What is the effect of Ground Defense?

[Truchses]

By studying the CORE tech tree, I noticed that MGs and mines increase the ground defense. I thought ground defense would absorb attack points. But MGs and mines cause casualties to the attacker instead of absorbing hits. It seems to me that the whole concept of ground defense is unrealistic. So MGs and mines should also get attack points.

 

[unmerged(16682)]

The whole notion of absorbing hits is artificial, save for bullet-proof vests :rofl:

No, I think the rationale is that mines are generally used by the defender in defending a fixed position, and as such a defensive bonus. The MG is more difficult, as it is efficient both as an offensive weapon (suppressing enemy strong points), and a defensive weapon in fixed positions against enemy advances. +1 attack and +1 defense?

But I guess if I would have to choose one, I would avoid lugging along a maxim HMG when advancing on the enemy

 

[Truchses]

Yes, of course I understand that MGs and mines are more advantages in defense than in attack. But as I also understand, the effect of these weapons might of course be to slow down and thus absorb attack points, but they are doing this in first place not like armour or energy shields, but by causing casualties to the attacker. If they get only a defense bonus then, as I understand, this will not cause casualties to the attacker.

I guess they should therefore 1 defense bonus and 1 attack bonus, though this looks strange for mines, but HOI works this way. MGs should of course get an additional SA point when attacking. The problem is that again it seems that HOI does not differentiate between attack and defense. At least as I understood, 'attack' does not mean 'attack' but 'causing casualties', and 'defense' does not mean 'defense' but 'armour' or 'energy shield'.

 

[Tamerlan]

Mg/mines are useful in a defensive purpose. Therefor U tend not (cancel) to attack a position full of MG/Mines.

AFAIK, the HA/SA/GD system is really a good, simple model.

 

[Truchses]

Yes, MGs and Mines are useful in a defensive purpose, as I stated several times. But why is the HA/SA/GD system realy good? I read posts of other people with different opinion. And you don't describe how specifically the GD works and how in HOI mines and MGs as defense weapons cause casualties to attackers.

 

[Mithel]

Well I certainly have a different opinion as additional GD points do NOTHING if you already have more GD than the attacker has attack points.

And if you have less GD than the attacker has attack points then you are taking horrendous losses and each additional attack point is extremely lethal.

- Mithel

 

[Tamerlan]

Well I certainly have a different opinion as additional GD points do NOTHING if you already have more GD than the attacker has attack points.

- Mithel

How do u know that ? If GD works as said in the manual (cough... ) GD should have an exponential effect when above the attacking value.

I just read on the GD forum somebody praising Mountain troops. Not from a theoritical point of view apparently.

 

[Truchses]

Is GD now Ground Defense or General Discussion or something else? Even if the GD works in a logical sense in decreasing casualties of the defender, which I doubt considering Mithel's post, then, again, that does still not mean that the GD causes casualties to the attacker. The only way, as I see it, is to give mines and MGs SA points. But that would again mean, that the unit gets this points also when it attacks, which further means, that HOI does not differentiate between attack and defense, and the words, as they are used in HOI, have gotten different meanings as I described.

 

[Tamerlan]

Even if the GD works in a logical sense in decreasing casualties of the defender, which I doubt considering Mithel's post, then, again, that does still not mean that the GD causes casualties to the attacker.

Sorry, but how do u know GD does not work ? It is not the first time I encounter people assuming such a thing. Where does it come from ?

If it works as it is supposed too, it is really a good thing to have a super-high GD value.

 

[Truchses]

Sorry, but you did not read my posts. I never wrote that GD does not work. Also I did never assume it. You are wrong in assuming that I assume that it does not work. I criticised how it works. You are right, if it works as it is supposed to, it is good to have a super-high GD value. But I think that is not the way how it should work. It drifts more off into fantasy. And if you assign 999 GD to a unit than what sense make the upgrades like GD +1 for mines? Also you have again avoided the part of my starting question, how the casualties that are inflicted by mines and MGs are featured with GD.

 

[Tamerlan]

Sorry, but you did not read my posts.

Truchses, I read ur posts.

I never wrote that GD does not work. Also I did never assume it. You are wrong in assuming that I assume that it does not work

U wrote u doubt it .

... how the casualties that are inflicted by mines and MGs are featured with GD.

The fact of increasing the GD does not directly increase the casualties inflicted, but overall it increases the ratio of damages inflicted/damages received in favour of the guy who has the mines/MG. And instead of speeding up the combat, they slow it down. Take WW1 with minefield/lots of MGs and u understand they are not blitzkrieg weapons. The current model seems normal to me.

If you speak of subMGs, then yes, they are an assault weapons, so then yes SA should be favoured.

 

[Mithel]

We have Mikel, Math Guy, my own tests and several other people that have spent nearly a year testing the combat engine. All GD points higher than the attacking strength do nothing. You can test this easily by playing the Starfire Historical Mod which has had all GD values at 999 or higher for over four months now.

GD certainly doesn't directly inflict any losses on the attacker, however by lengthing the combat it could effectively result in higher losses for an attacker.

Don't confuse this with "ground_def_eff" which impacts the enemy effectiveness of ground defense.

Or are we discussing changes that have been made in v1.06 that the rest of us are unaware of?

- Mithel

 

[Tamerlan]

We have Mikel, Math Guy, my own tests and several other people that have spent nearly a year testing the combat engine. All GD points higher than the attacking strength do nothing. You can test this easily by playing the Starfire Historical Mod which has had all GD values at 999 or higher for over four months now.

Hmmm, yeah I have seen a lot of work there, but silently, I was thinking that might result in speed of the combat problems... I mean I bet u can set the right ratio of killed received inflicted/received just with the SA through different tech level, but did you manage to arrange right speed of combat without taking into account the GD ? Tipically take (Tiger-Sherman)(JSII-Tiger), probably both have high casualty rates, but did you manage to have the speed of the combat not being screwed ?

GD certainly doesn't directly inflict any losses on the attacker, however by lengthing the combat it could effectively result in higher losses for an attacker.

I agree

Don't confuse this with "ground_def_eff" which impacts the enemy effectiveness of ground defense.

Yeah, but both are taken onto account isn't it? For 1 point GD there is (normally) ground_def_eff percent chance of a scored hit point cancelled?

Or are we discussing changes that have been made in v1.06 that the rest of us are unaware of?

- Mithel

I am speaking of 1.05c, even without the NDA, I would not be the right person to speak of 1.06. The big boss said that Math Guy is a beta for 1.06. Jump on him when 1.06 is release . I don't know if he raises the point or not.

 

[Truchses]

U wrote u doubt it .

No, I wrote in my first post, that I thought, that GD absorbs attack points, which seems to be the general agreement, and it is clearly contrary to that I would doubt that it works. My doubt referred to the realistic sense of the implementation of GD in HOI. This doubt seems to be supported by Mithel and some other guys, and by the problem with Mines and MGs, because of which I started this thread.

The fact of increasing the GD does not directly increase the casualties inflicted, but overall it increases the ratio of damages inflicted/damages received in favour of the guy who has the mines/MG. And instead of speeding up the combat, they slow it down. Take WW1 with minefield/lots of MGs and u understand they are not blitzkrieg weapons. The current model seems normal to me.

I agree that mines and MGs are not considered blitzkrieg weapons, though even german attack tatctics were based on the utilisation of MGs, and it was the main weapon of the mot troops. Still the effect you describe is the effect of armour and shields, not of MGs and mines.

If you speak of subMGs, then yes, they are an assault weapons, so then yes SA should be favoured.

The problem that I addressed is that mines and MGs inflict casualties, and that GD does not feature this. Therefore, as substitute, they should get also SA points. But another problem is that HOI does obviously not differentiate between attack and defense. So the attacker will also profit from MGs and even mines.

 

[Soapy Frog]

Still the effect you describe is the effect of armour and shields, not of MGs and mines.

GD represents a units ability to resist attack (not deflect bullts or whatever you seem to think it is for). There is no question that machineguns and mines are beneficial for this purpose. They reduce the effectiveness of, or even nullify, attacks by enemy forces.

Machine guns are not as useful in an offensive role, and mines are clearly defensive weapons.

 

[Truchses]

Yes, I know, I know, I know. And I did not question this. And, again, as far as I ever heard about MGs and mines, they reduce effectiveness of attacks by inflicting casualties to the enemy, not by providing an energy shield.

 

[Mithel]

Tamerlan, a major reason for raising GD to 999 (or higher) is to fix the speed of ground combat. Stock Vanilla HoI has very touchy combat timing. If you overwhelm the enemy the battle is over in blink. If you have small equal forces then battles will last a long time. In the Starfire Historical Mod battle length is a direct ratio of the relative strengths of the forces (the more you overwhelm the defender the faster the battle but overall we avoid any "instant" battles). The higher the GD values the more predictable the combat lengths.

- Mithel

 

[unmerged(13914)]

I am speaking of 1.05c, even without the NDA, I would not be the right person to speak of 1.06. The big boss said that Math Guy is a beta for 1.06. Jump on him when 1.06 is release . I don't know if he raises the point or not.

I won't risk the wrath of the Beta Police to talk about 1.06, but I haven't in the past proposed any changes to ground defense or the basic land combat mechanisms. I haven't even asked them to fix the ground_def_eff bug, because the consensus seems to be among modders that it is a fairly useful bug to have. I want to mod unit strengths in various ways, but I have never asked Paradox to change how combat fundamentally works.

Mithel is exactly right on how GD works. Our only two disagreements are that:

(a) So far I'm not keen to completely eliminate the "dog-pile" effect that occurs when one side has more SA points than the other has GD points. I want to increase GD but not to 999 points.

(b) I'm not sure that battles need to be a whole lot longer, because in practice a battle's full length is the march time plus the combat time. When you factor in marching time, battles seem close to the right length.

However, to give credit where it is due, Mithel has been moving ahead with his Starfire mod and getting results for the past several months, while I have mostly been waiting. He has gotten positive feedback from users with his approach, and I gather campaigns like the German-Soviet war are taking longer, as they should.

Version 0.1 of the Historical Statistics Pack will be based on 1.06. Once it is out, I will start serious testing on a variety of topics and see whether the results are truly historical. At that point you might see a version 0.2 that looks more like Starfire . . . or you might see me go to Paradox and start asking for changes to the combat system.

 

[unmerged(13914)]

Truchses,

It's tricky coming into the middle of a long "family argument". As Mithel said, we spent months trying to figure out what GD did. The manual is no help at all. Then once we knew how GD worked, we each had different ideas on what to change to make the game more realistic.

You're right that if you set GD = 999, a bunch of techs that now add GD become irrelevant. That is one reason why Mithel's approach is not universally popular. There are ways to fix that, but they're complicated.

You're right that machineguns and mines only have effect by killing people (obviously). But basically we only have two variables to play with in designing combat units -- SA and GD. A division may only have 5 SA and 7 GD, and that (plus the ground defense efficiency) is all the information we have about it. It's very abstract.

Now, a unit with high GD values can fight several enemy units at once and survive, if their combined SA is less than that unit's total GD. A unit with a lot of machineguns and a lot of minefields can cope with large numbers of enemies much better than a unit that has nothing but riflemen in trenches. So it makes sense to give it more GD. Call it "suppressing fire" -- the machineguns and mines do more than inflict casualties, they discourage the enemy from advancing at all.

In an ideal world, you could give the unit more SA (for the casualty-inflicting power of their machineguns) as well as more GD. But in HOI if you give SA for EVERY tech that inflicts casualties, units wind up being 3x stronger than they are now, and that means battles go 3x faster. To correct that, you have to also increase the value of ground_def_eff (the percentage of hits that one point of GD will prevent). And then if a "dogpile" occurs, the casualty rate jumps much more dramatically. So then you start thinking like Mithel has done, "I'll give units so much GD that dogpiles never happen." But now many techs give no benefit at all, so why build them . . .

It's a tough design problem, to give the right amount of SA and GD, portioned out in 50 different techs and a dozen unit types, and then have the results be historical. So far I don't think anyone has a perfect solution.

The one thing I am sure of is that standard HOI doesn't give enough GD. So I am doubling all GD values. I am also sure that HA (hard attack) needs to be higher because tank losses in HOI are far below historical. Before I go beyond that, I want to do more testing.

 

[Crusher Bob]

For those of you just joining up, here’s a description of how ground defense works:

Every ground defense point has a chance to stopping one attack point. ‘Extra’ ground defense points do nothing. The chance of a point of GD stopping an attack is based of the ATTACKERS ‘Ground Defense Efficiency’ (this is a bug, but a useful one, just think of it as ground attack efficiency instead, with lower values being better).

Sample:

A unit with 10 GD is under attack from a unit with 11 Attack. The attacker has a GDE of 80% (the defender’s GDE is irrelevant) which means that points of GD will stop attacks 80% of the time. So of the 10 points of attack ‘met’ by a point of GD, 8 (80%) of the attack points are stopped, with 2 getting through. The 11th point of attack that if not met by a point of GD automatically gets through. (3 attacks get through). Raising the attackers GDE to 90% would mean that 2 attacks get through. Raising the defending units GD to 11 means that ~2.2 attacks get through.

Note that in the first sample the 11th point of attack that is ‘unmet’ is worth 5 points of ‘met’ attack.

A units combat efficiency (what you see on the screen when battles take place) changes the GD and attack values. This is why certain combats (like air planes under night flyers) are massacres.

Sample:

Using the above units, it the defender is at 1% efficiency and the attacker is at 40% this means that the defender effectively has 1 point of GD while the attacker has ~4.4 points of attack. This means that the attacks that get through in this example are ~3.6.

The dog-pile threshold, this is where a ‘large’ number of attacks are getting through without being ‘met’ by points of GD, this leads to greatly increased casualties


Recap:

Anything which raises your GDE is bad (enemy defenses are MORE effective), while anything that lowers to GDE is GOOD (enemy defenses les effective).

‘Extra’ Ground defense is not used.

Combat efficiency affects both attack and GD.

A battle where one side is inflicting ‘unmet’ attacks is likely to be very quick (and a massacre).