phancucchin@aol.com Posted August 6, 2002 Share Posted August 6, 2002 hello Bastables, we must play again sometime. I think that some heat under the collar is present in some posts? I dont know why. Its possibly true that code may eb too hard to change for evry vehicle but something abstracted could be made? The BTS input is probably best. Link to comment Share on other sites More sharing options...
M Hofbauer Posted August 6, 2002 Share Posted August 6, 2002 Originally posted by TSword: Uhhh, i cannot believe what i heard in this thread ! One single additional variable enters the equation ! Take your normal distribution for a standard Tank, and then fill in a look up table with the deviations for each Major plate (and if memory is scarce take the front plates only 4 in our case with turreted tanks). Now once you were gone through the decision which section is "hit" -> Lower-, Upper-, Turret- Front call a probability to hit function which has your actual deviation value, if it's bigger than the default -> Same as in CMBO, else linear growing prob of miss (0.5 size turret has 0.5 chance to be hit then a 1.0 size turret). Where is the problem ??? Greets Danielthe problem is that the visible area ratio between turret and hull differs a lot depending on how the turret is facing w/r/t the hull. for illustrative purposes, lets take a Königstiger (use a three-sided view or CMBO for the following reading). from the front, with the turret facing dead ahead, the visible area is about 40 (turret) : 60 (hull) (turret area from the front includes the visible side areas, something CMBO neglects, too). from the side, the visible area is about 35:65. now, when the turret is facing 90° off to one side, these values and ratios change a lot. since from the front you are now comparing the huge side area of the turret vs the hull, the distribution is somewhere like 60 (turret) : 40 (hull). at the same time, the visible area from the side with the gun pointing at the viewer is also distributed different. the rather slim turret front is compared to the huge hull, making for a ratio of roughly 25 to 75. these are just very rough figures. my suggestion for a solution, which goes along your idea, is to give each area (lower hull, upper hull, turret) from each aspect a certain absolute size (not RELATIVE to the other areas of one heading, but an absolute value). that way, the distribution remains true even if the turret is rotating to the side, since the true value of the front is now compared to the true value of the side hull. Link to comment Share on other sites More sharing options...
TSword Posted August 6, 2002 Share Posted August 6, 2002 Hofbauer, Interesting, but i thought (Derived from the knowing that CMBO calculates hitangle) that the turrets facing also goes already in the equation. And my observations up to now has not revealed something strange in that respect. If it's the case that turretorientation is not taken into account, it would be a MAJOR ISSUE If it's the cast that it goes into the calculation as i'm quite sure off, the problem remains a simple one, you have to adjust hit probability by a single new variable... Btw. The impact of a Std. Distribution model is much larger then was mentioned in that thread. If i think of the IS-II, which had it's weak spot on the HUGE turret (the Hull being almost impervious, atleast later variants), or the T-34/85 with a similar huge Turret which again was THE weak frontal spot (The Hatch on the front plate was also but much smaller). In reality also has to take into account the effects of being Hulldowns with a huge turret vs. a small one.... So size does of course matter !!! (This time atleast...) Another aspect is the underrated importance of the principal form of a tank, or aspect ration of Height to Width. Question: Which plate is easier to be hit by a cannon with a 95 % hit chance on 1000 m on a 2 x 2 m target, when the plate is 10 m high and 2 m wide or the plate being 2 m high and 10 wide ?? Result: The High plate is a much easier target, because range estimation becomes much easier (errors in range estimation have a much lower impact). So of course a Sherman is as easy to hit as a KT or Panther and even Tiger front on, because the M4 has a smaller width but same Height, but still wide enough that shot distribution doesn't become a factor (Up to 1000 m). To the armor-quality thing, i read in some CMBB thread here, that they will not use an evenly distributed quality model, but instead more to the real thing of one tank being good, the other being bad --> Much, much better. So i'm quite optimistic looking forward to the CMBB release (Thanks to excellent posts recently made in this forum from first hand experience..). However the Distribution thingy would be hard to swallow..... Greets Daniel Link to comment Share on other sites More sharing options...
phancucchin@aol.com Posted August 6, 2002 Share Posted August 6, 2002 Originally posted by TSword: Another aspect is the underrated importance of the principal form of a tank, or aspect ration of Height to Width. Question: Which plate is easier to be hit by a cannon with a 95 % hit chance on 1000 m on a 2 x 2 m target, when the plate is 10 m high and 2 m wide or the plate being 2 m high and 10 wide ?? Result: The High plate is a much easier target, because range estimation becomes much easier (errors in range estimation have a much lower impact). So of course a Sherman is as easy to hit as a KT or Panther and even Tiger front on, because the M4 has a smaller width but same Height, but still wide enough that shot distribution doesn't become a factor (Up to 1000 m). This too is interesting observation. I read somewhere in search that length*width*height is used actually to judge the silhouette number seen in info window. The case of the long tank with short height then could be misrepresented in combat hits. Many turrets are of round or square dimensions when viewed from teh birds eye down. KT is not the best example but I value the point made above. A good point is that armor penetration calculates with angle of shot but how is the part hit determined? I am thinking of the glancing blow across the side of turret at extreme angular path. How? Many interesting things and I do agree, respectively, that maybe some parts of this nice game are more precisionally modeled than others? But it is just discussion because detailed inner workings arent well known. Thats all. Phan Link to comment Share on other sites More sharing options...
phancucchin@aol.com Posted August 9, 2002 Share Posted August 9, 2002 I think about distribution model and come up with this: 1. game determines that tank as a whole gets a hit (using size modifiers, etc) 2. game determines that part of tank (turret, upper/lower hull, gun, track) hit gets distributed to. 3. game determines the angle of 'glancing' at this point and calculates that into armor 4. effects then calculated in the case of the turret to one side, the game must determine front/side/back depending on the angle of shot. That is, you either hit front or side or rear. the determination being the one with the least angle perhaps? A rule for survival then for panzer IV is never face turret at other tanks. If given a choice, point the turret at other targets to sides and hope that the incoming rounds get the front of turret selected if hit and that the additional angle deflects shot. Hopefully, panther is there to help the survival! Link to comment Share on other sites More sharing options...
Olle Petersson Posted August 11, 2002 Share Posted August 11, 2002 Originally posted by John Kettler: ... Panzer IV players preferring to fight fully exposed because of the disproportionate number of turret hits when fighting hulldown. What's really disproportionate is the probability to spot a hull down tank. The absolute probability to hit the turret should increas with hull down, but the probability to see the turret in the first place is much lower. Hence; - Full view: high probability of being shot at, with little probability per shot to hit turret. - Hull down: low probability of being shot at, with higher probability of turret hits if the enemy shoots back. So it's sort of a gamble. The "best" should be to stay hull down until the enemy starts shooting back. Cheers Olle Link to comment Share on other sites More sharing options...
Michael Emrys Posted August 11, 2002 Share Posted August 11, 2002 Originally posted by Olle Petersson: The absolute probability to hit the turret should increas with hull down...Why is that? I'm afraid I don't understand. Michael Link to comment Share on other sites More sharing options...
Redwolf Posted August 11, 2002 Share Posted August 11, 2002 Originally posted by Olle Petersson: The absolute probability to hit the turret should increas with hull down As it does now. The vehicle spot system in CMBO is bejond repair anyway. Link to comment Share on other sites More sharing options...
M Hofbauer Posted August 11, 2002 Share Posted August 11, 2002 Originally posted by Michael emrys: </font><blockquote>quote:</font><hr />Originally posted by Olle Petersson: The absolute probability to hit the turret should increas with hull down...Why is that? I'm afraid I don't understand. Michael</font> Link to comment Share on other sites More sharing options...
Michael Emrys Posted August 12, 2002 Share Posted August 12, 2002 Okay. Got it now. Thanks. Michael Link to comment Share on other sites More sharing options...
Diceman Posted August 12, 2002 Share Posted August 12, 2002 Michael[/qb] Link to comment Share on other sites More sharing options...
phancucchin@aol.com Posted August 12, 2002 Share Posted August 12, 2002 If anything being hull down should make ranging more difficult, not less. You wouldn't be able to tell how high an over shot went. The turret is also a much smaller target so there is much less room for error due to inherent accuracy of the gun. Being hull down also makes guessing the range difficult because the vehicle height is not known. If the gaem does not make the spotting of hull down vehicles more difficult, it is using the wrong model. If the hit chance on a hull down vehicle is not less, it is also a wrong model. The game still will not use a gun down looking modeling either from what the search uncovers. Link to comment Share on other sites More sharing options...
Tarquelne Posted August 12, 2002 Share Posted August 12, 2002 Just curious: Anyone have any inkling of what a PzIV's QB point value would be if the turret was more-realistically hard to hit? Link to comment Share on other sites More sharing options...
Vanir Ausf B Posted August 12, 2002 Share Posted August 12, 2002 Originally posted by Tarqulene: Just curious: Anyone have any inkling of what a PzIV's QB point value would be if the turret was more-realistically hard to hit?5 points give or take a couple. Link to comment Share on other sites More sharing options...
Michael Emrys Posted August 12, 2002 Share Posted August 12, 2002 Originally posted by Vanir Ausf B: </font><blockquote>quote:</font><hr />Originally posted by Tarqulene: Just curious: Anyone have any inkling of what a PzIV's QB point value would be if the turret was more-realistically hard to hit?5 points give or take a couple.</font> Link to comment Share on other sites More sharing options...
Diceman Posted August 12, 2002 Share Posted August 12, 2002 The above screenshot is of a PZIV targeting two Shermans, one hull down, the other hull up. At 547 meters the PZIV has a 35% less chance of hitting the Sherman than if the Sherman were hull up. However, the hull down Sherman presents a 73.3% smaller target based on the numbers I calculated using a 1/76 scale diagram. The % chance of hitting the hull down target is 13.6%, which indecently equates to the % chance of hitting the turret hull up. Isn’t it nice when your math checks out? It is also interesting to note that the Sherman would have a 58.8% greater chance of taking a turret hit hull down than hull up. Now there is some truth to the argument that the odds of a turret hit would be increased due to it being the center of target, however I think that this would be off-set to a great degree by both the size of the target presented, and the greater difficulty in ranging the target both visually and mechanically. I don’t know how BFC came up with the 33% # but I think that the number is a bit high. I would suggest that the base # be much closer to the 13.6%, that is to say - the base chance of a hull down hit should be close to the % chance of a hull up turret hit, at least until the range to target is found. Based on these numbers, if the Sherman were a version with extra hull armor, it would be 59% more likely to survive hull up than hull down, presuming the apposing vehicle could not penetrate the supplemented hull armor. The same could be said for the PZIV. IMHO this does not agree with anecdotal evidence, nor does it agree with WWII or current armor doctrine. Now I don't want to start up a ruckus over this but I'm curious on how BFC handles hull down in CMBB. Statistically Speaking Diceman Link to comment Share on other sites More sharing options...
aka_tom_w Posted August 13, 2002 Share Posted August 13, 2002 Hi Diceman I might humbly suggest you have perhaps overlooked the possibility of an "upper hull hit"? Even hull down tanks in CMBO sometimes take a hit to the upper hull I think. Now it would be my guess (Completely unsubstantiated :confused: ) that the chance to hit the upper hull or the turret might be something like 40 % (upper hull) to 60% turret? But I'm ONLY guessing The possibility to hit the upper hull while in a hull down position must not be discounted. As far as I know Hull Down Status ONLY means your tank will not take a hit in the Lower Hull or in the tracks, when I play of think of my tanks behind an average "garden variety" 3ft tall (1 metre) stone wall, as that is ALL the protection the tank gets in ANY hull down position in CMBO. I mention this because pulling up to (behind) a stone wall in CMBO immediately confirs Hull down status for your tank, BUT it still can take a hit in the upper hull OR turret, (from a enemy unit targeting it from the front presumably, disregarding the loophole of the flank possibly being totaluy exposed to a flank shot from the unprotected sides). -tom w [ August 12, 2002, 09:10 PM: Message edited by: aka_tom_w ] Link to comment Share on other sites More sharing options...
aka_tom_w Posted August 13, 2002 Share Posted August 13, 2002 since this has come up before I did a search and found these threads searching for: "chance to hit hull down HD" this one is good: http://www.battlefront.com/cgi-bin/bbs/ultimatebb.cgi?ubb=get_topic;f=16;t=022294 this one is old but Still GOOD http://www.battlefront.com/discuss/Forum1/HTML/013585.html Date Forum TANK GUN ACCURACY AT SHORT/MED RANGE January 02, 2001 Combat Mission Archive #3 (2001) What reduction in silhouette associated with going hull down ? January 22, 2001 Combat Mission Archive #3 (2001) Max Hit Probability January 14, 2001 Combat Mission Archive #3 (2001) Small Quick Battle AAR January 15, 2001 Combat Mission Archive #3 (2001) Jumbo lies AAR January 23, 2001 Combat Mission Archive #3 (2001) Is the Sherman a tank? January 25, 2001 Combat Mission Archive #3 (2001) "Jumbo" AAR - Spoiler Alert February 21, 2001 Combat Mission Archive #3 (2001) Separate Hull down command March 09, 2001 Combat Mission Archive #3 (2001) SMG March 14, 2001 Combat Mission Archive #3 (2001) Eliminating The Ubertank Problem: Selecting The Right Parameters March 21, 2001 Combat Mission Archive #3 (2001) Hit %: When to Fire that AT-gun? April 04, 2001 Combat Mission Archive #3 (2001) M4 Ubertank Stealth Technology May 02, 2001 Combat Mission Archive #3 (2001) Defending Against the Assault, a CM Guide June 01, 2001 Combat Mission Archive #3 (2001) PzIV tactical suggestions? October 23, 2001 Combat Mission Archive #3 (2001) Schürtzen October 28, 2001 Combat Mission Archive #3 (2001) Hull down - is it worth it? November 15, 2001 Combat Mission Archive #3 (2001) Xerxes' Examples - two AARs December 19, 2001 Combat Mission A [ August 12, 2002, 09:20 PM: Message edited by: aka_tom_w ] Link to comment Share on other sites More sharing options...
aka_tom_w Posted August 13, 2002 Share Posted August 13, 2002 Found Charles' comment on this issue: here: http://www.battlefront.com/discuss/Forum1/HTML/013585-3.html "Big Time Software Moderator posted 12-13-2000 02:51 PM Combat Mission makes an assumption about gunnery: crews are assumed to fire at the "center of mass" of a target, and not to have detailed knowledge of varying armor thicknesses of the different parts of its target. By that I mean that a crew is not assumed to know that, for example, a vehicle has a thinner turret front armor than front hull (which is the case for the Pz IV). Crews do understand basic things like flank armor being thinner than front, and cases where an opponent outmatches them or cannot be killed (e.g. a Stuart faced with a King Tiger). So this can lead to results which might seem odd at first, but make sense when you think about it. Consider the case of a Sherman firing on a Pz IV. If the Pz IV is in the open, the Sherman will fire at its "center of mass", which means that a hit on the hull is quite likely (which let's say will not penetrate, for the sake of argument). If we assume that a hit on the thinner turret armor will penetrate, then the overall chance of penetration is only moderate, as has been observed. If the Pz IV is hull-down, the Sherman is forced by circumstance to fire at the turret (note that some of the upper hull is considered exposed so occasional shells can strike the upper hull of a hull-down target). Most hits will therefore strike the turret, (nearly) guaranteeing penetration. The chance of hitting the Pz IV is lowered, of course. So we have a situation in which the Sherman is actually better off firing at the turret rather than the Pz IV's "center of mass", penetration-wise, because the turret has thinner armor. Given that CM assumes that the Sherman crew does not know this, however, the Sherman will not take advantage of this fact when the Pz IV is out in the open. Only when the Pz IV is hull down will the Sherman do the "right thing" even though it's not aware of it. This is an intentional part of the design. A very, very early version of CM (in the alpha stage, I think) allowed crews to know all the weak spots of all enemy vehicles and make snap decisions about where to aim. And the results were problematic, because we suddenly had "robotic" tank crews that made "perfect" decisions far too often. It also had a strong imbalancing effect because it especially increased the effectiveness of Allied tank crews, since several German tanks have the "varying armor" characteristic in place (Pz IV has 'weak' turret, Panther has 'weak' lower hull, etc.) We were seeing Sherman crews picking off German tanks right in their 'thin' spots at a rate which seemed far too high to be historically accurate. So we took out that "perfect" knowledge. I think it was the right decision, but of course you do get what seems to be the strange result of a PzIV having to think twice before going hull-down in the face of 75mm weapons. It does actually make sense in the big picture though. Charles " Link to comment Share on other sites More sharing options...
aka_tom_w Posted August 13, 2002 Share Posted August 13, 2002 more from the distant past on this issue: http://www.battlefront.com/discuss/Forum1/HTML/013585-3.html Big Time Software Moderator posted 12-14-2000 01:19 AM Well, Mr. Non-Confrontational... let me tell you a thing or three... oops... sorry! I thought you said Mr. Confrontational quote: Steve, crucially it has been demonstrated by a dedicated gamers series of tests at the beginning of this thread that a tank with weaker turret armour than hull armour (King Tiger in that instance) has it's chances of surviving an engagement DECREASED not increased if it is in a Hull Down position. Perhaps in that test. But it is a clinical test that only covered one possible instance which has little to do with a real game situation. The important part to keep in mind is the range. The chance of hitting at 500m is very good. Therefore, even though the HD tank is harder to hit, the chances of hitting are still very favorable. Now... move that tank back to 1000m. The chances of hitting, at least on the first or second shot, decreases to a large degree. And now the smaller target really begins to pay off, making a hit in the first couple of shots much more likely to be a miss. In a direct head to head confrontation, this is a distinct edge. But Simon laid out some other advantages that this test doesn't account for at all. These advantages increase chances of surviving an engagement. Again, not surviving a hit. And that is the critical point of this discussion. The purpose of acheiving a Hull Down position is to get an advantage over the enemy vehicle so you are NOT HIT at all. But like any position a tank takes on the battlefield, there are risks involved. And if they go against you, they will most likely outweigh the benefits. Since the test examaple does not look at the full range of benefits, the risk of a turret hit therefore appears to outweigh the benefits. It is an unfair test in that regard. Oh... and RoF is not lowered for engaging a HD tank. I'll mention it to Charles as it does make sense. However, I don't think there is a way to change the way it works at the moment. Steve IP: Logged Big Time Software Moderator posted 12-14-2000 01:42 AM Just did a VERY quick test. 4xPzIVH vs. 4xM10s (two Hull Down) Range was 1000m, clear weather, all other things equal. Ran the test 4 times. Results were 6/8 non-hull down M10s were killed. Many of them on the first or second shot. In contrast, 2/8 of the hull down M10s were killed. Several of them survived 5 or 6 misses. So... by this quick test I have shown the opposite of the previous test. And that is, being hull down significantly decreases the chance of being hit. Three times more likely in fact. And that is more than just a little bit of a bonus Note that my test is not much more "realistic" than the first test. Instead I did it to illustrate that different circumstances, even on a "test range" yield different results. Meaning... large, generalized conclusions based on a few tests is most often the wrong way to figure out what is what. Steve [This message has been edited by Big Time Software (edited 12-14-2000).] IP: Logged BloodyBucket Member posted 12-14-2000 01:49 AM As an infantryman, I will always try to stay "Hull Down", no matter what anyone says to the contrary! ------------------ "Roll on" IP: Logged Madmatt Moderator posted 12-14-2000 02:03 AM Just to put things in a slightly different light, i have often conducted repeated tests to prove a point or make a suggestion to Charles about the game (yes, some of what you see in-game came from my demented brain BWAHAHAHAH!). How many tests do I usally run before I weigh in on something? 100! Any less just won't produce a good sample at all. Yeah, it's takes a large amount of time to run tests like this but its really the best way to make sure you aren't getting too may outliers in your numbers. Madmatt IP: Logged Big Time Software Moderator posted 12-14-2000 02:31 AM It is also necessary to point out that Madmatt also takes the time to conduct tests that are representative of the types of situations the subject requires to prove his point. Otherwise Charles tells him to go get bent The problem with this hull down test is that it is so HIGHLY variable. Range and vehicle types are a huge factor, but other elements which are harder to do in a "lab test", play a critical role in the risk:benefit ratio. Really, the best test would be to watch 100 different playings of the same scenario and record all the hit/miss info by vehicle type, range, etc. In other words... this one is a difficult thing to simulate in a lab environment. No matter how many times it is repeated. Steve more: Mr. Clark Member posted 12-14-2000 10:42 AM I think that one of the problems here is that there are two ways of looking at this "situation." One is that it is better to be hit non-lethally, and the other is that it is better to not be hit at all. I honestly think that people are getting worked up over nothing here, now that Charles and Steve have explained it. The general purpose of being hull down is to expose less of your vehicle to the enemy, and this is accomplished in the game with a lowered chance to be hit. Obviously, if you ARE hit, it will probably be in the turret. Look at the Infantryman example! If you were sticking your head out of a foxhole, and you got shot, it would be in the head... and thus much more likely lethal. However, if you are running at the enemy in the open, you could be hit in any body part, and thus the hit would have less a chance of being lethal. (EDIT: Yet, men still used foxholes for the cover it provided, lessening the chance of being hit) I've not been in the army, but I know that when in police handgun training, we were told to fire for the center of mass. (Double tap, move to next target.) I'm betting that in the thick of combat, most tankers simply fire quickly at enemy tanks, rather than always trying to zone in on weak spots. (This is only a "guess", as I'm not a tanker that has ever been in combat.) [This message has been edited by Mr. Clark (edited 12-14-2000).] IP: Logged Big Time Software Moderator posted 12-14-2000 01:06 PM MichaelU, correct. Sometimes it is very difficult to look at one factor objectively in a test situation, at least relevant to the game in generally. However, the more a factor is affected by tactics (good and bad), the less likely such a lab test will have real meaning. Basically, you can't draw too many conclusions from such a test. As we can see in this discussion, the observations you made were accurate. The conclusions were also accurate (after a small discussion about probability and how the code works). What was not accurate was its relevance to a real game. The reason is there are many more factors that make hull down beneficial which were not tested for (such as quicker escape, easier to reposition, harder to spot, etc.). So saying that hull down is somehow a disadvantage vs. being in an open field, based on your tests, is unsupportable using your test results alone. Contrast this with an accuracy test. That part of your test was easy to simulate. You really needed to do things at different ranges to base conclusions on the results, but the test itself is a fair representation of battlefield accuracy in static conditions. Why? Because tactics and other dynamically variable circumstances play little role in simple accuracy measurements. So as long as conclusions are drawn from such a test that only speak about accuracy, then the test is valid. Using an accuracy test to say one vehicle is "better" or "worse" than another runs into the same problem as the hull down conclusions. Ben, RoF should be slower for smaller targets. It matters not if the target is a smaller vehicle of the top half of a larger vehicle in hull down mode. I have passed this on to Charles and it has gone onto The List for future inclusion (sometime in the next century ) Mr. Clark has FINALLY put into words what many of us have been dancing around. Excellent example! This is exactly the case. Think of an infantry man instead of a tank. Would you rather be positioned behind a stone wall or standing in an open field? The answer is, of course, behind a stone wall. Yet the chance of dying from a hit (or a near miss even - stone splinters!) behind the wall is FAR greater than dying from a hit in an open field. So for the same reasons an infantry man would rather be behind a wall, a tank would rather be hull down. All things being equal. Steve Big Time Software Moderator posted 12-15-2000 12:10 AM Tom, RoF is already dependent on crew Experience. The only question is, how much TIME should it take to lay the gun correctly for the first shot. This should be calculated based on the crew's experience (as nearly everything is already), distance to target, and visible area of target. I don't know when we will add this, but it will eventually go into the game. Not for CM2 though. Treeburst155, "to hit" is based on a large number of factors. Velocity, shell type, range, a bunch of flight physics, etc. are all included with basic crew Experience modifiers (i.e. Green is bad, Veterans are good ). Optics are not simulated. Don't even ask about this until you do a search There is at least 2 FIVE HUNDRED+ threads on this subject Rex: quote: Mr. Clark all I was trying to suggest is that a Hull Down tank should be more likely to survive an encounter than one in the open. A hull down tank DOES stand a better chance of survival in a battlefield encounter. However, depending on the tank and the shooter, it might not stand a better chance of surviving a hit. Totally different things. As stated several times by several people, these artificial tests are totally irrelevant for assessing hull down effectiveness. See Mr. Clark's point again. Then, reread it again The benefit of being "hull down" is reducing the chance of being hit, not reducing the chance of surviving a hit. You keep pointing to a fact, which is totally realistic and not in dispute, and drawing the wrong conclusions from it. The only way to simulate what you are advocating is to give the hull down vehicle some magical armor bonus. Based on what... I have no idea because it is totally unrealistic to do this. Again... a man standing in a field has a greater chance of being hit, but a greater chance of surviving if hit vs. a man in a foxhole, who has a lesser chance of being hit but a greater chance of being killed if hit. What you are advocating here is that we should give the man in the hole Three Lives or an armor plated head If you still feel that your point is valid, you first need to show why Mr. Clark's point is invalid since it is the exact same principle. Then you would need to conduct tests as I described, which include tactics and not just shooting range conditions. Science is the key here. Without any insult intended Rex, you are not being scientific. Steve [ August 12, 2002, 09:40 PM: Message edited by: aka_tom_w ] Link to comment Share on other sites More sharing options...
aka_tom_w Posted August 13, 2002 Share Posted August 13, 2002 Steve Again: http://www.battlefront.com/discuss/Forum1/HTML/013585-5.html Big Time Software Moderator posted 12-15-2000 06:59 PM M. Hoffbauer wrote: quote: However, it clearly contradicts Steve's opinion that "And that is, even when all other factors are ignored, it is still better to be hull down than to be in an open field." No, it doesn't. As others have pointed out, you have removed a key point. The KTs can not fire back. So eventually the M-18s will kill them all. It is that simple since the KTs are defenseless. The other problem is that you chose one of the largest targets in the game at a very close range (500m). The chance of the M-18 hitting, and killing, is very good no matter what position the KT is in. So coupled with the point above, the results you got are not surprising at all. But they do not contradict what I said in the context in which I said it. MichaelU wrote: quote: Couldn't agree more, but if the reduction in chance of being hit is outweighed by the increase in the chance of a hit penetrating, then you are worse off being hull down. Unscientific conculsion. You can't say this without weighing in the other advantages of being hull down (listed many times in previous posts). In other words, you are bringing in outside factors for one side and not for the other, then drawing a conclusion from this lopsided look. Therefore, your point is invalidated right there. Treeburst wrote: quote: If you are close enough to the enemy, such that your turret alone presents a fat target then I would say being hull down is not so much of an advantage. He will hit you anyway, and it will be in the turret. Well put. I said this above, but I thought it would be good to say it again. The KT test does not measure hull down advantage in any meaningfull way. Kinda like testing out the strength of a bulletproof vest at 3 meters and allowing the shooter to use a high powered rifle and fire as many rounds as he likes. It doesn't matter what he is wearing, the guy with the vest will die, period. So much for testing the value of the vest Rex wrote: quote: This is the kind of situation we are discussing, we need tests where you are unlikely to penetrate the hull but you can penetrate the turret. I can't be bothered to try to explain why again (Sorry it's late over here!) What you still fail to understand here is that penetration has NOTHING to do with the advantages, or disadvantages, of being in a hull down position. It is totally, and utterly, irrelevant. The advantages of being hull down are to avoid being HIT, not to avoid being PENETRATED. If you can not seperate these two elements then we are going to just argue forever. Please... address the point we have made over and over again about the soldier behind cover. So far you have not so much as attempted to explain why it would be better to stand out in a field than to be in a foxhole or behind a wall. It is critical that you do this to prove your point. Now... If the point is "in some circumstances it is less desirable to be Hull Down than in others" the answer is "of course, but only if you lose the gamble". In other words, statistically speaking, it is better to be hull down. My test and Treeburst's test confirm this, as does simple mathematical probability. But sometimes luck does not go in your favor, and therefore BECAUSE OF BAD LUCK being hull down might not be much of an advantage. But it is LUCK (and other circumstances) that are what leads to this, not the fact that you are hull down. And finally... Does anybody here really think there is some sort of need to change Combat Mission's coding? I certainly hope not since there hasn't been any case presented to suggest that we should Steve [This message has been edited by Big Time Software (edited 12-15-2000).] Link to comment Share on other sites More sharing options...
aka_tom_w Posted August 13, 2002 Share Posted August 13, 2002 ok last one sorry for all the old quotes: http://www.battlefront.com/discuss/Forum1/HTML/013585-6.html Big Time Software Moderator posted 12-16-2000 05:35 PM People are discussing the hit probabilities versus hull-down targets and how they relate to fully-visible targets. Jarmo stated that hull-down targets are roughly 30% harder to hit. Everyone please understand this: The increase in difficulty of hitting a hull-down targets is heavily dependent on range. In other words, at short ranges (say, less than 250m) being hull-down isn't much protection at all. A tank's turret (and a portion of its upper hull too - don't forget that!) is still a very big target when the range is short - so it's not difficult to hit it. But at 1000m the difference is signficant. And at 1500m or more, the chance to hit a hull-down tank is roughly halved. That's a huge advantage for the hull-down tank. You can verify these values for yourselves right in the game editor (just draw a line of sight and it tells you the to-hit%). As for our thoughts on hull-down in general, and why there are certain cases where being hull-down doesn't help much (at short range!) what Jarmo said sums it up quite well so I'll quote him here: quote: Suggestion: hulldown should always be favourable. The infantryman example: If the infantryman only shows his head, all hits will hit the head. If the whole infantryman is fully visible, the opponent will likely aim for the torso. Now let's suppose the infantryman has a bulletproof vest but a no-good helmet. From a bit longer range it's obvious you should remain down and hope the opponent misses. From a close range, where the enemy will hit anyway. It might be better to show yourself and sucker the enemy to shoot at your chest, and then pop him before he fires again. Now, just to clarify (not to antagonize because it sounds like many of you already agree with this, I just want to state it for the record): Some have based their arguments on the assumption that hull down should always be advantageous, and because certain field tests don't show this, they feel something is wrong with Combat Mission. But the assumption that hull-down is always good is incorrect. If we assume that Combat Mission's behavior which precludes gun crews from knowledge of whether the target's turret or hull is the weaker target is correct (and therefore aim for center-of-mass) then even in the real world there will be certain circumstances when being hull-down is not advantageous, and may even have a deleterious effect - because you're forcing the gunners to shoot at your (weak) turret when they might not otherwise have done so. There is room for debate on whether gun crews should or should not know the weak points on their targets. But it doesn't seem that anyone is arguing with this. So if you accept that gun crews aim for center-of-mass, then logically it follows that in some cases, at short range, with a tank that has a relatively weak turret, then being hull-down is not always advantageous. That's not a quirk of Combat Mission. That's the real world (assuming gunners fire at center of mass). Charles IP: Logged A Arabian Member posted 12-16-2000 06:00 PM Thanks, charles, for keeping tabs on this thread. I want to clarify - I suspect the BTS stance will be validated - and I COMPLETELY agree that the decrease in "to hit" isn't directly proportional to the decrease in shown SA. I just want to run some stats to convince myself the numbers work. I fully admit this as my problem - I have a hard time thinking of 500 meters as "close", but when you're shooting a 6 kg projectile and have good rangefinding etc., its right next door. Anyway, thanks again for addressing the issues. I've come around to the "company line" on this one I think, and I can be pretty damn hardheaded at times, so I think your case is pretty tight. [This message has been edited by A Arabian (edited 12-16-2000).] [This message has been edited by A Arabian (edited 12-16-2000).] IP: Logged Theron Member posted 12-16-2000 06:29 PM The tests look good. I beleive them, but I am suprised that the 1000m test didn't show that it is an advantage to be hull down. Given Steve's aiming explanation the 500m results make sense. Thanks to the members who did the tests. My computer is close to the minimal system so it takes a while to crank out a tank battle, especially when I need to do it repeatedly Theron IP: Logged Mr. Clark Member posted 12-16-2000 06:46 PM Thanks Charles! ... and to everyone else here, I cannot believe we argued through 6 pages of posts to come to the conclusion that was stated in the opening post... and that CM is fine the way it is, as the quirk (aiming for a weak, hull down turret) is a real life one. IP: Logged Chupacabra Member posted 12-16-2000 06:50 PM quote: Originally posted by Mr. Clark: Thanks Charles! ... and to everyone else here, I cannot believe we argued through 6 pages of posts to come to the conclusion that was stated in the opening post... and that CM is fine the way it is, as the quirk (aiming for a weak, hull down turret) is a real life one. Hrm, you've been on the board since September and you can't believe it? ------------------ Grand Poobah of the fresh fire of Heh. IP: Logged Mr. Clark Member posted 12-16-2000 06:59 PM Heh... Good Point Chupa! IP: Logged Pillar Member posted 12-16-2000 07:00 PM The best part about CM is that the players and the developers have regular opportunity to debate, and shed light on the truth in order to make CM a better product. Of course, the fact that the developers are always right is just more tribute to the work Steve and Charles have done on the game I'm impressed everytime I see this stuff going on. You don't see much of it in other game forums. IP: Logged Jarmo Member posted 12-16-2000 07:11 PM Charles quoted me. IP: Logged Jeff Duquette Member posted 12-16-2000 07:16 PM Shot dispersion at 500 meters should not change weather firing at a small or large target. So if a 90% dispersion pattern has a diameter of 1.5 meters over a target located at 500 meters it will still be 1.5 meters weather the target is 1 meter across or 10 meters across. Therefore if the dispersion pattern remains constant and target size decreases, a corresponding decrease in the probability of achieving a hit should result. What I seem to be reading is that even a target that is presenting a substantially reduced cross-sectional area (in the case of a hull down tank this would be reduced by as much as 60% to 80% of the targets hull-up cross-sectional area) the probability of hitting a substantially reduced cross-section is either not decreased at all – or is decreased only a cursory amount. These are numbers collected by British Army Operational Research Sections during WWII (summarized in WO 291/180) Ranges are in yards, report indicates that the target is assumed to be a approximately the size of a Tiger Ie. Hit probability also assumes no crew error in line or range estimation. Versus a Hull-Up static Target 6 pdr @ 500yrds…..100 percent chance of a First Round Hit (FRH: first round hit) 6 pdr @ 1000yrds…100 percent FRH 6 pdr @ 1500yrds…96 percent FRH 6 pdr @ 2000yrds…87 percent FRH 17 pdr @ 500yrds……100 percent FRH 17 pdr @ 1000yrds….100 percent FRH 17 pdr @ 1500yrds….100 percent FRH 17 pdr @ 2000yrds….98 percent FRH 17 pdr @ 2500yrds….93 percent FRH Probability of a hit on first round, hull down static Tiger Ie sized target, assumed no error in line or range by crew. Versus a Hull-down static Target 6 pdr @ 500yrds…..85 percent FRH 6 pdr @ 1000yrds…43 percent FRH 6 pdr @ 1500yrds…22 percent FRH 6 pdr @ 2000yrds…14 percent FRH Versus a Hull-down static Target 17 pdr @ 500yrds…..88 percent FRH 17 pdr @ 1000yrds…51 percent FRH 17 pdr @ 1500yrds…29 percent FRH 17 pdr @ 2000yrds…18 percent FRH 17 pdr @ 2500yrds…12 percent FRH Radical divergence in both cases from 500 to 1500 yrds [This message has been edited by Jeff Duquette (edited 12-16-2000).] IP: Logged Big Time Software Moderator posted 12-16-2000 08:26 PM Well, I for one still strongly disagree that Hull Down is something that should be avoided based on the turret and hull armor thickness vs. known enemy gun. The reason why is that these tests are far too clinical to carry over to the battlefield tactical assesement. These tests assume (friendly = hull down, enemy = not hull down): - friendly tank knows that it has been spoted - friendly tank knows what the attacking weapon is - the enemy tank is capable of penetrating the friendly's turret armor but not its hull armor - friendly tank has an even or less than even chance of hitting the enemy tank before it hits it - crew experience is either even or in favor of the enemy tank - friendly tank will take no corrective action after being fired upon - enemy tank is not being fired upon by any other weapon friendly to the hull down tank - enemy tank remains in position and doesn't take any corrective defensive action - friendly tank is incapable of shooting (in M. Hoffbauer's example at least) - both vehicles are facing each other dead on - enemy tank is not moving Unless all this situations above are in place, it is statistically better to be in a hull down position REGARDLESS of armor in a real battle in CM. To state this again very clearly... Being hull down is NOT just about surviving a shot fired in its direction. It is all about being in an overall advantageous position. Being hull down has the following advantages 1. Avoid being detected. Easier to see a tank in an open field than in a hull down position. If you see the enemy first, advantage to friendly unit. 2. Avoid being hit. Smaller target presents a lower chance of being hit (NOT penetrated). This allows the friendly tank more chances to hit. 3. Quick cover. Quite often, hull down terrain allows the friendly unit to withdraw and instantly kill the enemy's line of sight/fire. A tank in the open has no such luxury. A perfect REAL first hand example of this was posted a couple of weeks ago. It was a Panther in Normandy (SS div, forget which one). The commander used hull down positions to nail something like 5 Shermans which were out in the open. He utilized the three advantages above repeatedly throughout the engagement. However... at the range in question the Sherman's only chance of penetrating was the turret, not the highly sloped frontal hull armor. Yet, contrary to what some here wish to believe, the Panther commander knew that being hull down gave him the edge he needed. So in conlcusion... it is my strong opinion that avoiding hull down opportunities, simply because of these static tests, is tactically unsound. Steve [This message has been edited by Big Time Software (edited 12-16-2000).] [ August 12, 2002, 09:52 PM: Message edited by: aka_tom_w ] Link to comment Share on other sites More sharing options...
aka_tom_w Posted August 13, 2002 Share Posted August 13, 2002 well, as we can see, most of this has been covered before and there are very specific replies on this matter from Steve and Charles. -tom w Link to comment Share on other sites More sharing options...
Diceman Posted August 13, 2002 Share Posted August 13, 2002 Originally posted by aka_tom_w: Hi Diceman I might humbly suggest you have perhaps overlooked the possibility of an "upper hull hit"? Even hull down tanks in CMBO sometimes take a hit to the upper hull I think... -tom wI see said the blind man to his deaf daughter as he picked up his hammer and saw. So much for assumptions. This time I did a statistical analysis of over 400 hits at a range of 994 meters, 205 hull up, 200 hull down. At this range shot fall and range estimation would be much more critical than at the previously analyzed 547 meters. The method of analysis also relies far less on assumptions. In this case I used a statistically significant sample of over 400 hits. I think the below chart is rather interesting, and answers a lot of concerns including my own. For instance at this range the % chance to hit the turret hull down is only approximately 15% greater than hull up. This does not seem unreasonable to me. There are a couple of statistical anomalies on the chart worth mentioning. For example: hull up gun hits as a % of turret hits is 0.5% higher than the same stat for hull down. Also, there is a statistically significant difference between hull up weak spot hits and hull down weak spot hits. Although the modeled percentage may be different between the two positions, the incidence is low enough that one can assume that a sample of 405 is too low to get an accurate number. The difference in % of gun hits per turret hit can also be written off as sample error. At any rate the below chart should give all concerned a good approximation of hit chances . I also think it backs up Charles' comments on the issue. Link to comment Share on other sites More sharing options...
aka_tom_w Posted August 17, 2002 Share Posted August 17, 2002 anything further here? great thread That was ALOT of hard work on the Part of the Diceman to run all those test and post those charts. THANKS! nice work. -tom w Link to comment Share on other sites More sharing options...
Recommended Posts