Command Delays & Absolute Spotting

[Hamsters]

I brought this up before but rather than bump the old topic, I'll sum up. Ordering your troops to fire on a target does not suffer from a command delay, unlike ordering your troops to move.

The argument against this delay was that a squad or section leader would quickly give fire commands. I don't think this applies, however, as the squad and section leaders (And TCs, et cetera) are handled by the TacAI. Specific targeting orders are considered to have come from 'Up top', just like movement orders.

If command delays were implemented for targeting orders, it would solve the problem with absolute spotting, as your tanks, et cetera, would not be able to target unseen forces until receiving word from HQ about their existence.

Command delays should also be implemented in the act of spotting, itself, thus simulating the unit's ability to get on the horn and alert HQ about the presence of an enemy vehicle, unit, et cetera. Because CM turns are compiled and then displayed, it can allow for regressive display of enemy units that have been spotted and reported during the turn.

[Brer Pinkey]

Just a few problems I can see with this proposal.

If a command delay is added for targeting commands from the player as opposed to the AI then the following case can occur; Multiple targets move into LOS/LOF of a hiding unit late in a turn, next turn the hiding unit is ordered to unhide and fire on one of the ID'd units, command target delay allows all spotted units to safely crossing the open area into cover.

Also. The problem of absolute spotting is increased by adding a 'Spotting Delay'. Every instance of spotting would need to be buffered and timed out before displaying to the user/TacAI. I can only wonder what kind of problems that would add... Maybe squads running accross small open areas between cover unopposed cause nobody 'spotted' them in the 10 seconds they're visible.

Editied cause I'm a typing fool.

[This message has been edited by Brer Pinkey (edited 03-27-2001).]

[Hamsters]

Originally posted by Brer Pinkey:

Just a few problems I can see with this proposal.

If a command delay is added for targeting commands from the player as opposed to the AI then the following case can occur; Multiple targets move into LOS/LOF of a hiding unit late in a turn, next turn the hiding unit is ordered to unhide and fire on one of the ID'd units, command target delay allows all spotted units to safely crossing the open area into cover.

Also. The problem of absolute spotting is increased by adding a 'Spotting Delay'. Every instance of spotting would need to be buffered and timed out before displaying to the user/TacAI. I can only wonder what kind of problems that would add... Maybe squads running accross small open areas between cover unopposed cause nobody 'spotted' them in the 10 seconds they're visible.

Editied cause I'm a typing fool.

[This message has been edited by Brer Pinkey (edited 03-27-2001).]

Editied?

I agree with point two and but consideration must be made for more realistic gameplay. Ordering a unit to unhide is like any other command, it has to make its way down the command structure just like any other, otherwise units end up being selectively telepathic.

A simple and realistic way to avoid mistakes caused by slow commands is to update the Ambush command (Making it longer distance and perhaps more selective) thereby allowing you to give realistic preparatory orders to the unit in question.

I will not argue any resource issues involving a delay for spotting, as this is outside the realm of our knowledge and can only be commented on by someone with an in-depth knowledge of Combat Mission's code. I will say that it is a realistic aspect of Fog of War and, to ease the learning curve, could be added as a fourth FoW option in CM2 or CMII.

[Simon Fox]

I think there is some merit in what you have to say. I certainly agree that you have an unrealistic control over your troops. However, I disagree that the command process in CM simulates commands "from above" alone. The TacAI is not intelligent enough to simulate the actions of section or platoon leaders on its own. You as the player represent all levels of command which unfortunately gives you a degree of coordination not present in real life. If you really want to simulate the relaying of orders from the Btn or Co HQ down you should have a command delay of 5-10 minutes while the HQ runner finds the platoon/team/squad.

Since you seem to have misconceived the rationale BTS used to design the orders/delay structure in CM, I suggest you might like to get your collective rodent paws and click on that search button. It may be hard to find but BTS have stated their views in the past. Personally I think relative spotting will be the panacea for many of these problems and tinkering with the current system might open a can of worms. Then maybe Hamsters like worms in their lettuce?

------------------

"Stand to your glasses steady,

This world is a world of lies,

Here's a toast to the dead already,

And here's to the next man to die."

-hymn of the "Double Reds"

[Guest wwb_99]

Relative spotting is a great idea, but no one has come up with a method for getting around the fact that one brain (the player's) controls all of the actions of all of the men on the field.

I am as big a fan of realism as the next man, but if you take too much control away from the player, the game ceases being a fun interactive experience. And no one wants that.

WWB

------------------

Before battle, my digital soldiers turn to me and say,

Ave, Caesar! Morituri te salutamus.

Check out the Dogs of War CM Players Community

[Hamsters]

Originally posted by Simon Fox:

I think there is some merit in what you have to say. I certainly agree that you have an unrealistic control over your troops. However, I disagree that the command process in CM simulates commands "from above" alone. The TacAI is not intelligent enough to simulate the actions of section or platoon leaders on its own. You as the player represent all levels of command which unfortunately gives you a degree of coordination not present in real life. If you really want to simulate the relaying of orders from the Btn or Co HQ down you should have a command delay of 5-10 minutes while the HQ runner finds the platoon/team/squad.

Since you seem to have misconceived the rationale BTS used to design the orders/delay structure in CM, I suggest you might like to get your collective rodent paws and click on that search button. It may be hard to find but BTS have stated their views in the past. Personally I think relative spotting will be the panacea for many of these problems and tinkering with the current system might open a can of worms. Then maybe Hamsters like worms in their lettuce?

Simon, I have read BTS's rationale for the command delays and know that the player represents a nebulous 'command presence', I just think that if this command presence is made more universal (eg the targeting/hiding/et cetera) it would be rendered more realistic. BTS has publicly stated that absolute spotting is a problem, I am merely suggesting a simple solution to ameliorate it. We all agree that command delays are a good thing, why is there friction when related to the targeting and hiding commands?

[Simon Fox]

Friction?

Well there are negative aspects to what you propose and there's more than one way to skin a hamster, err I mean cat. What you propose is not relative spotting and I venture to say that would be plenty of complaining if it were implemented. The TacAI is simply not smart enough to rely on without the opportunity to correct it every 1min.

http://www.battlefront.com/discuss/Forum1/HTML/000342.html

http://www.battlefront.com/discuss/Forum1/HTML/000267.html

Charles wrote:

You've hit upon one of the most fundamental problems in wargaming: what I call the multiple hats problem. I call it that because, as a player, you wear many hats. You're the company commander. You're also the platoon commanders. And the sergeants. And even, to an extent, the corporals and privates!

Basically any wargame really is like taking Star Trek into WW2: you're the "Borg" because all your soldiers collectively think with one mind: yours.

Obviously this is unrealistic. <http://www.battlefront.com/discuss/smile.gif>

But there's no good way around it, unless you have about 150 buddies who are willing to play the game with you at the same time, and you each play one "person". It's possible, to an extent, to substitute AI for other "people" in the game, but this usually has unsatisfactory results if relied upon too heavily. And even if it did work, who would really want it? It's fun to wear lots of hats in a game, and see the battle from many perspectives. One minute you're in a tank, and the next you're jumping through windows with your Tommy gun. Fewer limitations and more "hats" means more fun.

But that takes us back to the "one mind" problem. Since all your troops have access to the information in the player's brain, they effectively have instant communications with one another. There is no simple way to avoid this.

[Hamsters]

Originally posted by Big Time Software

(Charles):

However, in Combat Mission we've taken pains to reduce the problem somewhat. For example, buttoned vehicles do not react quickly to events. There's usually a delay. If a threat pops up in the action phase, your infantry units generally react quickly, but a buttoned-up tank will tend to keep on doing what he's doing for another 5-10 critical seconds. Sometimes that's all it takes for a bazooka to hit home...

This was posted in June of '99, so things may have changed somewhat. In my own test,

http://www.battlefront.com/discuss/Forum1/HTML/017490.html

units did not ignore known bazookas for '5-10' seconds but rather entirely if the TacAI didn't know about them. The TacAI seemed to take a while to catch wind of the enemy 'zook and then blow it to smithereens, which is entirely realistic. If you feel it is not realistic for a unit that has no idea about another unit to not react to that unit, than we should debate this, please let me know.

This is patently not the case, however, under very specific circumstances, which we are all familiar with. Those circumstances are when a zook pops out at the end of a turn. The player can then order a buttoned, entirely-unaware tank, to instantly target this new threat, even though had this happened 3 seconds into the turn, the tank would have not reacted for a significant amount of time.

The same goes for infantry squads targeting units of strategic importance and any other targeting command given by the player. Thus we have unrealistic reactions based entirely on a gamey contrivance (1 minute turns) that are not consistent because of this. Everyone who plays this game breaths a sigh of relief when something surprises them at 59 seconds and gasps when something surprises them at 2 seconds because of this.

[This message has been edited by Hamsters (edited 03-28-2001).]

[Leonidas]

Originally posted by Simon Fox:

The TacAI is not intelligent enough to simulate the actions of section or platoon leaders on its own. You as the player represent all levels of command which unfortunately gives you a degree of coordination not present in real life.

Simon has this one nailed, and it's been discussed in a variety of contexts, pretty much anyone suggests a feature that would give players more or less control. I guarantee that we won't resolve it, but we can discuss it some more. The logical error by the relative spotting crowd runs like this:

1) Absolute spotting is unrealistic. It allows units to know more than they really would.

2) Therefore relative spotting would make the game more realistic, and therefore better.

More generally, the argument is:

1) Players have an unrealistically high level of control over their troops.

2) Therefore, reducing player control over troops would make the game more realistic, and therefore better.

The error is that relative spotting, or anything that reduces player control, would aggravate another realism problem: The TacAI isn't capable of simulating the decisions of dozens of subordinate commanders. Current AI technology simply can't handle the job.

Relative spotting would decrease the information available to the TacAI, which would highlight its unrealistically poor judgment. So it's just a balancing problem: Are you more aggravated by unrealistically coordinated forces or poor TacAI decisionmaking?

Personally, the TacAI bothers me more than the coordination. I seem to have a higher opinion than most CM players of the decisionmaking capabilities of low level commanders. The high unit coordination doesn't bother me so much because I think it's required in order to turn a combat simulation into a fun game. I want a fun game instead of a WWII combat movie generator.

I would rather shift the balance in the other direction by giving players the capability of giving much more detailed orders, such as a Hunt order for infantry, Move until you attain LOS to a given point for all units, and Ambush armor only for AT guns.

[Simon Fox]

This was posted in June of '99, so things may have changed somewhat.

It is not the specifics of what BTS had to say (ie 10 secs or whatever) but the underlying principles. You originally said

The argument against this delay was that a squad or section leader would quickly give fire commands. I don't think this applies, however, as the squad and section leaders (And TCs, et cetera) are handled by the TacAI. Specific targeting orders are considered to have come from 'Up top', just like movement orders.

Which is, in part, the basis for your argument and is wrong. The concept of "many hats" is alive and kicking The player and the tacAI do not have distinct roles nor do they represent orders emanating from different units. My inclination is that movement orders attract an additional delay because movement orders are likely to require the coordination of a higher HQ. In contrast targetting decisions are more likely to be controlled at the unit or sub-unit level. Obviously there are exceptions.

If you feel it is not realistic for a unit that has no idea about another unit to not react to that unit, than we should debate this, please let me know.

You misunderstand me. I am strongly in favour of relative spotting. But I think it needs to be implemented properly and that will be a difficult task, though not insurmountable. I feel your simple solution is too simple, a stop-gap measure which will result in possibly more unrealistic situations than it solves. I reiterate, the tacAI cannot be relied upon to always make the most reasonable targetting decision, even in the context of the individual unit alone and leaving aside wider tactical considerations. Any proposal which reduces player control (something I am not in principle against) must be implemented in as realistic a manner as possible.

This is patently not the case, however, under very specific circumstances, which we are all familiar with. Those circumstances are when a zook pops out at the end of a turn. The player can then order a buttoned, entirely-unaware tank, to instantly target this new threat, even though had this happened 3 seconds into the turn, the tank would have not reacted for a significant amount of time.

But this is relative spotting and has nothing to do with a targetting delay. I can equally think of circumstances where your proposal will cause problems. What happens when an infantry squad faced with mutiple targets, chooses to fire at a highly exposed vehicle crew close by instead of firing at an AT team further off about to shoot one of your tanks up the arse. Boy, I can just see the "BTS, please fix or somefink" now hehe.

There will always be a problem with absolute spotting because you will never get away from the fact that the player will know what the units see. I think the most likely and feasible solution will be that units will be unable to react too or fire at things they have not themselves spotted. Even so some inter-unit communication of spotted threats will have to be allowed especially for close infantry units... gets more complicated the more you consider it hehe

------------------

"Stand to your glasses steady,

This world is a world of lies,

Here's a toast to the dead already,

And here's to the next man to die."

-hymn of the "Double Reds"