RHS Maneuverability Review: Data [ALL Data Done]

[witpqs]

ORIGINAL: Herrbear

ORIGINAL: witpqs

Except for the case of multiple engines along the center axis (very rare), multi-engine planes in WWII have their engines on/in the wings. Because the engines are heavy, this means that they have a lot more inertia to overcome when they try to roll. It's basically the physics concept of angular momentum.

This is one big difference between single and multi-engine planes in WWII.

I understand about the concept of angular momentum. But I am only discussing max speed and rate of climb. I don't think that angular momentum plays a part in this.

This thread is about the maneuverability rating. Max speed and ROC are used in the formula to derive maneuverability because they affect it. Same thing regarding multi-engine - it affects angular momentum, which affects roll - so it is part of the formula to derive maneuverability, because roll is part of maneuverability.

[Herrbear]

What are you using for gross weight. Loaded or Max?

[el cid again]

ORIGINAL: Herrbear

ORIGINAL: witpqs

Except for the case of multiple engines along the center axis (very rare), multi-engine planes in WWII have their engines on/in the wings. Because the engines are heavy, this means that they have a lot more inertia to overcome when they try to roll. It's basically the physics concept of angular momentum.

This is one big difference between single and multi-engine planes in WWII.

I understand about the concept of angular momentum. But I am only discussing max speed and rate of climb. I don't think that angular momentum plays a part in this.

Well - as far as it goes - you are right. But the function under construction/analysis here is NOT about speed or ROC per se. They are ELEMENTS of the function. So is number of engines. The function is maneuverability. Worse - it is a composite maneuverability - both horizontal and vertical - and at all altitudes. So we must put in all the elements that matter. We are NOT saying speed or ROC are related to the number of engines (except, of course, they really are in this sense: for any given size of engine, the more you have, the more speed and ROC you will get - which is why they have more engines!). The multiple engine solution comes at a penalty - you DO get more speed and ROC - but less maneuverability. The function says that. Wether or not it does so properly requires statistical analysis. But it is in a crude sense quite right - 4E is worse than 3E is worse than 2E is worse than 1E. It is all I hope for in a simple "quick and dirty" approximation.

[el cid again]

ORIGINAL: Herrbear

ORIGINAL: el cid again

ORIGINAL: Herrbear

Something I just thought of concerning the dividing by the number of engines. Now that Wing Loading and Power Loading is taken into account do you penalize multi engine planes for speed and ROC. Wouldn't all planes, in theory, be consistent for those two factors. For example, shouldn't a 2-engine plane with 1000 hp engines with a ROC of 3000 and weight of 10,000 lbs work the same as a 1-engine plane with a 2000 hp engine also with a ROC of 3000 and a weight of 10,000 lbs.

Why would the 2-engine plane be penalized in those factors?

The multi engine penalty - it is not just for twin engine planes - but all of them - is meant to address the problem of conservation of angular momentum. It is meant to say that it is a severe restriction on maneuverability. It is a nice part of the function - whatever the number of engines - it goes there. There is, however, this caviet: multiple engines on one axis count as one engine. This can occur in 3 ways (and does NOT occur in WITP so far).

1) Two engines coupled to act as one.
2) Two engines with a common shaft - driving counter rotating screws on the same hub.
3) Two engines on a single axis - one driving a tractor and one a pusher.

Anyway - the number of engines is in the deonominator because that is the right place for it to be. It divides what otherwise would be the maneuverability value - a composite function - by N. It is not meant to say any particular element of that function is so divided.

Sid -- I understand why the penalty is used as it impacts wing loading and power loading. But does the "problem of conservation of angular momentum" have any impact on max speed or ROC. If it doesn't, why penalize those factors?

The problem is you are being too analytical: I am NOT penalizing those factors. The number of engines is meant to affect the composite function - and it is not a comment on any particular other factor.

[el cid again]

ORIGINAL: Herrbear

What are you using for gross weight. Loaded or Max?

Until now we did not use weight at all. Now we have loadings, we are using two weights. Wing loading uses empty equipped weight and power loading uses all up weight. This isn't what I would have done - I would have used normal weight for both. But in playing with the numbers this is what seemed to work best - and I did delegate this task to a volunteer. One must respect volunteers - or one won't have any.

[Mifune]

"What are you using for gross weight. Loaded or Max?" Gross weight is "loaded" weight, not "max" weight.

[el cid again]

Well - what he meant was "normal load" or "maximum load" - I am sure.

[Herrbear]

Mifume answered fine. Normally stats list empty, loaded and max weights. He clarified that loaded weight and not max weight was used. Thanks.

[Nemo121]

So, with all of this being said are we actually ending up with any major changes in the aircraft manoeuvrability scales?
 

Question: I note the Ki-21 Sally doesn't have any armour in RHS EOS 4.46 ... Is this correct?

[el cid again]

ORIGINAL: Nemo121

So, with all of this being said are we actually ending up with any major changes in the aircraft manoeuvrability scales?


Question: I note the Ki-21 Sally doesn't have any armour in RHS EOS 4.46 ... Is this correct?

They are significant changes. They are not really major - except perhaps in the case of the later P-38 (and maybe the F7F) - which went up by 43% if we do not give it a bonus! Since addressing the concerns of P-38 fans was the reason for the review - this is not at all bad an outcome - although it took three tries and bringing in another person to crunch endless numbers.

As far as I know - Sally does not have any armor. This was introduced by the Ki-49 Helen for JAAF service. See

The Nakajima Ki-49 Donryu in Japanese Army Air Force Service. It appears someone gave Ki-21 armor in WITP - and I talked to Joe about it when we were reviewing Japanese planes for CHS. We decided it was probably a JFB somewhere - and could find no justification for it - so we took it out.

[Nemo121]

Ok, cool. Just wanted to know since I noted the disparity... If that's what the facts show then that's how it should be.

[el cid again]

First of all, we have a superior rating system to what was previously available.

It keeps heavy bombers (and similar 4 engine transports and flying boats) anchored about 4 (3 to 5); it lets high performance planes get rated way up in the sky; and it spreads twin engine planes way out - almost down to big plane levels if dogs and better than some single engine types if hot - and mostly in between - just right.

Second, we have a data exchange problem: I don't have the Japanese data (done first) - and cannot use it until I do - so there will be no 5.10 today.

Third, Houston - we have a problem:

RHS tried to do several things at the same time re aircraft - including address operational attrition - AAA attrition and air combat attrition. Two of these - the first two - required we use a lower range for durability. The third is hurt by lower durability - but we compensated for that by weapons changes AND BY a lower maneuverability scale. I set the limits as "3 or 4 at the low end and just over 30 at the high end" because - at 36 - planes become excessively lethal in our system.

The present - vastly improved - scale has planes in the 40s - up to 47 I have seen (for the P-80), 42 (F4U-1D), etc.
We probably will make "uber cap" worse than ever it was if we let such ratings in.

The solution is some kind of data compression of the range. Normally I would use a K factor - but to get 47 to below 36 is way too big a correction to apply to many cases.

Seeking ideas.

[Herrbear]

ORIGINAL: el cid again

First of all, we have a superior rating system to what was previously available.

It keeps heavy bombers (and similar 4 engine transports and flying boats) anchored about 4 (3 to 5); it lets high performance planes get rated way up in the sky; and it spreads twin engine planes way out - almost down to big plane levels if dogs and better than some single engine types.

Second, we have a data exchange problem: I don't have the Japanese data (done first) - and cannot use it until I do - so there will be no 5.10 today.

Third, Houston - we have a problem:

RHS tried to do several things at the same time re aircraft - including address operational attrition - AAA attrition and air combat attrition. Two of these - the first two - required we use a lower range for durability. The third is hurt by lower durability - but we compensated for that by weapons changes AND BY a lower maneuverability scale. I set the limits as "3 or 4 at the low end and just over 30 at the high end" because - at 36 - planes become excessively lethal in our system.

The present - vastly improved - scale has planes in the 40s - up to 47 I have seen (for the P-80), 42 (F4U-1D), etc.
We probably will make "uber cap" worse than ever it was if we let such ratings in.

The solution is some kind of data compression of the range. Normally I would use a K factor - but to get 47 to below 36 is way too big a correction to apply to many cases.

Seeking ideas.

Not to sound like a dunce, but are you saying that the lower durability is causing an increase in the air combat attrition?

[Nemo121]

Herrbear,
 
Yes... and also vs AAA and in terms of ops losses.
 
 
Why not apply a K factor? It preserves relative manoeuvrabilities thus fulfilling the goal of internal consistency. I don't see how any non-K solution is going to keep the relationship of the manoeuvrabilities of planes to the same ratio as currently pertains.

[witpqs]

Sid,

I'm also at a loss to understand why you feel a K factor would mess things up? If the highest you've got so far and anticipate is 47, then use a K factor or 1/3 (0.75). All the calculated Maneuver ratings get multiplied by 0.75. What would this mess up?

[el cid again]

We have one data anomoly - the P-80 - and it is way up there at 47.
We make a special correction for jets. This is valid: in my day we found that speed was not
all - and we were DEFEATED by ancient MiG 17s - out turning us - UNTIL we engaged them
with A-1 Skyraiders - which out turned them! A plane is not always better due to speed - too
much speed = reduced ability to engage in tactical combat. Dog fights are never at high speed
for that reason: you may be in a Mach 2.2 figher - but NOT if you are in a dog fight.

So we divide by 3 for all points above 29 for jets: 47 (18 points higher becomes 6 points higher) = 35
acceptable.

[el cid again]

Accepting the jet plane modifier, we have reduced the scale of the issue: worst case (as far as I know) is 42.

So we use similar logic: high speed beyond a certain point is of diminishing returns in terms of value in
maneuver:

we say "every 2 points above 29 is needed to up the score by 1" -

so 41 (12 points above becomes 4) = 33; 42 (13 points above becomes 4.5 rounding up to 5) = 34

both acceptable.

[el cid again]

ORIGINAL: Herrbear

Not to sound like a dunce, but are you saying that the lower durability is causing an increase in the air combat attrition?

[/quote]


I am saying it should tend to do that. More than that - I am saying that something awful happens when you reach or exceed maneuverability = 36. I want to avoid that.

[el cid again]

ORIGINAL: Nemo121

Herrbear,

Yes... and also vs AAA and in terms of ops losses.


Why not apply a K factor? It preserves relative manoeuvrabilities thus fulfilling the goal of internal consistency. I don't see how any non-K solution is going to keep the relationship of the manoeuvrabilities of planes to the same ratio as currently pertains.

It is the obvious answer - but if applied to the whole range K would be about 0.75

and that is way too much at the low end. We will force all big planes into 3s and you won't know the difference between a dog and a good one maneuverability wise. We will make the twin engine types recompress and be much closer to the same. I think I have a better idea - leave most alone - and just address the statistical anomolies at the very top end. Relatively - this favors the 2E types - which should win points for those who think they always get short changed.

[Nemo121]

Sid,
 
Is it possible to give us a list of the ratings, as turned in, for planes with manoeuvrability above 30, listing which will be reduced and how much they'll be reduced to...
 
Then you could get some good feedback ( you'd probably also need to put in the manoeuvrability for the P38s and Stan and Toryu-Ki-45).
 
Your explanation re: modding the top manoeuvrabilities seems reasonable but a lot would depend on the exact numbers.
 
Ps. From what you are saying I am presuming there is an in-code "knee" at 36? That's pretty poor modelling ( conceptually I disagree with all knees. relative relationships must be conserved without any "knees" interfering).