SUBSIM Radio Room Forums

SUBSIM Radio Room Forums (https://www.subsim.com/radioroom/index.php)
-   Silent Hunter III (https://www.subsim.com/radioroom/forumdisplay.php?f=182)
-   -   Please give suggestions for Realism (https://www.subsim.com/radioroom/showthread.php?t=149092)

Otto Heinzmeir 03-09-09 02:56 AM

Quote:

Originally Posted by RoaldLarsen

Does anybody know what factor constrains the longest range at which a target can be seen?

I don't use a 16k environment mod because of my ancient graphics card. If I reduce the Visual range factor to 0.4, will the maximum range at which I can detect aircraft be increased, or will I just have a higher chance of seeing an aircraft which is far away but still within some maximum range that is established elsewhere?

I would think this would be the Visual range factor setting. I havn't tried modding this in a 8k environment. I suspect that in an 8K environment the max sighting range is 8k. I haven't playing in an 8k setting in awhile. IIRC correctly the furthest away I would see a ship is about 7600m. Lowering the Visual range factor may give you 300 or 400 extra meters. Probably would not have any effect plane sightings. I am curious about the visual enemy speed setting. Since planes are fast changing this may be of some help. Though not sure which way to go with it. I suspect in the stats I list if I increased it from 0.2 to 0.3 I might have a better chance to sight planes sooner.

You might try testing it out in a game that you aren't playing as a career.


;Visual.
Visual range factor=1.0 ;[>=0]
Visual fog factor=1.1 ;[>=0]
Visual light factor=5.0 ;[>=0]
Visual waves factor=0.8 ;[>=0]
Visual speed factor=0 ;[>=0]
Visual aspect=0.9 ;[>=0]
Visual enemy speed=0.2 ;[>=0]
Visual noise factor=0 ;[>=0]
Visual sensor height factor=0.4 ;[>=0]
Visual already tracking modifier=600 ;[detection probability modifier], most accurate, once a contact is detected it will lose it very hard
Visual decay time=200 ;[>0] already tracking bonus decay, in seconds
Visual uses crew efficiency=true ;[true or false]

I like the idea about having to abandon your external torpedoes.

Otto Heinzmeir 03-09-09 03:25 AM

Quote:

Originally Posted by Oneshot/Onekill
@Otto Heinzmeir. You can also try this sometime, its what i do. If i recieve a contact on my map and its more than 100km away as a rule of thumb i do not attempt to track it down, UNLESS its on a general course heading to allow me an easy intercept. ie using very little fuel.

I just had something similar to this in my current mission. Was tracking a merchant on Hydro but he was going away and the contact was faint. Probably 12k to 15k. My top speed in rough seas at the time was 10kts so I would have used tons of fuel and maybe not even have caught him if he as doing 10kts. Then I get a radio contact from only 17k away, moving almost directly towards me. So I broke off my pursuit of the one going away and took the one coming towards me. Which ended up being 12 tons.

Murr44 03-09-09 11:55 AM

Sure the number of ship contacts & sightings is greater than would happen in RL but if we only saw as many ships as RL U-boat commanders did I think that a lot of us (myself included) would quickly lose interest in the game.

Sailor Steve 03-09-09 11:57 AM

But others of us would like it even more. I think it needs to be one of the realism options.

RoaldLarsen 03-09-09 12:08 PM

Quote:

Originally Posted by Otto Heinzmeir
Quote:

Originally Posted by RoaldLarsen

Does anybody know what factor constrains the longest range at which a target can be seen?

I don't use a 16k environment mod because of my ancient graphics card. If I reduce the Visual range factor to 0.4, will the maximum range at which I can detect aircraft be increased, or will I just have a higher chance of seeing an aircraft which is far away but still within some maximum range that is established elsewhere?

I would think this would be the Visual range factor setting. I havn't tried modding this in a 8k environment. I suspect that in an 8K environment the max sighting range is 8k. I haven't playing in an 8k setting in awhile. IIRC correctly the furthest away I would see a ship is about 7600m. Lowering the Visual range factor may give you 300 or 400 extra meters. Probably would not have any effect plane sightings. I am curious about the visual enemy speed setting. Since planes are fast changing this may be of some help. Though not sure which way to go with it. I suspect in the stats I list if I increased it from 0.2 to 0.3 I might have a better chance to sight planes sooner.

You might try testing it out in a game that you aren't playing as a career.
...

I agree with the bolded parts. Each of the factors affects the chance of detecting the aircraft each cycle. The range factor doesn't change the maximum visual range, it just determines how much effect the range to target has on your chance of seeing the aircraft. That's why I think the absolute limit must be defined somewhere else, such as the environment limit.

Unfortunately, I don't know how to mod that, and I'm not going to move to an existing 16K mod w/o first buying a new graphics card, which I won't do without buying a whole new computer to go with it. That's at least a year away. In turn that really limits my ability to play type IX boats after 1942. I have started 9 patrols in type IX boats in 1943. Six resulted in the sub being lost. In the other three the sub made it back to port with severe damage. All due to aircraft. I think that's just about double the appropriate power of the air forces.

I will definitely experiment with tweaking the enemy speed factor, as you suggest, and possibly reducing the chance of air attack as suggested in an earlier post.

If I understand how the air part of the game works, there aren't actually any aircraft out there flying patrol routes, the way surface ships do. Rather, the game caculates the chance of you being attacked by aircraft and when that chance is hit, spawns aircraft some random distance and bearing from your sub already heading to attack you. The problem with this is that it means just about all enemy aircraft sightings end with an attack by the aircraft. That is not how it hapened in real life. U-boats would often detect the aircraft and dive without the aircraft ever seeing the u-boat. I wish there was a better way to simulate this in the game. The amount of aircraft sightings is about right, but the number of aircraft attacks is too high.

Perhaps one approach would be to reduce the maximum speed of aircraft, so it took them more time to get to the sub. A downside would be that this would make it easier to shoot them down.

Otto Heinzmeir 03-09-09 01:01 PM

Yea the setting to actually expand the 8k limit may be in perhaps a dat file, or some other type that can't be opened by a text editor. There is a program called 3sd that can open and edit the other files. In the 16x environment mod I use, the files seam to all pertain to scenery. I don't reallly know which file it could be or if you can have a 16k viewing radious without a 16k environment which is what you need at present. We need to get the modding experts in here. :DL

The idea about reducing the planes speed would work I bet. Just don't shoot at them I guess. I notice they also have size parameters. You could half there speed and half there size to make to make them smaller targets to compensate. I'm pretty sure that they would still appear the same in game and just be harder to hit. Maybe halving there size would then make then 2 hard to hit.

This sounds like a lot of work. I'm going to take a nap:up:

Oneshot/Onekill 03-09-09 01:30 PM

Quote:

Originally Posted by RoaldLarsen
Quote:

Originally Posted by Oneshot/Onekill
Personally i dont think there is an abnormal amount of ships in the game. You have to understand that most of the time that contact was made whether by hydrophone or visual, there were many factors that had to be accounted for before a U-boat commander could even think about making an attack or attempting an intercept. Here is just a small sample of what went into the equation.

(sea conditions)would affect a U-boats surface speed and thus their ability to overtake or intercept a target. Not accuratley moddeled into the game IMO.

(time of day) If during daylight hours had a much higher probability of being spotted on the surface.

(Fuel remaining)depending on how far away you are patrolling from your base had a dramatic influence and impact on how far away a Kaptain was willing to track distant contacts, especiallly if you were already mid patrol!

these are just a few, there are literally dozens of other factors that came into play.

You have to understand that in reality, during a typical patrol U-boat Kaptaind did recieve quite a bit of information about ship movements or convoy traffic its just that more than 50% of the time they were not in a favourable positon to prosicute these reports!

I'll have to respectfully disagree, though admit I do so on the basis of limited research.

All the factors you list are modelled by the game, though perhaps they don't have quite as much effect as they did in real life. In the game, sea state affects a U-boat's speed, light conditions affect chance of being visually spotted and fuel affects range.

I do not disagree very much about the number of radio contact reports. The number might be a bit high, but German naval SigInt (XB-Dienst?) and reconnaisance from other boats and aircraft resulted in a lot of radio contact reports. In fact I don't like mods that remove the colour of contact reports from the map, because in reality, those contacts would often be reported as enemy or friendly.

You seem to be arguing that in real life, u-boats saw as many ships as we do in-game, but they sank a much lower proportion. I take the position that they made fewer sightings than we do. Perhaps they also sank fewer per sighting. I have no opinion on that.

As for the number of sightings, I have read a couple KTB's and a few books. The number of sightings per day on my in-game patrols is higher than what I have read was actual experience. Also, while the number of convoys spawned in the campaign may be similar to historical, in real life the allies were using Ultra to route convoys around u-boats, so fewer were seen than is the case in-game.

You missundertood me. I was only referring to U-boat surface speed in high waves as not being realistically modeled in the game, although GWX did a very good job trying to. For example i can go to flank speed in 10+ meter per second winds and still maintain 12-14 knots. In reality waves that high at that speed would beat you to death and your watch wouldnt stand a chance.

Again i think i didnt clarify or maybe i miss spoke, but when talking about # of contacts spotted during a patrol, i was more reffering to other U-boats, aircraft, surface vessels, and radio intercepts. Not personally spotting them yourself!

RoaldLarsen 03-09-09 02:36 PM

Quote:

Originally Posted by Otto Heinzmeir
Yea the setting to actually expand the 8k limit may be in perhaps a dat file, or some other type that can't be opened by a text editor. There is a program called 3sd that can open and edit the other files. In the 16x environment mod I use, the files seam to all pertain to scenery. I don't reallly know which file it could be or if you can have a 16k viewing radious without a 16k environment which is what you need at present. We need to get the modding experts in here. :DL

Let's hope one of them notices this thread and offers some info.

Quote:

Originally Posted by Otto Heinzmeir
The idea about reducing the planes speed would work I bet. Just don't shoot at them I guess. I notice they also have size parameters. You could half there speed and half there size to make to make them smaller targets to compensate. I'm pretty sure that they would still appear the same in game and just be harder to hit. Maybe halving there size would then make then 2 hard to hit.

This sounds like a lot of work. I'm going to take a nap:up:

Is the chance to hit a target proportional to its radius or to its area? Is the chance to hit a target inversely proportional to speed or inversely proportional to the square of the speed? Intuitively, I'd say that target area not radius, matters, and that doubling the perpendicular speed of something reduces the chance to hit it by more than half, but doubling the speed of something coming straight at you doesn't affect your accuracy very much at all. However, that is for a single shot - for continuous firing the amount of time you can shoot at the target directly affects your chance of hitting, so multiply the effects by 1 over speed. So, what I might try is to reduce the speed of aircraft by about 30% (that's what I guess I need to give my subs a better chance of diving before an attack), assume this increases my chance to hit them by about 117% (about halfway between (1/0.7)cubed and (1/0.7) and then decrease the dimensions by about 1/3 to compensate (square root of 1/2.17 is about 0.67). The only question is: does changing the MaxSpeed, Length and Width parameters in data\Air\<aircraftname>.cfg actually change in-game performance, or does it just change data displayed in the museum? I suppose an easy way to find out would be to build a test that changes these values to extremes and increases the chance of an airstrike. If an aircraft has an effective length and width of 200 metres and min and max speeds of 30knots, I think we would notice.

You're right, it does sound like a lot of work. Before I do that I'm going to try the following:

Keep a log of all occaisions I surface a boat, to see how long I run before an aircraft shows up, and whether I am attacked. Then, try tweaking the Sensors.cfg file to decrease Visual range to 0.4 from 0.5, increasing the Enemy speed factor to 0.3 from 0.2 (but setting use of crew efficiency to true), and then, in AirStrke.cfg, reducing Default Air Strike Probability from 10 to 7 for 1943 and later, and log the results and compare.

Otto Heinzmeir 03-09-09 02:57 PM

Quote:

Originally Posted by RoaldLarsen
Keep a log of all occaisions I surface a boat, to see how long I run before an aircraft shows up, and whether I am attacked. Then, try tweaking the Sensors.cfg file to decrease Visual range to 0.4 from 0.5, increasing the Enemy speed factor to 0.3 from 0.2 (but setting use of crew efficiency to true), and then, in AirStrke.cfg, reducing Default Air Strike Probability from 10 to 7 for 1943 and later, and log the results and compare.

The thing to watch out for is not to change too many parameters all at once or you won't know which one is having the greatest impact. But sounds like you have done these types of testing before. I might suggest making a save in an area of high concentration of allied planes. Setting the airstrike probabilty high. Is 100 max? Then change just one of the values, but go overboard. For example change enemy speed factor from .3 to 1.0 just to see if anything changes dramatically or maybe its not noticable. If it is dramatic but in the wrong direction then try 0.1

The same with the other factors. You can load the same save game in allied airspace and just change one parameter at a time. When you want to test stuff like this is usually when no planes ever appear :O: Good luck. I'd be interested in how it turns out.

RoaldLarsen 03-09-09 03:27 PM

Quote:

Originally Posted by Otto Heinzmeir
Quote:

Originally Posted by RoaldLarsen
Keep a log of all occaisions I surface a boat, to see how long I run before an aircraft shows up, and whether I am attacked. Then, try tweaking the Sensors.cfg file to decrease Visual range to 0.4 from 0.5, increasing the Enemy speed factor to 0.3 from 0.2 (but setting use of crew efficiency to true), and then, in AirStrke.cfg, reducing Default Air Strike Probability from 10 to 7 for 1943 and later, and log the results and compare.

The thing to watch out for is not to change too many parameters all at once or you won't know which one is having the greatest impact. But sounds like you have done these types of testing before. I might suggest making a save in an area of high concentration of allied planes. Setting the airstrike probabilty high. Is 100 max? Then change just one of the values, but go overboard. For example change enemy speed factor from .3 to 1.0 just to see if anything changes dramatically or maybe its not noticable. If it is dramatic but in the wrong direction then try 0.1

The same with the other factors. You can load the same save game in allied airspace and just change one parameter at a time. When you want to test stuff like this is usually when no planes ever appear :O: Good luck. I'd be interested in how it turns out.

All good advice, thanks.

If I had no idea what the various paramaters did, that's exactly what I'd do. In this case, I have a pretty good working hypothesis of what's going on, and while I am changing four parameters simultaneously, only two will actually interact with each other. My test crews will have 100% efficiency, so that change will have no effect during the tests. The change to Airstrike will affect whether aircraft show up at all, while the other two changes will affect how well my crew spots them when they do show up. Since I am not too concerned about the relative impacts of the two sighting parameters, I'll only test them separately if something unusual is the result.

Also, the sensor changes will affect ability to see ships, too, and I don't want to affect that too much, so I will have to test against surface vessels as well.

gigel_escu 03-09-09 04:33 PM

Of course first thing you can reduce the number of ships or randomize their routes on chart, maybe it is hard (programming I mean). I know a little bit of programming and I think it is very hard or the SH make 10 GB n HDD and took 3 days only for load the mission. It is annoying when I must patrol EG87 an example, and the most convoys and naval traffic are in DS15 or DG47. Of course you can divert you route, you don't take the striaght course on the patrol area (in real most uboats go in straight course, because of sortage of fuel), but its is quite impossible to find a convoy outside of convoys routes. Normally tou can find a struggler or something.

In the other hand, I had read somewhere here that the "damage meter" that percent who indicates the damage level of the sub was deleted because some wise guys do that or do bla bla. That is a stupid idea, I am very angry. In middle of 1942 with VIIC in mission on Venezuelean coast (why so far away) I was damage by a Hudson an Avenger almost dead in the water, not very hard flooding or something. I reload the saved game, I met few Hudson who make damage, flooding (about 5 minutes, the boat went down uncontrollable until 120 m), attack periscope destroyed and few damage not critical. I arrived in Lorient and surprise 76% damage, in conclusion not so hard. I don't know the real damage status, I don't understand what "hull damage" means, my CE officer indications not readeble (I played with English subtitles and German speak). Not important, but I must know about my sub are heavy, medium or light damage. I image Erich Topp asked his chief engineer about damage and chief engineer answered: "Her Topp it is a secret I will tell you in the base". Get serious.

Another stupid thing, in WW2 you hadn't be officer to maned a AA gun on the submarnes. At my type VIIC with type 2 connig tower with 2 AA's I must to put 2 officer at the gun, I put 1 just one guns shots, I put 2 sailor man no gaun shots. That is f.... stupid, who think that guys he or she can try that:damn:.

Otto Heinzmeir 03-09-09 10:43 PM

Quote:

Originally Posted by gigel_escu

In the other hand, I had read somewhere here that the "damage meter" that percent who indicates the damage level of the sub was deleted because some wise guys do that or do bla bla. That is a stupid idea, I am very angry. In middle of 1942 with VIIC in mission on Venezuelean coast (why so far away) I was damage by a Hudson an Avenger almost dead in the water, not very hard flooding or something. I reload the saved game, I met few Hudson who make damage, flooding (about 5 minutes, the boat went down uncontrollable until 120 m), attack periscope destroyed and few damage not critical. I arrived in Lorient and surprise 76% damage, in conclusion not so hard. I don't know the real damage status, I don't understand what "hull damage" means, my CE officer indications not readeble (I played with English subtitles and German speak). Not important, but I must know about my sub are heavy, medium or light damage. I image Erich Topp asked his chief engineer about damage and chief engineer answered: "Her Topp it is a secret I will tell you in the base". Get serious.

If I understand you correctly your playing one of the mods that doesn't show hull integrity? The SH3 Commander program has an option that allows you to show hull intergrity. You just check the option and your %damage will appear again. I like to have the % show because there is no way in game to inspect your damage.

gigel_escu 03-10-09 11:06 AM

And that option is??

Sailor Steve 03-10-09 11:35 AM

'Show/Hide Hull Integrity'.

mookiemookie 03-10-09 12:35 PM

Another suggestion for realism: when reloading external reloads, only do it in fair weather and relatively calm seas at slow speed. No diving while reloading either. In other words, if you see a plane, you're screwed.

gigel_escu 03-10-09 03:18 PM

Quote:

Originally Posted by Sailor Steve
'Show/Hide Hull Integrity'.

Gotcha Einstein, but I don't have the mod:woot:

Sailor Steve 03-10-09 03:50 PM

:rotfl:

Okay, you win. You do need SH3 Commander, for that but also for a dozen other reasons, all of them good. The fact that when you sink a merchant it gives it a name is among them.:rock: Also the ability to adjust many of the parameters in the game with a single mouse click. The ability to edit and write your own patrol reports. Realistic crew transfers. Realistic career length. Malfunctions and sabotage aboard your boat if you want it. Randomized gramophone songs. Randomized loading screens.:rock: Randomized anything you want to add in.

It's the ultimate tool for SH3. Period.

gigel_escu 03-11-09 04:16 PM

Yep. You sayed brother, good point:yeah: Thanks

Schöneboom 03-12-09 11:07 PM

To bump up realism, I ignore single ship reports, and convoy reports, too, when the weather's rotten (it never improves by the time I make contact anyway). In addition, my submerged shots are with manual targeting, the "notepad" version. I usually shoot anywhere from 1000 - 3000 meters. As the miss rate increases, the eels get used up sooner.

Even so, on my latest patrol in 1942, I bagged 8 ships, which is rather a lot for a month's work. I don't expect that kind of tonnage for much longer. Now I gotta admit, if I really want to make it harder, the first thing to do is never return to previous saves when the current attack goes awry. This is distinct from the DID approach, in that I'm not referring to impending doom or actual game death, but just when things don't go according to plan. "No going back" is really key, I think; that means, once the attack starts, don't go back, and don't save again till after you've disengaged (the enemy is out of range). If you're alive, and the boat still floats, keep going forward, never back.

Sailor Steve 03-13-09 12:44 PM

I think GWX reduces the 'Range To Opportunity Targets' in the Main.cfg by a bit, but it can be reduced further. A long time ago there was a mod that reduced the number of single ships by 80% and then reduced the number of reports on them by 80%, making it very unlikely to encounter single ships in the game. Unfortunately the author had to change each individual ship, and with several dozen new ships in the game since then it doesn't work anymore.

Me, I never chase down any single-ship report, on the assumption that they came from an unreal source anyway.


All times are GMT -5. The time now is 06:10 PM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
Copyright © 1995- 2025 Subsim®
"Subsim" is a registered trademark, all rights reserved.