Announcement: Historical Statistics Pack

[unmerged(12760)]

those guys got to fight as infantry replacements at places like Hurtgen Forest and Bastogne, their trucks were handed over to supply units, and the AA guns were just piled up by the side of the road and abandoned.

you forget that AA is mostly very effective in the ground combat role, too.

americans used the 4x0.50cal.AA with devastating effect. it was called the "meat grinder". also the germans used most of their AA ( 20mm,37mm,88mm ) in ground combat. further the other main allied gun 40mm bofors also was used against all kinds of targets - can easily destroy a ligt armored tank.

note that german gepard crews ( 2x35mm) in the bundeswehr also trained to some extend ground combat, but they only get handed a few AP rounds for emergencies.

hint: play SPWAW ( www.matrixgames.com ) to see the effectiveness of some of these weapons.

i also read combat reports from american units in france, they often used their AT guns as light arty ! also 75mm shermans served this way

 

[unmerged(2833)]

you forget that AA is mostly very effective in the ground combat role, too.

snip

He did not actually:

(...)
Those guys got to fight as infantry replacements at places like Hurtgen Forest and Bastogne, their trucks were handed over to supply units, and the AA guns were just piled up by the side of the road and abandoned.
(...) Of course, heavy AA was a very useful ground unit in itself, as Rommel proved with his 88s.
(...)

 

[unmerged(12760)]

yes, but the US didn´t pile their guns on the roadside.

or do you have any reference for this, math guy ?

 

[unmerged(13914)]

yes, but the US didn´t pile their guns on the roadside.

or do you have any reference for this, math guy ?

Sure, Shelby Stanton for one. There are actual pictures somewhere of the stuff lying by the roads.

It's not all that hard to understand. When the Allies were setting up their tables of organization in 1942 and 1943, they expected a fierce ground-to-air struggle against the Luftwaffe. They couldn't guess at that point how many planes the Germans would actually produce, or how many would be diverted to the Eastern Front or defence of Germany.

When they landed in Normandy, they still had all the AA guns (and searchlights, and rangefinders, and radar, and so on) as planned, but the Luftwaffe had very few planes. They had over-estimated what level of AA protection would be needed.

The Allies also dramatically under-estimated the need for infantry replacements. Their statistical models were flawed. AND they had tremendous challenges in finding enough trucks to haul supply to the front. So they did indeed abandon a substantial number of AA guns, some heavy, some light, and all the related equipment. (Eventually, I suppose the stuff was gathered up and shipped back to England, but not during the buildup and breakout.) They handed the AA troops machine guns and BARs and grenades, gave them some refresher training in basic infantry skills, and used them to fill holes. It wasn't that they didn't understand how a 90mm AA gun could be used in the antitank role . . . it was that they had enough tank destroyers and tanks already, but didn't have enough men with rifles. You can't storm a pillbox with a searchlight.

Armies are forever doing this kind of thing. Standard policy is to bring everything "just in case". When the Soviets first arrived in Afghanistan, they took all their 23mm quad AA guns and SA-7 missiles, even though the Afghan resistance had no aircraft of any kind. So after a short while, those cunning Afghan mujahideen were equipped with 23mm quad AA and SA-7s and Soviet helicopters were dropping from the sky . . .

 

[unmerged(12612)]

BTW, as I understand it, aircraft speed is the same for all aircraft: one province per hour. The speeds in the unit description don't do anything, only range matters.

Speed does work and its easy to test. I've tested a pre-war fighter with a range of 1000 and speed of 1000 versus a pre-war fighter with a range of 1000 and a speed of 75. The faster fighter flew from Moscow to Baku in 8 hours. The slower fighter flew the same route in 18 hours. Why there is such a small difference in flight times for such a large disparity in speeds is beyond me, but speed does work. A note of caution: when testing speed (and range) you must exit HOI completely before the changes take effect. If you simply tab to your desktop, make the change in the unit file, then return to HOI (which is how I do a great number of my edits) the change will not take effect.

Regarding the manpower issue, I return to my original question: "I suspect that support staff for infantry divisions actually did sustain higher casualties [than support staff for aircraft "divisions"]...but high enough to justify the large disparity in manpower between aircraft and infantry that current exists?"

I understand you to be arguing that it is appropriate to model ground unit support personnel, but not air unit support personnel. That seems contradictory to me. Your approach may very well be valid, if it can be shown that casualties of air unit support personnel were, proportionately speaking, so vastly smaller than casualties of ground unit support personnel, as to be irrelevant. The Bomber Command numbers you've provided are helpful in that regard, but not entirely convincing for the reasons I stated earlier.

Perhaps my skepticism stems from my own views of HoI, which admittedly is not particularly relevant to your goals for HSP. But to me, HoI is an abstraction in many ways, and manpower is no different. An oil point doesn't really represent an actual barrel of engine lubricant, just as a fighter unit doesn't really represent just the actual aircraft. Instead, these concepts are abstracted. That is why technologies such as "sheltered airfield doctrine" improve the aircraft unit itself. The unit's statistics encompass more than the specs of the actual planes. Manpower, in my view, should be treated in a similar fashion. If you view manpower for aircraft as representing only the actual pilots and aircrews, then you would be inclined to make air units have a very low manpower cost. This means, in turn, that manpower is an extremely limited consideration when a player goes to build or reinforce an air unit...which strikes me as quite ahistorical. My understanding is that pilots, aircrews, and ground crews were highly trained and highly valuable individuals who were difficult to replace when lost, perhaps more so than the men who made up the infantry and armor units. Air units with very low manpower costs creates odd game dynamics as well. It can create a situation where a country that is basically drained of manpower, can nevertheless field (and repeatedly reinforce) a massive air armada...a tactic I have seen used with great effect in an MP game by a Germany that was basically tapped for manpower after several years of combat on the eastern front.

 

[unmerged(13914)]

Speed does work and its easy to test. I've tested a pre-war fighter with a range of 1000 and speed of 1000 versus a pre-war fighter with a range of 1000 and a speed of 75. The faster fighter flew from Moscow to Baku in 8 hours. The slower fighter flew the same route in 18 hours. Why there is such a small difference in flight times for such a large disparity in speeds is beyond me, but speed does work. A note of caution: when testing speed (and range) you must exit HOI completely before the changes take effect. If you simply tab to your desktop, make the change in the unit file, then return to HOI (which is how I do a great number of my edits) the change will not take effect.

Wow. Now that's something! I know a number of people, including beta testers, who would raise their eyebrows at that. Suddenly all kinds of interesting possibilities come to mind . . . does a plane that moves two provinces per hour get AA fire from both? Can we give rockets speed 10,000 and get them to target in an hour? I will look at this.

Regarding the manpower issue, I return to my original question: "I suspect that support staff for infantry divisions actually did sustain higher casualties [than support staff for aircraft "divisions"]...but high enough to justify the large disparity in manpower between aircraft and infantry that current exists?"

I understand you to be arguing that it is appropriate to model ground unit support personnel, but not air unit support personnel. That seems contradictory to me. Your approach may very well be valid, if it can be shown that casualties of air unit support personnel were, proportionately speaking, so vastly smaller than casualties of ground unit support personnel, as to be irrelevant. The Bomber Command numbers you've provided are helpful in that regard, but not entirely convincing for the reasons I stated earlier.

No, actually I can't see any way to model ground unit support personnel, OR air unit personnel. Neither took casualties at anything like the rate that aircrew and combat units did, and so there's no way to model the full extent of either one without creating excess casualties. The only ground personnel that HOI models or that HSP will model are the troops in the divisions and brigades themselves -- not the engineers, field hospitals, signals, etc. that supported them. When a division is surrounded, or retreats from a lost battle, you lose the men in the division, not their line of communications troops.

My previous note showed a way that we could boost air unit manpower requirements so as to roughly simulate the LOSSES that air unit support troops took. You could do the same with divisions and brigades, boost their size slightly to capture LOSSES. But the Allies, for example, had about 550,000 combat troops in Italy and another 1.1 million in air and ground support units. There's no way to model the 1.1 million, that I know of.

 

[Soapy Frog]

Wow. Now that's something! I know a number of people, including beta testers, who would raise their eyebrows at that. Suddenly all kinds of interesting possibilities come to mind . . . does a plane that moves two provinces per hour get AA fire from both? Can we give rockets speed 10,000 and get them to target in an hour? I will look at this.

I get the impression from DogRed's experiment that the absolute maximum speed is 1 province per hour. Baku is 7 provinces from Moscow...

Perhaps the reason why it has generally been assumed that all air units move at the same speed (1 province per hour), is because the speed threshold to attain one province per hour is pretty low... especially when you are flying around western europe where provinces are quite small.

I suspect in the end we are limited to 1 province per hour as the max speed for aircraft (or any unit for that matter). After all HoI is a turn based game (not an RTS), with turns processed every "hour", so there is likely going to be a hard limit to the granularity we are able to acheive.

 

[unmerged(13914)]

Ah. Yes, that sounds right. So it's not so useful as all that, unless you want to model Zeppelin travel.

 

[Soapy Frog]

Well it is more useful if you are willing to abstract air sorties a little more.

i.e. you do not need to model the exact speeds of the planes. Rather you would use the speed stat as a way of increasing or reducing the frequency of attacks.

I am generally in favour of abstracting air combat even more since: a) it doesnt really work satisfactorily, and b) from a playability standpoint it's too much needless micromanagement.

If you were to slow planes down a great deal, and make it so they might fly effectively only one or two missions in a 24 hour timeframe, then you might start to see an air combat system which is a little more manageable. Presumably you would have to increase the hitting power and org of bombers to ensure they do proportionally the same damage, and reduce the effects of AA, or otherwise find a way to slow the pace of air combat.

I realize you won't do this for HSP, of course, just musing out loud.

 

[unmerged(13914)]

No, no, I see your point and I would do it for HSP, if I could make it work out properly.

However, I'm not sure I can, at least not the way I would most like to. The parameter I would much rather control is the length of a tactical air attack, which is 5 hours.

The Germans were masters of achieving high sortie rates per day. The key was to keep their air units well forward, so that flying time was held to a minimum. Then pilots could literally shuttle back and forth, landing & refueling & rearming & taking off again without even turning off the engine.

-- In France 1940 the fighters and dive bombers managed 4 sorties each per day.
-- At the siege of Sevastopol, some dive bomber units were based literally within sight of the city (like a scenario from Battlefield 1942) and could fly a dozen sorties per day.

That's impossible in HOI, obviously, because of the "granularity" between neighboring provinces and the 5-hour raid structure. The most you can hope for in a long summer day is 3 raids of 7 hours each, and even that requires that you arrive at the target just before dawn, and leave it after dusk.

Now, it MIGHT still be interesting to set the aircraft speed very low for the Soviets, for example, to represent lower serviceability and poorer ground crews. Then while the Germans were getting their 2 or 3 raids per day, the Soviets (who managed 2 raids per day during the Arctic winter in my Winter War AAR, arriving before dawn and leaving after dusk) would only fly one. Their interception capability could be held lower too.

I was satisfied in the Winter War AAR that the total effect of the Soviet raids was reasonable, so not much tweaking would be needed to bombing ratings.

I like this, actually. For the Chinese, both sides in Spain, all sorts of second-tier air forces, it might work out very well.

One big consideration: this won't work for aircraft with ranges much above 300 km. They'll wind up "hovering" in highly unrealistic fashion in enemy-held provinces, and they'll be cut to ribbons by the AA or by interceptors. So the aircraft that we would most like to hold to one sortie per day -- the heavy bombers -- we can't really mess with.

I also recall someone saying that speed affects interception. This might be another tool for fine-tuning performance, if we knew that an extra 25 kph difference in speed between bomber and fighter might prevent an interception.

 

[unmerged(12612)]

I get the impression from DogRed's experiment that the absolute maximum speed is 1 province per hour. Baku is 7 provinces from Moscow...

Perhaps the reason why it has generally been assumed that all air units move at the same speed (1 province per hour), is because the speed threshold to attain one province per hour is pretty low... especially when you are flying around western europe where provinces are quite small.

I suspect in the end we are limited to 1 province per hour as the max speed for aircraft (or any unit for that matter). After all HoI is a turn based game (not an RTS), with turns processed every "hour", so there is likely going to be a hard limit to the granularity we are able to acheive.

I tend to agree with Soapy in this area. I first began playing with air unit speed because of my concern that air units don't use oil properly. Unlike ground units, aircraft only use oil when moving; so if you have high air unit speeds, then they spend very little time moving, and hence, consume very little oil. Add this to the fact that aircraft don't consume oil during combat, and aren't affected by oil shortages, and you have an incredibly "cheap" unit, as far as oil is concerned...something which is hard to fathom.

As I continued to experiment, I found other advantages to lowering air unit speed. To begin with, it makes technologies that grant speed increases more relevant, which in turn opens more interesting possibilities for modders. In the mod I'm working on, for example, there are several techs dealing with high octane aviation fuel. Second, it makes upgrading to better models of air units have more meaning. As it stands now, there is almost little to no difference in terms of game play between improved fighters and rocket/jet aircraft when it comes to speed...one moves very very fast, the other very very very fast.

Mathguy is probably right in his observation that lowered speeds will cause increased air unit casualties, but air units are such a bargain in so many other very important respects (oil consumption, manpower, ground attack power) that increased air casualties don't trouble me too much. Of course, it may be inconsistent with HSP's goal of obtaining historical levels of aircraft combat casualties, but it might work well from a game-play perspective.

 

[saintsup]

Well it is more useful if you are willing to abstract air sorties a little more.

i.e. you do not need to model the exact speeds of the planes. Rather you would use the speed stat as a way of increasing or reducing the frequency of attacks.

I am generally in favour of abstracting air combat even more since: a) it doesnt really work satisfactorily, and b) from a playability standpoint it's too much needless micromanagement.

Support that !!

 

[saintsup]

One big consideration: this won't work for aircraft with ranges much above 300 km. They'll wind up "hovering" in highly unrealistic fashion in enemy-held provinces, and they'll be cut to ribbons by the AA or by interceptors. So the aircraft that we would most like to hold to one sortie per day -- the heavy bombers -- we can't really mess with.

Is this possible to reduce AA efficiency accordingly ?

 

[unmerged(14683)]

Quick question - I've noticed in HoI 1.06 that their technologies rising Ground Defense Efficiency got still positive values. From one of the posts by Max I also have the impression that they "fixed the problem".

Is that mean that GDE works differently now? Anyone tested that already?

 

[unmerged(14603)]

Quick question - I've noticed in HoI 1.06 that their technologies rising Ground Defense Efficiency got still positive values. From one of the posts by Max I also have the impression that they "fixed the problem".

Is that mean that GDE works differently now? Anyone tested that already?

Yes it has now been "fixed".

I've tested it and GDE now adds to defense. For example if Germany researches a tech with GDE +5 then it will block 80+5 = 85% of all opponent's shots that are matched by its GD.

Although now WAD I found the old GDE logic more useful for modding.

 

[unmerged(14683)]

Damn.

 

[Mithel]

Argh, it sure would be nice if they documented this stuff clearly! These little tidbits are critically important to know!

So now we need to design with ground_def_eff as properly improving the nation's defense instead of offense.

- Mithel

 

[unmerged(14603)]

Argh, it sure would be nice if they documented this stuff clearly! These little tidbits are critically important to know!

So now we need to design with ground_def_eff as properly improving the nation's defense instead of offense.

- Mithel

I plan to keep testing. Lastest news: fort_attack is still bugged. Only works in increments of 100 just like in 105c.

 

[Mithel]

Mikel, I certainly appreciate your testing. Please keep us informed.

I think we'd better double check everything. Many of the v1.06 readme items were already in v1.05c. I'm going to start a thread "v1.06 Fact or Fiction" to discuss strange things I'm discovering.

- Mithel

 

[unmerged(13914)]

Various observations:

In the beta forum I requested that fort_attack be fixed and cited Mikel's bug report, but too late . . . the design was "frozen" the same week.

Too bad about the GDE bug being fixed. Nobody told me, I found out when you all did. There's a bunch of clever planning apparently wasted. But let's make sure it really is WAD now before we conclude that!

Saintsup -- it is possible to modify AA effectiveness but the whole idea of making aircraft go slow needs some more thought. Basically what DogRed said, the goal of HSP is to get the historical numbers.