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

[skwasjer]

Originally Posted by Anvart

I already wrote... do not need to buy SH5 for watching of SH3's files.
RB ... oops, sorry TDW, as always, offers a convoluted and irrational way... it's his way of thinking ...
For most modders his sentence sounds very strange (to put it mildly)...
S3D is an excellent tool that solves the vast majority of tasks in SH3... This tool allows you to edit existing and add missing classes and parameters very easy.
This is the best tool in the history of SH.
If TDW does not like S3D and he does not use S3D to view the files, even (which I doubt) ... it's his choice ... may be why some of his mods don't have absolute purity and perfection.

@ skwasjer

S3D excellent tool!
S3D excellent tool!
S3D excellent tool!
S3D excellent tool!
S3D excellent tool!

And that means a lot coming from you. Thank you Anvart!

[Stiebler]

@H.sie:

I'm running through a career in NYGM with your V15G1 installed. Currently in January 1942. No bugs, so far.

However, there are a few issues, the most important first.

1. There is definitely a problem with the usage of the oxygen supply in submerged U-boats. I have made the following timings:
Type IIA: 17% oxygen reached after 5 1/4 hours. Then oxygen compressed supply 50% consumed after a further 15 1/4 hours.
Type VIIB: 17% oxygen after 5 1/2 hours. Oxygen supply 50% consumed after a further 20 hours.
Type IXC: 17% oxygen reached after 6 1/2 hours. Oxygen supply 50% consumed after a further 25 1/2 hours.

Let us consider these figures compared with standing orders in mid-1943 for the Bay of Biscay:
'Remain dived during passage. Surface only long enough to recharge your batteries.' (Affects Types VII and IX.)
The time to recharge batteries depended naturally on how far they had been exhausted. However, 4 hours on the surface would be an above average time for recharging in the Bay of Biscay.

This means that the oxygen supply of a VII U-boat in 1943 would involve 4 hours (or less) on the surface, 5 1/2 hours consuming natural oxygen in the U-boat, and then 14 1/2 hours using oxygen from the limited compressed air supply in V15G1, which could not be replaced. Thus, after just three days (maximum), all the compressed air reserve would be used up and the U-boat would have to return to base before it had even left the Bay of Biscay!

Therefore, there is definitely an error in the algorithm in V15G1. I do not know what the error is, but the oxygen supply does not comply with actual standing orders for U-boat submerged-passage across the Bay of Biscay in mid 1943.

Perhaps you have under-estimated the amount of compressed air cylinders, carried aboard a U-boat, or the oxygen content of each cylinder. Or maybe the natural air in the U-boat lasts longer than you think. Or maybe air could be partly compressed for release in the boat later - the U-boat used air compressors to drive out the water when surfacing, so the technology certainly was present.

This problem of the oxygen supply has other effects in the game too:
1. It is impossible to stay submerged for more than a few days in any patrol area which is close to aircraft (such as off the British coast). The air is exhausted too quickly.
2. Submerged attacks on a convoy can result in necessary use of the compressed air tanks in V15G1 before the U-boat can escape, even if it is not detected.
3. Submergence to avoid heavy fog results also in rapid consumption of the compressed air supply in V15G1.

I thought the earlier model of air supply in the earlier versions V15E was about right.

2. Problem with BdU messages.
The bands for the tonnage responses seem to be incorrect. For example:
a) Sending a radio status message to BdU with 0-10,000 tons sunk results in the message 'Don't sink neutral ships, amateur!' This message should be sent with negative tonnage sunk. Or, with 7,000 tons, the message is sometimes 'Show more courage - charge at the enemy!', which should be for very low tonnage sunk.
b) The message 'Return to base as directed' should be given a reason (ie, 'due to low fuel', 'due to heavy damage', 'out of torpedoes'), which would require three entries in en_menu.txt to replace the original one message. Actually, the 'out of torpedoes' message is useless, the player must already know that, so it would be more encouraging to have a message of congratulations instead.

These problems are easy to fix, I think. But the oxygen supply probably requires further discussion in SubSim.

However, no other problems seen, so far!

Stiebler.

[Stiebler]

In connection with the discussion of wolf-packs:

1. It may indeed be possible to modify aircraft from an airbase to behave like AI U-boats. However, I found some years ago that *scripted* aircraft behave like flying merchant ships. They fire guns, but will not carry out any other offensive action.

2. If you can beam-in U-boats close to a convoy's position, then presumably H.sie's code can be modified so that, with each passage of the code, the U-boat comes a little closer to the position of the convoy (or rather, to the position of the player U-boat after it reported position of the convoy). That is a standard algorithm, reduce distance between two latititudes and two longitudes. The player position report is definitely accessible in the SH3 code, I have seen it myself. (But is there a way of finding the AI U-boat's position?)

3. As I've mentioned earlier, you don't actually need the U-boats to fire torpedoes, just strike down merchants randomly by increasing their damage. How will anyone know the difference?

4. A related idea:
If you can beam U-boats close to the player's U-boat, we have a method to refuel a U-boat stranded because of lack of fuel. Code sequence: player U-boat makes signal; (perhaps a message from BdU); pause of 2-3 days; AI U-boat is beamed in; player U-boat's fuel is magically increased. Just as in real life (well, almost).

Stiebler.

[h.sie]

@Stiebler: Thanks for the issues report. I will look into it.

Oxygen: I am very sure it's not a bug in the algorithm itself, it's only a question of quantities (how much time takes O2 to go down to 17%, and how much O2 supply is stored in the sub). There are 2 constant numbers for each sub type that control these times and that could easily be changed / enlarged. A good idea to discuss this topic the in the forum.

BDU-Messages: Will look into it. Easy to fix.

h.sie

[h.sie]

@Stiebler: According to your initial idea, I divided the amount of total tonnage sunk into several "tonnage groups":

0 - 10000 t : group 1
10000 - 20000 t : group 2
20000 - 40000 t : group 3 and so on.

I assigned every group a certain chance for a motivation message, which gets lower for higher groups:

50% for group 1,
25% for group 2,
12,5% for group 3
6,25% for group 4 and so on

Would you suggest to completely disable these motivation messages for other groups than group 1 (0 - 10000 t) ??

[reaper7]

Been doing some test to see if I could get something working on the sub AI side of things.
As it stands the stock sub stops when in range of a merchant and then sits there doing nothing
Setting the aggressive nature to Elite seems to have no effect.
So I set the sub from type 200 (submarine) to type 4 (Destroyer) and moved it to the ship folder as mentioned earlier in the thread - effect was no CTD but sub still stopped and did nothing
So going into the sim file for that uboat (SSTypeIID) i deleted the cmdr_AIHumanSub enteries and created a new cmdr_AIShip entry, now on going to mission the sub changed course straight for the merchants at full speed but didn't last long before it either Dove straight to the bottom and stayed there or was sunk due to merchant fire
Sub did not release torpedoes or fire its flak gun at the merchants.
Is there a link to Sergbuto AI sub, as stock don't fire as mentioned that I can do some more trials with


Does the Torpedo boat fire torpedoes, if so could we clone that one and replace the ship model with a uboat model - It means it wont dive but it can at least manoeuvres and do torpedo runs ?
I'll have a go at that see what the results are

[SquareSteelBar]

Originally Posted by reaper7

...Is there a link to Sergbuto AI sub, as stock don't fire as mentioned that I can do some more trials with

As always: Plissken's ftp server....

Look for WolfpackMod_2.0

[h.sie]

@reaper: Thanks very much for trying to help. On my mediafire page you can find Sergbutos AI Subs reloaded.

[Hitman]

but didn't last long before it either Dove straight to the bottom and stayed there or was sunk due to merchant fire

You mean it actually can take the decission to dive?

Does the Torpedo boat fire torpedoes, if so could we clone that one and replace the ship model with a uboat model - It means it wont dive but it can at least manoeuvres and do torpedo runs ?

IIRC there is no Ai capable of firing torpedoes in SH3 (unlike in SH4, where torpedo ariplanes can and do throw them), but it's one of those things said in the past that could be worth rechecking again

[Wolfstriked]

Kinda embarrassed to ask this....when a player fired torpedo scores a hit on a ship it triggers the explosion and damage.Could this trigger be spawned with no actual vehicle doing the firing?

[reaper7]

Originally Posted by Hitman

You mean it actually can take the decission to dive?

IIRC there is no Ai capable of firing torpedoes in SH3 (unlike in SH4, where torpedo ariplanes can and do throw them), but it's one of those things said in the past that could be worth rechecking again

Not sure if it dove or not, did it on each mission load Could be due to damage from marchants, but sub did not dissapear from bottom.
But this is likely due to a conflict from using AIship and not AIHumanSub but worth investigating more.

Pity we have no AI torp firing units, did TDW get something like this working for SH5 with Planes dropping Torps

[Hitman]

Kinda embarrassed to ask this....when a player fired torpedo scores a hit on a ship it triggers the explosion and damage.Could this trigger be spawned with no actual vehicle doing the firing?

The idea is not bad, but our main objective from our player's point of view should be to have the escorts scatter and pursue for some time other "uboats", even more than creating damage to the convoy. Many times escorts got the uboats before they could fire and simply tried to push them far enough to preclude any shooting oportunity.

[LGN1]

@all: I don't think it's possible to 'beam' a unit inside the rendered area. It only works outside (approx. 36km from the player). Please correct me if I'm wrong, h.sie.

@Stiebler: Without snorkel no u-boat could be able to stay submerged for a few days. Please read the whole discussion we had in this thread. H.sie's numbers are derived/estimated from original documents about O2 supply, O2 consumption,... if you think they are wrong, please provide values that you think are more realistic (and your sources).

I don't know if you are aware of this, but you should use 'silent running' when being submerged. The oxygen will last longer in this case! Concerning the crossing of the Bay of Biscay: Maybe boats just surfaced and recharged the batteries twice a day I can imagine several reasons for this:

- No oxygen supply needed
- In general a higher battery status gives you more safety because if a destroyer shows up you have more power left
- I'm not sure about this, but maybe it's better to be on the surface for shorter times and more often than for longer time and only once

In addition, IIRC, Hitman once mentioned that u-boats equipped with snorkels would snorkel twice a day to recharge the batteries, in the morning hours and evening hours. So, maybe they had the same strategy before the snorkel arrived.

Regards, LGN1

[Wolfstriked]

Originally Posted by Hitman

The idea is not bad, but our main objective from our player's point of view should be to have the escorts scatter and pursue for some time other "uboats", even more than creating damage to the convoy. Many times escorts got the uboats before they could fire and simply tried to push them far enough to preclude any shooting oportunity.

Is a spawned hit possible then???If so the then the sub spawned 3K away....wouldn't the dd track if its not at silent running?

[h.sie]

Regarding O2-Supply / Diving times

The algorithm works as intended. I choose the O2-supply/diving times as follows:

A = Diving time with renewable O2 from air in the sub.
B = Diving time with non-renewable O2 from bottles.
Sum = A+B = Max. total diving time

Type II: A = 5,50 h / B = 36,50 h / Sum = 42,00 h
Type VII: A = 6,25 h / B = 44,00 h / Sum = 50,25 h
Type IX: A = 7,00 h / B = 49,00 h / Sum = 56,00 h
Type XXI: A = 7,75 h / B = 98,00 h / Sum = 105,75 h

These values are for standard crew number and Standard operation mode (no silent-running nor repairs). For silent-running, the O2 consumption is only 70% compared to normal operation, so that we get the following max. total diving times:

Type II: Sum = 60 h
Type VII: Sum = 72 h
Type IX: Sum = 80 h
Type XXI: Sum = 150 h

The 72 h for the Type VII are according the value LGN1 found in an document about an VII Sub (don't remember the exact name). The other values are more or less extrapolated from the Type VII value.

@Stiebler:

1) The diving times I measured weeks ago differ a little bit from your measured values. Could this be related to NYGM/GWX differences?

2) A short surfacing with the type VII to load batteries and get fresh air, and you can dive for another 6 hours without using the O2 supply. If we won't find a consense here, I'll give you the Hex-Adressses of the values determining the times, so that you can adjust them to fit your personal taste.

3) Or, if you don't like the O2-Mod at all, I could provide you the source-code of the V15E version, which you like more. You could use that and build your own exe. This gives the people here an opportunity to choose between 2 different approaches. I, personally, prefer the current O2-supply solution, but I am flexible regarding fine-adjusting the diving-times, if we find some information about that.