Commanders: Original Vision, Crippled Default, Future ?

[iap]

About sacreficing a solider to save the king:
This is the problem with calculating the hit chance when the shot is being shot, and there is no way to change the precalculated results.
However, maybe its not possible to catch the bullet, but it is possible to draw fire, and in some situations I do this.

But for this I need to slow the game to 1/3.

By the way, to get the "chess" effect I really recommend the game "Battle for Weanoth", if you havn't heard of it then now is the time!

[Rman Virgil]

.

I like The Battle for Wesnoth.

But the critical diff here is TB vs RT. I believe these "chess effects" can be achieved in RT with a few additional command behaviors, new Command UI and linkage to a re-vamped MM. But these changes are not trivial & are indeed on the complex side of the Mod Continuum. I'm guessing a couple years away from feasibility.

.

[iap]

By the way another huge problem that I sew in commanders is the lack of autonomy in units decisions, in particular artillery units. This is happening in several situations.
For example - when units go to repair and come back, or coming from the factory witch is far away, they ignore enemies on the way. They do not shoot them, they don't even turn the turret.
Same thing is when all the units are near the commander, but I didn't give the commander a specific target. It does (sometimes) find a target to shoot - but the artillery will not shoot until I specifically click a target!
When I order to retreat, I'm in troubles - I have the direct units in the front and the artillery in the back, but when I tell them to go back, the artillery just sits there and waits that the commander and all the other direct and repair units will pass them. This is not a path walking problem - this is autonomy movement problem,
I need them to act as if I just unassigned them all ant ordered to retreat. well this is what I do, and reassigning them is so much time consuming.

So what I'm saying (And I'm aware to the fact that this might change something fundamental) is that the commanders will work in "override" control mode, instead of "explicit" control, I mean, let all the units act normal (like in group) but when the commander is giving an order - all of them will react.
Or at least... Make them fire at will when the commander is set to fire at will...

[Rman Virgil]

By the way another huge problem that I sew in commanders is the lack of autonomy in units decisions, in particular artillery units. This is happening in several situations.

For example - when units go to repair and come back, or coming from the factory witch is far away, they ignore enemies on the way. They do not shoot them, they don't even turn the turret.

Same thing is when all the units are near the commander, but I didn't give the commander a specific target. It does (sometimes) find a target to shoot - but the artillery will not shoot until I specifically click a target!

When I order to retreat, I'm in troubles - I have the direct units in the front and the artillery in the back, but when I tell them to go back, the artillery just sits there and waits that the commander and all the other direct and repair units will pass them. This is not a path walking problem - this is autonomy movement problem, I need them to act as if I just unassigned them all ant ordered to retreat. well this is what I do, and reassigning them is so much time consuming.

So what I'm saying (And I'm aware to the fact that this might change something fundamental) is that the commanders will work in "override" control mode, instead of "explicit" control, I mean, let all the units act normal (like in group) but when the commander is giving an order - all of them will react.

Or at least... Make them fire at will when the commander is set to fire at will...

All duly noted problems. I even spoke to WZ Creators about it in thier WZ bbs and was told they would not be dealing with it in WZ 2100 but rather in WZ 2120, the full blown sequel.

That said, the first step in dealing with these issues, in the here and now, is by exposing Commander Turret A.I. to modifying (it's hard-coded).

.

[iap]

Oh, I didn't realize it is by design and for these reasons.

Do you mean to be able to program commanders in JS? This would be awesome!
(On a side note, I did searched for JavaScript files last night to see how everything is built and if I can change... but I couldn't find )

[Rman Virgil]

Oh, I didn't realize it is by design and for these reasons.

Do you mean to be able to program commanders in JS? This would be awesome!
(On a side note, I did searched for JavaScript files last night to see how everything is built and if I can change... but I couldn't find )

Exactly! The JS API is yet under development with the focus still on CAM & SKI migration. WZ Creators efforts along these lines with WZ Script were a convoluted, gnarly, mess (a good discription for their entire original C source) which you can follow in their original source.

.

[Rman Virgil]

.

Here is what Pumpkin Studios AI Guru, Alex Lee, had to say in July 2009 about the original design and implementstion of Commanders during WZ 2100's development 1996-1999:

~ QUESTION: Was there many things that you were planning to do with commanders that you never got around to? The source code certainly gave that impression

~ Alex Lee's ANSWER: Yes, commanders did a bunch of extra things during development, but they got scaled back, both the control mechanism and ui got very complicated.

We wanted them to order units realistically, not just become a grouping with a bonus and a common goal.

viewtopic.php?f=2&t=122&p=31591#p31591

.

[Stratadrake]

About sacreficing a solider to save the king:
This is the problem with calculating the hit chance when the shot is being shot, and there is no way to change the precalculated results. However, maybe its not possible to catch the bullet, but it is possible to draw fire, and in some situations I do this.

I agree that once WZ has performed its "accuracy check" there is nothing further that can be done - but the simplest implementation I can think of is that if your target unit is obstructed by an enemy, then the game will silently and invisibly direct your attacks at the obstructing enemy until the target is clear, THEN direct your attacks at the designated target. If your target is obstructed by a friendly - the unit will refrain from attacking at all until it's clear.

As an example I see personally, if you order a group to strike a factory on the opposite side of the concrete wall, what do they shoot first? The wall itself. Their assigned target is still the factory behind it, they are just removing an obstruction from their line of fire. That's a perfectly reasonable behavior, but it is one that does NOT exist when mobile groups of units are engaging one another.

Forcing mobile units to obey stricter and unobscured line-of-sight rules would change the balance of the game, I argue for the better. People complain about mass rushes of groups having too much power - the same tight group would also suffer from occluded lines-of-sight, leaving only the units around the edges free to attack without mixing up friendly fire.

By the way another huge problem that I sew in commanders is the lack of autonomy in units decisions, in particular artillery units. This is happening in several situations.

For example - when units go to repair and come back, or coming from the factory witch is far away, they ignore enemies on the way. They do not shoot them, they don't even turn the turret.

I agree these are situations where a unit needs more autonomy - if I have a group of mortars/howitzers assigned to a commander they become sitting ducks to anyone else - they WILL NOT FIRE until and unless the commander explicitly gives them a designated target. By contrast, the same weapons implemented as defensive emplacements will happily defend themselves from whatever encroaches upon their personal space - but mobile units will not? These guys have standing orders of FIRE AT WILL - if there is currently no target from the commander they CAN and SHOULD be allowed to pick their own.

[aubergine]

I agree that once WZ has performed its "accuracy check" there is nothing further that can be done - but the simplest implementation I can think of is that if your target unit is obstructed by an enemy, then the game will silently and invisibly direct your attacks at the obstructing enemy until the target is clear, THEN direct your attacks at the designated target. If your target is obstructed by a friendly - the unit will refrain from attacking at all until it's clear.

That's going to be weird in the case where the attack has an indirect fire weapon - why would it not shoot directly at it's target? If the described method was limited to direct fire weapons, but does not affect indirect fire weapons, then it would be OK.

[aubergine]

Exactly! The JS API is yet under development with the focus still on CAM & SKI migration. WZ Creators efforts along these lines with WZ Script were a convoluted, gnarly, mess (a good discription for their entire original C source) which you can follow in their original source.

IMHO allowing command turret to be scripted by JS will allow a vast array of experimentation for new ideas. However there are still lots of things need adding to the JS API, most notably I think for a reliable commander AI we'd need access to map terrain data so the commander can make useful decisions about where to place units.

[Rman Virgil]

.

Exactly! The JS API is yet under development with the focus still on CAM & SKI migration. WZ Creators efforts along these lines with WZ Script were a convoluted, gnarly, mess (a good discription for their entire original C source) which you can follow in their original source.

IMHO allowing command turret to be scripted by JS will allow a vast array of experimentation for new ideas. However there are still lots of things need adding to the JS API, most notably I think for a reliable commander AI we'd need access to map terrain data so the commander can make useful decisions about where to place units.

+1

While it is no more than an excercise in prophecy, I do believe that if ever there is an Ender - Bolo Commander Mod, WZ 2100's popularity will blow through the roof like never before. And the beauty of this will be that it would be a realization of this games unique identity in the RTS genre and the fullfillment of WZ Creator's truely original vision. Original to this day, some 15 years after conception.
.

[Iluvalar]

http://en.wikipedia.org/wiki/Omega_%281 ... er_game%29 is love ^^

[Rman Virgil]

http://en.wikipedia.org/wiki/Omega_%281 ... er_game%29 is love ^^

That brings back good memories. Imagine creating captivating games with a 64k kernel. Cool game and so original and daring too.

.

[Stratadrake]

That's going to be weird in the case where the attack has an indirect fire weapon - why would it not shoot directly at it's target? If the described method was limited to direct fire weapons, but does not affect indirect fire weapons, then it would be OK.

Indirect-fire weapons can be exempt from checking for a clear line of sight, yes. Now VTOL weapons are direct-fire themselves so perhaps they should also have a clear-line-of-sight check, but taking into account the fact that they are striking from an above angle making it less likely for something to get in the way.

[iap]

So indirect fire should avoid VTOL.