SUBSIM Radio Room Forums

SUBSIM Radio Room Forums (https://www.subsim.com/radioroom/index.php)
-   SHIII Mods Workshop (https://www.subsim.com/radioroom/forumdisplay.php?f=195)
-   -   IMPOSSABLE to escape DD's (https://www.subsim.com/radioroom/showthread.php?t=112152)

Umfuld 04-18-07 10:34 AM

About DDs knowing where you shot from, yeah, that's something I've always noticed. But two convoys - 5 in close approaches so far in 1.03 and each time I've slipped away without any hint of detection.


Also, something I saw long ago. Not sure if it was stock SHIII, or maybe the first release of GW, but it blew my mind.
I hit a merch in a convoy. Good weather, fairly long shot (I'm getting bolder now, but back in the day I shot everything from a distance - it can be done effectively, darn it). And I swear the escort steamed over to the ship I hit as if it was inspecting the damage to see where the torp came from.

Anyone else ever see something like that?

gabeeg 04-18-07 06:26 PM

I agree that they do know pretty much where the torp came from. The worst example I can think of was when I fired a LUT and missed, it turned back around and hit the opposite side of the ship (away from me)...but to my surprise the escort and merch searchlights were pointed my way and I had to go evasive right away...was hoping they would head the opposite direction and I could get off a reload or two :(

To me the must frustrating escort cheat is that they always seem to know exactly what depth I am at when dropping DC's....no matter how radical my changes. Aircraft on the other hand do not exhibit this behavior and always seem to set their charges pretty shallow.

fpg909 04-18-07 11:14 PM

Manually working the hydrophone is a huge help for me. Your sonar man will generally tell you which bearing is closest but that is about it. By using the hydrophone yourself you can hear the charges hit the water and what bearing they were dropped helping you to isolate which DD dropped them. Using this information will tell you exactly when you can break silence and speed up, perform an evasive turn, which way to turn, and when to dive deep to get under them. Generally ill start at 150 meters and work my way deeper ascending and descending as necessary to keep them guessing as to my actual depth.

CaptainAsh 04-19-07 08:59 AM

From my experience in 41-42, if you aren t able to dive at at least 160m and if the escort succeed once in having an accurate ping possition, you re dead. Once they know exactly were you are, if you can t go to more than 140m you re just a sitting duck :). If they don t succeed in locking their ping on you, between 80 and 100m is good enough...

sandbag69 04-19-07 02:55 PM

Quote:

Originally Posted by CaptainAsh
From my experience in 41-42, if you aren t able to dive at at least 160m and if the escort succeed once in having an accurate ping possition, you re dead. Once they know exactly were you are, if you can t go to more than 140m you re just a sitting duck :). If they don t succeed in locking their ping on you, between 80 and 100m is good enough...

Yeah liek that was realistic in 41 -42 :down:

GWX has been uber'd up to much.

Spytrx 04-19-07 03:31 PM

I finally figured out why by doing some of my own research into the sonar capabilities of the Allied Forces during WW2

For a start, the area covered was a cone heading away from the ship - the method used in GWX is that of a sphere... Basically what this means is that the DD's which have their sonars at the front quater of the ship with their tight turning circle will never loose a boot once it is in their range (and the range is far greater than historical data shows to be capable of, even under ideal test conditions) since the sphere of detection starts off at a right angle to the ship and extends from there. R/l however was flawed with blindspots and little nags like the limited angle (between 25-60 degrees) off the sonar which then extended, giving the U-boots a fair crack at hiding right underneath or at the side of the DD looking for them (if they so wished and the ship trawled that slow). Other factors like speed and noise of the ship and sea condition as well as the sea floor were also important factors in this matter - all of which have been left out (as well as the vanilla version, but GWX is said to have improved on that) - basically it is lazy mans maths to blame: The method used is far easier to implement and calculate then the real model

Kpt. Lehmann 04-19-07 05:27 PM

Quote:

Originally Posted by Spytrx
I finally figured out why by doing some of my own research into the sonar capabilities of the Allied Forces during WW2

For a start, the area covered was a cone heading away from the ship - the method used in GWX is that of a sphere... Basically what this means is that the DD's which have their sonars at the front quater of the ship with their tight turning circle will never loose a boot once it is in their range (and the range is far greater than historical data shows to be capable of, even under ideal test conditions) since the sphere of detection starts off at a right angle to the ship and extends from there. R/l however was flawed with blindspots and little nags like the limited angle (between 25-60 degrees) off the sonar which then extended, giving the U-boots a fair crack at hiding right underneath or at the side of the DD looking for them (if they so wished and the ship trawled that slow). Other factors like speed and noise of the ship and sea condition as well as the sea floor were also important factors in this matter - all of which have been left out (as well as the vanilla version, but GWX is said to have improved on that) - basically it is lazy mans maths to blame: The method used is far easier to implement and calculate then the real model

Sandbag and Spytrx, you are both flatly incorrect. Read the entries on this link closely: http://www.subsim.com/radioroom/showthread.php?t=104377

The speed of the enemy hunter and of the hunted was indeed modelled and tested extensively over months. Furthermore, limitations of SH3 sensor capability prevent one from simply adding in real-life historical data... and expecting that a logical outcome be produced in simulation.

Reasonable blind spots and sea state are also modelled.

Reading the above article will increase your understanding of what you feel to be "uber."

From stock SH3 to GWX... yes as a player you must adapt to harsher surroundings and always remember that most of the U-boatmen who went to sea, did not return. Indeed many crews were sunk on their first patrol.

Something that we can never remove from the player as an advantage... is the ability to gain experience as a result of simulated destruction. In real life they did not have that advantage... it was permanent. You do have that advantage... and will need to adapt to a more realistic environment to be successful in GWX.

Spytrx, your analogy that the enemy AI adjustments were simply "lazy man's math" is entirely incorrect and uninformed. You weren't a part of the months of work that went into getting rid of enemies that were dumber than dirt.

If you want arcade style play... I suggest you uninstall GWX. As a develoment team, our hunt is about the attempt to produce a better realistic environment and simulated behavior as best as is possible... be it code, graphics, or sound.

sandbag69 04-19-07 08:12 PM

Quote:

Originally Posted by Kpt. Lehmann
always remember that most of the U-boatmen who went to sea, did not return. Indeed many crews were sunk on their first patrol.

.


You are thinking of the whole of WW2. up to April 1943 the U-Boat losses were not that great. In fact they were pretty low. After April 1943 they were slaughtered.

There is definately something wrong with the AI in GWX.

Reece 04-19-07 08:16 PM

What you say is correct Kap, but you should have options for those who don't want full 100% of your Ideals or realism, it is after all a game to enjoy and 100% realism or harder is not everyones "cup of tea", would be nice to see on next version, if there is one, some options to suit all players.:yep: This would keep everyone happy & you wouldn't have to keep defending your ideals for the game play.:up: No offence meant.

sandbag69 04-19-07 08:20 PM

Quote:

Originally Posted by Kpt. Lehmann
Quote:

Originally Posted by Spytrx
I finally figured out why by doing some of my own research into the sonar capabilities of the Allied Forces during WW2

For a start, the area covered was a cone heading away from the ship - the method used in GWX is that of a sphere... Basically what this means is that the DD's which have their sonars at the front quater of the ship with their tight turning circle will never loose a boot once it is in their range (and the range is far greater than historical data shows to be capable of, even under ideal test conditions) since the sphere of detection starts off at a right angle to the ship and extends from there. R/l however was flawed with blindspots and little nags like the limited angle (between 25-60 degrees) off the sonar which then extended, giving the U-boots a fair crack at hiding right underneath or at the side of the DD looking for them (if they so wished and the ship trawled that slow). Other factors like speed and noise of the ship and sea condition as well as the sea floor were also important factors in this matter - all of which have been left out (as well as the vanilla version, but GWX is said to have improved on that) - basically it is lazy mans maths to blame: The method used is far easier to implement and calculate then the real model

Sandbag and Spytrx, you are both flatly incorrect. Read the entries on this link closely: http://www.subsim.com/radioroom/showthread.php?t=104377

The speed of the enemy hunter and of the hunted was indeed modelled and tested extensively over months. Furthermore, limitations of SH3 sensor capability prevent one from simply adding in real-life historical data... and expecting that a logical outcome be produced in simulation.

Reasonable blind spots and sea state are also modelled.

Reading the above article will increase your understanding of what you feel to be "uber."

From stock SH3 to GWX... yes as a player you must adapt to harsher surroundings and always remember that most of the U-boatmen who went to sea, did not return. Indeed many crews were sunk on their first patrol.

Something that we can never remove from the player as an advantage... is the ability to gain experience as a result of simulated destruction. In real life they did not have that advantage... it was permanent. You do have that advantage... and will need to adapt to a more realistic environment to be successful in GWX.

Spytrx, your analogy that the enemy AI adjustments were simply "lazy man's math" is entirely incorrect and uninformed. You weren't a part of the months of work that went into getting rid of enemies that were dumber than dirt.

If you want arcade style play... I suggest you uninstall GWX. As a develoment team, our hunt is about the attempt to produce a better realistic environment and simulated behavior as best as is possible... be it code, graphics, or sound.

I will keep perceivering with my mission in GWX.

Dont understand why my watchmen can't see a convoy taht I can clearly see through the bins.
Has the range that the Watchcrew see been reduced?

Umfuld 04-19-07 10:24 PM

Quote:

Originally Posted by sandbag69
Dont understand why my watchmen can't see a convoy taht I can clearly see through the bins.
Has the range that the Watchcrew see been reduced?

I just loaded up the 'GWX Stay Alert crew fix v1c'. Can't say how well it works yet but it's meant to address this issue.

I'd link it but not sure where I got it from. (I just d/l it today. Heh. Short term memory. I remember what that was like.)

Reece 04-20-07 03:15 AM

Well for a test I tried with a "65meters under keel", waited for a good target to the rear, waiting at periscope depth, all silent, scope down checking hydraphones for most to pass, then raised scope, setup final solution (speed & course calculated before the escort was even close), open tube 1, fired, open tube 2, fired!:yep: Set depth to 67 meters and 2 knots still silent, this time I headed off at 25 degrees to Port! Both Torpedo's hit & 1 destroyer (last in group), headed for where I launched the torpedos, this time he didn't even get near to me & I got away without detection!:up:
Whew, good Rush!!:D

Kaleun Cook 04-20-07 03:54 AM

I'm only in 1941 so I can't say anything about the later war AI. I'm playing since a few weeks so I guess I'm still rather a noob, playing with 80 % realism (can't aim manually).

I've experienced it the other way around. I played one patrol in vanilla sh3 and tried to attack a convoy. I had no chance to get near to it - the escorts spotted me far off each time allthough it was deep night. In GWX it worked better for me.

Lately it happened two times, that I set an interception course and then left the PC to do something else. By the time I got back, I had accidently sailed in front of the convoy - still the escorts didn't see me. Hung up at the peri and infiltrated the columns successfully both times.

When it came to withdrawal it worked out for me what's said here over and over again: deep and slow - 180 meters or deeper and silent running, the course set through the convoy to make it harder for the escorts to follow the boat. One can escape in shallow water the way reece did but it's of course more likely to be spotted. And even if the escorts run out of DC they might follow you for several miles waiting for you to surface and calling in each destroyer on the way. Actually I never got away once caught in shallow water. :shifty:

Spytrx 04-20-07 04:04 AM

Quote:

Originally Posted by Kpt. Lehmann
Spytrx, your analogy that the enemy AI adjustments were simply "lazy man's math" is entirely incorrect and uninformed. You weren't a part of the months of work that went into getting rid of enemies that were dumber than dirt.

The "lazy man's math" is a quote taken from the Ubi chief designer about the sensor model used in SH3 and applied here as that, a quote - if you start off with the wrong set of values and formulas given by the game and partly hardcoded into it you will not be able to change much. Furthermore the cone model is far more complex to claculate and model and requires a very good understanding of physics, particularly since half the data is still classified and needs to be worked out by the programmer themselves...

You can take offense in criticism all you like (and it seems to me you do even though there is none intended) - 1939/40 were not the years that DD's were so effective to kill every U-boot they encountered (which was my point right from the onset (just on the by since I am asked to read something clearly over and over) and even later when the DD's became more sophisticated and the crews were better trained/prepared the wolfes changed their tactics and hunted in packs.

I have taken off GWX days ago so simply because there are other Mods out there that go for the realism flavour without hitting you with a sledgehammer the first time you get into the game and if I wanted to play arcade games, I would, thank you. There is nothing wrong with GWX if you like that sort of thing - just don't advertise it for beginners as something better than the vanilla!

down and out 04-20-07 05:10 AM

Quote:

Originally Posted by sandbag69
Quote:

Originally Posted by Kpt. Lehmann
always remember that most of the U-boatmen who went to sea, did not return. Indeed many crews were sunk on their first patrol.

.


You are thinking of the whole of WW2. up to April 1943 the U-Boat losses were not that great. In fact they were pretty low. After April 1943 they were slaughtered.

There is definately something wrong with the AI in GWX.

9 sunk in 3 months of 1939 out of 57 total boats
23 U-boats lost in 1940 (643 men died and 331 men survived those losses)
34 U-boats lost in 1941 (887 men died and 645 men survived those losses)

The number of boats built during that time
1939 (18)
1940 (50)
1941 (199)


84 U-boats lost in 1942 (3221 men died and 818 men survived those losses)
235 U-boats lost in 1943 (10053 men died and 1842 men survived those losses)
215 U-boats lost in 1944 (7976 men died and 2349 men survived those losses)
124 U-boats lost in 1945 (3963 men died and 396 men survived those losses)

Number of boats built during that time
1942 (237)
1943 (284)
1944 (229)
1945 (91)


Aside from the fact the allies got better at hunting uboats technical advances such as radar in 42 made it easier as well plus more uboats at sea will also lead to more sinkings
Green crews late war didnt help either

I think that SH3 in general and GWX in particular mirror RL

The skippers with patience and understanding are succesful and return from patrols

We are lucky in that we can reload and learn from our mistakes
Sadly for 30,000 that wasnt the case


All times are GMT -5. The time now is 12:26 AM.

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.