADE NUMEROUS (MORE THAN 20) TESTS:

What is your experience regarding the lost messages? Does it happen often to you that important messages (wolfpack assembled / wolfpack cannot reach convoy in time / wolfpack damaged or sunk) got lost???

If you decide to answer, please tell me the approx. number of tests you made. This is important, because we are talking about a statistical problem here, and a low number of tests cannot be considered as relevant.

(Background: The BDU response to your, say, 5th contact report will get lost, if you send the 6th contact report BEFORE you have received the response to your 5th report. The chance of lost messages could be reduced by lowering the response-time (time between contact report and BDU response)).

[makman94]

Originally Posted by Flyingsub

After a mornings testing I can confirm Mep 3 is the issue with the mission in NYGM. I also tried MEP 2.1(16km) and LSH3 MEP (16km) with negative results. In all cases the the other U-Boat shows up on the map but can't get the orders. I have not tried MEP in campaign mode yet.Regards

hello Flyingsub,

MEP (it doesn't matter what version) or ANY environment mod can't be an issue for the wolfpacks mod . the env mod files are totally indepedent and are not effecting the wolfpack 's critical procedure by any way that i can think.
of course this game is full of 'surprises' so if ,indeed, it is an env issue then ,surely , this will be a big...'wow' !

i believe that you were just ...unlucky or you didn't do something correct at procedure becuase it works also on NYGM (at least on my pc)

bye

[Flyingsub]

Originally Posted by makman94

hello Flyingsub,

MEP (it doesn't matter what version) or ANY environment mod can't be an issue for the wolfpacks mod . the env mod files are totally indepedent and are not effecting the wolfpack 's critical procedure by any way that i can think.
of course this game is full of 'surprises' so if ,indeed, it is an env issue then ,surely , this will be a big...'wow' !

i believe that you were just ...unlucky or you didn't do something correct at procedure becuase it works also on NYGM (at least on my pc)

bye

@ Makman, Thanks for the input. I Agree that MEP should not affect the mod unless its some issue with scene.dat and NYGM being sensitive to it's change in my install ,which is unlikely. Regards

[Dani]

Tested with NYGM 3.5a. Total of seven tests.

1. Attack alone response
2. Attack alone response
3. Attack alone response
4. 3-4 shadowing responses , then fog problem for the pack
5. 3-4 shadowing responses , then fuel problem for the pack
6. Attack alone response
7. 5 shadowing responses, between the first and the second one 3 or 4 (not sure) unanswered reports. Between the forth and fifth 2 unanswered.
The pack attacked, sunk 4 ships(1 escort) and damaged 2 more.

[h.sie]

thanks dani. very helpful. i am astonished about the high amount of attack-alone orders, since i programmed a chance of 20% for them for that specific mission date. but according to probability calculation, your results are not unlikely.

you conclusion about lost messages ?

[Dani]

I forgot to say that I was playing with less contact reports checked. From first contact report to the attack 11-12 hours passed. So I'm guessing at least 1 day of shadowing with the harder version is required.

I don't know for sure,yet. The lost messages are giving the chase a "realistic" taste and should stay in the patch. Maybe reducing the response time just a little bit.
I like the extra feel it gives to the shadowing, the anticipation, will BDU response? Do they got my message? What are the orders?
I like it.
(I probably will regret these words when facing later in war, increasing number of aggressive escorts with radar.)

[fitzcarraldo]

Originally Posted by h.sie

thanks dani. very helpful. i am astonished about the high amount of attack-alone orders, since i programmed a chance of 20% for them for that specific mission date. but according to probability calculation, your results are not unlikely.

you conclusion about lost messages ?

I continue to have high amount of attack-alone orders. I made 23 tests with Wolfpack Mission, in only seven I obtain the order of shadowing, and four with pack attacks.

Regards.

Fitzcarraldo

[h.sie]

@Dani: Thanks for your helpful comments. I am currently thinking about making the response time shorter (-> less lost messages) for the time until end of 1942 and using the standard response time (more lost messages) for 1943 and later.

-> in accordance to urfisch's input:

• early war: Uboats have no problem relaying messages

• late war: Uboats forced to dive more often - messages can get lost.

[h.sie]

@fitzcarraldo: This shows that the mod works in principle, but I am astonished about the high amount of "attack-alone" orders.

I shall try to find the reason for that behaviour.

Thanks for your patience.

[Stiebler]

I have tested an uncounted number of wolfpack requests (probably around 50) in campaign mode with NYGM 3.5A.

The testing has been south of Iceland or near the Canary Islands, mostly in late 1942, some in late 1940, early 1941, early 1943.

As expected, there are very few wolf-pack attacks in late 1940/early 1941. The player is ordered to attack alone.

However, in late 1942, it seems that a wolf-pack is ordered to attack nearly every time that the weather is good (no fog/light fog and winds <8 m/s). However, if the weather is worse, wolf-pack attacks are rare.

In early 1943, the number of wolf-packs ordered to attack seems to be fewer than in late 1942. (But not much tested.)

In particular, in all cases, if first contact with the convoy is reported late in the afternoon (close to darkness), then the player is always (? - in my experience) ordered to attack alone. Contact reports do not seem to be required during the night in this circumstance.

I am using the 'less contact reports' option for testing purposes.

Most of the above has been in connection with testing of an upgraded version of my Surrender-Smoke-CAMship overlay patch for H.sie's V16A3.

Stiebler.

[h.sie]

Hi Stiebler,

thanks for your numerous testing and the report. All seems to work as expected, except:

Originally Posted by Stiebler

In particular, in all cases, if first contact with the convoy is reported late in the afternoon (close to darkness), then the player is always (? - in my experience) ordered to attack alone.

This behaviour is not intended and I have not (knowingly) programmed it. But such a behaviour is of course possible (but unlikely) due to the usage of random numbers. ATM I cannot say whether this behaviour is really only the result of (bad) random numbers, OR, the result of a methodical error in the algorithm of the BDU-AI code.

I shall look into it.