|
Post by nimrod on Mar 28, 2022 7:28:38 GMT -6
The purpose of my describing the damage was to point out that the Japanese AP rounds, not even HE, caused critical systems damage, and one shot, *missing its target* at ~30 km nearly split a CVL in half. The ship was rushed out of action for repairs and would be relegated to aircraft transport for the rest of its life. Expendable, I apologize for miss construing your remarks. Thank you for clarifying.
For the record, I see comparable types of AAR rather often.... Yes it is a 7,000 ton CL, but 3 15" (supposedly AP) hits didn't do any real damage other than taking out a torp launcher and a BE hit perforated the funnels. Most actual damage was from the 6" hits such as the flotation damage from a BE waterline hit. She stayed in the fight despite having a -1 crew, guns pretty much stayed in operation but the 5" mains only scored one hit (mostly due to targeting DDs at 4-6 thousand yards away rather than the BCs at 600-2800 yards), and there was no fire or flooding events noted in her log:
Enemy coal powered DD, so some hits were soaked by the coal storage (significant number of near misses in the log as well), but 21 4" hits didn't do significant structural damage to an 1,100 ton DD (she sank due to flotation damage and she had only lost 30-40% structural strength):
Little better, but again coal powered DD, 9 meaningful 5" and 10 4" hits and she sunk due to flooding.
These and similar AARs indicate to me that HE values / effects are probably low or that the damage models on light ships should be re-looked at and finessed.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 29, 2022 5:00:13 GMT -6
Generally speaking, the damage model may be considered some kind of abomination. As is the accuracy of fire.
|
|
|
Post by williammiller on Mar 29, 2022 9:36:51 GMT -6
<button disabled="" class="c-attachment-insert--linked o-btn--sm">Attachment Deleted</button> Generally speaking, the damage model may be considered some kind of abomination. As is the accuracy of fire. We have internal tools where we can attack test ships tens of thousands of times and get very detailed results on damage to said test ships: they match expected historic result ranges very well, thank you. Frankly speaking, outliers and single ship examples are just that - outliers and single ship examples.
|
|
|
Post by nimrod on Mar 29, 2022 12:15:47 GMT -6
Williammiller, I do see a lot of smaller ships CL and smaller taking significant amounts of heavy hits and surviving in rather good fighting shape. I'll throw a couple additional images up later as I'm away from my gaming computer, but I just had an enemy 4,000 ton AMC get hit by something like 15 15" shells before finally sinking. I totally get that over 1000s of simulations that outliers exist and as we play the game we see and latch on to them more often than not. The issue for me is what I think should be "outliers and single ship examples" happen about once every battle on average... Maybe its due to the crews inability to be impacted by damage and thus the ships stay in fighting order when historically the -1 crew would have abandoned ship with the loss of their captain and 15-25% of the crew... I admit that is my view and as such it may be ahistorical or tainted by what I've read. Additionally, I don't have details on the in-game damage model or what shell tech my enemy is using, etc (maybe my issue is that the code for determining AP vs. HE shell usage is problematic, or its the crews inability to surrender the ship, or something else I haven't thought of...). Having disparaged my own views, I'll humbly suggest that you might want to consider an outer band limit (an example would be, 10"+ pass throughs cause a min. amount of damage, while 5-9" pass through hits cause a different amount of min. damage) to help limit atypical results (Just had a battle this morning in which a captured CA with 4.5" of belt armor had a pass through from a 4" or 5" gun). I can't really state the exact mechanism that I'm having trouble with, but, at the end of the day, I consider this a game rather than an historical simulation - heavy gun hits and significant numbers of medium gun hits (on light ships in particular) pulls me from the experience and leave me scratching my head just about every major battle.
Just for fun... I'm guessing their is a really good bakery there...
|
|
|
Post by sagaren on Mar 29, 2022 17:17:47 GMT -6
Williammiller, I do see a lot of smaller ships CL and smaller taking significant amounts of heavy hits and surviving in rather good fighting shape. I'll throw a couple additional images up later as I'm away from my gaming computer, but I just had an enemy 4,000 ton AMC get hit by something like 15 15" shells before finally sinking. I totally get that over 1000s of simulations that outliers exist and as we play the game we see and latch on to them more often than not. The issue for me is what I think should be "outliers and single ship examples" happen about once every battle on average... Maybe its due to the crews inability to be impacted by damage and thus the ships stay in fighting order when historically the -1 crew would have abandoned ship with the loss of their captain and 15-25% of the crew... I admit that is my view and as such it may be ahistorical or tainted by what I've read. Additionally, I don't have details on the in-game damage model or what shell tech my enemy is using, etc (maybe my issue is that the code for determining AP vs. HE shell usage is problematic, or its the crews inability to surrender the ship, or something else I haven't thought of...). Having disparaged my own views, I'll humbly suggest that you might want to consider an outer band limit (an example would be, 10"+ pass throughs cause a min. amount of damage, while 5-9" pass through hits cause a different amount of min. damage) to help limit atypical results (Just had a battle this morning in which a captured CA with 4.5" of belt armor had a pass through from a 4" or 5" gun). I can't really state the exact mechanism that I'm having trouble with, but, at the end of the day, I consider this a game rather than an historical simulation - heavy gun hits and significant numbers of medium gun hits (on light ships in particular) pulls me from the experience and leave me scratching my head just about every major battle. Just to throw in my own two cents on this topic. I wonder how much of this is not that these ships don't take less hits to sink than we're discussing here, but that just because a ship has taken enough hits to cause it to sink, doesn't mean that is going to happen immediately. For instance, just recently I had a small light cruiser I attacked with torpedo planes and it took something like fifteen hits, at least eight of those were before it stated "ship sinking". Now I'm pretty sure a small old 4000 cruiser was dead after the second, or at least third torpedo, but it took a round or two before all that damage actually took full effect and listed the ship as sinking. I wonder if that's where a lot of these cases of over-survivability are actually coming from. I'd be curious what other people think.
|
|
|
Post by nimrod on Mar 29, 2022 18:13:59 GMT -6
Williammiller, I do see a lot of smaller ships CL and smaller taking significant amounts of heavy hits and surviving in rather good fighting shape. I'll throw a couple additional images up later as I'm away from my gaming computer, but I just had an enemy 4,000 ton AMC get hit by something like 15 15" shells before finally sinking. I totally get that over 1000s of simulations that outliers exist and as we play the game we see and latch on to them more often than not. The issue for me is what I think should be "outliers and single ship examples" happen about once every battle on average... Maybe its due to the crews inability to be impacted by damage and thus the ships stay in fighting order when historically the -1 crew would have abandoned ship with the loss of their captain and 15-25% of the crew... I admit that is my view and as such it may be ahistorical or tainted by what I've read. Additionally, I don't have details on the in-game damage model or what shell tech my enemy is using, etc (maybe my issue is that the code for determining AP vs. HE shell usage is problematic, or its the crews inability to surrender the ship, or something else I haven't thought of...). Having disparaged my own views, I'll humbly suggest that you might want to consider an outer band limit (an example would be, 10"+ pass throughs cause a min. amount of damage, while 5-9" pass through hits cause a different amount of min. damage) to help limit atypical results (Just had a battle this morning in which a captured CA with 4.5" of belt armor had a pass through from a 4" or 5" gun). I can't really state the exact mechanism that I'm having trouble with, but, at the end of the day, I consider this a game rather than an historical simulation - heavy gun hits and significant numbers of medium gun hits (on light ships in particular) pulls me from the experience and leave me scratching my head just about every major battle. Just to throw in my own two cents on this topic. I wonder how much of this is not that these ships don't take less hits to sink than we're discussing here, but that just because a ship has taken enough hits to cause it to sink, doesn't mean that is going to happen immediately. For instance, just recently I had a small light cruiser I attacked with torpedo planes and it took something like fifteen hits, at least eight of those were before it stated "ship sinking". Now I'm pretty sure a small old 4000 cruiser was dead after the second, or at least third torpedo, but it took a round or two before all that damage actually took full effect and listed the ship as sinking. I wonder if that's where a lot of these cases of over-survivability are actually coming from. I'd be curious what other people think. Sagaren,
Good question, and I would be interested in what others have to say. I'm going to bow out of this conversation with this post as I'm just repeating my-self and I've moved the conversation far-away from the OP's Torpedo Speed question. I do apologize for that Christian.
I'm looking and posting screen shots of the post battle results and I'm cherry picking ships that sunk without additional post sinking hits (so ships sunk in the middle of fight melee aren't generally represented. These cherry picked ships are end of battle sinkings or survivors, so I'm actually posting a very small set of ships from what I can access). The AMC and AH CL in the above attachments sunk due to flooding, but what concerns me is the lack of structural damage, how many guns are still operational, what speed they can maintain. The ships in these example are getting pretty well pounded by large and medium caliber guns and yet still maintain warefighting capability (reasonable speed, guns are generally intact / repairable rather than knocked-out, etc.) until the end. Yes ships lose accuracy due to damage and maneuverability, but they still launch torps and can be deadly with their guns.
Common example below - top left image has all the heavy hits, but the right and then bottom image shows the damage progression of the medium 6" hits to the structural damage and gun / torp mountings (hint the disabled main guns come back online). Please note that there was 5 actual 16" hits and a 16" near miss that didn't seem to cause any notable damage in the top left image, additionally, the flooding damage comes from the 6" medium hits (this is pretty typical in that the heavy hits don't seem to cause much flotation damage on the smaller ships).
Other examples of things that make me go huh from a recent battle... A supposedly 36800 ton CVL with 11.5" of belt armor that took 31 heavy hits before sinking due to fire (also has two funnels in the flight deck so I'm guessing my intelligence personnel got into the Champagne again; or the British have gone crazy again and tried for a proof of concept Habbakuk), a CA with 4.5 belt armor that took 37 heavy hits to sink, a 5400 ton CL with 2" of belt armor that took 24 medium and about 32 4" hits to sink due to flooding with a bit over half its structure still in-tack and only one main gun permanently out of action...
Thinking through the damage model, maybe the larger caliber guns should be capable of causing multiple critical events per hit, especially on sub 10,000 ton ships... Like taking out the bridge, funnels and two adjacent turrets / torp launchers in a hit...
Another, just for fun screen shot - CVL on AF intercepts raider... Yes the CVL and DD escorts prevailed against the AMC...
|
|
|
Post by mdesanta on Mar 30, 2022 0:12:14 GMT -6
The results obtained by torpedoes in the game are hugely influenced by how you, personally, handle destroyers in the game. Aggressive manual control of destroyers will result in a percentage of hits that is far above what was ever achieved historically. This advantage is so extreme that experienced players can probably win most battles using just their destroyer force under manual control. On the other hand, allowing the AI to handle destroyers probably results in fewer hits than were historically typical, though there is so much variation in battle situations it is quite hard to evaluate this. Why not just simulate a torpedo shot like you would a gun shot? Take into account all the variables and roll a die, maybe with a few ticks of delay depending on the distance, and a chance to hit other ships in the way (also die roll). That way there would be no difference between AI torpedoing and human micromanagement at the lowest level of simulation, thus allowing you to reach historical levels of performance without worrying about balance. It is, in my opinion, the most realistic way. The battlefield is chaotic and die rolls represent all those countless small variables not taken into calculation.
|
|
|
Post by seawolf on Mar 30, 2022 11:22:46 GMT -6
The results obtained by torpedoes in the game are hugely influenced by how you, personally, handle destroyers in the game. Aggressive manual control of destroyers will result in a percentage of hits that is far above what was ever achieved historically. This advantage is so extreme that experienced players can probably win most battles using just their destroyer force under manual control. On the other hand, allowing the AI to handle destroyers probably results in fewer hits than were historically typical, though there is so much variation in battle situations it is quite hard to evaluate this. Why not just simulate a torpedo shot like you would a gun shot? Take into account all the variables and roll a die, maybe with a few ticks of delay depending on the distance, and a chance to hit other ships in the way (also die roll). That way there would be no difference between AI torpedoing and human micromanagement at the lowest level of simulation, thus allowing you to reach historical levels of performance without worrying about balance. It is, in my opinion, the most realistic way. The battlefield is chaotic and die rolls represent all those countless small variables not taken into calculation. The problem is that it would make it most efficient to just sail in strait lines, which the AI refuses to do but players often do.
|
|
|
Post by tbr on Mar 30, 2022 16:32:46 GMT -6
The results obtained by torpedoes in the game are hugely influenced by how you, personally, handle destroyers in the game. Aggressive manual control of destroyers will result in a percentage of hits that is far above what was ever achieved historically. This advantage is so extreme that experienced players can probably win most battles using just their destroyer force under manual control. On the other hand, allowing the AI to handle destroyers probably results in fewer hits than were historically typical, though there is so much variation in battle situations it is quite hard to evaluate this. Why not just simulate a torpedo shot like you would a gun shot? Take into account all the variables and roll a die, maybe with a few ticks of delay depending on the distance, and a chance to hit other ships in the way (also die roll). That way there would be no difference between AI torpedoing and human micromanagement at the lowest level of simulation, thus allowing you to reach historical levels of performance without worrying about balance. It is, in my opinion, the most realistic way. The battlefield is chaotic and die rolls represent all those countless small variables not taken into calculation. The time to target for a torpedo is a multiple of the "turn time" (RTW is a WEGO turnbased game) in RTW's tactical layer. That means course changes of the target etc. over the turns since firing can outright eliminate a torpedo's chance to hit but another ship can be endagered etc. This cannot be abstracted within a single RTW tactical turn. However, for the "end game" of a homing torpedo I actually recommended exactly this, i.e. doing such a calculation when the torpedo enters a surface ship's range that represents torpedo homing sonar acquisition range and would be crossed within one or two more "minute turns":
I think resurrecting my post from two years ago on how to implement torpedo homing and countermeasures is timely here: On fusing it is important to take into account that there are two separate types of "magnetic" proximity fuses. The passive magnetic fuse relied on the disturbance caused by the target ship in the earths's magnetic field and therefore was very unreliable, e.g. the localized magnetic conditions around Norway practically rendered German passive magnetic fuses useless during Weserübung. However, there is also the active magnetic fuse which generates its own magnetic field for detection of the ship's keel above it. This one is far, far more reliable and still in use today. The German Navy introduced it in the later guided G7's like the Zaunkönig II. In guidance the IBIS wake homing torpedo, prototype fired in 1944 at the Torpedoversuchsanstalt (TVA), would have been the most dangerous to surface ships, especially in combination with the active magnetic fuse. Then there was passive and active/passive acoustic homing, also combined with wire guidance and/or swiveling "searchlight" sonar. All in trials in 1942-44 and some of it in early frontline use in WWII. Earliest electric wire guidance was in WWI coastal defence torpedoes, but that was with optical guidance where a flare was put on the torpedo. That concept was pursued a bit in the TVA ("Spinne") in the late 1920's to early 1930's. True wire guidance of homing torpedoes, i.e. two-way communication between torpedo and firing unit with at least a "lock on" signal from the torpedo and the command guidance option from the unit to the torpedo was trialled by the TVS in 1944 as well. I would need to look at some documentation to quote the project name though. Electric propulsion has only recently become fully superior to thermodynamic propulsion with the power/energy density jump on the 1990's. Within the RTW timeframe electric propulsion is markedly inferior to thermodynamic (within this timeframe wet heater/air, perhaps oxygenized, with piston or turbine engine) in regards to speed and tactical range, albeit with the advantages of "optic stealth" and cheap construction as well as suitability for homing torpedoes. The electric WWII torpedo had lead-acid batteries which were a hassle for the crew due to the need for on-board maintenance and charging but were far, far quicker and cheaper to produce than thermodynamic propulsion units, especially by "non-arsenal" wartime contractors. With electric propulsion it is also easier to operate homing torpedoes due to lower self-noise. And there is one exception to the range, the longest ranged torpedo used in WWII, the "Dackel", was electrically propelled, fired over-the-horizon by E-Boats at the Normandy invasion. But even with the pattern runnning mechanism it was a relatively ineffective weapon. And we even had the first two rounds of the countermeasure-countercountermeasure evolution cycle there during WWII as reflected by the steps from Falke (first passive homing torpedo that was susceptible to evasive throttling as it could home only on escorts when they were within a certain speed segment) to Zaunkönig I (which led to the Foxer decoy) and Zaunkönig II. Arguably, Ibis (wake homer), Geier (active-passive homer) and the wire guided Lerche also were part of this cycle, as even if they did not enter frontline use due to the 1944 war emergency program cut-off at the TVA, their technology was used and implemented in the Cold War.
In RTW2 terms it will be very difficult to implement the different homing torpedo mechanisms while respecting the countermeasure-countercountermeasure cycle. Kinematic and sensor simulation of torpedo performance is difficult, extensive and mostly done only in the classified realm. Nothing I have seen in games, that includes Dangerous Waters and Cold Waters, truly does it "right".
Perhaps it will be easiest if homing and countermeasure effectiveness get abstracted a bit. Early homing mechanisms and proximity fuses were unreliable even when not faced with countermeasures. Just have a tech step in torpedo homing provide two different values, one is initiation distance (in yards), i.e. the distance at which the torpedo detects and begins homing on a target. The second value would be initiation probability, i.e. the probability that a given torpedo's homing mechanism will work until hit. "Work" in this case, to simplify implementation, would therefore mean a hit against the closest surface ship (so that no complex kinematics need to be simulated, let alone target-loss and reattack manoeuvres, interference, self-noise vs. speed etc.). If the homing mechanism and proximity fuse does not "work" due to a failed probability roll (easiest to do this roll in the launch turn) the torpedo would act as a straight runner with an impact fuse. So an early homing mechanism could have a 250yds 30% capability, i.e. the torpedo would home (and hit) in 30% of the cases it enters within 250yds of a ship. Countermeasures would impose negatives on these stats, e.g. -50 and -20% with the first generation Foxer. At the beginning of the scenario the respective techs would be compared and the values for the scenario assigned, e.g. if both sides have 1st gen homing and countermeasures homing torpedoes would only home in 10% of the cases when they enter 200yds of a surface target. If one side does not have the countermeasure tech its enemy's torps would home at 250yds in 30% of cases. Oh, and do not forget a 1 or 2 minute/turn safety distance before turning "homing" live to avoid sui- or fratricide.
Another advantage of this approach is that the AI would not need to "learn" using homing torpedoes since it would continue to use them the same as straight runners. This also would mean that the final "home run" distance would in most cases be covered by the torpedo within two "minute turns" anyways, so there is really no huge point not to abstract this stretch as simulated torpedo himong behaviour would not cover more than those two tunrs anyways (the rare exception of the approach directly from aft at a high speed target should be ignored here as the AI will not willingly shoot for it anyways). Further generations in torpedo homing and torpedo countermeasures, in their abstraction, would incorporate general homing mechanism improvement, countermeasure and countercountermeasure development by raising both stat types (positive for homing tech, negative for countermeasure tech). To avoid "gamey" effectiveness of homing torpedoes implement a "hard" ceiling on initiation range and initiation probability (e.g. 750yds and 60%), even if the "consolidated tech level" of a scenario is higher due to one side having late gen homing and the other no countermeasure tech.
|
|
|
Post by mdesanta on Mar 31, 2022 1:15:14 GMT -6
Why not just simulate a torpedo shot like you would a gun shot? Take into account all the variables and roll a die, maybe with a few ticks of delay depending on the distance, and a chance to hit other ships in the way (also die roll). That way there would be no difference between AI torpedoing and human micromanagement at the lowest level of simulation, thus allowing you to reach historical levels of performance without worrying about balance. It is, in my opinion, the most realistic way. The battlefield is chaotic and die rolls represent all those countless small variables not taken into calculation. The problem is that it would make it most efficient to just sail in strait lines, which the AI refuses to do but players often do. One of the variables could be whether and by how much the target's heading changed after the torpedo is fired.
|
|
|
Post by mdesanta on Mar 31, 2022 1:37:00 GMT -6
Why not just simulate a torpedo shot like you would a gun shot? Take into account all the variables and roll a die, maybe with a few ticks of delay depending on the distance, and a chance to hit other ships in the way (also die roll). That way there would be no difference between AI torpedoing and human micromanagement at the lowest level of simulation, thus allowing you to reach historical levels of performance without worrying about balance. It is, in my opinion, the most realistic way. The battlefield is chaotic and die rolls represent all those countless small variables not taken into calculation. The time to target for a torpedo is a multiple of the "turn time" (RTW is a WEGO turnbased game) in RTW's tactical layer. That means course changes of the target etc. over the turns since firing can outright eliminate a torpedo's chance to hit but another ship can be endagered etc. This cannot be abstracted within a single RTW tactical turn.
This can be abstracted over the course of several ticks. For example, you fire a torpedo or a spread of torpedoes. You don't decide on the angle of fire, or the width of the spread. All you do as the player (and as the AI) is decide when to fire and how many. You shouldn't even be able to designate a "target", unless the target's stationary. The location of each individual torpedo is abstracted. After each turn, a dice is rolled on ships within the potential zone. The limiting angles of the potential zone would be the maximum angles of rotation of the torpedo launcher. Basically imagine a quarter pizza pie that extends from the side of the launching ship (for a side-firing deck launcher). Among the modifier to the dice roll would be course change of ships, speed of ships, etc. The side that the ship's on is also accounted for as a modifier (your crew wouldn't be deliberately aiming at friendly ships). If you hit, it means the ship's crew had fired at the correct angle and/or had correctly predict the target's evasion, or just got lucky. The only way for a ship to be 100% safe is if it is outside the abstracted potential zone on every turn.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 31, 2022 10:02:24 GMT -6
Let's play a game. Its called what portion of this destroyer (1942, Z-51 class, 36 knots) is vitals?
|
|
|
Post by nimrod on Apr 7, 2022 10:17:39 GMT -6
The time to target for a torpedo is a multiple of the "turn time" (RTW is a WEGO turnbased game) in RTW's tactical layer. That means course changes of the target etc. over the turns since firing can outright eliminate a torpedo's chance to hit but another ship can be endagered etc. This cannot be abstracted within a single RTW tactical turn.
This can be abstracted over the course of several ticks. For example, you fire a torpedo or a spread of torpedoes. You don't decide on the angle of fire, or the width of the spread. All you do as the player (and as the AI) is decide when to fire and how many. You shouldn't even be able to designate a "target", unless the target's stationary. The location of each individual torpedo is abstracted. After each turn, a dice is rolled on ships within the potential zone. The limiting angles of the potential zone would be the maximum angles of rotation of the torpedo launcher. Basically imagine a quarter pizza pie that extends from the side of the launching ship (for a side-firing deck launcher). Among the modifier to the dice roll would be course change of ships, speed of ships, etc. The side that the ship's on is also accounted for as a modifier (your crew wouldn't be deliberately aiming at friendly ships). If you hit, it means the ship's crew had fired at the correct angle and/or had correctly predict the target's evasion, or just got lucky. The only way for a ship to be 100% safe is if it is outside the abstracted potential zone on every turn. That sounds like a pretty decent idea. I'm not sure on, "You don't decide on the angle of fire, or the width of the spread." but otherwise it makes sense to me. I especially like the inability to select a specific ship - currently if a ship is turning or has a damaged rudder the code doesn't allow for a continuation of the turn... So being able to select a target division and have it go into the target envelope sounds like a win to me.
Also, just because I can't help it a couple of good late game damage examples of questionable damage modeling of lighter ships.
This is stuff I can't grasp, so if somebody would. Please explain how a 10700 CL took a 2000lb bomb hit which destroyed a turret but the ship is showing no structural damage? Flooding is from a 1000lb bomb near miss.
Or this 1400 ton DD that took a 1000lb bomb hit to a turret and a 1000lb bomb near miss yet has just two bars of structural damage.
Or this 900 tonner that has no structural damage after 3, 1000lb bomb hits and two 1000lb near misses (one post sinking due to flooding). Post sinking a 1000lb bomb hit was limited by the coal bunker, so I'm again pretty confused as I'm pretty sure a coal bunker on a ship that small shouldn't absorb a 1950's 1000lb bomb...
My understanding is that these hits should have distorted the hull, slowed down the ships by a good amount or otherwise caused some significant to serious structural damage.
|
|