Does anybody know ...

[David Chapuis]

Does anybody know the exact rules on when/why engineer squads will throw satchels. I cannot express the extreme frustration I get when I order my engineers to area fire and use explosives, and they dont throw satchels. Or when they are less than 30m from armor with a target order, and they dont throw satchels.

It drives me NUTS!!!!!!!

And I know it has to be 29.5 meters or less. I know they wont throw if friendly are too close. There just seems to be a lot of randomness, and I am hoping somebody has figured out how to not make it random.

[bitchen frizzy]

There's been lots of discussion of that over the years. Here's one thread:

mine clearing

You can find a lot more with searches of the CMx1 Forums.

[David Chapuis]

I did search the cmx1 forums and found what that thread is saying - that it is pretty random.

Mostly I am concerned about the time it takes to throw a satchel - maybe it is random between 5 seconds and 90 seconds. I was hoping somebody could get a little more specific than "it is random".

[JasonC]

Think "pre armor attack task check". They are "rolling morale", and until they pass, don't have the brass balls to place the charge.

Unsuppressed, unmolested, veterans, in cover, stationary, good commander - all help.

Any suppressing incoming, much much less likely.

But it is random because it is random, you can't make it deterministic any more than you can make every tank gun shot fired into a perfect direct hit on the right plate.

This goes for "grenade" attacks on tanks by infantry squads, too - which are actually close assaults.

[The Coil]

Ran a quick test, lining up a whole bunch of engineers of varying experience levels, and let 'em have at it. I learned...nothing much. Granted there probably weren't enough of them for any kind of statistical significance, but they all started chucking satchels pretty quick, starting, oddly enough, with the conscripts. Everyone (talking 30+ squads here) chucked at least one satchel in the first turn, most chucked 2, a few chucked 3, except for one green squad who held on to their satchels. Draw whatever conclusions you like.

JC: How do you know this stuff? Not doubting, just curious. Is it extrapolation from ASL or did you crack the code or something?

Dave: Did you get the setup I sent you? Should I expect engineers?

[JasonC]

I am extrapolating from ASL and from lots of CM experience waiting for infantry to deliver close assaults or throw DCs. In my experience, you have to throw a realistic mix of movement and suppression states into a test to get a sense of how random it actually is.

30 seconds is typical to see a throw, but it can happen in 6, or not happen in 60. To me that says "geometric distribution" and PAATCs under the hood.

[David Chapuis]

Originally posted by The Coil:

Dave: Did you get the setup I sent you? Should I expect engineers? [/QB]

yeah I got the file. Why would you expect engineers from me. You know I prefer tank swarms!

Actually this question has nothing to do with our game. Do you know how aggrivating it is to actually sneak an engineer squad within satchel range of a tank or building - not the easiest feat under most circumstances - and then give the target order, and they just sit there with their head up their Leto shooting their rifles. In my current game squads went an entire turn without throwing one satchel at least 5 times.

[SgtMuhammed]

I tend to find my engineers rather satchel happy. I have had several throw charges while in a building and bring the entire structure down on their own heads.

[Ted]

When targeting around 30 meters it should say "Area Fire Use Explosives".

Usually within 40 seconds Regular Engineers should throw their charges.

Outside their throwing range, when targeting, it will just say "Area Fire"

[ February 27, 2008, 05:59 PM: Message edited by: Ted ]

[David Chapuis]

Originally posted by Ted:

When targeting around 30 meters it should say "Area Fire Use Explosives".

Usually within 40 seconds Regular Engineers should throw their charges.

Unfortunately it seems that nobody knows what makes this only "usually". Maybe it is just completely random - if nobody here knows something else I will just assume it is.

[JasonC]

Which part of "geometric distribution" is giving you trouble?

There are only two ways to time spacing between shots. Either it is a deterministic function or it has a random component. If it has any random component, then it is possible for a period of time to pass in which one unit fires ("throws"), and another unit does not.

If there is a probability of throwing p per unit time dt, then the probability of any throw in a time t goes as the cumulative distribution function (CDF) of the geometric distribution with parameter p, evaluated at t.

If there is a 2% throw chance per second then there is a 47% throw chance in 30 seconds and a 71% throw chance in 60 seconds. If dt is 6 seconds instead but the chance of a throw is 10% in each, then the chances of a throw by seconds 6, 12, 18, 24, 30, 36, 42, 48, 54, and 60 are 10%, 19%, 27%, 34%, 41%, 47%, 52%, 57%, 61%, 65%, 69%, 72%, and 75% respectively.

Let quality and suppressing conditions drop the per dt chance to 5% and those chances drop to 26% at 30 seconds and 49% at a minute. Raise them instead to 15% and the chances at 30 and 60 seconds are 62% and 88%.

Why is this any harder to understand intuitively than a hit probability, and your chance of KOing an enemy tank over a period of multiple shots? Do you get really frustrated that you don't know how to point the tank and change the covered arc settings so your tanks always hit with their first shot?

[bitchen frizzy]

It helped me a lot to understand that the "throwing" action is an abstraction of the actual process of placing and detonating the satchel charge, and the randomization covers the imaginary frustrations and delays your pixellated pioneers might be experiencing.

[David Chapuis]

Originally posted by JasonC:

Which part of "geometric distribution" is giving you trouble?

No reason to be surly.

I understand geometric distribution well enough. What I am not fully convinced of is that that is the only answer. I am wondering if there are any other "rules" that affect the geometric distribution.

For example, elite pioneers will throw sooner than conscripts - that is pretty intuitive. I was wondering if there was something that wasnt so intuitive.

another example not related to satchels - I just recently learned that for a leader's combat bonus to take affect the leader has to have LOS to the target. That is a "rule" that one has to learn. Another example - to ensure a tank is buttoned throughout an entire turn you have to unbutton and rebutton in the orders phase.

So, again, I am wondering if anybody knows of some "rule" that prevents or causes a satchel to be thrown. I assume there isnt, but it is always worth asking.

Originally posted by JasonC:

Do you get really frustrated that you don't know how to point the tank and change the covered arc settings so your tanks always hit with their first shot?

No. but I would get really frustrated if I had frequently managed to sneak a piat to within 40m of an unsuspecting enemy tank and, after given the piat a target order, it sits there for an entire turn and doesnt fire.

[ February 28, 2008, 11:48 AM: Message edited by: David Chapuis ]

[The Coil]

For the record, my purely anecdotal experience is that the biggest influencing factor is morale state - they'll almost never let fly if not at 'OK'.

Um...is there a difference in the frequency of satchel chucking if you area target a unit v. directly targeting them?

[JasonC]

Not being surly, I honestly don't get the difference or understand why it bothers you more than your tank's first shots missing their targets.

As for other factors that reduce the throw chance, several have been explained to you numerous times. Suppression is the single largest factor, quality a secondary one that does matter. I will add that friendlies right next to the likely blast location can also interfer with a throw. But all these have already been explained to you, and all of them are going to leave a random resolution component.

Throwing is pretty much hitting, so seeing the throw is seeing the attack succeed. If there is any remaining chance to not deliver, you will see typical binomial delays, which means sometimes you will get a cold streak and not see them throw for a minute or more, and sometimes a hot one, where they throw in 15 second or less.

If your piat missed, might you be frustrated? Sure, but there is no mystery in it. If your T-34 hits a Tiger at point blank from the side and penetrates, but does "no significant damage", is that frustrating? Sure - it is at least as hard to get the T-34 there. But it takes on average 3-5 penetrating hits to KO a Tiger - and sometimes you will get a string of misses and bounces and partials that only get crew member and fulls that still do "no significant damage" - while sometimes you will get it on the first solid hit and they'll bail.

Same with DCs, no different. You can't control the remaining variation, it is just there. Since obvious tests in conditions in which nothing else is wrong, have been reported to you in great detail, and show there is substantial variation of exactly the expected binomial kind even without suppression or interference or anything, what other evidence do you require, to see it is just random?

[Leprechaun]

Originally posted by David Chapuis:

I just recently learned that for a leader's combat bonus to take affect the leader has to have LOS to the target. That is a "rule" that one has to learn.

I've never seen this discussed before. I thought that the leader only needed to be in range of his subordinate units for the combat bonus to apply. Nice to know, thanks

[The Coil]

To be fair, there are a lot of events that appear random, but in fact are not. I think DC's inquiry was along the lines of "this appears random, I sort of expect it to be random, but was just wondering if anyone had stumbled across anything to suggest there was some non-random trigger, even though I expect there not to be".

[JasonC]

I get that, that is fair enough. What I didn't understand was what I (perhaps wrongly) took to be an insistence that somehow it had to not be random at bottom, that there just had to be knobs to make it work deterministically somehow. The test evidence is, without doing anything wrong, the time to throw still has a random element, just what you'd expect from a PAATC under the hood every n seconds.

[Silvio Manuel]

Originally posted by Leprechaun:

</font><blockquote>quote:</font><hr />Originally posted by David Chapuis:

I just recently learned that for a leader's combat bonus to take affect the leader has to have LOS to the target. That is a "rule" that one has to learn.

I've never seen this discussed before. I thought that the leader only needed to be in range of his subordinate units for the combat bonus to apply. Nice to know, thanks </font>

[hoolaman]

That's not what he said. The commander doesn't need LOS to its subordinate, but to give a bonus the HQ must be in LOS to it's subordinate's target. I have never noticed this effect but it may be true.

As for satchels, I noticed it is always the squads not taken under fire that throw em. A period of not being shot at will always do the trick. Obviously this is supression/morale related, but I find only fresh-as-a-daisy squads can do it.

[The Coil]

Originally posted by Silvio Manuel:

I believe that is incorrect, unless it's a change they made for CMAK, as I don't think it was in CMBO or CMBB. Having LOS greatly increases the distance a unit can be from an HQ and still be "in-command". But it certainly can be in-command while being out of LOS, and all bonuses "should" still be applied.

Just to be clear, the effect is: Unit, say a gun, in command of a leader with a combat bonus. Gun shoots at a target unit. If the leader has LOS to the target unit as well, combat bonus applies. If the leader does not have LOS to the target unit, the combat bonus does not apply. I'm all but certain it applies to both CMBB and CMAK, not sure about CMBO. It's easily demonstrated if you set up a test with a gun firing on a tank - watch the hit chances change based on whether or not the HQ can see the tank.

[David Chapuis]

Originally posted by Hoolaman:

That's not what he said. The commander doesn't need LOS to its subordinate, but to give a bonus the HQ must be in LOS to it's subordinate's target. I have never noticed this effect but it may be true.

I had never heard that until about a month ago. Go to the 7th post of this thread to see screen shots of the command bonus hit% jump only when the commander has LOS to target (http://www.theblitz.org/message_boards/showthread.php?tid=43829)

From the screen shots I am assuming that for squad FP to also increase the HQ needs LOS to the target - but that might be a bad assumption.

[The Coil]

Originally posted by David Chapuis:

From the screen shots I am assuming that for squad FP to also increase the HQ needs LOS to the target - but that might be a bad assumption.

I think maybe that is a bad assumption - a quick test showed no change in the stated FP in or out of command. Of course, another quick test showed no change in stated FP no matter what combat bonus I gave the HQ. Are you sure that combat bonuses increase FP? If so, it doesn't show on the stated FP number when you target a unit...

[David Chapuis]

Originally posted by The Coil:

</font><blockquote>quote:</font><hr />Originally posted by David Chapuis:

From the screen shots I am assuming that for squad FP to also increase the HQ needs LOS to the target - but that might be a bad assumption.

I think maybe that is a bad assumption - a quick test showed no change in the stated FP in or out of command. Of course, another quick test showed no change in stated FP no matter what combat bonus I gave the HQ. Are you sure that combat bonuses increase FP? If so, it doesn't show on the stated FP number when you target a unit... </font>

[David Chapuis]

This link http://www.battlefront.com/discuss/ultimatebb.php?ubb=get_topic;f=23;t=010589#000009

says that I was seeing an experience bonus difference and that HQ combat bonuses dont show up.