Determining the cost of the Armor Save for a model that uses a Hit Location Template takes four steps. These steps are:

A. Base Armor Save value
B. Lethality
C. Miss Chance
D. Final Value


A. Base Armor Save value

This is determined as normal for a model with only one Armor Save, except that you must find the total value for all hit locations. To do this, count how many locations the chart has with each Armor Save rating and sum these. For example, a hit location chart with 5 locations total has 3 locations with an Armor Save of 1+ and 2 with 2+. As a 1+ is worth 6 points and a 2+ is worth 5, this means the cost is three times six plus two times five (3*6+2*5). As multiplication is always done before addition, that means the total is 28 (18+10).


B. Lethality

This step adjusts the value by how likely a hit to each hit location is to kill the model, or otherwise adversly affect it. For each Hit Location you must rate each result on it's damage chart and add together these values. The Lethality of a hit location is rated by three factors:
_B1: Location status
_B2: Model status
_B3: Secondary effects

_B1 Location status measures whether the location survives the hit and whether or not the hit is repairable. Damaged locations can be repaired by Self-Repair systems. Destroyed or Crippled locations cannot be repaired. Blown off means that the location is physically removed from the model by the hit, and it lands on a random spot causing a hit to any model at that spot. See also Secondary Effects.
__ Location status:
__ No effect __________ 0.2
__ Damaged ____________ 0.1
__ Partial Destroyed __ 0.05 (IE, not destroyed but Secondary Effects cannot be repaired)
__ Destroyed/Crippled _ 0
__ Removed/Blown off _ -0.1

_B2 Model status measures whether the model itself is destroyed by the hit, and how directly.
Model status:
Continuing __________ 0.2 _ Model is not destroyed. May or may not be usable depending on other effects.
Minor Blowback ______ 0.1 _ Roll something (generally Aim dice) to see if adjacent location takes a hit at TSM X.
Minor Flashback _____ 0 ___ Roll on chart that can roll on chart that can destroy model.
Lesser Blowback ____ -0.1 _ Automatic hit on one adjacent location at TSM X.
Lesser Flashback ___ -0.2 _ Roll on chart that can directly destroy model after X locations destroyed.
Blowback ___________ -0.3 _ Automatic hit on two or three (or 1d3) adjacent locations at TSM X.
Critical Flashback _ -0.4 _ Roll on chart that can directly destroy model OR model is destroyed after X of this location are destroyed.
Major Blowback _____ -0.5 _ Automatic hit on all adjacent locations at TSM X (& Penetrate +1 per unused Plasma counter if used).
Destroyed __________ -0.6 _ Model is destroyed and VP are awarded. Model may or may not remain on table.
Values of X are generally -4, but sometimes -3. If you use a value of 0 to -2, increase cost by 0.05 (IE, a -0.3 would become -0.25). Decrease cost by the same amount for every 2 TSM above -3 or -4.

_B3 Secondary Effects are any effect on the model that do not destroy it, though a few arise because of such destruction. Value is per instance for those that can be.
Secondary effect(s):
none _______ 0.1
rolls +1 ___ 0
temporary _ -0.05 (effect on a Damaged result)
Permanent _ -0.1 (effect on a Destroyed/Crippled/Blown Off result)

Note that if the model is Destroyed, most Secondary Effects are ignored (movement penalties, no firing, etc). The only ones that would matter are those that happen as a result of the model being destroyed (Falls and Reactor Explodes! for example).

CAF halved: 1
roll 4+ to give orders: 1
Titan falls: 2 (and damages other models fallen on)
Movement restricted: 1/ [Move reduced, cannot turn, cannot enter difficult terrain, etc]
looses Command: 2
No Movement: 2
Fire at penalty: 1
No firing: 2
Shields down: 1
Area explosion: 1 per d6cm diameter & 1 per TSM
spot attacked: 0.5 per d6cm distance & 0.5 per TSM
Ethereal Psychic: 2 (only applies to some Area Explosions)
Models Transported in this location hit on 5+: 0.5
Models Transported in this location hit on 3+: 1
Models Transported in this location destroyed: 1.5
Models Transported in this model hit on 5+: 2
May not add Plasma to location: 1
Reactor generates less (energy) per turn: 2
Reactor generates no further (energy) : 5
Psychic Save reduced: 1

For example, the Imperial Titan Leg Location chart.
_ 1) Damaged, Move halved and cannot enter difficult terrain. Value is 0.1 (location damaged) +0.2 (model continues to exist) +2*-0.05 (2 temporary effects) = 0.1 +0.2 -0.1 = 0.2
_ 2) Damaged, Move halved and cannot enter difficult terrain. Value is 0.1 (location damaged) +0.2 (model continues to exist) +2*-0.05 (2 temporary effects) = 0.1 +0.2 -0.1 = 0.2
_ 3) Damaged, leg may snap if Titan had orders that allowed movement. Value is 0.1 (location damaged) +0 (I'm counting this as Minor Flashback because it depends both on a roll AND on what orders the player placed on it, thus two steps) -0.1 (half of Falls as it is possible but not likely) = 0.1 +0 -0.1 = 0
_ 4) Damaged, leg may snap if Titan had orders that allowed movement. Value is 0.1 (location damaged) +0 (I'm counting this as Minor Flashback because it depends both on a roll AND on what orders the player placed on it, thus two steps) -0.1 (half of Falls as it is possible but not likely) = 0.1 +0 -0.1 = 0
_ 5) Destroyed. No move or turning. Value is 0 (location destroyed) +0.2 (model continues) -0.3 (permanent No Move and No Turning (move restricted)) = 0 +0.2 -0.3 = -0.1
_ 6) Destroyed and Falls. Value is 0 (location Destroyed) -0.6 (model destroyed) -0.2 (model falls) = 0 -0.6 -0.2 = -0.8
Total value is thus: 0.2 +0.2 +0 +0 -0.1 -0.8 = -0.5 per Leg location.

If a particular model had 8 Leg locations, the value for those would be -4 (8*-0.5).


C. Miss Chance

This adjusts the cost by a factor determined by how likely it is to deviate from the model's hit locations. For this rating, having seven empty adjacent squares is considered baseline with adjustments for every additional occupied hit location. To determine this rating, count up how many hit locations there are on the chart that have the same number of empty adjacent squares and multiply that number by the value for that many empty squares.

_ Empty
Squares _ modifier
__ 0 ______ -3.5
__ 1 ______ -3
__ 2 ______ -2.5
__ 3 ______ -2
__ 4 ______ -1.5
__ 5 ______ -1
__ 6 ______ -0.5
__ 7 _______ 0

In other words, the more actual hit locations that surround a given location bring down the effective cost, as a hit is more likely to actually hit and not deviate into an empty square.

For example, a model with 9 locations is arranged neatly as a three by three grid. This means that there is one location with zero empty squares (the center one), four with five empty (NE, SE, SW, NW), and four with three empty (N, E, S, W). Thus (1 * -3.5) + (4 * -1) + (4 * -2) = -3.5 -4 -8 = - ( 3.5 + 4 + 8 ) = -15.5


D. Final Value

To find the Final Value, first add up the totals for parts A, B, & C. Next, take the total number of hit locations that the model has and divide by 5. [Why 5? Based on a limited sample, five seems to be the average number of hits needed to destroy a Titan or Praetorian.] Then take the sum of A, B, & C and divide that by the hit locations divided by 5. Mathematically, that would be: (A + B + C) / (HL / 5)

The final value is then inserted into the overall Points Formula as that model's Armor Save value.
__________________

Full Example.

Imperial Warlord Titan

Well, nearly full example. I'm not going to go through every result for every damage chart just now. I may later if requested.

A. This model has 10 locations with 1+, 31 with 2+, & 3 with 3+. Thus 10 * 6 + 31 * 5 + 3 * 4 = 60 + 155 + 12 = 227
B. This model has 3 Head (-1.1), 1 Reactor (-3.5), 22 Weapon (0.1), 13 Leg (-0.5), & 5 Carapace (-0.5). Thus 3 * -1.1 = -3.3, 1 * -3.5 = -3.5, 22 * 0.1 = 2.2, 13 * -0.5 = -6.5, 5 * -0.5 = -2.5; -3.3 + -3.5 + 2.2 + -6.5 + -2.5 = -13.6
C. This model has 6 locations with 1 Empty Adjacent (EA: -3), 6 with 3 EA (-2), 10 with 4 EA (-1.5), 12 with 5 EA (-1), 4 with 6 EA (-0.5), and 6 with 7 EA (0). Thus 6 * -3 = -18, 6 * -2 = -12, 10 * -1.5 = -15, 12 * -1 = -12, 4 * -0.5 = -2, 6 * 0 = 0; thus -18 + -12 + -15 + -12 + -2 + 0 = -59
A + B + C = 154.4
Total number of locations is 44, 44 divided by 5 is 8.8, 154.4 divided by 8.8 is 17.5454 for the Final Value.

Placing this in the Armor Save entry for the Warlord gives it a model cost of 434.6.

The Adeptus Mechanicus models I've processed so far:
__ Model _ Final Value _ Model cost _ Company _ Special _ Support
_ Warlord ___ 17.5454 ____ 434.6 ______ 1643 ____ 487 _____ 608
_ Reaver ____ 16.119 _____ 376.4 ______ 1423 ____ 422 _____ 527
_ Warhound __ 18.7272 ____ 326.2 _____________ 365 / 731 _ 457 / 913
_ Psy Titan _ 16.8152 ____ 675.4 ________________ 756
ImpDropship _ 17.5391 ____ 562 __________________ 629
Armageddon __ 13.0048 ____ 420.4 ________________ 471
AMO Golgotha_ 10.575 _____ 271.5 ________________ 304
_ AMO Mars __ 13.615 _____ 986.9 _______________ 1105
_ Imperator _ 40.089 ____ 1276 ________ 1608

_________________
Net Epic Coordinator

Post reserved in case additional space is required to explain whatever system is built.

_________________
Net Epic Coordinator

The system has to take three factors into account. Side, Lethality, and Chance to Hit/Miss.

Side:

I am considering a couple of options here. One possibility is giving a minor discount to Right/Left side locations and a larger discount to Rear ones. Another possibility is considering Right/Left to be baseline and then increasing costs for Front and lowering for Rear.

Lethality:

Right, when I began looking at this, I figured I could just quickly note down the details of the few hit location types and generate values for them. Boy, was I wrong. There is a great variety of hit locations. So far, just from the Adeptus Mechanicus (and not even all of those), there are: [roughly in order starting with most destructive chart to least destructive]

1-2 (1-5?) destroys location and erratic landing; 6 destroy location and crash landing:
1-2 damaged; 3 crippled; 4 crippled more; 5-6 destroy model:
1-2 damaged; 3-4 crippled; 5-6 destroy model:
1-2 damaged; 3 damaged; 4 crippled; 5-6 destroy location, defeat model:
1-2 damaged; 3 damaged; 4 damaged; 5-6 destroy model:
1-2 damaged; 3-4 damaged; 5-6 destroy model:
1-4 damaged; 5-6 destroy model:
1-2 damaged; 3-4 damaged, small chance to destroy model; 5 destroy location; 6 destroy model:
1-2 damaged; 3-4 destroy location; 5 destroy location, small chance to destroy model; 6 destroy model:
1-3 damaged; 4-5 crippled; 6 destroy model:
1-2 damaged; 3-6 destroy location, cumulative (3) destroy model:
1-4 damaged; 5-6 destroy location, cumulative (2) destroy model:
1-2 damaged; 3-4 destroy location; 5 removes location; 6 remove location and flashback:
1-3 damaged; 4-5 destroy location; 6 destroy location and flashback:
1-5 bonus to future rolls; 6 flashback:
1-2 penalty; 3-4 damaged; 5 destroyed; 6 lesser flashback:
1-2 damaged; 3-4 destroy location; 5-6 minor flashback:
1-2 damaged; 3-4 crippled, chance of minor flashback; 5 destroy location; 6 destroy location and minor flashback:
1-2 damaged; 3-5 destroy location; 6 minor flashback:
1-3 damaged; 4-5 destroy location; 6 remove location:
1-3 damaged; 4-5 destroy location; 6 destroy location:
1-3 damaged; 4-6 destroy location:

Flashback = roll on chart that can directly destroy model.
Lesser = roll on chart that can directly destroy model after X locations destroyed.
Minor = roll on chart that can roll on chart that can destroy model.
____

Once I got a ways into doing that, I realized that it was not going to be simple and I would essentially have to generate values for each chart entry anyway, so why not just do that to begin with. On that note, at the moment, I have the following:

Location status:
No effect __________ 0.2
Damaged ____________ 0.1
Destroyed/Crippled _ 0
Removed/Blown off _ -0.1

Non-lethal Secondary effect(s):
none _______ 0.1
rolls +1 ___ 0
temporary _ -0.05 per
Permanent _ -0.1 per

Model status:
continuing __________ 0.2
Minor Flashback _____ 0.1
Lesser Flashback ____ 0
Critical Flashback _ -0.1
Destroyed __________ -0.2

Each location would be assessed for each of the three factors. The above list of factors may not be exhaustive, as the other armies' models have yet to be examined in detail. I started into assessing Imperial Hit locations, but have not gotten far. As an example, the Titan "Head" location, result of 1 has a value of 0.2 [+0.1 for Damaged, 2x -0.05 for temporary effects, +0.2 since the Titan continues of as (mostly) functional].

Miss Chance.

At the moment, my thoughts on this are that a location that has other locations in all 8 possible adjacent spots is baseline value, with an added amount per missing adjacent spot. I'm not quite sure about values yet, but probably about half a point (IE, 0.5) per empty spot. The baseline could easily be moved.

I expect to be getting more details into this over the next couple of days. I just have not had a lot of time the last few days.

_________________
Net Epic Coordinator

While typing that up, I actually had an idea that could possibly simplify things a bit.

Cubes.

Instead of thinking about a Titan or Praetorian as having locations on each of four sides, perhaps we should be thinking of them as being constructed out of cubes. Each cube would be a single hit location and is essentially what otherwise constitutes a single model, having it's best save at the front and lower on the sides and rear. We do not charge extra for normal models' Armor Save for the sides and rear, so why should we for Titans or Praetorians?

This approach would mean assessing only two factors. The Front Armor Save and the Lethality of that cube's damage chart. Only the Front Armor Save would matter, as with a normal model - unless the cube has the 'all around armor' ability as a normal model can have.

Well, actually there would be a third thing that would modify the cost. That would be when a given cube cannot be hit from one or more sides.

_________________
Net Epic Coordinator

Thinking of them as being constructed out of cubes would also allow the definition of what composes the "interior" cubes, as well as possibly allowing for those inner cubes to be shot at and damaged once the cube(s) blocking them are destroyed.

For example, on a Warlord Titan the Rear Reactor location could be replaced with an "Armor" location that when destroyed has no effect other than allowing for the Reactor, which is the next cube in, to be targeted.
_____

Just in case it is not obvious, I'm posting these as I'd like feedback from people. Which systems do you like or dislike, and why?

_________________
Net Epic Coordinator

Hi!

I like this approach.

I would agree the the frontal valuations is all that matter, that and lethality. Everything else should not be charged "extra" for.

Primarch

_________________
Primarch


The Primarchload
Magnetized Titans Tutorial
Net Epic Gold
Heresy Rules

Er, I made a typo. The factors that would matter with Cubes would be lethality, miss chance, and whether or not it can be targeted from all four sides, as changes in value for side would be included in the base cost. So I guess there are three factors after all.

_________________
Net Epic Coordinator

Small update on the Lethality factor. I've updated the Secondary Effects costs slightly and added a basic list of effects. So far, this is just from Imperial Scout & Battle Titans, so the list of effects will expand.

Secondary effect(s):
none _______ 0.1
rolls +1 ___ 0
temporary _ -0.05 per level
Permanent _ -0.1 per level

CAF halved: 1
roll 4+ to give orders: 1
Titan falls: 2
Movement restricted: 1 level
No turning: 1
No Movement: 2
Fire at penalty: 1
No firing: 2
Shields down: 1
Area explosion: 1 per d6cm diameter & 1 per TSM
spot attacked: 0.5 per d6cm distance & 0.5 per TSM
Ethereal Psychic: 2
cannot enter difficult terrain: 1

As a detailed example of applying this, the Imperial Titan Head location's damage chart:
1 [0.2]: Damaged (+0.1): CAF halved (temp, -0.05) and roll 4+ to give the Titan orders (temp, -0.05). (Titan otherwise continues existing: +0.2) for total of +0.2.
2 [0.2] same as 1
3 [0.1]: Crippled (0): CAF halved for rest of game (Perm, -0.1). (Titan continues: +0.2) for total of +0.1.
4 [0]: Crippled (0): both secondary as result 1, for rest of game (each -0.1). (Continues: +0.2) for total of 0.
5 [-0.4]: Destroyed (0): Titan falls (-0.2). (Model destroyed: -0.2) for total of -0.4.
6 [-0.4]: same as 5.
Total cost for Head location is (0.2+0.2+0.1+0-0.4-0.4) -0.3.

Using these values, Imperial Titan location Lethality values are:
Head: -0.3
Reactor: -2.1
Weapon: +1.3
Leg: +0.1
Carapace: +0.2
Psy Cannon: +1.4
Psychic Resonator: -1.7

How does this adjust Titan values? Well, they don't just yet. Next I'm going to look at Miss Chance adjustments and see what I can come up with there.

_________________
Net Epic Coordinator

In thinking about the Cubes concept a bit more, I think I'm not going to pursue that at the moment. Part of the reason for that is that looking at the hit location templates shows that the majority of locations that would be differing sides of the came cube do not vary in save value based on which side it is on. Some do, but most do not. Also, some vary from one direction (usually the Rear) but not the others (L/R), which makes that concept more difficult to work with. Thus it seems to me at this time that applying a modifier for 'all around saves' to the majority of locations is more complex than just figuring out cost from all four sides, as previous versions have done.

Also the number of factors being the same means there is no simplification from changing systems, thus changing would probably be a bad idea.

That said, I think a few of the individual location templates need a bit of revising. I'll bring up specifics of that in the threads for those armies.

_________________
Net Epic Coordinator

Another option might be to use the averages for each facing.

The average save on a given armour facing can be easily worked out.
Then you can combine these for the average save of the model.

For example, a warlord has an average save of:
Fa 1.75
Sa 1.8
RA 2
For a combined average save (1xfront + 2xside + 1xrear)/4= 1.84

Then you need to work out the chance of an unsaved hit killing it.

Head 33%
Reactor 33%
Leg 33%
Weapon 6%
Carapace 0.166+(0.166x0.66)=27%

So you can average this out based on the number of hit locations on a facing for:
Front=23%
Sa=16%
Ra=23%

So a warlord titan is a unit with an armour save of 1.84, with each unsaved hit having a 19.5% chance of killing it. Or to put that another way, 5 wounds.

How would that look in the points formula?

_________________
.'.
http://ragged-they-kill.blogspot.com.au/

That is an interesting idea. I'm not putting a lot of time into hit locations just now, as adjusting Template weapons is filling most of my spare time, but when I do get back to this I'll give that some thoughts.

However, I should have posted this some time ago. I did work out some thoughts for Miss Chance, and a bit more on Lethality. What I have so far:

C. Miss chance

Each location has eight possible adjacent boxes, from a two-dimensional perspective.

If we say that three empty adjacent boxes is baseline, with an adjustment of 0.2 per, we get the following chart.

_ Empty
Squares _ modifier
__ 0 ______ -0.6
__ 1 ______ -0.4
__ 2 ______ -0.2
__ 3 _______ 0
__ 4 ______ +0.2
__ 5 ______ +0.4
__ 6 ______ +0.6
__ 7 ______ +0.8

Next I've been counting up the number of locations (of all types) that has any given number of Empty Squares adjacent to it and multiplying that number by the modifier from the chart just above. The total of these numbers I've added together, then divided by the total number of locations on that hit location chart to find the final Miss Chance value.

As a specific example, an Imperial Warhound.
Has no locations with 0, 1, or 2 empty adjacent squares.
Has 2 locations (Leg) with 3 empty adjacent, but that is default so value 0.
Has 2 locations (leg) with 4 empty adjacent, so 2 * 0.5 = 1.
Has 4 locations (Head, Reactor, 2 Weapon) with 5 empty adjacent, so 4 * 1 = 4.
Has 8 locations (weapon) with 6 empty adjacent, so 8 * 1.5 = 12.
Has 6 locations (leg) with 7 empty adjacent, so 6 * 2 = 12.
Total is (-0-0-0+0+1+4+12+12) 29. Warhound has 22 total locations, so 29 / 22 = 1.31818

So as to have something to compare to, the Imperial Warlord.
Has no locations with zero empty adjacent, so 0.
Has 6 locations (1 Head, 1 Reactor, 2 weapon, 1 leg, 1 carapace) with 1 empty adjacent, so 6 * -1 = -6.
Has no locations with 2 empty adjacent, so -0.
Has 6 locations with 3 empty adjacent, so 6 * 0 = 0.
Has 10 locations (6 weapon, 4 carapace) with 4 empty adjacent, so 10 * 0.5 = 5.
Has 12 locations (6 weapon, 6 leg) with 5 empty adjacent, so 12 * 1 = 12.
Has 4 locations (weapon) with 6 empty adjacent, so 4 * 1.5 = 6.
Has 6 locations (leg) with 7 empty adjacent, so 6 * 2 = 12.
Total is (-0-6-0+0+5+12+6+12) 29. Warlord has 44 locations, so 29 / 44 = 0.65909

Oddly, the Reaver also comes out to a total of 29. It has 42 locations, so it is 0.69048

At this time, I'm planning on using the Miss Chance value as a multiplier. Most likely, I'll add together the Armor Save value and the Lethality value and multiply that total by the Miss Chance value, but I haven't quite gotten that far as yet. Most my little spare time has been spent adjusting Template values for the adjustments there and posting on various Facebook threads. It's starting to look like I may not like NE6 very much, even if it does use the formula, but that's how things go.

_________________
Net Epic Coordinator

Scattering on the template is pretty tricky, especially if using the average method I mentioned for comparison above. You could work out the average chance of missing, and decrease the average lethality proportionally. That doesn't take into account the chance of scattering on to another location, so you could work out the chance of that, and then the average lethality.
I have a long flight coming up, so I'll have a couple of red wines and scrawl on a napkin and see what I come up with.
I'm not a Facebook guy, is there a rundown here somewhere on NE6?

_________________
.'.
http://ragged-they-kill.blogspot.com.au/

Hi!

I post summary updates here from time to time, but for this particular topic, your at the "epicenter" of the debate on this particular topic.

Primarch

_________________
Primarch


The Primarchload
Magnetized Titans Tutorial
Net Epic Gold
Heresy Rules

Got it. Cheers

_________________
.'.
http://ragged-they-kill.blogspot.com.au/

Yeah, I haven't had any significant time to continue this for the last few days. Once I finish typing up and posting the results of V0.32 & V0.325, I'll start looking at this again. I'm not planning on using the averaging method you mentioned above, as that is factored into how many empty squares are adjacent to any given location. At least, I feel that it is.

My comment on NE6 was slightly misplaced. At the least, it should have been a different paragraph. I said that, because, at the time, several people were / are pushing for NE6 to be made in certain ways that I don't care much for. I guess I should not be surprised at that, as I've often found that my opinions on things tend to be in a minority group. Well, as Primarch often says, my opinion does not outweigh that of everyone else. If the majority want things a certain way, then it should be that way, even if I feel it is wrong.

_________________
Net Epic Coordinator