RE: Aircraft Maneuverability (Japan added/Updated) (Full Version)

All Forums >> [Current Games From Matrix.] >> [World War II] >> War In The Pacific - Struggle Against Japan 1941 - 1945 >> Scenario Design



Message

Drongo -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 4:05:00 AM)

quote:

ORIGINAL: el cid again
Examination of the database for aircraft indicated two patterns were used:
Neglecting the divisor, these were either

Speed/10

or

Speed/10 + ROC/500.


quote:


There is NO case where a SINGLE ENGINE plane EVER had a maneuverability less than 10% of its speed. Many cases have more. This is the simple case: the divisor is one for a single engine plane.


El Cid,

Could you clarify which d/base you are talking about here? I assume it wasn't the stock d/base as the bulk of the single engine fighters there have stats where their listed manuverability rating < 10% of their listed max speed.

Cheers



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 4:24:07 AM)

quote:

Could you clarify which d/base you are talking about here? I assume it wasn't the stock d/base as the bulk of the single engine fighters there have stats where their listed manuverability rating < 10% of their listed max speed.


I looked at stock and also CHS scenario 155 to see patterns. Good guess.



Drongo -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 7:48:46 AM)

quote:

ORIGINAL: el cid again
I looked at stock and also CHS scenario 155 to see patterns. Good guess.


OK, thanks for that clarification. I was getting a bit confused with your discussion of what the Matrix designers came up with and then seeing you state that your two base manuverability formula had been produced from reverse engineering the database. It sounded like you were saying that they were derived from a direct analysis of the stock Matrix database (only).

Since what I've seen of the aircraft manueverability factors in the stock Matrix database, your formulas don't appear to produce many matches at all, indicating that the designers most likely calculated their manuver values in some other way. So is it correct to assume then that it is the CHS database that produced your stated formula when it was reverse engineered (I haven't yet had the chance to look at that one)?

BTW, just one caveat. WitP Beta testing of air combat did strongly suggest that the listed max speed rating of an aircraft has a sizeable influence on air combat results (in addition to manuever, durability, etc).

This is in line with what is stated (and described with an example) in the manual. If you are aware of some additional knowledge about the game that discounts the max speed factor as having any influence, feel free to ignore my caveat.

I just thought it was worth mentioning as building your manuever rating primarily around speed and ROC might then be expected to double up on the effectiveness of these two factors as they are (supposedly) designed to be an influence on air combat results in their own right (ROC is one of the factors influencing CAP intercepts).

Cheers



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 1:15:07 PM)

quote:

WitP Beta testing of air combat did strongly suggest that the listed max speed rating of an aircraft has a sizeable influence on air combat results (in addition to manuever, durability, etc).


IF my understanding is correct, that makes perfect sense. "Maneuverablity" is really mostly speed. It is either a fraction of speed divided by a factor (for plane size) or it is that plus a fraction of ROC divided by the same factor. Of course testing would seem to show speed is critical - that is how the routine works - and I think it is correct.
And of course durability would matter too - and so does firepower. I always assumed that speed was used directly in the routine - and so I am in harmony with everyone else who ever assumed that. I was surprised to learn otherwise - but it did help me to understand how to rate planes so we could address the air air combat problem. Someone wanted to make sure I understood - so I got told what matters.



Drongo -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 2:21:11 PM)

quote:

ORIGINAL: el cid again

quote:

WitP Beta testing of air combat did strongly suggest that the listed max speed rating of an aircraft has a sizeable influence on air combat results (in addition to manuever, durability, etc).


IF my understanding is correct, that makes perfect sense. "Maneuverablity" is really mostly speed. It is either a fraction of speed divided by a factor (for plane size) or it is that plus a fraction of ROC divided by the same factor. Of course testing would seem to show speed is critical - that is how the routine works - and I think it is correct.

I think you misunderstood. When I said testing strongly suggested that the max speed factor influenced air combat, I meant that it did so not as a component of the aircraft's manueverability factor but on it's own. That is, changing the max speed of an aircraft while leaving it's manuever rating untouched will tend to alter the aircraft's effectiveness in air combat in the same direction as the speed change, ie increase an aircraft's listed max speed (not manuever) and it's effectiveness in air combat will tend to increase.

Now as to your mention of how manuever is calculated, I'm getting confused again. You describe how manuever is calculated but these calculations will rarely match the stock WitP database aircraft manuever ratings. So are you talking about the CHS database or are you actually saying that the game routines themselves will ignore the aircraft's listed manuever rating and instead generate one based on (speed/10 or speed/10 + ROC)?

quote:


I always assumed that speed was used directly in the routine - and so I am in harmony with everyone else who ever assumed that. I was surprised to learn otherwise - but it did help me to understand how to rate planes so we could address the air air combat problem. Someone wanted to make sure I understood - so I got told what matters.


I obviously can't comment on what you've been told but your statement above would seem to put the routine at loggerheads with the way the WitP manual describes how the game uses the manuever and max speed factors in air combat.

It would almost seem possible that you may have been told details of the way one part of the routine works but not neccessarily all the parts.

Anyway, I'm sure whoever told you knows best. [:)]

Cheers



herwin -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 2:44:56 PM)


quote:

ORIGINAL: el cid again

quote:

WitP Beta testing of air combat did strongly suggest that the listed max speed rating of an aircraft has a sizeable influence on air combat results (in addition to manuever, durability, etc).


IF my understanding is correct, that makes perfect sense. "Maneuverablity" is really mostly speed. It is either a fraction of speed divided by a factor (for plane size) or it is that plus a fraction of ROC divided by the same factor. Of course testing would seem to show speed is critical - that is how the routine works - and I think it is correct.
And of course durability would matter too - and so does firepower. I always assumed that speed was used directly in the routine - and so I am in harmony with everyone else who ever assumed that. I was surprised to learn otherwise - but it did help me to understand how to rate planes so we could address the air air combat problem. Someone wanted to make sure I understood - so I got told what matters.


There are a number of elements to maneuverability, but the major one is sustained rate of turn. This is driven by the interaction of power, wing loading, and drag.

About 25 years ago, I decided I wanted to design an air-to-air combat game that modelled the technical factors correctly. I found it had to keep track of energy, but it allowed me to model maneuvers (in both horizontal and vertical dimensions) and performance realistically. Later, when I did my PhD research into how bats capture insects in flight, that work helped me develop an accurate model of how bats fly and maneuver. Insects can react much faster than bats, but bats respond by using scissors maneuvers to turn inside the insects. The usual action of an insect on hearing a bat cry is to dive, but it has to time its dive very carefully. If it dives too early and picks up airspeed, it lacks the power to maneuver, and the bat can chase it down. If it delays its dive too long, it's dinner.



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 9:58:48 PM)

quote:

I think you misunderstood. When I said testing strongly suggested that the max speed factor influenced air combat, I meant that it did so not as a component of the aircraft's manueverability factor but on it's own. That is, changing the max speed of an aircraft while leaving it's manuever rating untouched will tend to alter the aircraft's effectiveness in air combat in the same direction as the speed change, ie increase an aircraft's listed max speed (not manuever) and it's effectiveness in air combat will tend to increase.


I understand (and understood) your meaning. I am treading on thin ice and must be guarded in my response: that is not how the routine works.
My knowledge is not based on analysis - it is based on technical fact about the routine. The speed field is NOT used at all for this purpose.
[It may have some other function, such as when calculating the chance of an intercept to see if there is air combat at all. It may only be so the plane data charts look right to players. I don't know. It is similar to the "max load" field. This field IS used by transport planes - apparently we "cheat" and ALWAYS carry maxload to transfer range when we air transport! But the max load field is NEVER used by bombers for bombload! It is used to decide about the base required for bombers to work as bombers, but it is not used for bombing itself! It is the way it is, right or wrong, like it or not, it just is and we cannot change it.] I will say this again - and I am not guessing - speed is in the routine ONLY insofar as it is a component of maneuerability.



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 10:08:11 PM)

quote:

Now as to your mention of how manuever is calculated, I'm getting confused again. You describe how manuever is calculated but these calculations will rarely match the stock WitP database aircraft manuever ratings. So are you talking about the CHS database or are you actually saying that the game routines themselves will ignore the aircraft's listed manuever rating and instead generate one based on (speed/10 or speed/10 + ROC)?


I did a "quick and dirty" reverse engineering of the field values. My work is based on an unissued version of CHS - with spot checks back into CHS scenario 155 and stock scenario 15. You cannot look at the data I used because Andrew has not quit issued it yet. [It is in proof release to the CHS team now - I got my copy last night and have not yet examined the latest form]. But this IS based on analysis - and a very fast and limited one at that. I made some guesses - and found many points of agreement with both - and decided this probably was correct. Regardless, these guesses led me to a method that is very simple and works remarkably well. I don't think I invented it - I think it was invented at Matrix - and if that isn't so probably by someone at CHS - and I have tried to say how impressed I am such a simple value works in such a simple routine to show the differences for all planes so well. But maybe I was totally confused and invented it all by mistake and accident? [I did run into the theory there was NO method at all - that it was all "seat of the pants guesses" - and I think any modder must have done that because there was no stated definition to use. But maybe some modder invented his own definition and I just decoded it? Does it matter? No longer - we have a way to get the value for each plane that works to show in a reasonable way the differences for all types of planes. I at least do not care. For political reasons - and to keep my ego in line - I prefer to think Matrix is peopled by brilliant programmers who came up with a very clever idea - and if that is false - that CHS has another genius besides Andrew the mapmaster - who invented it. I am not believing I came up with this.]



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 10:12:34 PM)

quote:

I obviously can't comment on what you've been told but your statement above would seem to put the routine at loggerheads with the way the WitP manual describes how the game uses the manuever and max speed factors in air combat.


In US Navy ET school they tell you "You may end up alone, in a submarine cut off from communcations or some other isolated place, working on something with NOTHING but the manual - but that is enough." I found it to be true. But WITP does NOT have a manual (four inches or so thick) like I used to use on Navy gear. It is very frustrating for me. I think if something is wrong in the manual there is a moral obligation to fix it. But obviously my idea is not profitable or otherwise germane: Matrix thinks it is fine to say "you can mod this game" without telling you how it works! I proposed that they create an editor's technical manual and charge for it - but either they have not done - or it is in the works and a secret from me. MANY things in the manual either were wrong in the first place, missing altogether, or have changed without a revision of the manual being made. I wish it were not so. I would pay for it not to be so. But it is so and we must live with it. Probably forever, but at least for now. Fact of life.



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/2/2006 10:17:59 PM)

quote:

There are a number of elements to maneuverability, but the major one is sustained rate of turn. This is driven by the interaction of power, wing loading, and drag.


Try not to confuse reality with WITP! Names of fields are often quite misleading. "Accuracy" is NOT accuracy, for example. I didn't name it and it isn't my fault. Similarly, "maneuverability" is not exactly maneuverability as I would define it - and I used to care for aircraft simulators costing 7 figures you could climb inside (nice ones with moving scenery out the windscreen, and the whole thing turns to simulate the way it feels in a cockpit). I think you need to think about power loading, wing loading, rate of turn, lots of things. This has nothing whatever to do with the value called maneuverability in the game - except to the extent those things are related to ROC - and that only if I decoded the field values properly. Clearly many cases are PURELY speed divided by 10 for single engine planes and divided by 20 for twin engine bombers/transports. and divided by 80 for all four engine planes.

When I said "maneuverability is" I mean, here, in this code, for our purposes, not "maneuverability really is." I understand what you are saying, and I have a different meaning in my own simulations.



Drongo -> RE: Aircraft Maneuverability (Japan added/Updated) (3/3/2006 10:09:59 AM)

quote:

ORIGINAL: el cid again

quote:

I think you misunderstood. When I said testing strongly suggested that the max speed factor influenced air combat, I meant that it did so not as a component of the aircraft's manueverability factor but on it's own. That is, changing the max speed of an aircraft while leaving it's manuever rating untouched will tend to alter the aircraft's effectiveness in air combat in the same direction as the speed change, ie increase an aircraft's listed max speed (not manuever) and it's effectiveness in air combat will tend to increase.


I understand (and understood) your meaning. I am treading on thin ice and must be guarded in my response: that is not how the routine works.
My knowledge is not based on analysis - it is based on technical fact about the routine. The speed field is NOT used at all for this purpose.

I fully accept you have been told something about that particular routine but I still strongly suggest that an aircraft's max speed rating on it's own is very influencial in determining air combat results.

From what I've seen of your comments, you imply that a manuverability factor (possibly a differential) is used to determine whether an aircraft can attack another aircraft and possibly the conditions of that attack.

Without knowing the actual details of what you were told, I am wondering whether the manueverabilty factor used in the routine is actually a different value to the listed manueverabilty rating from the d/base and is instead a calculated variable passed to the routine that is based on multiple factors, only one of which is the manueverability rating for aircraft as listed in the database. If that is not the case, then perhaps max speed is one of the factors used to influence conditions prior to the determination of whether an aircraft can attack and then again immediately afterwards when determination is made whether an aircraft can evade an attack.

The reason I'm suggesting this is because testing of air combat in WitP (up to and including the current release) has shown when all other factors are equal (including listed manueverability for opposing aircraft), that a difference in listed max level speeds between opposing aircraft will show a favourable trend in results for the faster aircraft. That means that an aircraft's listed max speed rating plays its part in determining the outcome of air combat in WitP, just like an aircraft's listed manueverability rating does. I would strongly recommend you take this on board when deciding on your aircraft stats and not just assume that an aircraft's listed manueverability in the game is the primary factor of air combat.

quote:


[It may have some other function, such as when calculating the chance of an intercept to see if there is air combat at all. It may only be so the plane data charts look right to players.....I will say this again - and I am not guessing - speed is in the routine ONLY insofar as it is a component of maneuerability.

As I outlined above, an aircraft's listed max speed may or may not be one of the factors used in the calculation of the "manueverability" variable used in the routine you mention.

If you are suggesting listed max speed is not used in that particular routine then I'll accept that but then suggest it must be in other air combat routines that you have not seen. As I said, my testing experience indicates it's there somewhere.

On the other hand, when you talk about about speed being a component of manueverability, do you mean (as you suggested with your formula (speed/10 etc)) that the speed has already somehow been built into the listed manuever ratings of aircraft in the database? Your formula certainly failed to show any connection between an aircraft's listed max speed and listed manuever for the stock Matrix database, a database designed specifically with the game routines in mind.

Either way, I feel that you may be making some assumptions without all the relevent information (much like I have to [:)]) and would certainly suggest some basic testing occur before you go further. Perhaps something simple like combat between two identical aircraft types and then run it again but this time give one of the aircraft a big increase in it's max speed (something like 100 extra speed should prove whether or not listed speed plays any direct role in air combat). You might be surprised.

Good luck anyway.



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/3/2006 12:35:33 PM)

quote:

I fully accept you have been told something about that particular routine but I still strongly suggest that an aircraft's max speed rating on it's own is very influencial in determining air combat results.


I will stipulate this is correct and concur in detail with you with this one qualification:

This is true in the EXCEPTIONAL case BOTH sides are aware of each other. The really critical factor in air combat is surprise: it wins both offensively and defensively 90% of the time. IF you see him and he does not see you, you either run away and are not seen at all (90% of the time), or you engage in a way that gives you an advantage before you are seen (90% of the time).

But in the situation where you both know about each other - or where you have to let him know about you (say you are closing a target defended by CAP and/or interceptors) - it is clear speed matters. So does maneuverability in a non speed sense - but not as much as being able to move fast (mostly).

I am not really interested in a debate on this point. We agree - so there is nothing to debate. I am only interested in getting this particular game to work as well as possible. The routine is what it is - and until I get to see the code I am going to accept it and work within that context.



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/3/2006 12:43:42 PM)

quote:

Without knowing the actual details of what you were told, I am wondering whether the manueverabilty factor used in the routine is actually a different value to the listed manueverabilty rating from the d/base and is instead a calculated variable passed to the routine that is based on multiple factors, only one of which is the manueverability rating for aircraft as listed in the database. If that is not the case, then perhaps max speed is one of the factors used to influence conditions prior to the determination of whether an aircraft can attack and then again immediately afterwards when determination is made whether an aircraft can evade an attack


I am not sure why you want to go into this? I am only dealing with a narrow module involked in a rather complex routine dealing with possible air combat. All I want to do is get the data right in the fields of the records used by the air combat module itself.

But you are correct: wether or not there is air combat at all is a function of other modules - and those modules undoubtedly use speed in some sense. [I personally think that often it is cruising speed that will matter - but it may be both crusing speed and maximum speed can be used in different situations]. Other things matter too - for example visibility - is it day or night? And the WITP philosophy almost guarantees that the proficiency of the pilots, and their state of tiredness, will also matter. But may I point out that to the extent either kind of speed - max or cruising - is used in these routines which determine if you engage or not - they will work perfectly if we put in correct speed data in the fields.



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/3/2006 12:50:10 PM)

quote:

The reason I'm suggesting this is because testing of air combat in WITP (up to and including the current release) has shown when all other factors are equal (including listed manueverability for opposing aircraft), that a difference in listed max level speeds between opposing aircraft will show a favourable trend in results for the faster aircraft. That means that an aircraft's listed max speed rating plays its part in determining the outcome of air combat in WITP, just like an aircraft's listed manueverability rating does. I would strongly recommend you take this on board when deciding on your aircraft stats and not just assume that an aircraft's listed manueverability in the game is the primary factor of air combat.


I am confused by your words here: what more can I do than insure the speed data is perfectly correct? It was astonishingly wrong in many cases too - like the more than 60% error in the Empire flying boat max speed I found today. Fixing it can hardly make things worse. Actually, you have the situation backwards: I was ignoring maneuverability - I didn't know the definition and I didn't know if it was used at all - and I assumed it would have a minor role (secondary to speed) if it was. I got told to mend my ways - that speed is not used at all in the critical module - and getting maneuverability right was going to be critical to getting air air combat ratios right for different planes. When the advice is informed, I always listen. I do not understand what else you think I can do? I need to make a maneuverability rating that is better than "seat of the pants."
And I already have correct speed data (or will fix any case I do not when it comes to my attention).



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/3/2006 1:05:22 PM)

quote:

On the other hand, when you talk about about speed being a component of manueverability, do you mean (as you suggested with your formula (speed/10 etc)) that the speed has already somehow been built into the listed manuever ratings of aircraft in the database? Your formula certainly failed to show any connection between an aircraft's listed max speed and listed manuever for the stock Matrix database, a database designed specifically with the game routines in mind.


I mean exactly that: "speed has already somehow been built into the listed manueverability ratings of aircraft in the database." I have no clue to your belief my formula "failed to show any connection" -- first this is bad English - formulas do not "show connections" - and second if I follow your meaning rather than what you said - the data says (within the limits of all WITP data - which is somewhat sloppy) exactly what I say it says.
More than that, if my formula is the product of insanity (that is, not dervived from the data I derived it from), it does not matter a whit: it works and better than I would have regarded possible for any such simple device. Since we are limited to a single two digit field, we do not have the option of doing anything very much more in the first place. At this time the critical thing is to show the air air combat routine is better than it used to be and that it does give advantages to good planes over bad ones. IF it works well, THEN I will consider a more sophisticated formula to get it - limited by the kind of data available. [We do not often get sustained rate of turn data to work with - and we can hardly go measure it - so we may not be able to use that - even if we want to. We need it for ALL plane types - or we cannot use it.] I am perfectly willing to use a different formula which I do not write too - if it produces a result in the range from about 3 to 60. If it can be shown it differentiates by type better than the formula I am using, I have no shame at all, and I will use it. But since the routine I am concerned with does not use speed directly, we must insure that it is in the value that is used.



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/3/2006 1:11:33 PM)

quote:

Either way, I feel that you may be making some assumptions without all the relevent information (much like I have to ) and would certainly suggest some basic testing occur before you go further. Perhaps something simple like combat between two identical aircraft types and then run it again but this time give one of the aircraft a big increase in it's max speed (something like 100 extra speed should prove whether or not listed speed plays any direct role in air combat). You might be surprised.


Here I have what you may regard as good news. I am a former professional test technican. I also have an actual test team of volunteers.
I believe in testing in a sense that has not been done by either Matrix or CHS. And I am at this time in process of testing the values of the aircraft which are tentatively assigned. If anyone wants to test using his own ideas they only need to send me their address and I will sent the particular files needed. The problem is that we are forced to test in what I regard as an unprofessional context: we are not allowed to know all we should know to design difinitive tests. Lots of factors we do not understand will mask the results. For this reason, the greater the variety of tests we run, the more likely we will understand what the situation may be. But, and for the first time, we have a test oriented module in process.



Drongo -> RE: Aircraft Maneuverability (Japan added/Updated) (3/3/2006 6:21:11 PM)

quote:

I mean exactly that: "speed has already somehow been built into the listed manueverability ratings of aircraft in the database."

Which is why I was questioning what you were saying earlier. I was seeking clarification as to whether you were talking about the ratings in a particular database or all of them and why you thought it was as you described.

You had mentioned that you had been told about what happened in a particular air combat routine. That's fine but you were also making statements about how the listed aircraft manuever ratings were established in the database, which is a seperate issue. Since you hadn't mentioned being told by someone in the know about this, I naturally wanted to know why you felt so confident that speed was being represented within the listed d/base manuever rating.

quote:


I have no clue to your belief my formula "failed to show any connection" -- first this is bad English - formulas do not "show connections" - and second if I follow your meaning rather than what you said - the data says (within the limits of all WITP data - which is somewhat sloppy) exactly what I say it says.


Actually, it would be bad terminology rather than bad English but the use of the latter by you does imply that it has been taken out of context.

The context I was suggesting is that your formulas could be somehow used to prove that speed was already part of listed manuever in the databases and that it was a very sizeable component of that factor, even to the point of being the sole determinant in some cases.

In the absence of further clarification from you at the time, I assumed you may have been also referring to the stock Matrix database when discussing the formulas and listed manuever components. When the formulas are used with the listed max speed and ROC for an aircraft in the stock database, they rarely produce a result to match the same aircraft's listed manuever rating. I thought that was important to mention (given your statments in this thread) in case you were suggesting that your formulas, when used with the stock database would produce results that matched an aircraft's listed manuever rating.

Anyway, you just stated above that speed is already factored into the listed aircraft manuever rating "within the limits of all WitP data". In relation to the stock database, I obviously disagree but it matters not a whit.

quote:


More than that, if my formula is the product of insanity (that is, not dervived from the data I derived it from), it does not matter a whit: it works and better than I would have regarded possible for any such simple device.


My intention was not to question your sanity but rather to seek enlightenment on certain higher matters.

I think that has been achieved.

Cheers







el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/4/2006 2:13:52 AM)

For stock Scenario 15

My data matching routine produced the following points of perfect agreement for Formula A: MaxSpeed/10 for single engine case or
MaxSpeed/20 for twin engine case:

Type........MaxSpeed....Maneuverability....Formula Result

N1K1-J.....363..............36.......................36
D3A.........240..............24.......................24
G4M1.......266..............13.......................13
Fulmar.....280..............28.......................28
F4F3........325..............33.......................33
F4F4........320..............32.......................32
Mohawk...302..............30.......................30
Hawk75...302..............30.......................30
CN21B.....314..............31.......................31
Brewster..288..............29.......................29
P36A........302.............30.......................30
B-26B......310..............16.......................16

It produced the following exact matches for Formula B:
MaxSpeed/10 + ROC/500 (this total divided by 2 for twin engines):

A6M2N........270............32.....................32
D1A............192............21.....................21
B4Y............173............19.....................19
SeaGlad......245............29.....................29
Lkhed212....214............12.....................12
Lysander.....219............25.....................25
Whirraway...205............24.....................24
BlenheimIF..285............16.....................16
TIVa...........161............18.....................18
BlenheimI...285............16.....................16
BeaufortI....265............14.....................14
BeaufortV....265............14.....................14
B18A...........215............12.....................12
IL4C............261............15.....................15
IL4..............261............15.....................15

The number of points within one of the formula results was significantly higher than the number of exact matches. I concluded the results from the formulas, derived from CHS data, were in the same range as much of the stock data.

However, on visual examination of stock values, I see that many
cases involve maneuverability values less than MaxSpeed/10 for single
engine planes or half that for twin engine planes - which cases could not
be correct by my formulas (although they might be within a point in some cases). It is indeed possible that the values in the table are "seat of the pants" (as was suggested as a hypothetical possibilty by a mathmetician on the basis of what he expected game designers to do) or are based on some other algorithm. It is equally clear they used something completely ficticious for twin engine fighters and night fighters - in the sense that they are not similar to otherwise similar bombers or transports with twin engines. I also note that cases of four engine aircraft I manually checked (B24D, PB4Y and LB30) indicate that where my calculated value (in all cases) was 8, the value in the field in the table was 4. I assumed this was deliberate, to give heavy bombers a greater chance of being shot at, and little chance to close successfully and shoot offensively. Again, this was an assumption, and it may not have been correct. It no longer is germane. I decided I liked formula B since it said something besides speed and engine count matters. I decided to cut the result in half to reduce air air lethality. And then I decided to double the relative value of the non-speed factor - in effect creating a new formula:

Speed/20 + ROC/500 (this total always divided by engine count, no exceptions unless two engines are tandem, reducing the net drag, which is hypothetical: no such case in our data set).

The results are a superb but simple relationship which testing seems to indicates works well in this routine. I no longer care if the reasoning used in determining it was entirely correct or accidental. We have a way to calculate the field value which is easy, fast, consistent, and yet shows differences for performance and configuration which are reasonable.




akdreemer -> RE: Aircraft Maneuverability (Japan added/Updated) (3/4/2006 4:27:21 AM)

quote:

ORIGINAL: el cid again


Speed/20 + ROC/500 (this total always divided by engine count, no exceptions unless two engines are tandem, reducing the net drag, which is hypothetical: no such case in our data set).

The results are a superb but simple relationship which testing seems to indicates works well in this routine. I no longer care if the reasoning used in determining it was entirely correct or accidental. We have a way to calculate the field value which is easy, fast, consistent, and yet shows differences for performance and configuration which are reasonable.




However, for high performance twin engined fighters like P-38, F7F and the like what would the results be if you did not devide by 2?



witpqs -> RE: Aircraft Maneuverability (Japan added/Updated) (3/4/2006 4:33:45 AM)

Just a thought about where the data might have originally come from.

1) They used a formula (either exactly or similar to what you described).

2) They then made adjustments for things like 2 engine fighters.

3) They then made adjustments on a case by case basis to achieve what were believed to be historically accurate capabilities (tweak this one up, that one down, etc.).

4) Someone else came along, without knowing about steps 1-3, and defined new aircraft by comparing them to existing aircraft and adjusting as needed.

5) CHS Modders came along, without knowing about steps 1-4. They made adjustments (as in step 3) according to thier beliefs, and the added aircraft (as in step 4) again according to their beliefs.

6) Step 5 continues up to CHS release 1.6 (current release).

Now, you come along and try to decode the data.

Also, at every step of the way, each person had too little time to do what they really wanted in terms of testing, vetting data, etc.

This is just my theory, but it makes loads of sense to me having plenty of experience in business that parallels it. Without fail someone comes along 'today', after years or even decades of work already put in by others, and gets a headache trying to figure it out!



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/4/2006 5:48:16 PM)

quote:

However, for high performance twin engined fighters like P-38, F7F and the like what would the results be if you did not devide by 2?


They would be about twice as high (neglecting the impact of fractions which might subtract one from a doubling in some cases). Note, however, this is not just true of "high performance twin engined fighters" - it would be true for ALL twin engined (and presumably four engined) planes. The RELATIVE advantage of a twin engine fighter vs a twin engine bomber would be the same. The relative advantage of a twin engine fighter vs a four engine bomber would only be HALF. This would distort the reality insofar as twin engine night fighters really were effective against four engine bombers, even if they were only marginally better than twin engine bombers (which mostly they derived from). A high performance twin engine plane is modeled here more or less correctly: it is BETTER than a twin engine bomber - because it has a better power ratio and better ROC. Note I have increased the proportion of the non-speed component in the maneuverability rating, and this in particular benefitted high performance twin engine fighters. Heavier planes stayed anchored at the same values, but these gained points.

And similarly a 4 engine plane would be four times as high.

This results in preposterous values. Further, if you take the cases of planes which were used in more than one role - in particular bombers that were able to operate as night fighters - or night fighters that were intruder bombers - you will find that the maneuverability of these planes is in the same class as other twin engine aircraft, not single engine aircraft. Yet in the stock system, such aircraft were rated entirely differently - they divided a B-25 by 2 but not a twin engine fighter or night fighter. This was deliberately to address the concerns of those who wanted better performance for fighters I am sure. But it is the wrong way to proceed: Joe (who knew he was going to be able to do code) said "Get the database right first and we will get the code right next and then everything will be right." An aircraft is represented by a set of values - not one value. You may not correctly think of a plane if you are not considering ALL its fields at the same time. P-38s are going to do well because (a) their maneuverability is good relative to other twin engine aircraft and very good vs four engine aircraft; (b) they have great firepower - when they do get to shoot head on it is not nice to be in the way; (c) they have higher durability than any single engine plane - if they get hit they are much more likely to stay around for another round - or at least to go home and survive. PV-1 Ventura night fighters are much more marginal because their maneuverability is less - in spite of also having fair firepower and durability ratings.

That said, it would be better to have a different algorithm for maneuverability. Since we cannot really test all these 240 planes to measure values, we probably cannot use the ideal data set: sustained rate of turn data is not recorded for most of them. But we can easily use weight, power and wing area to calculate wing loading and power loading - and we could combine these with initial ROC data in some way. HOW to do that well could take a long time to understand - but in principle this is better than the simplistic algorithms I have discussed here - and it is my plan to consider this for the next round of RHS. Right now I want a system that is based on real data and says, approximately, that maneuverability is better for small planes than big ones, etc. This to permit testing of the entire revised set of weapons, durability etc.



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/4/2006 5:56:44 PM)

quote:

) They used a formula (either exactly or similar to what you described).

2) They then made adjustments for things like 2 engine fighters.


I think you are correct. I think they made adjustments for night fighters and also day fighters with 2 engines. Also for biplanes. And in the opposite direction for at least some 4 engine planes.





el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (3/4/2006 5:58:41 PM)

quote:

1) They used a formula (either exactly or similar to what you described).

2) They then made adjustments for things like 2 engine fighters.

3) They then made adjustments on a case by case basis to achieve what were believed to be historically accurate capabilities (tweak this one up, that one down, etc.).

4) Someone else came along, without knowing about steps 1-3, and defined new aircraft by comparing them to existing aircraft and adjusting as needed.

5) CHS Modders came along, without knowing about steps 1-4. They made adjustments (as in step 3) according to thier beliefs, and the added aircraft (as in step 4) again according to their beliefs.

6) Step 5 continues up to CHS release 1.6 (current release).

Now, you come along and try to decode the data.

Also, at every step of the way, each person had too little time to do what they really wanted in terms of testing, vetting data, etc.

This is just my theory, but it makes loads of sense to me having plenty of experience in business that parallels it. Without fail someone comes along 'today', after years or even decades of work already put in by others, and gets a headache trying to figure it out!



This analysis duplicates almost exactly that of Joe Wilkerson about a month ago. And I concur in detail (using formal military parlance).



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (5/10/2006 10:59:14 PM)

quote:

However, for high performance twin engined fighters like P-38, F7F and the like what would the results be if you did not devide by 2?


The result would be exactly the same as for single engine planes. I find this appropriate ONLY in the case where the twin engine plane has coupled engines (doubling output on one shaft) or dual-counter-rotating propellers driven by different engines or one engine forward and one aft.
Engines on the wings result in a considerable change in the performance of the aircraft and, while it is not purely a case of divide by engine count, it is so close as to be remarkably good as a rule of thumb.



witpqs -> RE: Aircraft Maneuverability (Japan added/Updated) (5/10/2006 11:42:57 PM)

Understood - just a quick note - the P-38 had counter-rotating props so their torque cancelled out.



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (5/11/2006 6:41:28 AM)

quote:

Understood - just a quick note - the P-38 had counter-rotating props so their torque cancelled out.


That is of course a good thing. The problem is that you have two separate sets of propeller discs and THREE sources of drag (other than the wings and tail): the main body where the pilot sits, and both engine naciles. A single engine plane has only ONE such source of drag - for both engine and main body. Similarly, if the two engines are in the body, and drive one hub - or even a prop aft and one forward - it is only one source of drag. Different proposition than a multi-engine plane of the classical sort.



witpqs -> RE: Aircraft Maneuverability (Japan added/Updated) (5/11/2006 7:38:10 AM)

Absolutely, they could not make it as good as a single-engine fighter. My point is that they designed in factors so that the P-38 was at least better off than a two-engine bomber.



RevRick -> RE: Aircraft Maneuverability (With Corrected Data) (5/11/2006 2:55:42 PM)

It should be pointed out that a P-38 with a takeoff weight of 19,500 pounds had either two 2000 lb bombs or two 310 gallon drop tanks on board.



RevRick -> RE: Aircraft Maneuverability (Japan added/Updated) (5/11/2006 3:26:06 PM)

quote:

ORIGINAL: el cid again

quote:

Understood - just a quick note - the P-38 had counter-rotating props so their torque cancelled out.


That is of course a good thing. The problem is that you have two separate sets of propeller discs and THREE sources of drag (other than the wings and tail): the main body where the pilot sits, and both engine naciles. A single engine plane has only ONE such source of drag - for both engine and main body. Similarly, if the two engines are in the body, and drive one hub - or even a prop aft and one forward - it is only one source of drag. Different proposition than a multi-engine plane of the classical sort.


However, applying the same algorithm to the P-38 that you apply to the B-25 is a little bit ridiculous on the surface if you even take more than a casual look at the aircraft. One is obviously not going to have the drag coefficient that the other has, or it would not have had 150+ of the top speed. It may have something to do with the coding of what type of AC it was as well. I would hope that Matrix would have figured out that applying the same code to a twin engine fighter that they would to a twin engine bomber would be a little absurd, sort of like being a little bit pregnant.

BTW - only data I have so far on climb rates are
P-38F - time to 20K feet - 5.94 min.
P-38J - rate of climb (SL) 4000 fpm
P-38J - rate of climb (critical altitude) 2900 fpm @ 23,400'
time to critical altitude - 6.19 min.
Service Ceiling - 40K feet



el cid again -> RE: Aircraft Maneuverability (Japan added/Updated) (5/11/2006 3:36:13 PM)

quote:

Absolutely, they could not make it as good as a single-engine fighter. My point is that they designed in factors so that the P-38 was at least better off than a two-engine bomber.


Yes they did. And it was superb. It had the things that mattered, in particular in PTO - where it was the highest scoring Allied fighter of all.
Not because it was better at fighting than a Corsair or a Hellcat - but because it could get there and outperform enemy planes of many kinds - and at least compete with fighters if it took advantage of things like diving and then running at high speed. A free shot - with good firepower - is a nice thing.

But I think it is asking a lot of as simple algorithm to know the difference between one 2 engine plane and another. In this case it DOES know the difference - compare the P-38 with ANY 2 engine bomber - it is much more maneuverable - it has much more firepower - it has more speed and ROC - and it is as durable. Not really bad for a simple system I think.



Page: <<   < prev  1 [2] 3   next >   >>

Valid CSS!




Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI
1.25