Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Please post here for questions and discussion about scenario design and the game editor for WITP.

Moderators: wdolson, Don Bowen, mogami

User avatar
JWE
Posts: 5039
Joined: Tue Jul 19, 2005 5:02 pm

Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by JWE »

The “Name” of the device is not particularly relevant. It is just a text string that the program plugs into the weapon suite on the game’s ship screen. In the example, to illustrate the point, the DC names were changed to ‘cans of soup’. Just imagine the fun you could have with a PBEM opponent who plays with your mod when his subs are sunk by cans of tuna fish, and yours are sunk by lobs of hamachi sashimi.

Image

The relevant parameter is the device “type”: 20 – ASW. Only certain device type weapons are looked for in the different combat algorithms. You can put a 1000lb bomb in a sub, but the code won’t see it because it doesn’t have one of the “correct” device types.

Our little DD, the Wolfgang Puck, has 3 ASW weapons, Wpn 8, 9 and 10, each with “Num” of 2, and all having various facings, turrets and “Ammo”.

An ASW TF (as a whole) does not detect subs; the individual ships, comprising the TF, each have a chance to detect the sub and, if the ship detects, each individual ship may then attack. Ships attack in “passes”; a ship makes a “pass” over the sub and fires its weapons (Wpns). Each Wpn invokes the hit calculation “Num” number of times.

On her first “pass” the Puck fires Wpn 8 which lobs 2 (Num) cans of chicken soup. Each can of soup individually attacks and has its own ‘chance’ to cause damage. Then the Puck can fire Wpn 9, which lobs 2 cans of tomato soup, each with its own ‘chance’. Then the Puck can roll 2 cans of Spaghetti O’s, again each with its own ‘chance’. The Wolfgang Puck gets 6 bites at the apple on a ‘pass’; 6 opportunities to cause damage.

A ship may make up to “Ammo” numbers of ‘passes’ over the target. A ship may not fire all available Wpns on a ‘pass’. As the “Ammo” parameter gets small, the effected Wpn may not choose to fire (a random calculation). Ciao.

John
Attachments
ASWDevices.jpg
ASWDevices.jpg (41.23 KiB) Viewed 236 times
User avatar
JWE
Posts: 5039
Joined: Tue Jul 19, 2005 5:02 pm

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by JWE »

Hedgehogs, in “game terms” are single installation launchers with multiple, substantially simultaneous, “rounds” in a launch. A typical Hedgehog fired maybe 20 projectiles in a salvo, each projectile having a 35lb torpex charge (somewhere earlier I wrote 65lbs, but that was the weight of the whole projectile – silly me).

Image

Due to how the code functions in terms of ‘passes’ and ‘chances’, and especially in terms of how Nik/Mike Wood/Michaelm tweaked the damage calculation routines, it seems reasonable to implement hedgehog, etc.. devices as follows:

On our exemplary DD, the BigB, Wpn 114 has a “Num” of 20 (this is the number of projectiles), turrets = 1 (single installation), “Ammo” of 6 (assuming a hard humping set of sailors doing the reloads).

If the BigB detects a sub, the first ‘shot’ from Wpn 1 will deliver 20 cans of soup, each with an ‘effect’ of 35, and each having a ‘chance’ to cause damage. It means you will have to sit through 20 iterations of the combat screen for each shot, but what the hey.

So bottom line, I think the best implementation, given how the code works, is to set “Num” to the number of projectiles in a shot, and keep each projectile’s effect at its stock value. Ciao.

John



Attachments
ASWHedge1.jpg
ASWHedge1.jpg (11.72 KiB) Viewed 236 times
herwin
Posts: 6047
Joined: Thu May 27, 2004 9:20 pm
Location: Sunderland, UK
Contact:

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by herwin »

ORIGINAL: JWE

Hedgehogs, in “game terms” are single installation launchers with multiple, substantially simultaneous, “rounds” in a launch. A typical Hedgehog fired maybe 20 projectiles in a salvo, each projectile having a 35lb torpex charge (somewhere earlier I wrote 65lbs, but that was the weight of the whole projectile – silly me).

Image

Due to how the code functions in terms of ‘passes’ and ‘chances’, and especially in terms of how Nik/Mike Wood/Michaelm tweaked the damage calculation routines, it seems reasonable to implement hedgehog, etc.. devices as follows:

On our exemplary DD, the BigB, Wpn 114 has a “Num” of 20 (this is the number of projectiles), turrets = 1 (single installation), “Ammo” of 6 (assuming a hard humping set of sailors doing the reloads).

If the BigB detects a sub, the first ‘shot’ from Wpn 1 will deliver 20 cans of soup, each with an ‘effect’ of 35, and each having a ‘chance’ to cause damage. It means you will have to sit through 20 iterations of the combat screen for each shot, but what the hey.

So bottom line, I think the best implementation, given how the code works, is to set “Num” to the number of projectiles in a shot, and keep each projectile’s effect at its stock value. Ciao.

John




It's a bunch of projectiles in a uniform pattern--if one hits, it's unlikely another will. The one that hits does its damage. The easiest way to model this is as 35 pounds of torpex with a reasonable probability of a critical hit. The use of contact fusing makes it harder for the sub to creep away and allows additional shots. The escorts are more concerned with keeping the sub suppressed (relatively easy once detected) than killing it (quite hard). The sub can't run, so it has to stay quiet and wait out the escorts.
Harry Erwin
"For a number to make sense in the game, someone has to calibrate it and program code. There are too many significant numbers that behave non-linearly to expect that. It's just a game. Enjoy it." herwin@btinternet.com
User avatar
Monter_Trismegistos
Posts: 1359
Joined: Tue Feb 01, 2005 8:58 pm
Location: Gdansk

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by Monter_Trismegistos »

Does "effect" plays a role during hit probability resolution? How "accuracy" works then?
Nec Temere Nec Timide
Bez strachu ale z rozwagą
User avatar
JWE
Posts: 5039
Joined: Tue Jul 19, 2005 5:02 pm

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by JWE »

ORIGINAL: herwin
It's a bunch of projectiles in a uniform pattern--if one hits, it's unlikely another will. The one that hits does its damage. The easiest way to model this is as 35 pounds of torpex with a reasonable probability of a critical hit.
Which is about as close as the proposed model can get it, given the algorithm.
The use of contact fusing makes it harder for the sub to creep away and allows additional shots. The escorts are more concerned with keeping the sub suppressed (relatively easy once detected) than killing it (quite hard). The sub can't run, so it has to stay quiet and wait out the escorts.
But since the code doesn't know contact fuses from horseradish, this isn't particularly useful.
herwin
Posts: 6047
Joined: Thu May 27, 2004 9:20 pm
Location: Sunderland, UK
Contact:

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by herwin »

ORIGINAL: Monter_Trismegistos

Does "effect" plays a role during hit probability resolution? How "accuracy" works then?

That's for the game designers to talk about. I'm only describing how we analyse it operationally.
Harry Erwin
"For a number to make sense in the game, someone has to calibrate it and program code. There are too many significant numbers that behave non-linearly to expect that. It's just a game. Enjoy it." herwin@btinternet.com
herwin
Posts: 6047
Joined: Thu May 27, 2004 9:20 pm
Location: Sunderland, UK
Contact:

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by herwin »

ORIGINAL: JWE
ORIGINAL: herwin
It's a bunch of projectiles in a uniform pattern--if one hits, it's unlikely another will. The one that hits does its damage. The easiest way to model this is as 35 pounds of torpex with a reasonable probability of a critical hit.
Which is about as close as the proposed model can get it, given the algorithm.
The use of contact fusing makes it harder for the sub to creep away and allows additional shots. The escorts are more concerned with keeping the sub suppressed (relatively easy once detected) than killing it (quite hard). The sub can't run, so it has to stay quiet and wait out the escorts.
But since the code doesn't know contact fuses from horseradish, this isn't particularly useful.

All it means is that hedgehogs and DC patterns function similarly except that DCs have a lower probability of hitting (one third or less), a higher probability of a critical hit (bigger charge), and a lower probability of multiple patterns. In the end, the pKill per charge was about 0.001 for both types, but you got more chances with a hedgehog (or squid).
Harry Erwin
"For a number to make sense in the game, someone has to calibrate it and program code. There are too many significant numbers that behave non-linearly to expect that. It's just a game. Enjoy it." herwin@btinternet.com
User avatar
JWE
Posts: 5039
Joined: Tue Jul 19, 2005 5:02 pm

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by JWE »

ORIGINAL: herwin

ORIGINAL: Monter_Trismegistos

Does "effect" plays a role during hit probability resolution? How "accuracy" works then?

That's for the game designers to talk about. I'm only describing how we analyse it operationally.
Harry, how the game designers did it is precisely what these threads' topics are about. Operational analysis is something that actually belongs elsewhere. Monter's questions are well taken and directly on topic.

"effect" does play a role, in that "hit probability" and "damage" are wrapped up into one big chunky calculation. I'm checking with the guys who did the latest code to see if they did anything additional to accuracy. Soon as I find out, we'll see how accuracy works in the current version. Ciao.

John
User avatar
witpqs
Posts: 26376
Joined: Mon Oct 04, 2004 7:48 pm
Location: Argleton

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by witpqs »

Did the Hedgehog (& Mousetrap) warheads have shaped charges? If so then maybe the 35 pounds should have an effect greater than the way the code would interpret 35 pounds.
User avatar
akdreemer
Posts: 1028
Joined: Sun Oct 03, 2004 12:43 am
Location: Anchorage, Alaska
Contact:

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by akdreemer »

ORIGINAL: herwin


It's a bunch of projectiles in a uniform pattern--if one hits, it's unlikely another will. The one that hits does its damage. The easiest way to model this is as 35 pounds of torpex with a reasonable probability of a critical hit. The use of contact fusing makes it harder for the sub to creep away and allows additional shots. The escorts are more concerned with keeping the sub suppressed (relatively easy once detected) than killing it (quite hard). The sub can't run, so it has to stay quiet and wait out the escorts.

I do not see how it would be unlikely more than one would hit. A 160 by 140 foot elliptical or a 200 foot circle would have a good probability of two hits, with three probably not uncommon, since most subs are over 200 feet in length. Indeed rereading the exploits of the USS England and her famous 6 subs in 12 days exploit, note that an explosion with a hedgehog meant a hit:

Sub one:

"On the second hedgehog attack, two explosions were obtained four seconds after the projectiles hit the water"

"On the fifth attack, two or three hedgehog explosions were heard twelve seconds after striking the water and a fathometer reading of 54 fathoms was obtained"

Sub two

"18.2 seconds after firing the second patterns three or more hedgehog projectiles detonated,"

sub three:

"On the second attack some 8 to 10 hedgehog hits were obtained 14.5 seconds after the charges struck the water."

etc, etc,
http://de635.ussengland.org/enclosure_b.htm

I would not change the code any for hedeghogs as it is obvious by operational experience that more than one hit was possible per pattern.






User avatar
akdreemer
Posts: 1028
Joined: Sun Oct 03, 2004 12:43 am
Location: Anchorage, Alaska
Contact:

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by akdreemer »

Hedgehogs were 24 round launchers, with Mk10 and Mk 11 differing only in the pattern, one a circle and one elliptical.
User avatar
akdreemer
Posts: 1028
Joined: Sun Oct 03, 2004 12:43 am
Location: Anchorage, Alaska
Contact:

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by akdreemer »

JWE

Be happy to share the hamachi sushimi with you..
User avatar
akdreemer
Posts: 1028
Joined: Sun Oct 03, 2004 12:43 am
Location: Anchorage, Alaska
Contact:

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by akdreemer »

ORIGINAL: witpqs

Did the Hedgehog (& Mousetrap) warheads have shaped charges? If so then maybe the 35 pounds should have an effect greater than the way the code would interpret 35 pounds.
Judging by the cut-aways I have seen they were not shaped-charges.

http://www.navweaps.com/Weapons/WAMUS_ASW.htm

cheers
herwin
Posts: 6047
Joined: Thu May 27, 2004 9:20 pm
Location: Sunderland, UK
Contact:

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by herwin »

If a scenario designer is concerned with getting the details right, he needs to know both how the game engine works and how things worked in reality.
Harry Erwin
"For a number to make sense in the game, someone has to calibrate it and program code. There are too many significant numbers that behave non-linearly to expect that. It's just a game. Enjoy it." herwin@btinternet.com
herwin
Posts: 6047
Joined: Thu May 27, 2004 9:20 pm
Location: Sunderland, UK
Contact:

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by herwin »

ORIGINAL: witpqs

Did the Hedgehog (& Mousetrap) warheads have shaped charges? If so then maybe the 35 pounds should have an effect greater than the way the code would interpret 35 pounds.

After getting told that the Hedghogs were designed to get multiple hits, I did a little research. My background is OR (and Cold War ASW, where individual warheads were designed to kill rather than damage the boat). My original reference was Morse and Kimball, Methods of Operations Research. The Squid mortar was the successor to the Hedgehog and functioned much like I described, with a pattern of three contact-fused mortar rounds. In 1944-45, it killed about one sub in three that was attacked (with one or two shots). Hedgehog fired a salvo of 24 charges, that landed in a 30m diameter circle at a fixed point relative to the ship. One to two hits were enough to sink a submarine, but there were no near misses, so it was an all-or-nothing weapon. Given that there was one sub sunk per about 1000 charges expended, one attack in 40 was effective. (The best kill rate seen was 25%.) That suggests the key factor was targeting accuracy, which is sensitive to crew/captain skill (on both sides) and sensor effectiveness. The right way to model this is to treat the weapons system as a single shot at a time, with the probability of hitting non-linearly sensitive to skill (most crews having low skill) and the probability of a critical hit given a hit being 100%. Hedgehog supplemented depth charges. Pictures here.

Mousetrap was an ASW rocket system that replaced Hedgehog. About 100 ships had it. No data on effectiveness.
Harry Erwin
"For a number to make sense in the game, someone has to calibrate it and program code. There are too many significant numbers that behave non-linearly to expect that. It's just a game. Enjoy it." herwin@btinternet.com
User avatar
vettim89
Posts: 3669
Joined: Fri Jul 13, 2007 11:38 pm
Location: Toledo, Ohio

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by vettim89 »

ORIGINAL: herwin

ORIGINAL: witpqs

Did the Hedgehog (& Mousetrap) warheads have shaped charges? If so then maybe the 35 pounds should have an effect greater than the way the code would interpret 35 pounds.

After getting told that the Hedghogs were designed to get multiple hits, I did a little research. My background is OR (and Cold War ASW, where individual warheads were designed to kill rather than damage the boat). My original reference was Morse and Kimball, Methods of Operations Research. The Squid mortar was the successor to the Hedgehog and functioned much like I described, with a pattern of three contact-fused mortar rounds. In 1944-45, it killed about one sub in three that was attacked (with one or two shots). Hedgehog fired a salvo of 24 charges, that landed in a 30m diameter circle at a fixed point relative to the ship. One to two hits were enough to sink a submarine, but there were no near misses, so it was an all-or-nothing weapon. Given that there was one sub sunk per about 1000 charges expended, one attack in 40 was effective. (The best kill rate seen was 25%.) That suggests the key factor was targeting accuracy, which is sensitive to crew/captain skill (on both sides) and sensor effectiveness. The right way to model this is to treat the weapons system as a single shot at a time, with the probability of hitting non-linearly sensitive to skill (most crews having low skill) and the probability of a critical hit given a hit being 100%. Hedgehog supplemented depth charges. Pictures here.

Mousetrap was an ASW rocket system that replaced Hedgehog. About 100 ships had it. No data on effectiveness.

I would agree with Herwin here in that Ahead Thrown Weapons (ATW) should be single shots with high critical hit occurance. I wonder though if the code can handle this. Does the ASW subroutine compute all weapons effects similarly? By that I mean does the engine consider ATH weapons rounds as mini DC's and computes for proximity vs. direct damage. Some one who has seen inside the dark regions of the WiTP code could better answer that.

For the record Mousetrap was not a replacement for Hedgehog but insted a modified weapon for use on smaller ships, specifically SC's. The torque generated by the spigot launchers on Hegehog would put undue stress on the superstructure of the small ships. The solution was to put the same weapon package on a rocket launched delivery device.
"We have met the enemy and they are ours" - Commodore O.H. Perry
User avatar
JWE
Posts: 5039
Joined: Tue Jul 19, 2005 5:02 pm

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by JWE »

ORIGINAL: AlaskanWarrior
JWE
Be happy to share the hamachi sushimi with you..
Oh gosh, send me some fresh Alaskan Halibut!! I'll be your Huckelberry forever!!

DC accuracy .. oh the horror .. the horror. It is not just a simple % chance to hit. There are some practical recommendations at the end of this, however.

The ‘accuracy’ numbers in the editor are evaluated very differently, depending on the “Device Type” of the weapon. For ASW weapons, the accuracy number gets evaluated against the target’s ability to evade (itself a function of speed, leader rating and maneuver), then gets multiplied by a hard coded number, then gets evaluated by combat range (which differs in shallow/deep water), then gets evaluated against cruise speed (if it’s a sub), then against target prior damage, then against day/night, and finally, after all the multiplication, division, subtractions, and comparisons, if the final number is too small, it is arbitrarily set to some nominal value, let’s just say, 2.

‘Accuracy’ is reduced in deep water, not ‘really’ by half, but this is a reasonable assumption. Combat range is a function of the sub’s depth (durability*10, for Class Type SS), but is arbitrarily set to 150’ for shallow water. So for a deep diver (300 footer), DC accuracy goes to half (sorta) in deep water, while for a shallow diver (maybe a 200 footer), DC accuracy only drops by 60-70% (sorta), in deep water. It’s ‘sorta’ because combat range is only one of the string variables, and is “adjusted” by the others.

Reasonable “editor” accuracy values for ASW weapons, thus don’t have much to do with IRL. IRL is useful, however, in determining “relative” values. If an IRL DC had a hit % of a billion to 1, the code promotes this to (effectively) 2%; so much for IRL.

Woof! So … the game code seems to like individual DCs with individual ‘accuracy’ figures between 4 and 10. The original game designers modeled things like Y and K guns and Hedgehogs, by adjusting the ‘accuracy’ parameters of the Devices (because you just can’t adjust the executable code). Not a bad way to go but, like everything else, it has implications. These will be dealt with in the next post.

Ciao. John
User avatar
JWE
Posts: 5039
Joined: Tue Jul 19, 2005 5:02 pm

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by JWE »

All right, given all the code horseradish, above, just what does a modder do to implement things like hedgehogs? Using the constraints of the code, there are two basic approaches; I use one of these, Harry Erwin suggests another as appropriate. Both are valid in the “game” context and both have their own internal implications; welcome to the wonderful world of programming.

There are only 2 realistic variables that pertain, here; 1 = (# of shots) and 2 = (% to hit per shot). My personal preference is to maximize “1”, with a nominal value of “2” per shot. My view is predicated on the individuation of individual warheads. Harry’s view is to maximize “2” while nominating load-out. His view is predicated on aggregate patterning.

Woof!! Both of these are so finely divided .. this is gonna have to be modder’s choice.

I can explain my choice upon request, but Harry’s gonna have to do the same.

Gosh, this is fun!! Ciao.

John
User avatar
vettim89
Posts: 3669
Joined: Fri Jul 13, 2007 11:38 pm
Location: Toledo, Ohio

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by vettim89 »

ORIGINAL: JWE

ORIGINAL: AlaskanWarrior
JWE
Be happy to share the hamachi sushimi with you..
Oh gosh, send me some fresh Alaskan Halibut!! I'll be your Huckelberry forever!!

DC accuracy .. oh the horror .. the horror. It is not just a simple % chance to hit. There are some practical recommendations at the end of this, however.

The ‘accuracy’ numbers in the editor are evaluated very differently, depending on the “Device Type” of the weapon. For ASW weapons, the accuracy number gets evaluated against the target’s ability to evade (itself a function of speed, leader rating and maneuver), then gets multiplied by a hard coded number, then gets evaluated by combat range (which differs in shallow/deep water), then gets evaluated against cruise speed (if it’s a sub), then against target prior damage, then against day/night, and finally, after all the multiplication, division, subtractions, and comparisons, if the final number is too small, it is arbitrarily set to some nominal value, let’s just say, 2.

‘Accuracy’ is reduced in deep water, not ‘really’ by half, but this is a reasonable assumption. Combat range is a function of the sub’s depth (durability*10, for Class Type SS), but is arbitrarily set to 150’ for shallow water. So for a deep diver (300 footer), DC accuracy goes to half (sorta) in deep water, while for a shallow diver (maybe a 200 footer), DC accuracy only drops by 60-70% (sorta), in deep water. It’s ‘sorta’ because combat range is only one of the string variables, and is “adjusted” by the others.

Reasonable “editor” accuracy values for ASW weapons, thus don’t have much to do with IRL. IRL is useful, however, in determining “relative” values. If an IRL DC had a hit % of a billion to 1, the code promotes this to (effectively) 2%; so much for IRL.

Woof! So … the game code seems to like individual DCs with individual ‘accuracy’ figures between 4 and 10. The original game designers modeled things like Y and K guns and Hedgehogs, by adjusting the ‘accuracy’ parameters of the Devices (because you just can’t adjust the executable code). Not a bad way to go but, like everything else, it has implications. These will be dealt with in the next post.

Ciao. John

Well this is what I was afraid of. It seems the way the code is set-up, it is very hard, but not impossible, for a modder to affect ASW. With so many variables we cannot get at, we are left with but 2 - accuracy and effect. So it does seem to be quite the challenge. What I did not see in your interation John was a variable for the firing platforms experience. Is this in there?
"We have met the enemy and they are ours" - Commodore O.H. Perry
User avatar
JWE
Posts: 5039
Joined: Tue Jul 19, 2005 5:02 pm

RE: Hedgehogs & Mousetraps & Bears, Oh My - Not RHS

Post by JWE »

ORIGINAL: vettim89
Well this is what I was afraid of. It seems the way the code is set-up, it is very hard, but not impossible, for a modder to affect ASW. With so many variables we cannot get at, we are left with but 2 - accuracy and effect. So it does seem to be quite the challenge. What I did not see in your interation John was a variable for the firing platforms experience. Is this in there?
Yes, it is difficult, but i will post relevant values and their spreads. A challenge, yes, but that is what these threads are devised to address. Once you really understand who does what to whom, it's a lot easier to figure out what you can do to tweak things your way.

To answer your explicit question, experience is a function of the detection routine, and not the firing routine. Six, hard humping gorillas, will fire the same as six Einsteins. However, six Einsteins will find the target way before six gorillas. Does this make sense???

Ciao. John
Post Reply

Return to “Scenario Design”