Page 1 of 1

some numbers

Posted: Fri May 29, 2009 10:25 pm
by damezzi
I was always curious as to what extent modifiers like those for flanking or cooperation would affect combat in Toaw, since the manual doesn't show numbers. So, I set up some combats for the following tests:

- artillery bombardment near peak -> since the manual doesn't state how this bonus is gained, I tried different configurations. Even things like enemy unit above peak, which may seem absurd, but I wanted to be sure.

- infantry flanking infantry - well, I didn't expect a difference here, since the manual states that flanking affects passive defenders that are exposed to combat.

- infantry with some art being flanked

- Three different reconnaissance level infantry units trying to disengage

- 2 stacked infantry units limited cooperative attack; non-cooperative attack with the same configuration and cooperative attack.

I have run the attacks 40 times each and have excluded extreme values for casualties, like those from evaporation, to avoid distorting the results.

I couldn't state any advantage for artillery based on it's position relative to peak. Does someone knows how this is gained?

And most important: absolutely no advantage for the more cooperative attacks. In fact, the manual states that non-cooperative units will tend to break-off earlier and that may explain the greater number of evaporations I have experienced with the full cooperative attacks. But in fact, at least at the odds I set this seems to bear a disadvantage to the cooperative attack, since most evaporations were from attacking units. Maybe that with favorable odds this would be inverted, but I don't see why cooperative units should experience a disadvantage when at low odds.

I want to test further the advantage of cooperation, since I think this is one of the most interesting concepts in Toaw and I would really be disapointed to find it doesn't work as expected. In game it is difficult to isolate those variables so I have run those tests.
If someone knows of a good setup for testing the effects of cooperation, please tell me. Maybe with more complex combats it will show more consistent results.



Image

RE: some numbers

Posted: Sat May 30, 2009 1:40 pm
by Curtis Lemay
ORIGINAL: damezzi

- artillery bombardment near peak -> since the manual doesn't state how this bonus is gained, I tried different configurations. Even things like enemy unit above peak, which may seem absurd, but I wanted to be sure.

...

I couldn't state any advantage for artillery based on it's position relative to peak. Does someone knows how this is gained?

There's also an advantage for bombarding down a major escarpment. In that case, the target must be adjacent to the artillery unit to get the benefit. I expect that it's the same for the peak bonus, too. Were they adjacent in these tests?

RE: some numbers

Posted: Sat May 30, 2009 7:53 pm
by damezzi
In fact they weren't adjacent. I remember the manual saying that artillery with a nearby peak would have this bonus, but I just can't find were in the manual. Perhaps I'm just crazy.
Well, I expected it to work in a different way, since peak can proportionate a 40km visibility.

I'll run the tests again with all units adjacent.

But the tests I really want to run are those with  cooperative and non-cooperative units. I'm not proving the system; I just want to measure the impact of those modifiers. I think that could be stated in the manual. Realistic or not, I like to know the rules and I think it can help in scenario design.

Has someone experienced a situation where cooperation vs non-cooperation in combat showed a significant influence? I mean in combat considering the participating units, since non-cooperative units won't support attacks and there are also supply and movement penalties. I just want to place some different setups for this, since the one I used didn't favor the cooperative units attack.

RE: some numbers

Posted: Sat May 30, 2009 9:21 pm
by rhinobones
ORIGINAL: Curtis Lemay
There's also an advantage for bombarding down a major escarpment.

If you know the math model, I would be interested to hear how much advantage there is for artillery firing from a higher elevation. I would expect that it is predicated on the ability of artillery spotters being able to pass along information that makes fire more accurate.

Regards, RhinoBones


RE: some numbers

Posted: Sun May 31, 2009 3:48 am
by el cid
I hope the experts give an answer to you questions as they are of much interest.

I do have a few comments/suggestions.

As far as I remember in TOAW you only get benefit for flanking when attacking from hexes that are 120º or more. But it does not say you would get more beneft for attacking from 180º than from 120º. The results for those should be the same.

In the test for cooperation, have you looked at how much of the turn is used? Are there units that keep on fighting, after others withdraw.?Would you care to run the test of attacking with pure infantry unit plus pure armor unit against a unit that has no AT capacity with different levels of cooperation?

RE: some numbers

Posted: Sun May 31, 2009 10:42 am
by Curtis Lemay
ORIGINAL: el cid

As far as I remember in TOAW you only get benefit for flanking when attacking from hexes that are 120º or more. But it does not say you would get more beneft for attacking from 180º than from 120º. The results for those should be the same.

That's changed under the New Flanking Rules.

RE: some numbers

Posted: Sun May 31, 2009 6:33 pm
by damezzi
I will do further tests with cooperation using different configurations. I could state a clear tendency of the cooperative units to stay longer in combat. I didn't count the rounds for each combat, however, since it would be too painful to do that. This tendency to not break off may be an advantage when one needs to break through and the odds are favorable, but with low odds and ignore losses, like in my setup, it can be a disadvantage, since, even if the numbers for casualties and dislodgement are the same, the number of attacker evaporation was greater for the cooperative unit and turn burning would be greater too. That doesn't mean that cooperation isn't advantageous, since there are other variables involved, like combat support, supply and movement penalties and with other setups results may be different. That's what I'll check.
But sure, if some veteran has this information, work could be spared.

RE: some numbers

Posted: Wed Jul 15, 2009 12:33 am
by Curtis Lemay
I was finally able to make time to check some of these items.

First, the artillery bonus for being on a peak:

The attached table shows that there is considerable bonus if the target is adjacent, a little bonus if it is at a 2-hex range, and by 3-hexes, it's negligible. This was with 2.5km hex scale, in case the bonus's range is a certain number of km instead of hexes. The guns had ranges of 7 hexes at this scale.

Image

RE: some numbers

Posted: Wed Jul 15, 2009 12:38 am
by Curtis Lemay
Second, the effects of flanking the defender:

The attached table shows that there is a very significant bonus for flanking the defender (the Germans in the table). There were three combats of each type in each trial, and I ran five trials, for a total of 15 tests of each type. Defenders and attackers were pure infantry (all active).

Image

RE: some numbers

Posted: Wed Jul 15, 2009 12:43 am
by Curtis Lemay
Finally, the effects of cooperation level on combat results:

The attached table shows that, indeed, cooperation penalties seem to be reversed. Full cooperation fared worst both attacking and defending. I only ran five trials of each type, and that may put some question on the results, but I think it's enough to suggest there's an issue worth some more investigation.

Image

RE: some numbers

Posted: Wed Jul 15, 2009 7:09 pm
by damezzi
The flanking bonus seem to be very consistent.

As for the peak bonus, I didn't try it further, but I suspected my setups weren't good. I didn't even try to put targets so far away, since I though it would only be active if adjacent.
What about the artillery unit being adjacent? Is there a bonus? I always thought those bonus represented the firing side ability to better regulate artillery fire from a good observation post. Are the artillery units always in the peak hex in your tests?

The cooperation bonus problem was the only one I had almost no doubt about, since I run a great number of tests and even if setups were wrong, I can't imagine a situation in which non-cooperative units should have an advantage. Anyway, after running a symmetrical scenario with Elmer vs Elmer and finding that non-cooperative units had a huge advantage gaining terrain, something must be wrong. It's nice you've run some tests to confirm it and to show how the peak bonus work.

RE: some numbers

Posted: Wed Jul 15, 2009 9:37 pm
by rhinobones
ORIGINAL: Curtis Lemay
The attached table shows that, indeed, cooperation penalties seem to be reversed. Full cooperation fared worst both attacking and defending.

This is really incredible to hear . . . sounds like in some formula an operator has been inverted. Funny part is that it has probably been like this since the beginning. If you have an older TOAW version it would be interesting to see if the anomaly is present.

If you haven’t brought this to Ralph’s attention, I will. Think this needs to be a priority for the next patch. The fix could delay Ralph’s release schedule, but I think a fix to this problem is worth the extra wait.

With the problem fixed, I bet there will be some noticeable dynamics changes in the scenarios which have a lot of non-cooperating formations.

Regards, RhinoBones

RE: some numbers

Posted: Thu Jul 16, 2009 12:57 pm
by Curtis Lemay
ORIGINAL: damezzi

The flanking bonus seem to be very consistent.

I was pretty sure about it from previous tests made when we modified the flanking rule.
As for the peak bonus, I didn't try it further, but I suspected my setups weren't good. I didn't even try to put targets so far away, since I though it would only be active if adjacent.
What about the artillery unit being adjacent? Is there a bonus? I always thought those bonus represented the firing side ability to better regulate artillery fire from a good observation post. Are the artillery units always in the peak hex in your tests?

As I thought was clear in post #8: if adjacent, artillery on a peak had a 29% bonus in losses inflicted over an adjacent artillery unit that wasn't on a peak. If firing at two-hex range, the bonus was 13%. At three-hex range it was only 0.25%. But, note that it may not be a certain number of hexes - it may be a certain number of kilometers - I didn't investigate that.
The cooperation bonus problem was the only one I had almost no doubt about, since I run a great number of tests and even if setups were wrong, I can't imagine a situation in which non-cooperative units should have an advantage. Anyway, after running a symmetrical scenario with Elmer vs Elmer and finding that non-cooperative units had a huge advantage gaining terrain, something must be wrong. It's nice you've run some tests to confirm it and to show how the peak bonus work.

Kudos for finding it.

RE: some numbers

Posted: Sun Aug 30, 2009 11:22 pm
by ralphtricky
ORIGINAL: damezzi

The flanking bonus seem to be very consistent.

As for the peak bonus, I didn't try it further, but I suspected my setups weren't good. I didn't even try to put targets so far away, since I though it would only be active if adjacent.
What about the artillery unit being adjacent? Is there a bonus? I always thought those bonus represented the firing side ability to better regulate artillery fire from a good observation post. Are the artillery units always in the peak hex in your tests?

The cooperation bonus problem was the only one I had almost no doubt about, since I run a great number of tests and even if setups were wrong, I can't imagine a situation in which non-cooperative units should have an advantage. Anyway, after running a symmetrical scenario with Elmer vs Elmer and finding that non-cooperative units had a huge advantage gaining terrain, something must be wrong. It's nice you've run some tests to confirm it and to show how the peak bonus work.
Can you please post or email that Elmer v Elmer scenario?

Thanks
Ralph

RE: some numbers

Posted: Mon Aug 31, 2009 1:47 am
by damezzi
Scenario sent Ralph.