Low CAP vs High SWEEPS test #1

[Yaab]

Interesting.

In combat animations, you can sometimes get a glimpse of attitude messages at which dogfighting takes place. I tested 100 feet ground attacks performed by fighters, and their approach altitude was 3000-5000 feet and exit altitude was 5000 feet. Thus, CAP patrolling at 10,000 feet minced them both at approach and exit altitudes.

Approach altitude is given in combat reports, but exit altitude I was only able to glimpse from combat messages i.e Zero diving at Warhawk at 5,000 feet.

In your tests, are the initial Franks climbing to 42,000 feet to meet the P-47? I don't think the P-47s can spot your low CAP from 42,000 feet, thus it must be the Franks guided by the radars.

[obvert]

ORIGINAL: Yaab

Interesting.

In combat animations, you can sometimes get a glimpse of attitude messages at which dogfighting takes place. I tested 100 feet ground attacks performed by fighters, and their approach altitude was 3000-5000 feet and exit altitude was 5000 feet. Thus, CAP patrolling at 10,000 feet minced them both at approach and exit altitudes.

Approach altitude is given in combat reports, but exit altitude I was only able to glimpse from combat messages i.e Zero diving at Warhawk at 5,000 feet.

In your tests, are the initial Franks climbing to 42,000 feet to meet the P-47? I don't think the P-47s can spot your low CAP from 42,000 feet, thus it must be the Franks guided by the radars.

I didn't see altitudes listed but I'll keep trying for that.

The 42k sweepers have never had problems finding he low CAP. I've experienced this in game and in tests. I think for sweeps things work differently than for a ground strike mission. The sweepers are looking and so go and find the CAP wherever they are, apparently.

The Franks are climbing though, and a lot seem to be arriving late to the battle (roughly half the number arrived after the P-47s had been whittled to the low teens in number). The overwhelming majority of attacks are being made by Franks diving. I'd like to know if these are diving from higher than the set CAP, or just diving on the next layer down from 9k to 7k for instance, so I'll keep trying to find some mention of altitude.

[Lowpe]

Look to see what altitude band the Franks are sent to. I find that gives a big clue to what ranges the fight is at.

Radar is horrendously important, as it gives a DL advantage to the defense, which the attackers usually don't get unless they have radar in that hex. Hard to overstate how important radar is I think.

[obvert]

ORIGINAL: Lowpe

Look to see what altitude band the Franks are sent to. I find that gives a big clue to what ranges the fight is at.

Radar is horrendously important, as it gives a DL advantage to the defense, which the attackers usually don't get unless they have radar in that hex. Hard to overstate how important radar is I think.

Yes. The Franks are being sent to between their assigned altitude and their max altitude in all of these battles. Not sure that gives an indication the actual battles are taking place up there though. How would I know other than seeing an actual note int eh animation?

The Allied radar has been working continually, and it didn't appear to help the Buffalos so much in that test. I'll run a few of those though to see if anything appears in the animation. The detection times for the Japanese here are not high, other than two at 20+ minutes. In the attacks I watched the P-47s had overwhelming majority in numbers early, but the Franks mostly hung on, and then the diving began. So yes, radar could be setting the stand-by fighters up high due to the original altitude of the P-47s, then they all dive throughout the battle once all P-47s are low.

So how will layering matter here if the majority of defenders are actually just climbing over those bands anyway, then getting the dive? Couldn't all groups be at 5k and get the same effects? That's what I want to see. That certainly seemed the case in the single George groups that were not layered in the mini-experiments on scrambling.

[obvert]

Here is something I hadn't seen before and I was excited to get the sweep cooperation. This is a different look, and should have meant that the P-47s had a massive overwhelming advantage early. Regardless, they did not do well on the day, as shown clearly in the report. Hugely surprising to me considering a few other runs did a lot better with groups arriving separately.

Also, I've noticed this time around that there are no instances in this run of ten tests of the small fragment groups arriving later (3 planes from each group).

I attached all groups this time to the 13th USAAF HQ at Port Moresby, so I think this is possibly having a direct effect on cooperation in the sweeps. That is good to see (or suspect anyway) and I'd like to try with and without that soon to see if there are noticeable differences. I've not usually given a hoot if one group is or is not attached to an HQ in range, as long as there is an HQ in range.

Because the situation is ever so slightly different with the current adapted test mod, I'll have to run another ten on this to see if 10, or 15, or 20 runs is sufficient for getting fairly consistent results.


[font="Trebuchet MS"]AFTER ACTION REPORTS FOR Aug 07, 42
--------------------------------------------------------------------------------
Morning Air attack on Lae , at 99,126

Weather in hex: Light rain

Raid detected at 80 NM, estimated altitude 46,000 feet.
Estimated time to target is 23 minutes

Japanese aircraft
Ki-84r Frank x 69

Allied aircraft
P-47D2 Thunderbolt x 54

Japanese aircraft losses
Ki-84r Frank: 6 destroyed

Allied aircraft losses
P-47D2 Thunderbolt: 11 destroyed

CAP engaged:
8th Sentai with Ki-84r Frank (0 airborne, 16 on standby, 0 scrambling)
7 plane(s) not yet engaged, 0 being recalled, 0 out of immediate contact.
Group patrol altitude is 5000 , scrambling fighters between 5000 and 34450.
Time for all group planes to reach interception is 26 minutes
8th Sentai B with Ki-84r Frank (0 airborne, 16 on standby, 0 scrambling)
7 plane(s) not yet engaged, 0 being recalled, 0 out of immediate contact.
Group patrol altitude is 7000 , scrambling fighters between 7000 and 34450.
Time for all group planes to reach interception is 21 minutes
8th Sentai C with Ki-84r Frank (0 airborne, 16 on standby, 0 scrambling)
7 plane(s) not yet engaged, 0 being recalled, 0 out of immediate contact.
Group patrol altitude is 9000 , scrambling fighters between 9000 and 34450.
Time for all group planes to reach interception is 24 minutes[/font]

[Lowpe]

ORIGINAL: obvert
I attached all groups this time to the 13th USAAF HQ at Port Moresby, so I think this is possibly having a direct effect on cooperation in the sweeps. That is good to see (or suspect anyway) and I'd like to try with and without that soon to see if there are noticeable differences. I've not usually given a hoot if one group is or is not attached to an HQ in range, as long as there is an HQ in range.

I don't believe that is the case.

I know some people swear by it, but I think the game doesn't care what HQ units are attached to.

[Lowpe]

Which planes are you using. Below is the Dababes modded planes.

Really, not a bad match up at all. Much more likely to see a Frank A fighting versus that P47 however.

Do you see any Franks evading messages at all?



Another really valid question is can you defend against Jugs with Tojo and Oscar down low, layered.

[obvert]

ORIGINAL: Lowpe

ORIGINAL: obvert
I attached all groups this time to the 13th USAAF HQ at Port Moresby, so I think this is possibly having a direct effect on cooperation in the sweeps. That is good to see (or suspect anyway) and I'd like to try with and without that soon to see if there are noticeable differences. I've not usually given a hoot if one group is or is not attached to an HQ in range, as long as there is an HQ in range.

I don't believe that is the case.

I know some people swear by it, but I think the game doesn't care what HQ units are attached to.

I've never known which way to go on that one, mainly because this is often so complex in game. Some units attached to a local HQ, some not. Or not a single one attached. Never all of them. So unless it's tested, how would one ever really know? [;)]

[obvert]

ORIGINAL: Lowpe

Which planes are you using. Below is the Dababes modded planes.

Really, not a bad match up at all. Much more likely to see a Frank A fighting versus that P47 however.

Do you see any Franks evading messages at all?

Another really valid question is can you defend against Jugs with Tojo and Oscar down low, layered.

Using stock Ki-84r. It goes 399mph with a 2502 climb.

I'm using this one because I want a tough match for the P-47, one that can in the right situation do damage. It also has great maneuver, consistent all of the way to 31k at 27.

Franks do evade, yes.

I'll be testing others after this. Once I've got the Ki-84r so much of the game is decided anyway. This set the stage though for this kind of test. I also ran a few with N1K1-J the other day, if you remember, and strangely, they did nearly as well as the Frank. They shouldn't by all indications of performance. I'm curious to look more there for sure.

[Yaab]

All my Franks in scen100 (historical scen1) have service rating of 3. This would influence Jap CAP number on subsequent days vs Thunderbolts with service rating of 1.

Did you mod the Frank's service rating for this test?

[obvert]

Here is another test of the same setup. So that's 20 runs now.

No intra-group cooperating sweeps this time, but still no inter-group small fragments either.

While this one was less violent, the two are close in ratio of losses (A:J). A difference of 0.08. Is that close enough?

1) 1.54:1
2) 1.62:1

I have that impatient feeling that I want to get onto more interesting tests, but I actually suspect I should run 20 more of these to see if they stay in this range.

[obvert]

ORIGINAL: Yaab

All my Franks in scen100 (historical scen1) have service rating of 3. This would influence Jap CAP number on subsequent days vs Thunderbolts with service rating of 1.

Did you mod the Frank's service rating for this test?

No need. I'll test CAP over duration of time later.

These tests are on repeated turns. Every few turns or so I push forward to change weather if added randoms are not changing it already.

[Lowpe]

I would move on, but you are calling the shots.

[obvert]

Here is a compilation of screens to tell a bit of the story of how these battles have been playing out. This is actually from one of the second sweep battles on the second run above where the score was 27-13 in favor of the Japanese.

First, the sweep is picked up by radar. Great note. Confirmation. Still on this battle, the TTT is only 14 minutes. The Franks open up first here, interestingly. Not a lot of diving messages early, if any. More of the kind like ... engages ... .

The stand-by CAP starts to arrive and engage the Allies. P-47s get most of the attacks here, with their numbers, but many times Franks evade.

After being outnumbered 20-16 in the middle of the battle, the P-47s are still diving on Franks. A lot of pilot names are given throughout and I'd like to check on those to see which groups were fighting in which sections. At the end a lot of Franks show up and dive on the remaining few P-47s. The final Allied plane continues to dive on 26 Franks and finally is shot down.

A few things. I bet this combat ends sooner with a less aggressive Allied group leader. The P-47s really hang in there, and I've seen a lot shorter sweeps in daily combat. This would mean a more even score as a lot of P-47s were shot down late. Something I can check on.

Also, no indications of how high dives are coming from or going to in the animations. The only altitude indicators here were about where the original sweep began and where the CAP was being sent when scrambled.

More in the next one. [:)]

[Lowpe]

Quite the graphic.[&o]

[Lowpe]

Perhaps there is more use for a low aggression fighter leader. Hmm...[:)]

[Lokasenna]

ORIGINAL: Lowpe

ORIGINAL: obvert
I attached all groups this time to the 13th USAAF HQ at Port Moresby, so I think this is possibly having a direct effect on cooperation in the sweeps. That is good to see (or suspect anyway) and I'd like to try with and without that soon to see if there are noticeable differences. I've not usually given a hoot if one group is or is not attached to an HQ in range, as long as there is an HQ in range.

I don't believe that is the case.

I know some people swear by it, but I think the game doesn't care what HQ units are attached to.

It definitely does. You probably don't notice much playing as Japan because you don't have 3 dozen air groups that are all going on a bombing raid in one day, but as the Allies... try a scattershot approach with your bombers and watch them arrive piecemeal over the target. Try assigning them to the same HQ (having an HQa at the base is also very important, but for THAT case it doesn't have to be the same one), and watch 200 bombers arrive over the target simultaneously.

I'm not sure if it works at all for sweeps, since sweeps are not coordinated but merely a chance arrival over the base at the same time. For that matter, I've seen "cooperating" sweeps where one of the groups that is cooperating just never even fights at all. For example, it will say 75 P-47's are sweeping in the combat report texts, but in the animation only 50 ever show up. Then at the end, it will show 3 groups sweeping at altitude. Is strange.

[Yaab]

I don't think leader aggro is the answer.

P-47D are the most durable single-engine Allied fighters. This, plus 90-99 morale pilots in the test air groups and there are no quitters - everybody fight until their death.

[obvert]

ORIGINAL: Yaab

I don't think leader aggro is the answer.

P-47D are the most durable single-engine Allied fighters. This, plus 90-99 morale pilots in the test air groups and there are no quitters - everybody fight until their death.

Maybe. Luckily, that is something I can test really easily. [:)]

[obvert]

ORIGINAL: Lokasenna

ORIGINAL: Lowpe

ORIGINAL: obvert
I attached all groups this time to the 13th USAAF HQ at Port Moresby, so I think this is possibly having a direct effect on cooperation in the sweeps. That is good to see (or suspect anyway) and I'd like to try with and without that soon to see if there are noticeable differences. I've not usually given a hoot if one group is or is not attached to an HQ in range, as long as there is an HQ in range.

I don't believe that is the case.

I know some people swear by it, but I think the game doesn't care what HQ units are attached to.

It definitely does. You probably don't notice much playing as Japan because you don't have 3 dozen air groups that are all going on a bombing raid in one day, but as the Allies... try a scattershot approach with your bombers and watch them arrive piecemeal over the target. Try assigning them to the same HQ (having an HQa at the base is also very important, but for THAT case it doesn't have to be the same one), and watch 200 bombers arrive over the target simultaneously.

I'm not sure if it works at all for sweeps, since sweeps are not coordinated but merely a chance arrival over the base at the same time. For that matter, I've seen "cooperating" sweeps where one of the groups that is cooperating just never even fights at all. For example, it will say 75 P-47's are sweeping in the combat report texts, but in the animation only 50 ever show up. Then at the end, it will show 3 groups sweeping at altitude. Is strange.

Have you noticed the HQ affecting those late fragments in sweeps?

I agree the cooperating sweeps behave oddly and I think many multiple groups is more effective to wear down CAP. The trick is getting them to arrive before bombing.