Discussion about unrealistic combat in TOAW III

[Mantis]

Can I get a save right before the combat, or a description of which hexes to attack with which units? I'll see if I can explain what's going on there.

Thanks,
Ralph

(Email address removed)

Ralph, just a quick word of advice - you should never enter your email addy straight out on a public website or bulletin board - it will be spidered by the search bots and your spam levels will rise... :angry: Use a few oddball conventions such as:

myemail *at* mydomain (dot) com type of thing, and then you can get your message acorss without getting spammed by the inevitable e-trash that will follow. (You might wanna go edit your previous post - your call!)

Just a friendly tip!

Best,

Shane

 

[ralphtrickey]

Ralph, did you open the save yet? I am looking forward to your conclusions, if any.

I've taken a look at it. My results mostly back up what Bob said based on the logs, but it did take a closer look at it, and found some interesting things.

Here's a copy of the units involved in the attack. The first number is the equipment count, the second and third numbers should be ignored right now. The TK means it's a tank, the AT means it's got AT and the AP means it's got AP capabilities.

The interesting thing is that code both sides only get 2 shots per round. There were a couple of bugs in the code that caused there to be only 2 shots per round. Norm had orginally written this to be between 1 and 10 with a slight benefit for the defense. If I modify the code to be what i think Norm intended, I'm seeing that the Russians get 10 shots for each 1 shot the Germans get.:bite:

The gang and I are going to have to kick the tires and see what other affects this may have, but this, or some variation will probalby end up in the next patch. It may also help with the ant problem, among other things, but I'm concerned that it's a large change. I'm pretty sure it's what Norm wanted to do, but there are some curiosities.:nuts:

I'm also not sure how far back this problem goes, and if it goes all the way back to COW or before. If anyone wants to research it, I'd be curious about the results. I think that you can look at the toaw log for the word max shots to see if there were any non '2' values there.:clown:

Anyway, I want to thank everyone for their persistence.:lier:

3/ 9/ 9 TK AT AP Wheeled 20mm SPAAG
7/ 21/ 21 TK AT AP Tracked MMG Scout
8/ 24/ 24 TK AT AP Tracked HMG Scout
15/ 45/ 45 Truck
18/ 54/ 54 TK Tracked tractor
9/ 27/ 27 TK AT AP SdKfz 251/1
21/ 63/ 63 TK AP PzKpfw II (flame)

6/ 18/ 18 AT AP Motorcycle Squad
60/120/ 60 AT AP Rifle Squad
21/ 42/ 21 AT AP Engineer Squad
20/ 60/ 40 AT AP Heavy MG
4/ 12/ 8 AT AP 45mm AT Gun
6/ 18/ 18 AP 76mm Gun
4/ 12/ 12 AP 122mm Gun
8/ 24/ 16 AP 50mm Mortar
3/ 9/ 9 AP 82mm Mortar
15/ 45/ 45 TK AT AP BA-10 Armored Car
40/120/120 Truck
15/ 45/ 45 TK AT AP T-26
197/591/591 TK AT AP BT-7
48/144/144 TK AT AP KV-I (early)
15/ 45/ 45 TK AT AP KV-II
86/258/258 TK AT AP T-34/76 (early)

Axis anti armor max shots per target : 2
Axis anti personnel max shots per target: 2
Soviet anti armor max shots per target : 2
Soviet anti personnel max shots per target: 2

 

[Olorin4]

Thank you Ralph (and Bob)!!

I will do some tests on ACOW and will get back soon with results.

 

[El Cid]

This is the sort of thing that I like most to see in a patch update, making combat more realistic.:clap:

 

[arclight]

This is the sort of thing that I like most to see in a patch update, making combat more realistic.:clap:

I concur - modeling the engine to preform to its best ability, and omiting the smell of smoke, cordite, and the normal enviroments of combat, would be great.

Ralph and co. have done a yeoman's effort to keep this gem chugging along.

 

[Fungwu]

I am glad to hear some light is being shined on this issue, but a few things in this thread left me scratching my head.

"Ok, fortified anti-armor values are multiplied by 5. It's the anti-personnel values that are multiplied by 8. The attackers do have some non-armored equipment."

There are a few problems with this. One is that flame panzers have an anti armor value of 0, and, if my math is as keen as ever, 5x0 = 0.

Another is that the best German at gun in this scenario is a 20mm cannon. The 20mm cannon cannot penetrate a soviet T-34 or KV-1 or KV-2 from any angle. If you multiply the defenders' anti armor value by 5 or by a million a 20mm cannon still cannot penetrate those tanks, so I don't understand how a defensive anti armor multiplier is relevant here.

How big of a pillbox do you build for your 20mm AA guns before it can penetrate more armor?

The only relevant types of fortifications that come to mind are anti tank mines and anti tank ditches, but those would benefit units of any size and make up equally, while only small fortified armor units seem to be especially invincible. And in any case those emplacements would still not allow the defenders fire to damage the attackers tanks.

"Regardless of combat strengths? The defender's defense strength was 5, the combined anti-personnel strengths of the two attackers was 34, and anti-armor strengths were 23. The defender was fortified. Combine the bonuses the defender was getting: 1.56 shock, 1.66 readiness, 2.14 supply, 1.5 proficiency and either 5.00 or 8.00 for being fortified. That adds up to a lot, resulting in very little, if any, actual attack margin."

This post leads me to believe that you are completely and totally missing the point of this example.

I felt something was wrong with the combat calculations so I set up a totally absurd situation that would irrefutably prove that the combats involving fortified armored defenders were broken.

On one side is the defender with tanks that had absolutely zero anti tank capability and whose armor plating could be pierced easily by the attackers tanks. The supporting units were confined to a few machine guns and light anti air guns with very little anti armor ability.

On the other side was an attacking force that not only possessed tanks whose armor rendered them nearly immune to any defensive fire, but also massively outnumbered the defenders.

Using Olorin's test the attacker had 366 tanks total and 151 that were almost invulnerable to the defenders fire. This vs. 21 tanks that are basically machine gun tanks with no anti armor ability at all. The defenders had only 11 guns that could reply with any kind of anti armor fire at all, and all these were heavy machine guns and 20mm aa guns.

No matter what numbers you crunch how could the results of this battle possibly be realistic?

What if the attacking force instead of T-34s, was 100 M1a2 Abrams tanks and the result were similar?

Would you sit here telling us that everything was okay, because of course panzer 2s can defeat Abrams tanks if they have a fortification AND a supply bonus!

If, after crunching the numbers, we have discovered that everything is all in order, imagine the "what if" scenarios we could come up with?

To think: instead of building tanks with thick armor and good guns, the Germans could have just built tanks with 30mm frontal plates and 20mm cannon, plopped one down every 500m and thrown a few sandbags on top. According to the insights we have developed here an arrangement of 2000 such tanks could easily hold of Soviet forces of 30,000 tanks or more! Thank god Hitler did not catch on to such wisdom or the Reich might have really lasted 1000 years!

 

[Bob Cross]

I am glad to hear some light is being shined on this issue, but a few things in this thread left me scratching my head.

"Ok, fortified anti-armor values are multiplied by 5. It's the anti-personnel values that are multiplied by 8. The attackers do have some non-armored equipment."

According to Ralph, the manual was wrong. Fortified anti-armor values aren't multiplied by 5.

Another is that the best German at gun in this scenario is a 20mm cannon. The 20mm cannon cannot penetrate a soviet T-34 or KV-1 or KV-2 from any angle. If you multiply the defenders' anti armor value by 5 or by a million a 20mm cannon still cannot penetrate those tanks, so I don't understand how a defensive anti armor multiplier is relevant here.

TOAW rates them as 7, 9, and 8 armor values respectively. There is a combat feature that gives a chance of exposure of side armor. That armor value would be some fraction of those numbers (I don't know the exact divider, but probably about 4). That would more than allow for the 20mm to kill some of the attacker tanks. The Soviet tank losses were small fractions of their total. That's not the main issue. It's the dislodgement of the unit.

"Regardless of combat strengths? The defender's defense strength was 5, the combined anti-personnel strengths of the two attackers was 34, and anti-armor strengths were 23. The defender was fortified. Combine the bonuses the defender was getting: 1.56 shock, 1.66 readiness, 2.14 supply, 1.5 proficiency and either 5.00 or 8.00 for being fortified. That adds up to a lot, resulting in very little, if any, actual attack margin."

This post leads me to believe that you are completely and totally missing the point of this example.

Ah. So combat strength is irrelevant? So let's change the Soviet prof, supply, & readiness to 1%, hit them with a 99% shock penalty, and leave the defenders fortified with a 200% shock bonus. They should still trounce those Flame Panzers, right? After all, all that matters is number of tanks, right?

As I pointed out, if you just increased the odds a little, the defender could be dislodged. The main reason the attack was failing was because the odds were too low - effectively below 1:1.

But, as Ralph pointed out, another factor was the bug he found that limited the number of shots the attackers were allowed. That's already been addressed.

 

[Olorin4]

So combat strength is irrelevant? So let's change the Soviet prof, supply, & readiness to 1%, hit them with a 99% shock penalty, and leave the defenders fortified with a 200% shock bonus. They should still trounce those Flame Panzers, right?

Pretty much. Unless all attacking tanks break down, or their crews are 15 year old girls, they should easily beat a few non-threatening vehicles, that are placed 200m from each other. I don't know if the bug discoverd by Ralph is responsible for the absurd results of this example, but if it's not, then we should seriously rethink the way battles are resolved. I hope the steps you mentioned earlier and the bug fix are enough to produce realistic results.

 

[Bob Cross]

Pretty much.

Was that a joke? We're actually trying to take this issue seriously, including the application of considerable effort by the team. I'm beginning to wonder why.

 

[Bob Cross]

Ok, here's the new combat system we're going to implement for TOAW III:

We're going to completely ignore unit proficiency, supply level, readiness, deployment mode, cooperation, terrain, and shock levels. All that will matter will be the equipment contents of the units involved. Defenders will be treated as if standing up, out in the open, shoulder-to-shoulder, on a flat, level, featureless plane. They will be presumed to remain cemented in their starting tactical locations as the attackers advance on them, being enveloped if not covering the entire width of the hex.

Because anything else would be absurd.

 

[Olorin4]

We're actually trying to take this issue seriously, including the application of considerable effort by the team.

If that is indeed the case, then that's all I want to hear. That the problem is acknowledged. But, save your statement above, the impression I got was exactly the opposite.

And no I wasn't joking. I still think that 20 flame panzers are not enough to stop a full strength Tank Corps, no matter the deployment, supply and other modifyers. :smoke:

 

[Bob Cross]

Now let’s take a “real world” look at this attack to contrast with the parade-ground fantasy expressed above.

First, the defense is fortified. That means there are plentiful entrenchments, pillboxes and bunkers for the troops and revetments for the vehicles across the defense and in-depth. Minefields and tank obstacles and ditches will tend to channel attacks into killing zones, where the attackers will be struck from multiple angles, and prevent their easy penetration of the defense.

The defender is high proficiency. It has exploited any terrain advantages available, including positioning the defense along the local high zones, leaving the attackers looking up at them. Their defensive deployment is well designed, with only a small fraction of the unit actually in the front lines – more as lookouts than as actual defenders. Most of the unit is held in reserve as a mobile reaction force, ready to respond to any attacks. They enjoy a shock bonus, perhaps reflecting exceptional intel about the attackers – so much so that they can anticipate their every move. They easily sniff out the coming attack and are waiting for it.

The attackers are low readiness. A significant fraction of their equipment must be left behind in need of maintenance. They are low proficiency. They rigidly adhere to doctrinal procedure with little room for initiative – just what the defense expects. They tend to be ill coordinated so that their attacks are launched piecemeal. The shock penalty they suffer from reflects the mass-confusion that their high-command is under. As a result, the attack orders they received were ambiguous, confusing, and even contradictory. Because of this, a significant fraction of the attackers will not participate in the attack – never leaving the starting blocks. The rest will be even more ill coordinated than normal.

Once the ill-coordinated attacks get started, the defense’s mobile reserve concentrates against the first part of it. Channeled into the kill zones, the attackers are blocked by tank obstacles and are suffering side hits from the defenders. The infantry supporting the attack would love to remove the obstacles, but whenever they expose themselves to do so, they are mowed down in a hailstorm of lead by the heavily over-supplied defenders, who have ammo to burn. The tanks try to knock out the defending vehicles, but due to the elevation disadvantages and the revetments, they can’t see more than the tops of their turrets. They weren’t very good shots to begin with, and they have to conserve ammo due to their shortage of it. The defenders, being high proficiency, are doggedly determined to hold their ground, despite losses (not that they’re taking many). The attackers, on the other hand, are low proficiency and soon get frustrated and give up. This happens with plenty of time for the defenders to shift their reserves to concentrate against the next part of the piecemeal attack.

After a few repetitions of this it all ends – having wasted a lot of time but accomplishing little.

 

[Bob Cross]

If that is indeed the case, then that's all I want to hear. That the problem is acknowledged. But, save your statement above, the impression I got was exactly the opposite.

A problem is acknowledged. Not the idiotic need to ignore all combat modifiers that you're touting.

And no I wasn't joking. I still think that 20 flame panzers are not enough to stop a full strength Tank Corps, no matter the deployment, supply and other modifyers. :smoke:

First of all, had it been the entire tank corps (including the mech division) then, as I showed, the attack would have likely succeeded. And it clearly can't be considered full strength if its readiness was 1% - 99% of it would be in the shop. And a 99% shock penalty would mean that only 1% of what remained operational would have participated! And the flame panzer unit had a total of 48 active vehicles, not 20.

 

[Olorin4]

There is no point arguing with you any more, You showed that your mind is already made up and no matter how absurd an example is presented, you will try to defend the results as realistic. They are not. It's funny to claim they are. Get over it.

 

[Heldenkaiser]

Now let’s take a “real world” look at this attack [...]

With no intention to take any side on the issue--as the nuts-n-bolts of TOAW are beyond me anyway--I simply have to say I thought this narrative was an exceptionally vivid and lucid illustration of how there's more to an attack, in the real world and hopefully in TOAW too, than "so many tanks meet so many tanks". :toast:

 

[Noxious]

There is no point arguing with you any more, You showed that your mind is already made up and no matter how absurd an example is presented, you will try to defend the results as realistic. They are not. It's funny to claim they are. Get over it.

And how about you start using a realistic example of non-realistic combat results ?
You're using a broken example to highlight a "broken" system ?
Do you understand anything about statistical behaviour ?
If you do, you realize your example is worthless on its own.
You need to prove that statistically, over a wide range of realistic combat setups, you get unrealistic results.
If you only take one extreme example that seemingly proves your point and stick to it, you're exhibiting the exact same behaviour that you're accusing Bob of

Cheers,
Nic

P.S. : I'm not saying anyone is wrong or right, I'm just saying you're not being rigorous, that's it.

 

[Bob Cross]

With no intention to take any side on the issue--as the nuts-n-bolts of TOAW are beyond me anyway--I simply have to say I thought this narrative was an exceptionally vivid and lucid illustration of how there's more to an attack, in the real world and hopefully in TOAW too, than "so many tanks meet so many tanks". :toast:

Thanks.

If I could just get him to wrap his mind around the - apparently novel - concept that the combat modifiers actually represent real and significant combat factors then the whole issue would clear up.

Because, when you apply those modifiers, what the combat boiled down to was an armored defender with a defense strength of 5 that was fortified (effective defense of 25). The attackers had a combined anti-armor strength of 23. In effect, the attack had odds of less than 1:1.

And, was that particularly unrealistic? After all, these units and circumstances were drawn from the first turn of Barbarossa!

TOAW has to base its combat model on the numbers. What the sizes of the units are is irrelevant. Now, can the designer make it so that those same forces would produce a different result? Sure. Just change all those combat modifiers. Should the game produce a victory for that corps regardless of his combat modifier choices? Of course not.

 

[Olorin4]

And how about you start using a realistic example of non-realistic combat results ?
You're using a broken example to highlight a "broken" system ?
Do you understand anything about statistical behaviour ?
If you do, you realize your example is worthless on its own.
You need to prove that statistically, over a wide range of realistic combat setups, you get unrealistic results.
If you only take one extreme example that seemingly proves your point and stick to it, you're exhibiting the exact same behaviour that you're accusing Bob of

Cheers,
Nic

P.S. : I'm not saying anyone is wrong or right, I'm just saying you're not being rigorous, that's it.

It's not my job to do the work. I am just one of several people, including probably the best scenario designer, that expressed a concern about a possible bug or bugs in the combat engine. I presented an absurd example, with repeatable results, that's only representative of the many other battles involving armored ants that resisted 50 times stronger enemies. I think it warrants a closer look and as Ralph said, he did take a closer look and he did found a bug. Despite this, Bob still thinks that all is well with the combat engine.

 

[Bob Cross]

I presented an absurd example...

The problem is that it wasn't an absurd example. It was an attack at less than 1:1 odds that failed. What's absurd about that?

...with repeatable results, that's only representative of the many other battles involving armored ants that resisted 50 times stronger enemies.

They weren't "50 times stronger". They were effectively slightly weaker. That's not an opinion. That's a mathematical fact.

 

[Noxious]

It's not my job to do the work.

You're kidding, right ? How can you say this and bother to come up with a repeatable example ?
Seems to me you're putting a lot of work and effort and passion into something that ain't your "job"

Yes, Bob is very strong headed, but he always documents his affirmations, and doesn't resort to childish behaviour : just because he disagrees with your interpretation of the facts doesn't mean he's wrong or that you're wrong...

Again, it feels more and more like you're sticking to your guns for the sake of sticking to them, when you've already achieved a code review by Ralph :nuts:

To finish, if you're not ready to do the work or be proven wrong (or right) with all it implies, don't bother posting bug reports : it's a lot of work for you and even more work for the devs (in direct inverse proportions to the work you've put in the report), however good your report is...

You forget that Bob is also one of those top scenario designers and a very active member of the small beta team.
Cheers !!
Nic