RE: Japanese Air to Air Combat Dropoff (Full Version)

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



Message

Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/13/2006 5:11:29 PM)

So pretty even the only thing I can think of is thats its got something to do with heavies as on pure fighter v fighter 42/44 makes no difference its radar difference that drive performance

I'll have to take a look at that. Still working on running this enough to eliminate some statistical anomoly.

I am running 5 tests now, a little of each at a time. Looking at the combat reports the increase Japanese losses are not really across the board (no kidding, since there is no code to produce this!) but rather is situational. Since all the radars are the same in both test (or not there at all in the second set) what other consistent factors are involved. As you point out, mission type and aircraft type involved are constants.



el cid again -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 1:05:08 AM)

Thanks for your testing and analysis.



Andy Mac -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 1:22:57 AM)

I am now baffled by the differing results My Sweep v CAP analysis was pretty conclusive (when I gave Japs the radar they had the advantage in same proportion as allies when allies had radar so on pure fighter combat there was no issue)

But these tests of escorted bombers assuming no radar at all are troubling. It is a pure guess but if radar is not a factor and all other things are equal between the tests I would not expect to see a variance but you are so all I can think of is some kind of allied box benefit in 43 but this is a pure guess.

My scenario is not set up to run bomber tests and it would take some faffing about to reset it so I will wait for Nicholas's analyis.

Andy



Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 3:00:42 AM)

I'm working on it [:)]



SamCole -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 4:54:48 AM)

Question for Andy Mac and Nicolas Bell. How are you treating op losses? Do you count them? Are you counting damaged aircraft in any way? I am using Andys test senario and I am seeing that the CAP numbers vary, are you taking this into consideration?



Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 5:15:41 AM)

I am not looking specifically at ops losses, although I am recording them. Focusing on air to air only - I am using the Intel screen losses, which are greater than the combat report losses. Some aircraft listed as damaged in air to air in the combat report become destroyed based on mission range (fatigue factor). Ground and FLAK losses are also not part of this study, although I am recording them.

Understand that sortie rates vary from test to test. Weather scrubs some missions entirely. I am running 30 tests for each year/radar condition each with 9 engagements each (ie 270 engagements each full test). The thought being this is a large enough sample (50,000+ sorties per year test) to even out these sortie rate fluctuations between the 1942 and 1944 tests.

Is there a statistician in the house who can help? [:D]

I also am putting the combat reports in spreadsheet format (as I have been) so individual engagements (all 30) can be compared to the same mission in a different year side by side.

I am running a test with no radars/detectors on map for 1942 and 1944, and also a test with some radars/detectors for both years too. Should have it done in the next day or two.



Sardaukar -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 3:54:25 PM)

Wasn't the heavy bomber defensive fire enhanced in one of the previous patches? There might be something in that with late war Allied heavies vs. Japanese fighers.



Andy Mac -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 4:04:12 PM)

I turn advanced weather off to minimise scrubs

Sardauker it was but I thought it applied equally to 42 and 44 so there should be no difference.

Ops losses in my test are counted but are rarely more than 1 but I only have 4 combats of 100 v 100 per run at max range 120 miles

Andy



Sardaukar -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 4:07:52 PM)

quote:

ORIGINAL: Andy Mac

I turn advanced weather off to minimise scrubs

Sardauker it was but I thought it applied equally to 42 and 44 so there should be no difference.

Ops losses in my test are counted but are rarely more than 1 but I only have 4 combats of 100 v 100 per run at max range 120 miles

Andy


Yes, I thought it was general enhancement too. But what if there is some sort of change depending on date ? It'd make some sense based on test results. Probably they'd have said if it was so, but it's one possibility.



el cid again -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 4:32:35 PM)


quote:

ORIGINAL: Sardaukar

Wasn't the heavy bomber defensive fire enhanced in one of the previous patches? There might be something in that with late war Allied heavies vs. Japanese fighers.



According to Joe, there is a "knee in the function" of combat results so an aircraft with a value of 40 or above (or above 40 - I forget which?) is much harder to kill. This because they didn't like their own durability rating system well enough to just use it. For this reason, RHS kept ALL durabilities below 40 - so we are not affected by the discontinuity of the function.

Still - in any normal system with high durability ratings for bombers - they will be almost impossible to kill. [Reminds me of UV - when we said "the best fighter is a B-17 - just send B-17s and watch enemy planes die"] When a plane cannot die, and when there are no ammunition limits, it can dish it out, round after round - and heavy bombers have lots of guns to use to dish it out too.



Nikademus -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 4:45:51 PM)


quote:

ORIGINAL: Sardaukar

Wasn't the heavy bomber defensive fire enhanced in one of the previous patches? There might be something in that with late war Allied heavies vs. Japanese fighers.


Yes. Applies to all LBA, both sides.

quote:


According to Joe, there is a "knee in the function" of combat results so an aircraft with a value of 40 or above (or above 40 - I forget which?) is much harder to kill


The DUR 40+ hard coding that restricted fighter attacks to Range 3 or farther was removed a long time ago.







Sardaukar -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 5:32:49 PM)

quote:

ORIGINAL: el cid again
Still - in any normal system with high durability ratings for bombers - they will be almost impossible to kill. [Reminds me of UV - when we said "the best fighter is a B-17 - just send B-17s and watch enemy planes die"] When a plane cannot die, and when there are no ammunition limits, it can dish it out, round after round - and heavy bombers have lots of guns to use to dish it out too.


Kind of reminds me what happened in one of the online air combat simulators, cannot remember if it was Air Warrior or Warbirds before they patched it. B-17 was called "Death Star" because one could fly it into furball and let computer-controlled gunners swat fighter after fighter ! [:D]



Mike Scholl -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 6:35:33 PM)

Truth is that Japanese fighters DID have a lot of trouble tackling B-17's until the more heavily-armed models started coming online. They lacked the firepower to bring down a "Flying Fortress" and to use what little they did have had to close well within the range of those defending .50 cals. Which considering the very light construction of Japanese early war fighters (you have to give up a lot to get that range and performance), was close to suicidal. Being able to "turn circles" around the bombers did them absolutely no good at all, as the bombers could pretty much outgun them from all approaches, especially if there were several flying in formation.

Basically you have a "Lightweight" with short arms trying to box with a slow but powerful "Heavyweight---with the additional dissadvantage that the flyweight is required to close in and try to "Knock Out" his larger opponant, meaning that Heavyweight "jab", "hook" and "uppercut" all get a chance to land.



wdolson -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 11:26:28 PM)

quote:

ORIGINAL: Mike Scholl

Truth is that Japanese fighters DID have a lot of trouble tackling B-17's until the more heavily-armed models started coming online. They lacked the firepower to bring down a "Flying Fortress" and to use what little they did have had to close well within the range of those defending .50 cals. Which considering the very light construction of Japanese early war fighters (you have to give up a lot to get that range and performance), was close to suicidal. Being able to "turn circles" around the bombers did them absolutely no good at all, as the bombers could pretty much outgun them from all approaches, especially if there were several flying in formation.

Basically you have a "Lightweight" with short arms trying to box with a slow but powerful "Heavyweight---with the additional dissadvantage that the flyweight is required to close in and try to "Knock Out" his larger opponant, meaning that Heavyweight "jab", "hook" and "uppercut" all get a chance to land.


Suburo Sakai talks about fighting B-17s in his book. The B-17s flew at altitudes where the Zero was very sluggish and then the Zero was very lightly armed for the task. He brought down a D early in the war with repeated tail end attacks, but the Zero's performance at that altitude was so poor, he only had a small speed advantage on the B-17. If he had been attacking a B-17E, or if there had been more than one, he probably would not have succeeded.

It probably wasn't until the Japanese had the Tony where they had a plane which could perform at the same altitudes the B-17s flew. I notice the ceiling for the B-17s is only in the low 20Ks. I know in Europe the B-17s were routinely flying around 30,000 feet, so I think the maximum altitude for American heavies is too low.

Bill



el cid again -> RE: Japanese Air to Air Combat Dropoff (9/14/2006 11:58:41 PM)


quote:

ORIGINAL: Nikademus


quote:

ORIGINAL: Sardaukar

Wasn't the heavy bomber defensive fire enhanced in one of the previous patches? There might be something in that with late war Allied heavies vs. Japanese fighers.


Yes. Applies to all LBA, both sides.

quote:


According to Joe, there is a "knee in the function" of combat results so an aircraft with a value of 40 or above (or above 40 - I forget which?) is much harder to kill


The DUR 40+ hard coding that restricted fighter attacks to Range 3 or farther was removed a long time ago.

That is good. Particularly as in RHS most fighters have no weapons effective at such ranges. Only some really big cannon and maybe rockets in late war planes.








el cid again -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 12:03:07 AM)


quote:

ORIGINAL: wdolson

quote:

ORIGINAL: Mike Scholl

Truth is that Japanese fighters DID have a lot of trouble tackling B-17's until the more heavily-armed models started coming online. They lacked the firepower to bring down a "Flying Fortress" and to use what little they did have had to close well within the range of those defending .50 cals. Which considering the very light construction of Japanese early war fighters (you have to give up a lot to get that range and performance), was close to suicidal. Being able to "turn circles" around the bombers did them absolutely no good at all, as the bombers could pretty much outgun them from all approaches, especially if there were several flying in formation.

Basically you have a "Lightweight" with short arms trying to box with a slow but powerful "Heavyweight---with the additional dissadvantage that the flyweight is required to close in and try to "Knock Out" his larger opponant, meaning that Heavyweight "jab", "hook" and "uppercut" all get a chance to land.


Suburo Sakai talks about fighting B-17s in his book. The B-17s flew at altitudes where the Zero was very sluggish and then the Zero was very lightly armed for the task. He brought down a D early in the war with repeated tail end attacks, but the Zero's performance at that altitude was so poor, he only had a small speed advantage on the B-17. If he had been attacking a B-17E, or if there had been more than one, he probably would not have succeeded.

It probably wasn't until the Japanese had the Tony where they had a plane which could perform at the same altitudes the B-17s flew. I notice the ceiling for the B-17s is only in the low 20Ks. I know in Europe the B-17s were routinely flying around 30,000 feet, so I think the maximum altitude for American heavies is too low.

Bill


Altitudes for planes in CHS and stock are completely inconsistent. This was deliberate - but only partially done - not consistently done - because of modders' views of the rarity of high altitude combat. Sometimes they took off 10,000 feet, sometimes 2000 feet, sometimes nothing, and sometimes they ADDED altitude - even in one case more than 100% more altitude (a mistake?). RHS adopted a different system - "operational altitude" - which is a fraction of "service celiing" - and the fraction depends on the engine type. Turbo supercharged planes do better than regular.
And jets better still. This was combined with realistic limits on AAA - so you can overfly it if your plane is good enough.



Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 5:43:02 AM)

Here's the raw data in an Excel spreadsheet. Rename the .txt extension to .zip to uncompress.

Note that the tests with radar and those without cannot be compared directly in aggregate because of some changes I made in missions. You can compare many of the missions directly, which is probably the way to go. The aggregate losses are from the intel screen as previously noted include in the air to air losses those aircraft which were lost on the return flight (crashed) which inflates the losses way beyond those shown in the combatreport for long range missions.

Only air to air losses are shown.

And for those who may not have been following this thread, the only difference between the 1942 and the 1944 sets of tests is the date. Numbers of aircraft, pilots missions, altitudes all remain the same. All pilots set to experience and morale of 75.

Each "test" consists of day with several aerial battles (7 engagements in the no radar tests, 8 engagements in the some radar tests - one reason why you cannot directly compare the aggregate numbers between the groups). Tests were repeated 30 times for each set.

The no radar test appears to confirm that there is no hard coded bias against the Japanese. Appears to, since the Japanese losses are higher by 16%. Perhaps someone who really knows statistics can confirm my hunch that the higher Japanese losses are within the margin of error for this sample size. If not, then further examination/comparision of individual raids is necessary to devine possible reasons for this.

The "Radar" test is somewhat of a misnomer. All bases have sound detectors except the following: (type 13: Hanoi, Lae, Wewak, Rahaeng SCR270 Kumning, Makin. Rangoon has no detection device) No detection device is in range to affect offensive operations. One might say that the Japanese radars are outside the range of defensive operations because they don't seem to help much at all (and yes they do have a penetration of 500).

What I do see based on the Japanese raid on Makin (with the SCR270) is an increase effectiveness in the number of F4U sorties flown. Perhaps the bounce probability is increased in 1944 when within range of radar?

This doesn't explain the 42% increase in Japanese losses since none of the other missions are in range of the other Allied radar. Oddly, the Rangoon Buffalos had an increased sortie rate in 1944 over 1942 in the radar tests accounting for some of the additional losses for the Japanese. Unfortunately I scrubbed this mission from the No Radars test, but I don't see an overall increase in sortie rate in the 1944 No Radars test over those of 1942 No Radar.

I need to study the individual mission combatreports more.




Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 5:43:40 AM)

Here are the aggregate No Radar losses

[image]local://upfiles/20401/64B71212AC1D47B2AA540E1C52460043.gif[/image]



Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 5:44:26 AM)

Here are the With Radar aggregate losses

[image]local://upfiles/20401/00A905EBF78F400F96E83415CF83BDBE.gif[/image]



Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 5:45:58 AM)

Here is the With Radar Makin Raid Comparision between 1942 and 1944. The increased effectiveness is apparent.

[image]local://upfiles/20401/57EFAA5CFC4642DD831B43C08E075923.gif[/image]



el cid again -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 2:45:10 PM)

Sound detectors in stock or CHS based tests may be very misleading. They have a very low effectiveness rating so they will normally NOT work at all - but when they do they will skew the results. It might be best not to use sound detectors at all - since the normal case with them is "no detection." The effect is 10 - meaning 10% - so 90% of the time it is no detect. So using none makes it even at 100% of the time.



Andy Mac -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 3:05:59 PM)

I agree with Cid sound detectors are weird and to random for testing. Give the Japanese a few CSP1's if you want to see the impact of radar.

I may run a few tests (for my own interest not to prove a point) testing various Japanese aircraft against Corsairs with all things being equal and no radar. Should be fun to find out which is actually the best Corsair Killer out there lets see if I run it for

Jacks
Georges
Franks
Tony 1
Tony 2
Tojo
Shinden
Zeke
A6M8

Anyone want to lay bets on who will do the best and on which will achieve a 1:2 loss rate against the bent wing bastard ?

Andy



Bliztk -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 3:21:05 PM)

I bet for Shinden



Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 5:27:05 PM)

Back to the original issue -

Do you think the sound detectors are causing the increased Japanese losses in 1944?

Matrix says there is no late war code bias against the Japanese. However, it appears that there is a dated related increase in Allied effectiveness - perhaps tied to detection systems? Doesn't seem to appear when there no detection systems.




Andy Mac -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 6:52:09 PM)

Yes and no.

I think Japanese losses at a base with sound detectors will be higher in 44 if in range of allied radar.

My reasoning is more fighters will scramble creating moretargets for allied fighters who are operating better fighters with better radar control and will therefore kill more.

So basically in 44 sound detectors increase the targets for allied fighters so yes sound detectors cause more casualties

I think we always need to remember the range of these allied radar systems I dont know the range of some of your tests but SCR 270's have a range of 5 hexes the CSP 1 a range of 8 hexes so any allied carrier or base force within 8 hexes of the air combat in 44 will give a bounce bonus.

What I am concerned about is the scenario where NO radar is in range for the allies and a sound detector is in range for the Japanese and Japanese losses are higher than they were with no sound detector this appears counter intuitive unless the allied planes have a large qualitative advantage.

Andy



Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 7:20:48 PM)

Any ideas on this?

US raid on Wewak - these are totals from the combat report

No Detection Devices
Ki-61 losses were:
1942: 157
1944: 285

With Type 13 radar at Wewak
the Ki-61 losses were:
1942: 140
1944: 253

In all four test the average raid looked like:
Ki-61 KAIb Tony x 39
P-38G Lightning x 21
B-24D Liberator x 60

So radar doesn't help the Japanese, but they suffer more losses in 1944 no matter what.




Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 7:26:12 PM)

Here is the combat report totals for the Japanese raid on Makin (shown above in graphic format)

No Detection Device:
Japanese aircraft lost (A5m5, & P1Y)
1942: 212
1944: 192

Looks statistically "equal" to me. Now...

With SCR270 on Makin:
1942: 312
1944: 567

Yikes! Improved radar?



Andy Mac -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 7:30:28 PM)

Sorry does this test have allied radar in 44 but not in 42 or was it a same scenario just change the date job ?

I am confused as it appears that adding japanese radar has reduced losses by about 10 aircraft in both tests.





Nicholas Bell -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 7:30:29 PM)

Here are the combat report results from the Japanese raid on Lunga (from Buin) against F4Fs.

No detection device:
1942: 115 Zero/119 Betty
1944: 108 Zero/133 Betty

Again this looks to be statistically equal.

With a sound detector on Lunga:
1942: 128 Zero/111 Betty (looks like no detector results)
1944: 200 Zero/173 Betty

Maybe the sound detector becomes a radar in 1944? [8|]




Andy Mac -> RE: Japanese Air to Air Combat Dropoff (9/15/2006 7:31:34 PM)

Dont know I am confused



Page: <<   < prev  3 4 [5] 6 7   next >   >>

Valid CSS!




Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI
0.984375