Aircraft ROC Review

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

Moderators: wdolson, Don Bowen, mogami

el cid again
Posts: 16983
Joined: Mon Oct 10, 2005 4:40 pm

RE: Aircraft ROC Review

Post by el cid again »

ORIGINAL: witpqs

Thank you. Based on the comparisons I saw earlier, I doubt that 1 point is adequate (for the L model), but how to convince you?

Start thinking in terms of the SET of fields which defines an aircraft - and testing the aircraft in special games where you set it up for determining results. It ought to clean the clock of ALL NON single engine types - and be very competative with single engine fighters. It should "win" vs other fighters because it survives when they do not - not because it outmaneuvers them. Try it - and I bet you like it.

If you ONLY focus on maneuverability - you will never like any rating that is fair.

And if you ONLY focus on subjective standards - you will never have a basis to say how many points are reasonable to add.

But here is one round of ammunition for your argument: the L has POWERED flaps - and that had SOME impact on maneuverability. It MIGHT be possible to justify a SMALL number of points due to them. In my view only if for some reason the current ratings are not working (e.g. one?). Just abstractly thinking about them is not going to tell us they are not working.
User avatar
witpqs
Posts: 26376
Joined: Mon Oct 04, 2004 7:48 pm
Location: Argleton

RE: Aircraft ROC Review

Post by witpqs »

ORIGINAL: el cid again
ORIGINAL: witpqs

We cannot devise a formula that will be "perfect" for all aircraft.

Of course not. That is why it is important to use the formula first, then make any meaningful modifications needed. Being slaved to the results of a formula is a mistake.


Actually, this is backwards: being "slaved" to the results of a formula that works reasonably well is a matter of principle, of fairness, and not being swayed by emotional perceptions of anyone is a very good idea.

My position is significantly different - but does allow us to agree - because you used the word "meaningful." For me - the burdon of proof is on an advocate of why this or that model is a "meaningful" enough exception to break the rule. It is a high hurdle - so we don't have every fanboy jumping in for his favorite just because he likes it. You need to have a basis for such a change.

Here - now I have given you the current ratings for all P-38 flavors we use - I suggest you review them - and see if you find them reasonable - which I suspect you will. If so - end of discussion. If not - well the burdon is on you - what is not reasonable - to what extent - and why? I don't think you can use these planes in WITP in any reasonable way and not regard them as effective. They have EXACTLY what P-38s ought to have - great protection - powerful armament for both air and surface targets - the best durability of all - and fabulous range. They also vie with each other for the title of "best maneuverability rating for a two engine aircraft" - as time passes the record simply passes from one P-38 variant to another. I don't think you can face them in WITP and not have your heart sink when they appear on your screen.


I will only be able to form an opinion of the new ratings when I see them in relation to the other aircraft that they were tested against and fought against.

BTW, no emotion involved in modifying the results of a formula. If you want a formula that can yield accurate results out of the box then you must mathematically model the entire aircraft, airflow, etc. Today they barely do it at Boeing and Airbus, and they still do wind tunnel and flight testing. Experience still shows them what those amazingly complex collections of formulas get wrong to various degrees.

With a simple formula you should expect that there will be likely exceptions. I agree the hurdle to make adjustments must be high. You will note that in addition to the anecdotal evidence other posters cited, one of them also cited actual test results versus other specific models of aircraft. That gives us a way to compare the formula ratings of the aircraft tested and see if the formula results are close to the results achieved in actual testing or should be adjusted. That was a very important post, worthy of re-reading.
el cid again
Posts: 16983
Joined: Mon Oct 10, 2005 4:40 pm

P-35A

Post by el cid again »

Max speed 310 mph, initial ROC 1920 fpm results in a maneuverability rating of 23 (down from 24). Endurance should be 219 minutes. Treated as eratta and backfitted because of max speed and range errors.

P-36A: Max speed 313 mph, initial ROC 2500 fpm results in a maneuverability rating of 26 (down from 30). Endurance should be 184 minutes. Treated as eratta and backfitted because of max speed error. We had far too high an initial ROC value at 3400 fpm before. But the increase in speed helped offset the reduction in ROC.

Hawk 75: Initial ROC unchanged at 2500 fpm, max speed still 302 mph, but recalculation yields maneuverability = 26 (up from 23) - so treated as eratta and backfitted. Guns increased from 4 to 6 x 30 cal - also treated as eratta.

Axis Hawk: Initial ROC unchanged at 2500 fpm, max speed still 302 mph, but recalculation yields maneuverability = 26 (up from 23) - so treated as eratta and backfitted.

Thunderbold II: Initial ROC 3120 fps changes maneuverability rating to 34 (up from 32).
User avatar
Nikademus
Posts: 22517
Joined: Sat May 27, 2000 8:00 am
Location: Alien spacecraft

RE: Aircraft ROC Review

Post by Nikademus »

ORIGINAL: witpqs

Thank you. Based on the comparisons I saw earlier, I doubt that 1 point is adequate (for the L model), but how to convince you?

Changing 1 mvr point does little. Roughly, if all other stats are equal, a 1 point mvr change alters the average kil ratio by 0.1.


el cid again
Posts: 16983
Joined: Mon Oct 10, 2005 4:40 pm

RE: Aircraft ROC Review

Post by el cid again »

What is the Kill ratio of P-38 (in various marks) vs various kinds of targets?

What should it be?

Is there any reason to believe that any particular plane is rated incorrectly in a relative sense? We set out to improve relative plane to plane ratings. Pretty clearly we have done that. How much farther can we go - at what cost in time? I think we have spent far too much time on the P-38 issue in the past - and are once more sucked into it here. But since we are sucked into it - lets get it right. Is there any problem with these new ratings? Any problem at all with any of them which is RELATIVELY wrong for P-38 vs other models - if so what - and to what degree? And how do we know that?

I read several books just for this review - and it looks to me like we are rating the P-38 variants just about dead on perfect - within the limits of the system. What could be better? I have no intention of perverting the system by pretending the P-38 isn't bigger than many twin engine light transports or bombers, or that it maneuvers like a single engine aircraft in a sense astrophysics says is impossible. [You want to talk about a two engine plane with two engines on the centerline - well we have a technical definition in place to address that case - if ever we have such an aircraft in the set: THEN you can treat it like a 1 engine aircraft. There are two ways to get there: put both engines on the same shaft, or on counter-rotating shafts, which seem to be a single engine to an observer; or put one engine fore and one aft, which you can see are two engines because there are two propellers - but both are on centerline.] A larger mass is harder to maneuver - because of the conservation of momentum principle. And engines offset from the centerline on the wings are harder to maneuver - because of principles discussed long ago on another thread. You may pretend both these principles do not matter, but IRL they do - and any fair rating system must account for them. We have a simple system and it works rather well. IF it is wrong for this aircraft - why and to what extent?
User avatar
witpqs
Posts: 26376
Joined: Mon Oct 04, 2004 7:48 pm
Location: Argleton

RE: Aircraft ROC Review

Post by witpqs »

ORIGINAL: el cid again
Is there any reason to believe that any particular plane is rated incorrectly in a relative sense? We set out to improve relative plane to plane ratings. Pretty clearly we have done that. How much farther can we go - at what cost in time? I think we have spent far too much time on the P-38 issue in the past - and are once more sucked into it here. But since we are sucked into it - lets get it right. Is there any problem with these new ratings? Any problem at all with any of them which is RELATIVELY wrong for P-38 vs other models - if so what - and to what degree? And how do we know that?

The only reason this is taking so much time is that you insist on ignoring the historical comparisons made in actual testing. Below is the post I referred to. At the bottom of this post I take the area in bold and break it out point by point.
ORIGINAL: Dili

Maneuverability: The subject aircraft was flown in mock combat against P-39D, P-40F, P-47C-1, and P-51 types of aircraft and the following results were obtained:




(a)
The subject aircraft could outclimb all other types used in the test.




(b)
The P-47C-1 was faster at all altitudes, and the P-40F and F-51 were faster up to fifteen-thousand (15,000) feet. The P-39D was considerably slower.




(c)
Against the P-39D, P-51, and the P-40F, the P-38F had a longer radius of turn below twelve-thousand (12,000) feet. From twelve-thousand (12,000) feet to approximately fifteen-thousand (15,000) feet, the radius was almost the same, and from fifteen-thousand (15,000) feet on up, the P-38F had a equal or shorter radius of turn. In the initial turn, due to the slowness of aileron roll of the P-38F, the other types could roll into a turn faster and close up the circle rapidly before the P-38F would reach its maximum radius of turn. It would then take the P-38F sometime, if ever, to overcome this initial disadvantage. The P-38F’s best maneuver against all types tested was to climb rapidly out of range and then turn and commence the combat from a superior altitude. Once gaining this altitude it should retain it, making passes and climbing again rapidly. Knowledge of the local enemy fighter performance will dictate the tactics to be used by the P-38F in the combat zone. It is doubtful if this aircraft will meet in combat any type of enemy aircraft in which close-in fighting will be its best offensive action.




(5)
Ceiling: The operational ceiling was approximately thirty-thousand (30,000) feet and the service ceiling approximately thirty-eight-thousand (38,000) feet, due to engine coolant and carburetor air temperatures becoming excessive.


http://www.wwiiaircraftperformance.org/ ... rials.html

Now how to use this? :

The maintenance difficulties experienced were greater than with any other standard type of American fighter.

The subject aircraft is easy to fly. However, a longer period of time will be required for a pilot to become familiar with the operations and maximum performances of the aircraft than is required for a normal single engine fighter.


Breakdown

Against the P-39D, P-51, and the P-40F, the P-38F had:

1) a longer radius of turn below twelve-thousand (12,000) feet.

2) From twelve-thousand (12,000) feet to approximately fifteen-thousand (15,000) feet, the radius was almost the same,

3) and from fifteen-thousand (15,000) feet on up, the P-38F had a equal or shorter radius of turn.

4) In the initial turn, due to the slowness of aileron roll of the P-38F, the other types could roll into a turn faster and close up the circle rapidly before the P-38F would reach its maximum radius of turn. It would then take the P-38F sometime, if ever, to overcome this initial disadvantage.

Analysis

Conclusion A) The P-38F should have a moderately lower maneuver rating than the P-39D, P-51, and P-40F.

Justification A) Characteristics 1), 2), and 3) were overall approximately equal between the P-38F model and the other aircraft. Characteristic 4) favors the other aircraft over the P-38F. Overall the P-38F was at a meaningful disadvantage to the other aircraft, in spite of having good overall MVR. Hence, the P-38F should be rated moderately lower in MVR than the other aircraft.

Conclusion B) The P-38L should have a slightly lower maneuver rating as the P-39D, P-51, and P-40F.

Justification B) Building on Conclusion A) and Justification A) - the P-38L model had powered control surfaces and hence did not suffer from the disadvantage noted in 4). As a result, we are left with results cited in 1), 2), and 3). It has been noted previously that air to air combat between fighters often winds up favoring lower altitudes. This being the case, slightly more weight is given to 1), resulting in a conclusion that the P-38L model MVR should be slightly lower than the other aircraft.


Sid,

This is my analysis of the historical evidence presented. If the pure formula is giving different results, then the formula is failing for a radically different configuration such as the P-38. We do not need to adjust the formula, we simply need to adjust the MVR ratings so they are in line with the historical evidence. No additional wasted time, just make a best-evidence estimate and be done with it.

The evidence tells us how the planes actually performed. Calculations must yield to evidence. If it were otherwise then bumblebees would have been prohibited from flying until only a few years ago when people finally figured out how they do it.
User avatar
Nikademus
Posts: 22517
Joined: Sat May 27, 2000 8:00 am
Location: Alien spacecraft

RE: Aircraft ROC Review

Post by Nikademus »

ORIGINAL: el cid again

What is the Kill ratio of P-38 (in various marks) vs various kinds of targets?

What should it be?

I didn't specify one. I was giving witpqs information on what can be roughly expected from a 1 point mvr difference. (which isn't much as he suspected)

Is there any reason to believe that any particular plane is rated incorrectly in a relative sense?

I have no idea. I've never tried your mod. I just know how mvr fits into the game fairly well as a variable.
We set out to improve relative plane to plane ratings. Pretty clearly we have done that.

If you say so.




How much farther can we go - at what cost in time? I think we have spent far too much time on the P-38 issue in the past - and are once more sucked into it here. But since we are sucked into it - lets get it right. Is there any problem with these new ratings? Any problem at all with any of them which is RELATIVELY wrong for P-38 vs other models - if so what - and to what degree? And how do we know that?

I read several books just for this review - and it looks to me like we are rating the P-38 variants just about dead on perfect - within the limits of the system. What could be better? I have no intention of perverting the system by pretending the P-38 isn't bigger than many twin engine light transports or bombers, or that it maneuvers like a single engine aircraft in a sense astrophysics says is impossible. [You want to talk about a two engine plane with two engines on the centerline - well we have a technical definition in place to address that case - if ever we have such an aircraft in the set: THEN you can treat it like a 1 engine aircraft. There are two ways to get there: put both engines on the same shaft, or on counter-rotating shafts, which seem to be a single engine to an observer; or put one engine fore and one aft, which you can see are two engines because there are two propellers - but both are on centerline.] A larger mass is harder to maneuver - because of the conservation of momentum principle. And engines offset from the centerline on the wings are harder to maneuver - because of principles discussed long ago on another thread. You may pretend both these principles do not matter, but IRL they do - and any fair rating system must account for them. We have a simple system and it works rather well. IF it is wrong for this aircraft - why and to what extent?
[/quote]
User avatar
Nikademus
Posts: 22517
Joined: Sat May 27, 2000 8:00 am
Location: Alien spacecraft

RE: Aircraft ROC Review

Post by Nikademus »

ORIGINAL: el cid again

What is the Kill ratio of P-38 (in various marks) vs various kinds of targets?

What should it be?

I didn't specify one. I was giving witpqs information on what can be roughly expected from a 1 point mvr difference. (which isn't much as he suspected)

Is there any reason to believe that any particular plane is rated incorrectly in a relative sense?

I have no idea. I've never tried your mod. I just know how mvr fits into the game fairly well as a variable.
We set out to improve relative plane to plane ratings. Pretty clearly we have done that.

If you say so.



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

RE: Aircraft ROC Review

Post by witpqs »

ORIGINAL: Nikademus

I didn't specify one. I was giving witpqs information on what can be roughly expected from a 1 point mvr difference. (which isn't much as he suspected)

Thank you. Actually, I assumed the difference is pretty small. My comment to Sid was along the lines that I felt a change of 1 point was much too small.
User avatar
m10bob
Posts: 8583
Joined: Sun Nov 03, 2002 9:09 pm
Location: Dismal Seepage Indiana

RE: Aircraft ROC Review

Post by m10bob »

ORIGINAL: witpqs
ORIGINAL: Nikademus

I didn't specify one. I was giving witpqs information on what can be roughly expected from a 1 point mvr difference. (which isn't much as he suspected)

Thank you. Actually, I assumed the difference is pretty small. My comment to Sid was along the lines that I felt a change of 1 point was much too small.

Why? Once you are already in the basement, bring up the extra beer.
Image

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

RE: Aircraft ROC Review

Post by witpqs »

Good idea!

el cid again
Posts: 16983
Joined: Mon Oct 10, 2005 4:40 pm

RE: Aircraft ROC Review

Post by el cid again »

ORIGINAL: witpqs
ORIGINAL: Nikademus

I didn't specify one. I was giving witpqs information on what can be roughly expected from a 1 point mvr difference. (which isn't much as he suspected)

Thank you. Actually, I assumed the difference is pretty small. My comment to Sid was along the lines that I felt a change of 1 point was much too small.


OK - are you guys just talking about one P-38 variant? RHS has FIVE DIFFERENT P-38 variants - and all five were presented above. Only the P-38L increased by a single point. The F-5 variant didn't increase at all (it decreased) - but except for an error it should have increased 2. The P-38J - which is a fighter type aircraft that appears early enough to matter big time - increased by 2. And the F-4 also increased by 2. So in the general case, we did not "increase by 1 point." Only in the last case - the least important because it was already highest and because it appears late - is the increase only 1 point.

When we modified the maneuverability system specifically to make you happy, P-38 has increased its rating by 67% - fully 2/3 - now that we have combined the new system with a proper definition of ROC (so it is consistent for all types). Going from 12 to 20 is not a trivial thing - and now we have gone one step farther than that - to 21 - so it is a 75% increase. That is certainly far from "in the basement" in my view. It begins to look like you want hundreds of % - and that is not reasonable.

el cid again
Posts: 16983
Joined: Mon Oct 10, 2005 4:40 pm

RE: Aircraft ROC Review

Post by el cid again »

ORIGINAL: m10bob

ORIGINAL: witpqs
ORIGINAL: Nikademus

I didn't specify one. I was giving witpqs information on what can be roughly expected from a 1 point mvr difference. (which isn't much as he suspected)

Thank you. Actually, I assumed the difference is pretty small. My comment to Sid was along the lines that I felt a change of 1 point was much too small.

Why? Once you are already in the basement, bring up the extra beer.

In the first place, as I just pointed out, we didn't increase by 1 point.

In the second place, this is the highest rated two engine aircraft we have - by far. [And also the second highest, and third highest, and so on.] To call that "in the basement" is pretty much to disregard the numbers - and indicates that there will never be a solution acceptable to you - and our time is being wasted trying for one (for the fourth time, may I say).

In the third place, the discussion was not ended - and I was considering adding yet another point to the L for its powered flap technology.

User avatar
m10bob
Posts: 8583
Joined: Sun Nov 03, 2002 9:09 pm
Location: Dismal Seepage Indiana

RE: Aircraft ROC Review

Post by m10bob »

When people are sleep-deprived and stressed out, and being pelted with rocks, they might tend to lash out at people coming to their defense.
My comment was in response to the question "Why JUST 1 point?"

If even that single point moves it closer to the realm of correctness, why would one not do it?
With the inaccuracies found thus far, I think we have had enough of "make do" and fudging of numbers.
Some of us are, anyway, and these "new" numbers are not "new" to some of us.
Further, these "new" numbers have a point of uniform referance, and can be checked by anybody willing to look in the same sources of info, as compiled in a public place and not using some "secret formula" which will apparently never get an official explanation, for whatever reason.
Image

el cid again
Posts: 16983
Joined: Mon Oct 10, 2005 4:40 pm

RE: Aircraft ROC Review

Post by el cid again »

That is fair - it is after midnight - I did just complete working an 8 hour day (preceeded by 8 hours working on data for WITP) - and it may be I missed the point.
el cid again
Posts: 16983
Joined: Mon Oct 10, 2005 4:40 pm

RE: Aircraft ROC Review

Post by el cid again »

ORIGINAL: witpqs
ORIGINAL: el cid again
Is there any reason to believe that any particular plane is rated incorrectly in a relative sense? We set out to improve relative plane to plane ratings. Pretty clearly we have done that. How much farther can we go - at what cost in time? I think we have spent far too much time on the P-38 issue in the past - and are once more sucked into it here. But since we are sucked into it - lets get it right. Is there any problem with these new ratings? Any problem at all with any of them which is RELATIVELY wrong for P-38 vs other models - if so what - and to what degree? And how do we know that?

The only reason this is taking so much time is that you insist on ignoring the historical comparisons made in actual testing. Below is the post I referred to. At the bottom of this post I take the area in bold and break it out point by point.


I don't think this is very helpful for us. We do not have (much against my theoretical wishes) a model that allows us to rate aircraft for different altitudes. If we MUST make assumptions about altitude as norms for this model, I am afraid the ONLY reasonable option is to consider LOW ALTITUDE maneuverability. And that is specifically the place your data says P-38 is outclassed by all the others. We must think in these terms because:

a) ALL game aircraft can reach low altitudes; This is a really big deal- a system of relative ratings is meaningless UNLESS all the members of the set can be rated on the same basis;

b) ALL interceptions must be based on low altitude maneuverability - and in particular initial ROC - considerations - which we have correctly modeled here;

c) All attacks against low fliers must by definition be at low altitudes; P-38 itself is a classic low altitude attack aircraft - and NOT to rate it properly for that altitude (instead to deliberately overrate it) would be to make it more effective than it should be in one of its primary roles;

d) All air battles involving extended combat maneuvering will lose altitude - so regardless of where they begin they will end at low altitude - and when they do - the critical thing for the system to be rating is low altitude maneuverability;

To these add that the P-38 was never considered a great performer at high altitudes, that it lacked the technologies required to be very effective way up there, and its operational significance was as a low altitude fighter, or in other roles entirely - the ground attack one also being dominated by lower altitude performance. This means there is specifically no justification for this aircraft to be uprated based on higher altitude performance - even if there might be were it an aircraft that focused on high altitude performance/missions.

Now this is entering the realm of necessary compromises. Technical compromises is a difficult art - one about which reasonable men might reasonably differ on this point or that one. It is also unfortunate - I for one would LOVE to rate planes at at least three different altitudes - and let players assign missions in the context of those ratings. But it is my judgement that there is little wiggle room about considering the LOW ALTITUDE performance band the most critical. As you go up, you reach a point where an aircraft's maneuverability approaches - then reaches - absolute zero. This point is different for different aircraft - and we only crudely deal with it. But at least RHS HAS a way to deal with that (however crudely) - a way invented by us (the "operational ceiling"). Instead of pretending an aircraft has full performance all the way to its Service Ceiling, we forbid operations above a certain ceiling defined in a way to give a greater fraction of SC to those planes that were better at altitude. It is not exactly the ideal best approach. But it is the system we have - for now.

Another dimension of this is your focus (and constant reference) to turning - while maneuverability as we are defining it is NOT mainly related to turning. It is dominated by speed and ROC, and turning is only a secondary feature related to loadings, wing loading in particular. Expecting a scale in which turning is not dominant to show relative values based ONLY on turning is a technical error on your part - and it cannot happen. We cannot justify a single case exception based on TURNING - we would instead have to make many exceptions for the same cause - or redefine altogether. Note there is no evidence any form of WITP before RHS ever considered turning (that is, loading) as part of maneuverability at all, so we are at least making a gesture in that direction. And note that my position is that we cannot really do this "right" until we are allowed to separate horizontal and vertical maneuvering. To go over to a system where turning dominates is to misunderstand WWII air combat - and to let biplanes rule over monoplanes. I don't think we should go that way, although I admit a biplane can out turn a monoplane in many cases.
el cid again
Posts: 16983
Joined: Mon Oct 10, 2005 4:40 pm

Brainstorming

Post by el cid again »

At work I had this thought:

we might keep the existing formula and simply substitute

dive speed

for max speed in the first element.

Dive speed is always greater than max speed, and so max speed is contained in it. And the faster the dive speed, the better the aircraft can maneuver in absolute terms.

The problem is - does my definition work? That is, my definition of dive speed. Since it is not recorded, we need to calculate it for all planes - and to be fair - do so on a standard basis.

That definition was

dive speed = maximum speed + (initial ROC * (wing loading / 2757))

[Converting the constant for English units - I do metric]

This change would tend to prejudice the maneuverability value in favor of planes that had a faster dive speed. It would also favor aircraft with a higher ROC.

A complication is that wing loading is in this factor - and in a way that penalizes aircraft with lower wing loading. We might want to compensate for that in the wing loading factor.

Other changes we could consider are changing the loading constants:

we do wing loading / 25

and power loading / 5.

Using lower constants would increase the significance of loading.

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

RE: Brainstorming

Post by witpqs »

proposed formula changes

My thoughts on that...

You are working very hard on the formula. IMO, the formula has served its purpose and it deserves no more of your effort. The objective is to assign a MVR rating to each aircraft that causes the game engine to make the planes behave according to their historical capabilities with respect to each other. You came up with a great formula that served well. It got you a lot of pretty accurate values (AFAIK). In the case of the P-38 the formula served much less well, but that was virtually expected considering the radical differences in design from other aircraft for which the formula worked well.

Basically, the formula is a 'hammer', and it worked great to bang in the 'nails'. But we have at least one 'screw', and using the 'hammer' to bang on the 'screw' is a sub-optimal approach.

For such an exceptional case the best approach is to forget the formula - forget even about 'adjusting' the result of the formula. Instead, use the available historical comparisons. The formula yielded decent MVR rating for the 'other' airplanes. Simply derive the P-38 ratings from the 'other' plane models ratings according to the available historical comparisons between the P-38 and the 'other' airplanes.

[Example with fake numbers: the historically performed tests indicated most P-38 models had moderately worse MVR than the P-47C, and the P-38L model slightly worse that the P-47C, and the formula yielded a MVR rating of 34 for the P-47C (this is a made-up example), then assign an MVR of 26 for all P-38 models except the L, and a MVR rating of 32 for the P-38L. Note this is just an example for illustration of the method. The actual numbers are probably wrong.]

A formula that works for all aircraft is not an objective. The objective is useful MVR ratings.
...I am afraid the ONLY reasonable option is to consider LOW ALTITUDE maneuverability...

Disagree. Placing increased emphasis on low altitude (as I did in my analysis) might be justified. Considering ONLY low altitude as you propose is entirely unjustified.
It begins to look like you want hundreds of % - and that is not reasonable.

Actually I don't care a whit about any value the formula came up with, so I do not want any given %. I only care that the MVR rating we use in game causes the game engine to make the planes behave according to their historical capabilities with respect to each other.
User avatar
m10bob
Posts: 8583
Joined: Sun Nov 03, 2002 9:09 pm
Location: Dismal Seepage Indiana

RE: Brainstorming

Post by m10bob »

ORIGINAL: witpqs
proposed formula changes

My thoughts on that...

You are working very hard on the formula. IMO, the formula has served its purpose and it deserves no more of your effort. The objective is to assign a MVR rating to each aircraft that causes the game engine to make the planes behave according to their historical capabilities with respect to each other. You came up with a great formula that served well. It got you a lot of pretty accurate values (AFAIK). In the case of the P-38 the formula served much less well, but that was virtually expected considering the radical differences in design from other aircraft for which the formula worked well.

Basically, the formula is a 'hammer', and it worked great to bang in the 'nails'. But we have at least one 'screw', and using the 'hammer' to bang on the 'screw' is a sub-optimal approach.

For such an exceptional case the best approach is to forget the formula - forget even about 'adjusting' the result of the formula. Instead, use the available historical comparisons. The formula yielded decent MVR rating for the 'other' airplanes. Simply derive the P-38 ratings from the 'other' plane models ratings according to the available historical comparisons between the P-38 and the 'other' airplanes.

[Example with fake numbers: the historically performed tests indicated most P-38 models had moderately worse MVR than the P-47C, and the P-38L model slightly worse that the P-47C, and the formula yielded a MVR rating of 34 for the P-47C (this is a made-up example), then assign an MVR of 26 for all P-38 models except the L, and a MVR rating of 32 for the P-38L. Note this is just an example for illustration of the method. The actual numbers are probably wrong.]

A formula that works for all aircraft is not an objective. The objective is useful MVR ratings.
...I am afraid the ONLY reasonable option is to consider LOW ALTITUDE maneuverability...

Disagree. Placing increased emphasis on low altitude (as I did in my analysis) might be justified. Considering ONLY low altitude as you propose is entirely unjustified.
It begins to look like you want hundreds of % - and that is not reasonable.

Actually I don't care a whit about any value the formula came up with, so I do not want any given %. I only care that the MVR rating we use in game causes the game engine to make the planes behave according to their historical capabilities with respect to each other.


I agree with witpqs and might go one further: We know there were certain 2 engine planes which were designed to be "fighters", and were never removed from that role, so their performance must not have forced them to be removed from that role.
Maybe it is because the 2 engines should NOT be considered an obstacle?
Recalculate, and just add the POWER of the 2 engines as a single functioning unit of power and compare that unit of power against the wing, (as if it were a single engine plane).
While as large as the A 20(which was the model it is likely being compared to,size-wise) it is certainly way more aerodynamic in appearance and offers much less drag, and with twin rudder fins for turning, of course it was a hot plane. It was DESIGNED to be.
Of course, having twin rudders and twin engines alone would not make a world-beater, just look at the ME 110, but that plane was never meant to be a "fighter" per se, but as an interceptor and long-range escort, and in the end, it performed poorly at both, except at night.It was all a matter of purpose-design and intent.
We can find a lot of referances to P 38's taking on Zeroes,Oscars, Fw 190's and Bf 109's, and emerging victorious, but we don't find too many accounts of the ME 110 emerging the victor ver Spits or Hurri's, or Mustangs, etc...
Besides, the P 38 started out 100 MPH faster then the Me 110, (just for comparisons' sake..)
Another plane which had good maneuverability as a twin engine fighter was the "Nick", but using just ROC as a measuring stick IMHO is not sufficient enough to give it justice.
IMHO, it might be closer to 22-24 based on wight to power and the lift from the wing area, and weight of the plane, (in its' more powerful version)..
Think of it as the "Japanese Mosquito" and you might understand my thoughts.
Image

Dili
Posts: 4742
Joined: Fri Sep 10, 2004 4:33 pm

RE: Brainstorming

Post by Dili »

Hmm i think the German unit with most kills in Battle of Britain was a Bf110 one.
Post Reply

Return to “Scenario Design”