x.7895 issues and 7.7896 uploading

[witpqs]

ORIGINAL: el cid again

ORIGINAL: witpqs

Sid, if someone else rejected hard test data (as presented in the scans of WWII test results documents) in favor of their own mathematical formula the way that you have been doing, you would cry foul. The combat results are corroborative, not mere anecdotes in a vacuum.

I try to understand the reports.

But it is like being a medical doctor. This is a very complex program - and it is hard to know what is going on. It takes focused tests to show what is going on - not wide open ones.

Since you have no specific example, I cannot comment specifically. Players have lots of ways to rig the reports - even without intention to do so. And lots of things impact what you seem to see. Sometimes reports are not even honest (code often "loses" data for example).

In this post I wrote about real life WWII test flight results and about real life WWII combat results (just referred to by others). You seem to think I was referring to results from the WITP computer game - I was not.

[goodboyladdie]

Hi Cid

Thanks for working so hard to get ship production sorted. Mike and I are discussing our lateset restart. Did you finish your deliberation on whether the DP main batteries throw off the AAA model? I personally feel they do, but I know you have more extensive experience of testing to inform your decision.

[el cid again]

ORIGINAL: 1EyedJacks

ORIGINAL: el cid again

ORIGINAL: 1EyedJacks




Hi Sid,

Where can I find a list of changes between this version and the last one?

Mainly in this thread - more detail on some in the thread on ship construction and AA.

Can I ask what the final solution is to ship production for Japan?

We found values I like for warships - these were slightly different for each scenario - and ONLY AIO and CAIO permit building everything.
So in other scenarios you either cancel some vessels - or you get some date delayed vessels - the less so if you build up capacity. I like it a lot.

I was unable to find any value whatever that works for non-warships - which includes a great deal more warships than I understood (I can tell by date behavior). I provided a workaround - it is simple, crude, very WITP like - and it mostly works. I gave you virtually every possible point - the most I think the fields will hold - every day - and let you add as many as you wish/can from the legitimate economy. It appears most ships get built - except for the highest slot numbers. For these (RTN mainly) I gave you the merchant fleet when the war starts - and I think I damaged it slightly - an AE solution.

Since then I have thought of several mechanisms which require time - and I am slowly working them into the next update:

1) Shipping Transportation Regiments (which take a long time to build) will appear at start - but be damaged. This is not good - it prevents you from selecting landing craft where you wish - except when they get sunk and reborn - but it saves a LOT of points.

2) Craft associated with points to be captured appear at start - in the wrong place. You must repair them, then sail them to the right place - or use them wherever you wish and never send them where planners thought they would go. Things like "Batavia Barge Group".

3) Ships that were taken over (Aramis as Tiea Maru for example, at Saigon) appear at start, damaged.

4) Some places will get more construction capacity.

If we get enough changes I will recalibrate and reduce the daily point allocation. I would like it to be about twice shipyard capacity - so you end up paying 1 HI point per merchant point - vice 3. This won't happen - but we can move toward it.

In AE we will revisit this matter - and see what can be done as well. There are more options in the editor there. I also am contemplating using a hex editor to do some of this in WITP 1 - adjusting things we cannot with Matrix editors.

[el cid again]

ORIGINAL: witpqs

ORIGINAL: el cid again

There is also a big difference in other ratings - which favor the P-38 - so why not think about them in the package?

Because the test documents (found scanned on line by a member of this forum and posted here a few months ago) clearly show that the maneuverability ratings of the P-38 models should be much closer to the fighters it was tested against. The document specifically commented on maneuver characteristics, and specifically on comparative turning at at various altitudes.

Historically documented performance (in controlled tests) must trump reconstruction calculations. The calculations are good faith estimates. The documented test results are actual reality.

You can do what you want and I shall not argue this further, but I have seen no facts that change my opinion.

The principle fact I believe you are ignoring is that "maneuverability" as a WITP game field is misnamed. Since it is dominated by speed and ROC, and since it must apply at all altitude regimes, it isn't exactly what pilots mean by maneuverability. RHS has a more sophisticated definition - which includes wing loading, power loading and other considerations as minority factors - but nevertheless it is not an ideal index of maneuverability as such. Using a single composite value - particularly one which must (for code reasons) include speed and ROC (which are fields in their own right - but which are not used in all the air combat routines in the way we would wish) - means that things are going to "look funny" compared to using a more sophisticated, multi field approach. The P-38 indeed had - and must have - a maneuverabilty disadvantage in some senses - due to its configuration - and the impact of angular momentum - also the impact of sheer mass (intertia - which resistes changing velocity and direction) - which are in stark contrast to other aspects of maneuverability. Crudely put, maneiverability is "the ability to change position in the sky in a minumum of time" - and a plane that has a huge turning radius at speed is simply not as maneuverable as one which does not. There is no doubt whatever that in absoute maneuverability terms lighter aircraft with lower wing loadings have decisive advantages. In the case of historical Japanese aircraft - the Ki-10 was probably the most maneuverable of all. [It SHOULD be in the game - as several regiments start the game with it] - and this extreme maneuverability dominated the politics of Japanese aircraft procurement in the run up to the war - to the detriment of Japanese fighter development in most expert opinon. Ki-27, Ki-43, A5M and its Ki cousin, A6M all are much more maneuverable aircraft than a P-38 - if "maneuverable" means "able to turn" and particularly "able to turn at full speed" where air combat usually begins. [Over time in a real fight you lose altitude and speed - until someone runs out of altitude or gets hit - or fearing either breaks out of the tangle - or tires to and fails] There are senses in which sheer speed and sheer ROC ALSO count as "abiity to change position in the sky" - and in situations where that matters - a P-38 is going to look very good compared to many opponents. The details of how one represents different abilities are important - but in the end not under our absolute control: we did not design the model, and we can not fundamentally change it by making it as complex as we would like. We need to guard against distorting it - or misunderstanding it - and I have diligently attempted to do that.

Opinions about figher planes are like opinions about guns and several other subjects. Quoting Joe Wilkerson - I cannot in the end please everyone - even though I want to - and try. This critical attitude - before anyone has even tried to find out the functional meaning of reclassifying the plane as a fighter - is a case in point: it is emotional rather than rational. We may even have a perfect model now - as near as we can get to in this system - but instead of looking to see if we do - or even if the P-38 is too effective now - it is just assumed it cannot be - because of a focus on a single value. I wonder why there is no similar focus on the durability field? Is it not unfair that it is significantly higher for a P-38 JUST BECAUSE it has more engines? Maybe it is too much of a boost - but since it does not oppose the overall performance of the plane you want to be great - you don't object to it. Yet if we spent a hundred hours of analysis we might conclude we should use some other value for engine impacts on durability. I am sorry you are unwilling to look at what code does with this new classification. I think the code system has way too much of a penalty for fighter bombers and P-38 might be too strong now. I know this: when a Japanese player sees P-38s - in any number - he is not happy - in the game - BEFORE this change.

[el cid again]

ORIGINAL: witpqs

ORIGINAL: el cid again

ORIGINAL: witpqs

Sid, if someone else rejected hard test data (as presented in the scans of WWII test results documents) in favor of their own mathematical formula the way that you have been doing, you would cry foul. The combat results are corroborative, not mere anecdotes in a vacuum.

I try to understand the reports.

But it is like being a medical doctor. This is a very complex program - and it is hard to know what is going on. It takes focused tests to show what is going on - not wide open ones.

Since you have no specific example, I cannot comment specifically. Players have lots of ways to rig the reports - even without intention to do so. And lots of things impact what you seem to see. Sometimes reports are not even honest (code often "loses" data for example).

In this post I wrote about real life WWII test flight results and about real life WWII combat results (just referred to by others). You seem to think I was referring to results from the WITP computer game - I was not.

You are correct. I did not understand that you were bringing up this matter again - something you did bring up in previous discussions when we evolved the maneuverability ratings - and which TWICE I (wrongly) thought YOU liked. I mainly redid them for you (and some others) - and I if I did not believe we had made some progress toward a better rating system - I might be upset you don't like the results (still). But THIS time I missed what you referred to - and I assumed you were writing in the context of THIS thread - which was about mod performance.

I don't think your data as presented is very germane. I think you are too narrowly focused on a misinterpretation of what maneuverability means. I begin to suspect you do not want to understand it objectively either: that somehow you think a plane that can turn in 100 feet or 100 yards is not more maneuverable than one that needs closer to 2000 yards than 1000 yards for the same maneuver. In some sense I know that you know there are many kinds of maneuverability, and I hope you know that when we are guided by programmers at Matrix we don't really have the option of making the field not represent what it needs to contain. But I begin to think nothing will ever be enough - except perhaps giving the P-38 a rating which is not even close to relatively correct. You start out with the "knowledge" it "should be near the value of fill in the blank" and - when it is not - then you are upset. Never mind we wanted an objective system based on real data - and got one that works better than I at least ever thought was possible (with so few fields) before we did it.

[el cid again]

ORIGINAL: goodboyladdie

Hi Cid

Thanks for working so hard to get ship production sorted. Mike and I are discussing our lateset restart. Did you finish your deliberation on whether the DP main batteries throw off the AAA model? I personally feel they do, but I know you have more extensive experience of testing to inform your decision.

I have not been able to show a case where they clearly are. The cases of objections from players do NOT involve such weapons at all - e.g. at PH. I thought they might be vs KB - but I was not able to show excessive AA losses - even with slow planes in awful attack profiles (e.g. dive bombing taking them down low at the end) - it appears every attack - however small - actually delivers bombs on capital ships.

But I am thinking of building a test bed for a clearer indication.

The critical question is effect - is the effect field used - or not? If not - I doubt this is going to be a problem.

[goodboyladdie]

ORIGINAL: el cid again

ORIGINAL: goodboyladdie

Hi Cid

Thanks for working so hard to get ship production sorted. Mike and I are discussing our lateset restart. Did you finish your deliberation on whether the DP main batteries throw off the AAA model? I personally feel they do, but I know you have more extensive experience of testing to inform your decision.

I have not been able to show a case where they clearly are. The cases of objections from players do NOT involve such weapons at all - e.g. at PH. I thought they might be vs KB - but I was not able to show excessive AA losses - even with slow planes in awful attack profiles (e.g. dive bombing taking them down low at the end) - it appears every attack - however small - actually delivers bombs on capital ships.

But I am thinking of building a test bed for a clearer indication.

The critical question is effect - is the effect field used - or not? If not - I doubt this is going to be a problem.

If effect is not used why is the AAA figure quoted in the ship and tf screens? This figure is the sum of effect (divided by two in the tf screen - not sure about the ship screen). Surely it makes no sense to use effect as a basis for this calculation if it has no bearing on AAA? All AAA devices are now fixed, calibrated and working, thanks to your hard work. The newly introduced DP main batteries are not necessary and use ammunition points that a player should have available for surface combat/bombardment. If operating from a forward base with no AE available, surely this could cause the Japanese player some problems? I hope you do have time to build a test bed. It will be too late for mine and Mike's game, which I hope will restart as soon as we get the new downloads and Mike can get the time to do his turn 0. I will be doing an AAR and will forward any interesting results we find.

Thanks very much for your hard work.

[Mifune]

All the latest RHS level 7 (7.7896) scenarios are now available at the RHS web site.

[goodboyladdie]

That's great news. Thanks Mifune. [:)]

Edit: Just went to download and EEO has updated, but EOS is still 7.7895?

[witpqs]

ORIGINAL: el cid again

The principle fact I believe you are ignoring is that "maneuverability" as a WITP game field is misnamed.

I assure you I am accounting for all of those things, I simply have reached different conclusions in my analysis. Caio.

[el cid again]

ORIGINAL: goodboyladdie

ORIGINAL: el cid again

ORIGINAL: goodboyladdie

Hi Cid

Thanks for working so hard to get ship production sorted. Mike and I are discussing our lateset restart. Did you finish your deliberation on whether the DP main batteries throw off the AAA model? I personally feel they do, but I know you have more extensive experience of testing to inform your decision.

I have not been able to show a case where they clearly are. The cases of objections from players do NOT involve such weapons at all - e.g. at PH. I thought they might be vs KB - but I was not able to show excessive AA losses - even with slow planes in awful attack profiles (e.g. dive bombing taking them down low at the end) - it appears every attack - however small - actually delivers bombs on capital ships.

But I am thinking of building a test bed for a clearer indication.

The critical question is effect - is the effect field used - or not? If not - I doubt this is going to be a problem.

If effect is not used why is the AAA figure quoted in the ship and tf screens? This figure is the sum of effect (divided by two in the tf screen - not sure about the ship screen). Surely it makes no sense to use effect as a basis for this calculation if it has no bearing on AAA? All AAA devices are now fixed, calibrated and working, thanks to your hard work. The newly introduced DP main batteries are not necessary and use ammunition points that a player should have available for surface combat/bombardment. If operating from a forward base with no AE available, surely this could cause the Japanese player some problems? I hope you do have time to build a test bed. It will be too late for mine and Mike's game, which I hope will restart as soon as we get the new downloads and Mike can get the time to do his turn 0. I will be doing an AAR and will forward any interesting results we find.

Thanks very much for your hard work.

WITP was originally done on the cheap. I think it likely the giant team working on AE is a bigger effort. If not, it is because of a lot of fundamentals necessary to a project of this sort (it isn't all about OB and combat mechanics). It is normal for things not to be properly implemented - and report values are high on the normal list. Nothing at all makes sense about the AAA values reported on player screens - except maybe they would make sense if one knew what the assumptions of the person who did those screens made at the time he designed them. I was astonished that AA has no range whatever in any situation in WITP - being an AA guy this is nothing like I have ever seen before - and nothing whatever like real life. It also appears that altitude has no meaning either - except "shoot - no shoot"! If you come in at 20,000 feet it may be it is as likely you will get hurt as if you come in at 5,000 feet - provided only the AA guns reach 20,000 feet. [Otherwise, altitude is a modifier - but I am doubtful this is the case - so simple is the routine - nor is it very likely anyone knows - due to the way code is written - and surely it is not documented] RHS long assumed range mattered for AA guns - and we had to revise when we learned it did not matter - range is SURFACE gun range only. I THINK the value used is "accuracy" and (dimly) it may be I "know" that (as in was told) - in which case effect is not going to hurt us.

I was surprised players wanted to include VH AA guns - but they did - and provided even cases I did not know about (notably Rechelieu's six inch battery). But I now believe that while the 8 inch (on RN and IJN) and 6 inch (RNN, FFN and IJN) matter, I fear the captial ship guns (IJN only) are not likely to be a good idea - in part for the reason you give (ammunition consumption).

[el cid again]

ORIGINAL: goodboyladdie

That's great news. Thanks Mifune. [:)]

Edit: Just went to download and EEO has updated, but EOS is still 7.7895?

My upload folder has 7.7896 in it - so a successful download should show 7.7896 in the comment file.
The scn file won't show that though - as it has not been updated.

[el cid again]

ORIGINAL: witpqs

ORIGINAL: el cid again

The principle fact I believe you are ignoring is that "maneuverability" as a WITP game field is misnamed.

I assure you I am accounting for all of those things, I simply have reached different conclusions in my analysis. Caio.

And - I note - you have reached that different conclusion without bothering to measure the impact of reclassification of the aircraft as a fighter. How it performs relative to other aircraft in the game is not a factor in your opinion on the matter. Indeed, if it racked up scores like Hartman or other German aces (which no Allied fighter pilot ever came close to) I suspect you would just say P-38 was getting its due.

But - assuming for a moment I have it wrong - what SHOULD the durability rating of the three different P-38 models in RHS be - and - critically - what objective system of calculation can we use which would yield those values and still yield relatively correct values for all other aircraft?
An RHS principle is "if it is better than we have now, it is in" - and I shamelessly throw out older ways of doing things in favor of even marginal improvements. And note that when I understood the classification issue re P-38 I did not hesitate to think about how to apply it.

I am profoundly skeptical your "analysis" has resulted in what we want - a way to rate P-38 in the context required by WITP AND rate all other aircraft fairly - objectively - based on easily obtained data (not something unavailable for three figures worth of models in our data set). IF you have such an analysis - please share it. I warn you - there is grave danger we will instantly adopt it.

[witpqs]

ORIGINAL: el cid again

Indeed, if it racked up scores like Hartman or other German aces (which no Allied fighter pilot ever came close to) I suspect you would just say P-38 was getting its due.

Why attack me by speculating out loud that I would hold some unreasonable position? This is ad hominem crap. Please don't.

- what objective system of calculation can we use which would yield those values and still yield relatively correct values for all other aircraft?

I contend that that this is the wrong objective. The objective is to arrive at historically accurate capabilities for each aircraft. The formula did a pretty good job of establishing relative ratings. However, for outlying designs - such as the P-38 - the formula does a poor job. If the formula were able to take into consideration all minutiae of each aircraft's design, it would be both incredibly complex and it would be used to design superior aircraft today. Investing effort in devising such a formula is way outside the scope of this effort. Instead:

Use the formula to establish a baseline rating for each aircraft. Then make any necessary adjustments according to historical evidence. This is what I recommended when the formula was devised.

IF you have such an analysis - please share it.

I did that already (back when the formula was devised). You rejected it. I didn't save any of the posts myself.

[el cid again]

ORIGINAL: witpqs

ORIGINAL: el cid again

Indeed, if it racked up scores like Hartman or other German aces (which no Allied fighter pilot ever came close to) I suspect you would just say P-38 was getting its due.

Why attack me by speculating out loud that I would hold some unreasonable position? This is ad hominem crap. Please don't.

OK - It was a suspicion probably not warranted - as indicated by your response. You seem almost unreasonable to me - in that you "believe" (in the religeous faith sense) we have P-38 wrong - when it may well be right - but I am willing to entertain the concept it may indeed be wrong. What I do not understand is exactly how to "know" it is wrong. And I do NOT see a value presented here - but I remember suggestions in the past which were clearly not in the ball park - clearly not showing the significant maneuveraility issues P-38 has. Even so - I retract the charge: if P-38 starts scoring 100 plus victories per pilot I expect you to object to it.

- what objective system of calculation can we use which would yield those values and still yield relatively correct values for all other aircraft?

I contend that that this is the wrong objective. The objective is to arrive at historically accurate capabilities for each aircraft. The formula did a pretty good job of establishing relative ratings. However, for outlying designs - such as the P-38 - the formula does a poor job. If the formula were able to take into consideration all minutiae of each aircraft's design, it would be both incredibly complex and it would be used to design superior aircraft today. Investing effort in devising such a formula is way outside the scope of this effort. Instead:

Use the formula to establish a baseline rating for each aircraft. Then make any necessary adjustments according to historical evidence. This is what I recommended when the formula was devised.

OK - I accept this too. We adopted this principle for durability - and still use it re one aircraft - the Sturmovik. We also applied it to the Stirling - no longer used because not a PTO aircraft. But it is a valid principle. It does, however, require CLEAR indications of WHY an exception should be made - and it does require a MODEST modifier (usually one or two in the ranges of values of these functions). When I tried to go this way for P-38 and maneuverability, you were not interested. Has that changed?

IF you have such an analysis - please share it.

I did that already (back when the formula was devised). You rejected it. I didn't save any of the posts myself.

Well - IF you feel that a P-38 is worthy of a special modifier in the context of current values - suggest what you think the modifier should be for each model - and why in objective terms this plane is an exception. ALSO explain if the SAME modifier would apply to any other similar aircraft - e.g. a Ki-102, or some other Allied aircraft. And on what basis we could know that?

[goodboyladdie]

The P-38 was a better fighter than any other early long range fighter. This fact is well known. The unique design was the main reason for this. I am happy to try out your redesignation, knowing that speed and ROC will also add to the mix, but witpqs has a point and relevant supporting documentation has been posted and lost in the depths of this forum in the past. Early issues effecting the F anf G in the ETO were medium altitude performance and reliability. In the MTO the Axis forces soon came to respect them, although they were less manoeuvrable than Axis single engined fighters. The J model introduced better peformance at all altitudes, chin radiators for better cooling which allowed the use of full take off power up to 26,000 ft AND hydraulically boosted powered ailerons. The L model had more powerful engines. I do not pretend to have your grasp of how the different factors come together in the WitP engine, but there is a case for treating the P-38 as an exceptional case when compared to other two engined fighters with a conventional lay out. The P-38 was the top scoring aeroplane of all US Army fighters in the PTO. This may in part be due to the fact that they could turn up where the Japanese were not expecting them and the better tactics employed after 1943, but it is still a fact to be considered when weighed up against the number of P-40s engaged in the PTO. In comparative terms you have the G model about equal to the P-40E, when it was a little better. The J and L are the most severely underrated models.

[m10bob]

ORIGINAL: goodboyladdie

The P-38 was a better fighter than any other early long range fighter. This fact is well known. The unique design was the main reason for this. I am happy to try out your redesignation, knowing that speed and ROC will also add to the mix, but witpqs has a point and relevant supporting documentation has been posted and lost in the depths of this forum in the past. Early issues effecting the F anf G in the ETO were medium altitude performance and reliability. In the MTO the Axis forces soon came to respect them, although they were less manoeuvrable than Axis single engined fighters. The J model introduced better peformance at all altitudes, chin radiators for better cooling which allowed the use of full take off power up to 26,000 ft AND hydraulically boosted powered ailerons. The L model had more powerful engines. I do not pretend to have your grasp of how the different factors come together in the WitP engine, but there is a case for treating the P-38 as an exceptional case when compared to other two engined fighters with a conventional lay out. The P-38 was the top scoring aeroplane of all US Army fighters in the PTO. This may in part be due to the fact that they could turn up where the Japanese were not expecting them and the better tactics employed after 1943, but it is still a fact to be considered when weighed up against the number of P-40s engaged in the PTO. In comparative terms you have the G model about equal to the P-40E, when it was a little better. The J and L are the most severely underrated models.

Here ya go, guys:

http://www.zenoswarbirdvideos.com/P38.html

[witpqs]

The document that GBL is referring to was first posted on the forums by someone else - I forget who. I am uncertain, but I think it was this site:

WWII Aircraft Performance

The site contains many actual reports from WWII. Happily, they are in two formats - nice clean text in HTML, and the original scans (accessed by a link on each corresponding HTML page).

They have lots of documents and I am still searching for the one that was posted here before. It was a record of a flight test comparison of a P-38 model versus a P-40 model, a P-47 model, and a P-51 model. It included several comments in the results that made specific comparisons. As such it provides a way to peg the MVR performance of the P-38 model to the other aircraft models.

Therefore, if one of the comparison models is rated (made up example here) MVR 32 and the P-38 model is best estimated at 90% of that, it would be rated a 28 or 29. This is just an example, not real numbers unless by coincidence.

Notice that this is about the real MVR performance of the airplanes - there is no consideration given to whether something is called a fighter or a fighter-bomber. The MVR rating should speak for itself. The F vs FB designation has only to do with the WITP game engine.

[witpqs]

Here is the original post by Dili that brought us the site:

ORIGINAL: Dili

Hehe. I have read everything about P-38 from the worst US fighter to the Uberplane. I suspect that is in part because of different versions but cant explain all divergence.

Post edit: see this http://www.rdrop.com/users/hoofj/

http://www.wwiiaircraftperformance.org/ ... l#p38-2338

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

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

[witpqs]

Here is the post were I responded to Sid's request for a specific recommendation:

Sid,

You asked that we look over the P-38 MVR versus other planes' MVR ratings in version the version that just became available. This is my own take on the situation. My approach is to compare the P-38 MVR rating to the MVR rating of other planes where real life comparison tests were made and reported here in the forum. Noted earlier in this thread, the P-38 was compared with P-39, P-40, P-47, and P-51 (see those posts for details).

In current RHS the MVR ratings are as follows.

P-38G 16
P-38J 17
P-38L 21

P-39D 28
P-39Q 32
average P-39 30.0

P-40B 28
P-40E 26
P-40N 26
average P-40 26.7

P-47D 32
P-47N 37
average P-47 34.5

P-51A 26 (not included in P-51 average)
P-51B 34
P-51D 36
average P-51 35.0


Here is how the P-38 models MVR stack up comparatively to the other planes noted.

_____________P-39______P-40______P-47______P-51
_____________30.0______26.7______34.5______35.0
P-38G 16_____53%_______60%_______46%_______46%
P-38J 17_____57%_______64%_______49%_______49%
P-38L 21_____70%_______79%_______61%_______66%


Going by the historical comparisons posted earlier, I have the impression that the P-38L model should be at about 95% of the P-47/P-51, and the other P-39 models should be at about 80-85% of the P-47/P-51.

I recommend we try:

P-38G MVR = 28

P-38J MVR = 29

P-38L MVR = 33




< Message edited by witpqs -- 10/2/2007 3:36:10 PM >

Now - do I know if these are the very best recommendations? No. You asked for my best estimate at the time and this was it. Maybe someone has a better idea for the percentages?

BTW, I have since learned that the P-38J model also had powered flight controls, so relatively speaking that would rate a little higher (closer to the L model than to the G model).