Phantom X (v1.8) Update Request

[Duck_42]

Ever since we introduced Paladin players into Phantom-X, I've been trying to think of a good way to address one of the majors flaws of the idea. One of the original intents was to introduce a "phantom-like" player on the innocent's team, but to do it in such a way that the innocents couldn't necessarily trust him (after all, he might be a Phantom masquerading as a Paladin). As Mead has pointed out, it's ALWAYS in the best interest of the innocents for the Paladin to announce himself immediately after the assignement.

If they're smart, the Phantoms will attempt no subterfuge because they will reveal themselves in the process. For example, if player A says "hey everybody, I'm the Paladin" and player B says "no wait, I'm the Paladin", then we've just narrowed the first Phantom down to 2 players. From that point, it's pretty easy for the innocents to figure out which one is the actual Phantom. If the Phantoms remain quiet (as well they should), the innocents now know they can completely trust the Paladin player (which was never the intent).

Since part of the point of Phantom is subterfuge, deception, misdirection, I've added an option called "Paladin Marks" to counter this behavior. A Paladin Mark is a single use "guess" that a Phantom can use to "mark" a player. If the player being marked is indeed a Paladin, they will imediately lose their Paladin bonus (yes Mead, I agree with you, having this kill the Paladin was a bit excesive ). It's not a major penalty, but it's sufficient to give Paladins a reason to remain hidden. It also gives innocent players a reason to claim to be a Paladin. If they can convince the Phantom to waste the Paladin mark on an innocent (on whom it will have no effect), it will prevent the mark's use on the real Paladin.

As with all the major options we've introduced so far, this one can be turned off if so desired (i.e. if you don't like it, dont use it).

EDIT: Paladin marks are set to none by default.


Changes in v1.8
1: Added Paladin Marking feature
Phantoms can now "mark" a player that they think is a Paladin. If correct, the Paladin will lose their Paladin bonus and all players will be informed that a Paladin has been marked. If incorrect (i.e. the marked player is not a Paladin), then nothing happens. Either way, the mark is spent and cannot be used again.

Paladin Marks are used via a new button on the phantom UI (crosshair icon). The button will be displayed next to the dump mass and dump energy buttons. If the Phantom has no available Paladin Marks, this button will not be visible.

2: Added The following Lobbly Options
Paladin Marks
-Available choices will be None, One per Phantom, Two per Phantom, One per Paladin, One, Two, Three, or Four.
-This option controls how many Paladin marks will be given to each Phantom.
-Paladin Marks are distributed in a round robin fasion between the phantoms.
(In some scenarios, not every Phantom will get a Paladin Mark)


EDIT: Removed file for further testing and modification.

[mead]

First off, thank you Duck for continuing to work on Phantom X. Regarding the changes in 1.8, I like the marker idea in principle. However, don't you think this should be tested a bit like the last versions were, before being uploaded to FAF? I don't doubt 1.8 works as advertised, but for such a change, I think some feedback would be nice before it is released.
One main concern I have is that the balance in games with paladin is a lot better than in games without. So giving the Phantoms 2 markers (one for each) with no time limitations / drawbacks on their use seems like a step back to the suboptimal balancing of earlier versions. Phants could either mark the two best players, or they could wait until it becomes obvious who the paladin is. I don't like the idea of them being able to eliminate the paladin as a major threat without any risk to themselves. I propose that if they place the marker wrong, their bonus should decrease by a set amount, for example by the same amount the actual paladin receives. This would still fulfill the original purpose, namely discouraging paladin self-reveal, while keeping a semblance of balance intact.

Edit: To put my concerns more succinctly: Good strategy is all about risk versus reward. With the marker feature, that risk element is absent.

[noobymcnoobcake]

I like the idea of making the phantom game more complex and more political but i think this feature is just asking to be abused.

[Duck_42]

I don't doubt 1.8 works as advertised...

I do hope so. It did take me a while to work out all the bugs in my own testing.

...for such a change, I think some feedback would be nice before it is released.

True. However, play testing changes can be a bit tedious (with everyone having to download the mod and all the other players who show up without it, etc). Since the Paladin Mark option can be turned off completely (set the number of Paladin marks to none), I didn't see the harm in going live with it.

So giving the Phantoms 2 markers (one for each) with no time limitations / drawbacks on their use seems like a step back to the suboptimal balancing of earlier versions

The default option is one mark for every paladin in the game. So, if there's one Paladin, there will only be one mark. It's given to a random phantom. The other phantom(s) in the game will have no marks. I felt this was a fair balance because the risk in using it is just that. Once it's used, it's gone, and if the mark is chosen incorrectly, the Phantom will have to contend the Paladin as normal.

they could wait until it becomes obvious who the paladin is

In most of the phantom games that I've been in, it's actually pretty hard to tell who the Paladin is. I'd say there's more of a chance that the Phantom(s) will just pick a mark at random. After all, 1/4 = 25% chance of guessing the Paladin. I don't really like that aspect (as that's not the intent of this change at all), so it's possible I may need to find some way to disuade guessing (perhaps some sort of penalty for guessing wrong might be in order). Time will tell on that one.

I like the idea of making the phantom game more complex and more political but i think this feature is just asking to be abused.

Nooby, What do you mean by "abuse"?

[noobymcnoobcake]

I mean if innos have it and just use it on the phantom because they hate the player. Or they could say give me that battleship and attack yellow or I will mark you and you will lose a large amount of your income.

[mead]

Please do not take my criticism the wrong way. You have done a lot of work for the phantom mod, and without you, it wouldn't be nearly as good as it is now.

The default option is one mark for every paladin in the game. So, if there's one Paladin, there will only be one mark. It's given to a random phantom. The other phantom(s) in the game will have no marks.

Perhaps this will make it clear why I don't like the current implementation, even though I consider the marker idea itself excellent. Let me briefly describe to you how I would use a mark as a phantom, without having to do any guesswork regarding the pala's indentity at all:
- verify who the other phantom is - as you know, this can be done easily enough in a number of ways
- if the best (highest skilled) non-phantom player is next to me, mark him
- if the best (highest-skilled) non-phantom player is neither next to me nor the other phantom, mark him
- if the best (highest-skilled) non-phantom player is next to the other phantom, make a judgement call if he
could overpower that phantom faster than I could kill my neighbouring innos; only mark him if the answer is yes
- if there are several innocent players of very high skill (given the small phantom fanbase, this does not occur too
often), prioritize the one next to me for marking
Note that this requires little to no attention at all. I don't have to watch anyone's base, trying to guess if he gets a paladin bonus. I just have to look at the player names and their position on the map, then mark accordingly.
This practically ensures that once i have a mark, I will almost never have to deal with a strong paladin. This in turn means we're back to pre-1.7 easy phantom wins.

I felt this was a fair balance because the risk in using it is just that. Once it's used, it's gone, and if the mark is chosen incorrectly, the Phantom will have to contend the Paladin as normal.

That is not a risk. The mark was given by luck, without any cost. It is a bonus that came completely free of charge, its use carries no risk. The balance of 1.7 is good, and this simply makes it worse for a considerable percentage of games.

I'd say there's more of a chance that the Phantom(s) will just pick a mark at random.

I've already adressed this above, but let me elaborate on the random point. So basically, you're correctly stating that even with completely mindless / random marking, 25% of games will have shit balance. In my opinion, this is unnecessary.

I don't really like that aspect (as that's not the intent of this change at all), so it's possible I may need to find some way to disuade guessing (perhaps some sort of penalty for guessing wrong might be in order).

The point isn't so much in dissuading guessing, its dissuading risk-free strategic marking as described above. The mark needs to come either with a significant one-time cost, or (preferably) a more severe cost if the guess is wrong.

As you have stated in your initial post, we both agree that there should be a more strategic element to revealing the paladin's identify. But I'm strictly opposed to sacrificing game balance to that end. Using the markers needs to carry a significant risk that is proportionate to the potential reward.

True. However, play testing changes can be a bit tedious (with everyone having to download the mod and all the other players who show up without it, etc). Since the Paladin Mark option can be turned off completely (set the number of Paladin marks to none), I didn't see the harm in going live with it.

Fair enough. However, I still maintain that unlike minor bugfixes, major changes should at least be discussed here beforehand.

Given the balancing concerns, I'm opposed to releasing 1.8 it in its present form unless the markers are turned off by default.

[Duck_42]

I really don't think it's quite as bad as you make it out to be, but fair enough. I've changed the default option to none. That way we can play test it (without all the tedious aspects of downloading a seperate mod).

The mark needs to come either with a significant one-time cost, or (preferably) a more severe cost if the guess is wrong

While we're on the topic, what do you think a reasonable cost is? Keep in mind that any risk to the Phantom should not be so great as to dissuade the reasonable use of a mark. Also, I don't think there should be a penalty if the guess is correct.


Nooby:

I mean if innos have it and just use it on the phantom because they hate the player. Or they could say give me that battleship and attack yellow or I will mark you and you will lose a large amount of your income.

I think you need to reread the first post, as it looks like you've missed a few important details.

[mead]

I've changed the default option to none. That way we can play test it (without all the tedious aspects of downloading a seperate mod).

Thank you.

While we're on the topic, what do you think a reasonable cost is? Keep in mind that any risk to the Phantom should not be so great as to dissuade the reasonable use of a mark. Also, I don't think there should be a penalty if the guess is correct.

A good way to do this would be that a wrong guess should come with a permanent reduction of the Phantom's bonus by the same amount that the paladin's bonus is. So, with standard settings, it would mean a Phantom bonus reduction down to 55% of its previous value (only for the phantom who guessed wrong).

[mead]

Having thought about this some more, I'm starting to think that even a correct marker guess should come with some sort of drawback. Otherwise it is just overpowered, and phantoms certainly don't need a buff compared to 1.7.
Here are some ideas regarding the cost of the marker use:

a) an upfront cost of 30.000 mass and 300.000 energy. An investment equivalent to an average t4 feels adequate. There is no refund if the marker guess is correct. Using the marker comes with a UI message visible to all: "Player xxx marked yyy" (substitute player names), and xxx being revealed as Phantom in the UI.
b) alternatively, as described above, if the guess is wrong, the phantom bonus is reduced massively by subtracting the paladin bonus percentage from it. Using the marker comes with a UI message visible to all: "Player xxx marked yyy" (substitute player names), and xxx being revealed as Phantom in the UI.

With both these variants, paladins will still learn not to reveal themselves publicly (which is the ONLY problem we want to fix in this release, right?). Because in most cases, neutralizing a paladin is easily worth 30k mass and / or being revealed as phantom, so its not like this makes the marker useless. It just strongly discourages random / uncertain marker use.
Personally, I'm starting to favor variant a), because it stays compatible with RTS philosophy and game design - you pay for something, then you can use it, and if you f*** it up, its your fault for wasting the investment.

[Duck_42]

a) an upfront cost of 30.000 mass and 300.000 energy. An investment equivalent to an average t4 feels adequate. There is no refund if the marker guess is correct. Using the marker comes with a UI message visible to all: "Player xxx marked yyy" (substitute player names), and xxx being revealed as Phantom in the UI.

b) alternatively, as described above, if the guess is wrong, the phantom bonus is reduced massively by subtracting the paladin bonus percentage from it. Using the marker comes with a UI message visible to all: "Player xxx marked yyy" (substitute player names), and xxx being revealed as Phantom in the UI.

1. The upfront cost might be reasonable. Still if we're going to a pay per guess model, it almost seems like the Phantoms should then be allowed to guess at will (as long as they have the resources).

2. Revealing the Phantom means these things are completely useless until after the player has revealed themselves as a Phantom. I intentionally left out names in my the mark messages so that no one would know for sure who marked who. Including the names takes away a big part of the intrigue and mystery.

3. Penalizing the Phantom by nearly half of his Phantom income for an incorrect mark means that no reasonable Phantom will ever use a mark. That's way too much of a risk. This would give the innocents (non-paladins) a huge incentive to actively deceive Phantoms in an attempt to get them to mark the wrong player. Also, this would tend to complicate the bonus calculation mechanics considerably.

With both these variants, paladins will still learn not to reveal themselves publicly

Probalbly true. Instead, you'd see the innocents (non paladins) attempting to use the marks as a weapon against the Phantoms. As an innocent, if I can get a Phantom to mark me early in the game, the odds of the Phantoms winning go to practically zero (unless the other Phantom just happens to be really, really good).

Why? Because...

a) It cost him 30k mass, so he's down an EXP (or two).
b) He now has half of his bonus, so he won't be able to build up any better than the real Paladin.
c) Everyone knows who he is and that the marked player is innocent (i.e. not a Paladin and not a Phantom).
d) The real Paladin can now reveal himself (unless there's more than one mark available).

If the innocents don't kill him, the other Phantom almost certainly will. So the cost for a Phantom guessing wrong early is almost certain death. Granted, later in the game the non rescource costs go down a bit, but so does the reward. So with all that in mind, if you were a Phantom would you ever use one of these things?


I agree some balancing will probably be necessary, but I don't want to make the cost so extreme as to render this feature useless. So, I suggest we try the mod out as is and go from there. I think we'll both get a better idea of how much balancing is needed after we play it a bit.