Cav/LightInf auto-foraging?

[NeverMan]

ORIGINAL: pzgndr

Marshall's take in post #15 above was clear and should be fine.

1. I didn't realize you spoke for Marshall/Matrix, though it would make sense if you actually were employed by this company.

2. Since post #15 much has come to light. There isn't a need to add clunky "checkboxes" to solve this problem, obviously.

[pzgndr]

Since post #15 much has come to light. There isn't a need to add clunky "checkboxes" to solve this problem, obviously.

Really? Jimmer's brought quite a bit to light but you're blind to it. It's six to one and half a dozen to another whether the system assumes foraging by default and the player explicitly checkboxes a unit to NOT forage and pay for normal supply, OR vice versa. Either which way, for players to have explicit control to not pay for supply or to pay for supply for individual units, as Marshall suggested, some type of order/checkbox or something needs to be implemented. It's not a simple binary choice, therefore player intervention is needed, clunky or no.

[NeverMan]

ORIGINAL: pzgndr

Since post #15 much has come to light. There isn't a need to add clunky "checkboxes" to solve this problem, obviously.

Really? Jimmer's brought quite a bit to light but you're blind to it. It's six to one and half a dozen to another whether the system assumes foraging by default and the player explicitly checkboxes a unit to NOT forage and pay for normal supply, OR vice versa. Either which way, for players to have explicit control to not pay for supply or to pay for supply for individual units, as Marshall suggested, some type of order/checkbox or something needs to be implemented. It's not a simple binary choice, therefore player intervention is needed, clunky or no.

You arguing with me? LOL, geez, didn't see that coming.

Anyways, hopefully Marshall has the chance to read this thread and figure out that he doesn't need more clunky buttons making the already horrible UI even worse.

[Jimmer]

ORIGINAL: NeverMan
No, it doesn't, not sure what you are talking about here... It only forages when it's the only choice left, the corps HAS to do something, it doesn't assume you want to forage at all... in fact, quite the opposite, it assumes you want to pay with normal supply and then only forages when you can't do that, not the other way around (as you suggest and per the current implementation).

But, it DOES do it.

I deliberately mis-quoted your statement here for effect: The two models BOTH have the problem you are trying to get rid of. You are just removing the auto-forage piece rather than both forage pieces. Marshall's model correctly includes it. I say "correctly" because your model would be vastly more expensive than his (as I showed in an earlier example).

[NeverMan]

ORIGINAL: Jimmer

ORIGINAL: NeverMan
No, it doesn't, not sure what you are talking about here... It only forages when it's the only choice left, the corps HAS to do something, it doesn't assume you want to forage at all... in fact, quite the opposite, it assumes you want to pay with normal supply and then only forages when you can't do that, not the other way around (as you suggest and per the current implementation).

But, it DOES do it.

I deliberately mis-quoted your statement here for effect: The two models BOTH have the problem you are trying to get rid of. You are just removing the auto-forage piece rather than both forage pieces. Marshall's model correctly includes it. I say "correctly" because your model would be vastly more expensive than his (as I showed in an earlier example).

Again, this is not correct. Models don't generally become "vastly less expensive" by adding more states. In fact, they generally become "vastly more expensive" by adding more states, so Marshall's model is not less expensive, it's actually more expensive (because of the redundancy involved in his/your model). I'm assuming we are talking about yours and his "new" model (not the current model since that is broken).

Let's make this simple: show me a state model.

[Jimmer]

Game expenses, not programming expenses.

[NeverMan]

ORIGINAL: Jimmer

Game expenses, not programming expenses.

Ah... ok, so I confess that I'm unclear how you are measuring game expenses?

[Jimmer]

Pounds, like everybody else.
 
When you remove the auto-forage step, you force the player to pay for (some) corps which could have foraged safely via auto-forage. This results in costs no lower than Marshall's model, and many times far higher.
 
Go back to your god in this: Original EiA. What happened at the end of a turn? People would figure out which corps could forage safely, and which corps could not. Then, they would make a financial/strength decision as to whether the cost savings by not foraging outweighed the risk of foraging.
 
THEN, on the next guy's turn (frequently), aplayer would say "Oh, I forgot to forage that corps", and promptly pay another buck or two or else forage the corps. Note that it was HIS choice. What your model would force, if applied to the board game, would be to force the player to pay for supply, no matter how stupid that might have been during the turn, simply because he didn't notice it while moving.
 
Marshall's model does it EXACTLY the way it was done in the board game: Cut the player a little slack (let corps forage, if they can do so safely). But, there's a problem: the computer can't be as free/loose with the rules; at the end of a turn, the computer MUST decide what to do. Marshall's decision was simple and obvious (and correct, IMO): Corps that can forage safely SHOULD forage safely.
 
What he missed was that some corps would now forage even though the player really would like to have payed for supply. In the board game, it was obvious where the attacks were, so the corps in the attacking stack would all get paid for. But, Marshall apparently didn't code to notice this, and applied the same process as he did with all of the other corps on the map. THIS is the problem that needs fixing. The main model MUST stay like it is, because it is exactly correct (i.e. mirrors what a human would do if presented with a choice). Only the end-case needs fixing.

[pzgndr]

What he missed was that some corps would now forage even though the player really would like to have payed for supply. In the board game, it was obvious where the attacks were, so the corps in the attacking stack would all get paid for. But, Marshall apparently didn't code to notice this, and applied the same process as he did with all of the other corps on the map. THIS is the problem that needs fixing. The main model MUST stay like it is, because it is exactly correct (i.e. mirrors what a human would do if presented with a choice). Only the end-case needs fixing.

Jimmer is spot on. This is all Marshall needs to fix here.

[NeverMan]

ORIGINAL: Jimmer

Pounds, like everybody else.

When you remove the auto-forage step, you force the player to pay for (some) corps which could have foraged safely via auto-forage. This results in costs no lower than Marshall's model, and many times far higher.

Again, this is just wrong. "you force the player to pay"... this is just plain WRONG.

It's not "forced" since the player has already had the option to auto-forage that unit!

How can something be "forced" if the player has already had the option to do it??? Your model, while needlessly complex, does allow the player to be lazy.

[NeverMan]

ORIGINAL: Jimmer

Marshall's model does it EXACTLY the way it was done in the board game.

Again, this is WRONG.

Marshall's current model FORCES the player to auto-forage EVEN if you want to pay for supply.

[bresh]

ORIGINAL: NeverMan

ORIGINAL: Jimmer

Marshall's model does it EXACTLY the way it was done in the board game.

Again, this is WRONG.

Marshall's current model FORCES the player to auto-forage EVEN if you want to pay for supply.

I dont experience this, though havent done any testing, but this turn in one of my pbm-games(1.06.03).
I moved attacked 3 areas, 1 i myself foraged, the other 2 where eligble for supply and did not forage, giving me the option to siege after winning the land battle vs 1 Corps who withdrew. No autoforage happend for any of my corps, since i did not use the autoforage all key.
The area where i laid siege after battle, one of my corps was able to "autoforage" if it wanted by using all its moves, but it did not.

Regards
Bresh

[NeverMan]

ORIGINAL: bresh

ORIGINAL: NeverMan

ORIGINAL: Jimmer

Marshall's model does it EXACTLY the way it was done in the board game.

Again, this is WRONG.

Marshall's current model FORCES the player to auto-forage EVEN if you want to pay for supply.

I dont experience this, though havent done any testing, but this turn in one of my pbm-games(1.06.03).
I moved attacked 3 areas, 1 i myself foraged, the other 2 where eligble for supply and did not forage, giving me the option to siege after winning the land battle vs 1 Corps who withdrew. No autoforage happend for any of my corps, since i did not use the autoforage all key.
The area where i laid siege after battle, one of my corps was able to "autoforage" if it wanted by using all its moves, but it did not.

Regards
Bresh

If this is in fact true then there is simply inconsistency in Marshall's algorithm, which is in itself a big problem.

[Jimmer]

ORIGINAL: NeverMan
It's not "forced" since the player has already had the option to auto-forage that unit!

I've already refuted this. This is the last time:

He ALSO had the opportunity to forage (non-auto). Why do you keep insisting that your arbitrary standard be applied to ONLY one of the two side-by-side buttons?

[Jimmer]

ORIGINAL: bresh
I dont experience this, though havent done any testing, but this turn in one of my pbm-games(1.06.03).
I moved attacked 3 areas, 1 i myself foraged, the other 2 where eligble for supply and did not forage, giving me the option to siege after winning the land battle vs 1 Corps who withdrew. No autoforage happend for any of my corps, since i did not use the autoforage all key.
The area where i laid siege after battle, one of my corps was able to "autoforage" if it wanted by using all its moves, but it did not.

This is a different situation (I think). If you move into the besieging "box" in an area, then the game knows you cannot be allowed to forage using movement.

What Neverman ran into was similar, but there was a battle first. As a result of having the battle, he could not invoke the game's automatic rejection of foraging with movement, because he wasn't in the besiege "box".

[Jimmer]

ORIGINAL: NeverMan
If this is in fact true then there is simply inconsistency in Marshall's algorithm, which is in itself a big problem.

I would certainly agree with this, if correct. However, see my other post in response to Bresh's post: I suspect his scenario did not include both a battle and a siege like yours did.

[Jimmer]

Which brings up a second point: It is a lot more likely that one will not see this if one is playing a nation other than France. Most nations tend to use up most of their movement just getting to combat, so an auto-forage (with movement) cannot occur "safely"; there isn't enough movement left! So, the game doesn't do it. But, it CAN occur to other powers; it's just not as likely.
 
Neverman ran into this playing Austria, but note that it was his Light Infantry corps which caused the problem. The LI corps has movement 5, making it a lot more likely to have enough movement left to "safely" autoforage.

[NeverMan]

ORIGINAL: Jimmer

ORIGINAL: NeverMan
If this is in fact true then there is simply inconsistency in Marshall's algorithm, which is in itself a big problem.

I would certainly agree with this, if correct. However, see my other post in response to Bresh's post: I suspect his scenario did not include both a battle and a siege like yours did.

Yes, I think you are correct about Bresh's post, this makes sense.

[NeverMan]

ORIGINAL: Jimmer

Which brings up a second point: It is a lot more likely that one will not see this if one is playing a nation other than France. Most nations tend to use up most of their movement just getting to combat, so an auto-forage (with movement) cannot occur "safely"; there isn't enough movement left! So, the game doesn't do it. But, it CAN occur to other powers; it's just not as likely.

Neverman ran into this playing Austria, but note that it was his Light Infantry corps which caused the problem. The LI corps has movement 5, making it a lot more likely to have enough movement left to "safely" autoforage.

I did, however, it became a problem for ALL my forces at that spot (which was actually a large force). The big problem is that it really did effect my strategy (since I had to waste a turn sitting there to re-siege).. (though it didn't since France surrendered anyways).

At this point I don't really care if the UI gets worse as long as it gets fixed, it is a problem IMO.

[Jimmer]

Say Marshall: This interchange with Neverman has given me another, simpler idea: When auto-foraging (ANY time, not just end of turn), check for battle areas where the non-phasing player has an enemy in the city. If that condition exists, abort the auto-forage and/or ask the player what to do. I would lean towards asking the player, because it's likely that he can't undo his move, and is stuck in this situation Neverman got into without recourse. So, the sequence would go:

1) Auto-forage initiated (end of turn or by button -- both methods should do this, similarly to how manual forages are executed).
2) Check all corps locations for a battle instance.
3) In the results of #2, check to see if any corps of the phasing player in that area are auto-forage candidates with positive movement remaining.
4) In the results of #3, check the city (if any) in that area for an enemy of the phasing player (does not have to be the same enemy as the force outside the city). These are the only areas which could result in the problem Neverman ran into: Areas which have both a battle coming up and a potential siege combat after battle.
5) If 4) above finds such a situation, pause the auto-forage and ask the user specifically what to do with each corps in that stack. A table would be appropriate here.
6) Allow options to forage with movement, forage without movement, draw supply (the table should show what the cost will be), or abort the auto-forage.
7) Gather up the information, but do not execute the forage/supply choice at this time. The player may find a battle later in the process which presses him to respond with "abort".
8) Repeat steps 2-7 as needed for all battle/siege areas.
9) Execute "orders".

This should be relatively easy to code, as it doesn't require changing any sigificant game element. It just adds a few steps to the auto-forage process.

NOTES:
Steps 2-4 can be iterated through prior to steps 5-7 being executed, if you are willing to maintain the results gathered in steps 2-4.

Table should have lines for each corps, plus columns showing

1- Movement points remaining.
2- The forage value of the area (and a note as to how many corps are present; if only the attacker, foraging could change).
3- The known quantities in the city (if a corps, just note "a corps", but garrison strengths should be listed).
4- Cost to supply this corps.

Other columns should allow the user to

A- Forage with movement.
B- Forage without movement.
C- Draw supply at cost listed.
D- Abort the whole auto-forage process.

At the bottom, the total cost of supply for all corps present should be noted. Also, there should be an indicator of current cash available.