32 replies to this topic

[Serpentine Shel Serpentine]

Koniving, on 25 May 2017 - 08:30 PM, said:

Some people value one more than the other.
For some, armor seems to vaporize really quickly, making it seem worthless.
For some, structure quirks aren't worth much since all the weapons, equipment, etc. is getting destroyed. There's also the 15% of crit damage goes into structure damage as bonus damage, making structured die quicker for weapons with high crit successes.

The "numbers" on your link, is the minimum amount of armor you get from quirks to make armor hardening worth the investment in skill points.
The same is true of structure and skeletal density skills, if you get a certain amount of structure from quirks for X weight, it's worth investing whole heartedly into structure.

Kinda like if your lasers get a reduction in beam time with quirks, the four laser reduction skills pay off far more than if you didn't have that quirk.

Thanks Koniving .. I do understand that metamechs is setting a quirk threshold for structure and armor by weight. What I don't understand is why there is a threshold and why it's not just either-or. For example since the choice or armor or structure is hit point neutral if there are no quirks, then any quirk in either category should tip the balance if there's no quirk in the other (and so far as I know it's always either structure or armor not both). But there are complications in many cases: (1) most people don't take full armor, especially on the legs, head, and "dead" arms and other places, reducing the impact of armor buffs, whereas structure is not reducible; (2) armor prevents critical hits and structure does not. Those variables are hard to quantify and weigh against each other. Metamechs doesn't mention that (I think) and doesn't explain why the weight class should affect the relative value of armor and structure. So I see what he is saying about when to use either one (although that could be more clear and I think he has a typo lurking in there somewhere making it a little inconsistent ). But I don't understand in principle why the thresholds are where he puts them. So if someone gets that and you have time to explain, it would be much appreciated. Thanks!

[Koniving]

Ever heard the phrase most value for your buck?

Consider this:
Pepsi is selling at this place nearby for $1.25. Coke is selling for $1.30 cents. It costs nothing to go there and come back home.

Considering that they both give you 2 liters, which is the better buy? Obviously the cheaper one, unless you really prefer the other one.
(This is why people have said "don't get armor hardening", as for example the values I see on my 65 ton mech are 1.7% armor hardening or 3.4% structure density. Ironically, 1.7*2 = 3.4 and armor has twice the value of structure when quirks are not involved so the cost is identical, in this metaphor both Pepsi and Coke is actually valued at 1.30. This means they should be the same, but if Pepsi is Skeletal Density and I have a quirk for structure, I'm gaining more with 'Pepsi' and thus, Pepsi is cheaper.. [though a more correct analogy would be to say I'm getting 2.5 liters for the price of 2 liters.... but lets not mess with the vibe here.])

Another place, which will cost you about a dollar in gas to go to and come back home is selling Pepsi for $1.00. Coke is $1.20 cents.
Now obviously there's a better deal here regardless of which one you actually prefer. But the cost of going there and back is a factor.
In order to offset the cost of going there and coming back versus the closer store with no cost to get there (because you can walk to that one), you must get at least get at least 4 Pepsi 2 liters to break even and 5 Pepsi 2 liters to make the trip worth while in order to 'save' money versus the closer more expensive option.
To save money on the Coke, you must buy at least 13 to break even for what you would spend to what you would gain compared to walking to the local store as anything less and you're losing money.. 14 is the minimum to buy in order to save a little money,

Thus, thresholds. Cost versus effect.
At said threshold, the gain value of the skill point being spent on that element is high enough to be considered a 'better buy'.

Edited by Koniving, 26 May 2017 - 08:47 AM.

[Serpentine Shel Serpentine]

Koniving, on 26 May 2017 - 08:41 AM, said:

Ever heard the phrase most value for your buck?

Consider this:
Pepsi is selling at this place nearby for $1.25. Coke is selling for $1.30 cents. It costs nothing to go there and come back home.

Considering that they both give you 2 liters, which is the better buy? Obviously the cheaper one, unless you really prefer the other one.
(This is why people have said "don't get armor hardening", as for example the values I see on my 65 ton mech are 1.7% armor hardening or 3.4% structure density. Ironically, 1.7*2 = 3.4 and armor has twice the value of structure when quirks are not involved so the cost is identical, in this metaphor both Pepsi and Coke is actually valued at 1.30. This means they should be the same, but if Pepsi is Skeletal Density and I have a quirk for structure, I'm gaining more with 'Pepsi' and thus, Pepsi is cheaper.. [though a more correct analogy would be to say I'm getting 2.5 liters for the price of 2 liters.... but lets not mess with the vibe here.])

Another place, which will cost you about a dollar in gas to go to and come back home is selling Pepsi for $1.00. Coke is $1.20 cents.
Now obviously there's a better deal here regardless of which one you actually prefer. But the cost of going there and back is a factor.
In order to offset the cost of going there and coming back versus the closer store with no cost to get there (because you can walk to that one), you must get at least get at least 4 Pepsi 2 liters to break even and 5 Pepsi 2 liters to make the trip worth while in order to 'save' money versus the closer more expensive option.
To save money on the Coke, you must buy at least 13 to break even for what you would spend to what you would gain compared to walking to the local store as anything less and you're losing money.. 14 is the minimum to buy in order to save a little money,

Thus, thresholds. Cost versus effect.
At said threshold, the gain value of the skill point being spent on that element is high enough to be considered a 'better buy'.

Well I'm not asking you to explain this again since you've already done a lot and I appreciate that. So feel free to ignore this question. But .. I understand transaction costs and the very clear example you give. But I don't see how it relates to the skill tree since there are no transaction costs once you have a skill point and an available node in the tree: at that point the only cost is the opportunity cost of not using another node. And in this limited example we are only trading off between structure and armor and ignoring all the rest (hypothetically - obviously you can't do that for real). And this example is before you have chosen either one -- there are transaction costs to switching a node of course, but that's not the example. So in the hypothetical example the only question is the relative marginal benefit of structure vs. armor. I think we agree that on a mech without quirks to either structure or armor the two different nodes are hit point equivalent (because structure nodes are double armor in % value, and base armor is double base structure) and that equivalence scales across all weight classes -- the two nodes are always equivalent in hit point value absent quirks (or the additional variables I mentioned before: armor reductions and indirect crit chance effects). Quirks change things because they are independent of each other -- a quirk to structure does not affect available armor and vice-versa. Quirks break the equivalence. But my question remains: in this example what is "cost of gas" (per your example) that is different for the two options? And how does that relate to weight class as Metamechs believes it does?

[Koniving]

If it helps, consider Pepsi and Coke to be armor and structure (doesn't matter which is which). The second store is a mech with hefty quirks in Pepsi and a miniscule quirk in coke.

In reality, the amount of soda -- the 2 liter is what actually represents the skill point which I know makes the analogy confusing and I hadn't thought of that. A better way to go about it was to compare maybe 16 oz to 20 oz at identical costs but then it'd be hard to represent a different mech with a different store.

So in the above example, consider a 2 liter to be your "skill point" and the money saved to be your reward (less money = more 'gain').
In the first store, you can pick either coke or pepsi due to preference. There isn't much difference and the gain differences are minimal if they exist at all. Ultimately, if you like coke you're gonna get coke. Apples to oranges. Whether it's one 2-liter or 10, you'll spend as much skill points (have as many 2 liters) as you want and you can tally what you gained (money saved) if you really care.

In the second store (the second mech), the cheaper prices are because the there's sales incentives (in other words, the mech already has some structure and armor quirks; but the significantly higher savings for one over the other means investment into the one will net significantly higher returns [represented in the analogy as savings]. If you only need five 2 liters to save money versus 14 2 liters to save money... think of it like you only need 5 skill points to reap higher than normal benefits, versus having to spend 14 skill points to reap higher than normal benefits.

To take it out of the analogy...

Lets say X mech of whatever size has +5 armor and +10 structure.
If you get 1.7% better armor with each SP or 3.4% better structure with each SP... right there, structure is better at face value. Right?
Actually it isn't that easy. It's completely identical here as you pointed out. Armor is 2x as plentiful as structure, so if armor = structure as HP, both are either 3.4 or if structure = armor both are 1.7... 5*2 armor = 10 structure, it's identical.
It's another apples to oranges. when really both are just fruit.

But what if it has +7 armor and +10 structure.
You'll get more out of armor per skill point.

What if it has +6 armor and +12.5 structure.
The gain is almost non existent, but you'd get more out of structure.

The list goes on. The analogy was trying to say for every skill point you spend (2 liter you buy), you can get X savings (X benefits), but the more you save the better (the more you get out of those SP, the better).
It then proposed the second store (second mech), which had incentives. The 'cost to get there' is actually just there to create the threshold you must hit in order to gain superior benefits from choosing one option or the other for your total skill points.

So for some mechs, quirks make more value in one way or the other.

In certain weight classes, however, the percentage given isn't actually 'identical', either.

Edited by Koniving, 26 May 2017 - 10:45 AM.

[Rogue Jedi]

to add to what Koniving has said, in the above example, with getting exactly the same number of hit points extra weather you choose armor or structure, the armor is clearly superior.
why? you have to take into account critical damage and critical hits.

basicly when you suffer any damage to an unarmored componant 1 random piece of equipment takes damage, ^{(in the case of a single impact weapon that is all taken by 1 componant, for missiles or LBX pelets or Clan ACs that is recalculated for each projectile, for lasers you get a number of ticks, or pulses of damage over the lasers burn time)}, if the equipment takes damage in excess of its health it is destroyed, some weapons have an improved chance of doing more damage to that piece of equipment, you can do 1, 2 or 3 times as much damage as it is supposed to, this is called a critical hit, and will not only do increased damage to the equipment but also the Mechs structure.

here is a more specific example if your Mech has 6 armor left on the CT and is hit by an IS AC20 shell you will suffer 6 damage to the armor, then 14 damage to the structure, that 14 damage will probably destroy something, perhaps a weapon, or a heatsink or some ammo, then there is a possibility you could recieve up-to an extra 28 damage.
In that example there the worst case is the AC20 does 48 damage, and also takes out a full ton of ammo for an extra 200 damage to your CT structure.

if however you had 21 armor left worst case you suffer 20 damage leaving you with 1 armor.

I hope I am not confusing things further.

Edited by Rogue Jedi, 26 May 2017 - 10:57 AM.

[Koniving]

That rabbit hole goes even deeper than Rogue described. If one were to use the Force (of math), Luke, there's also the fact that there's a mechanic where 15% of whatever crit damage is dealt is then delivered to the structure as bonus structure damage... Yeah. It's bad to (take bullets to) the bone.

If an IS AC/20 hits you in a spot with no armor, then in the rare event it nets 3 crits... 60 damage in critical hits is delivered (20 to each slot chosen at random), instantly destroying almost anything they touch. 15% of that 60 (9 damage) is then added on to the base 20 damage to your structure, so that shot did 29 damage to your bones and 60 damage to your guts. You just got screwed.

Of course, armor will keep that 20 damage AC/20 as 20 damage, provided you have at least 20 armor to block it off with.

Note: I understated "Rare."
Try super-duper-ultra-mega-rare...
or as unlikely as Half-Life 3 being confirmed.
<.<;
Or is it really that rare?

Edited by Koniving, 26 May 2017 - 11:03 AM.

[Rogue Jedi]

Koniving, on 26 May 2017 - 11:00 AM, said:

Note: I understated "Rare."
Try super-duper-ultra-mega-rare...
or as unlikely as Half-Life 3 being confirmed.
<.<;
Or is it really that rare?

for the AC20 yes it is, but for the Machine Gun or LBX ACs ^{(yes including the LB 20-x AC)} which have higher crit chances, I do not think it is that rare

[Koniving]

They're also not delivering 20 damage per crit and instantly destroying equipment. Though I do recall when the AC/20 once had 18 health, and twin MGs took it out in less than 2 seconds after getting the armor away (accompanied with 2 LB10-X and 4 flamers, but the LB-X were cycling so it was just the 4 flamers and 2 MGs).

[Serpentine Shel Serpentine]

Koniving, on 26 May 2017 - 10:32 AM, said:

In certain weight classes, however, the percentage given isn't actually 'identical', either.

Re This point above -- sorry to drill down on this relentlessly, but it seemed like the examples you gave were weight invariant. The quirks seem to determine the preference to the same degree now matter what the total armor or structure may be. Unless I missed something.

I totally get that there is a strong reason to prefer armor even if the hit point yield is the same or a bit lower -- although it seems very difficult to quantify where the break-point is that one should choose structure instead.

But above you say that in certain weight classes the percentage isn't "identical" (meaning there is no hit point equivalence at base values?) -- but why not? That is where I keep getting lost and where the Metamechs chart seems mysterious. What difference does it make to the armor/structure relative value whether the mech is 30 tons or 100 tons? Looking at Metamechs' charts the weight makes a large difference in the structure/armor point value that tips the balance -- but I cannot discern the principles or rules at work in these threshold numbers: See http://metamechs.com.../survival-tree/

I would ask "GMan129" (the player behind Metamechs) except I don't see a way to reach him on Metamechs. Perhaps he is listening to this and could chime in?

(To state the obvious: I understand that the weight matters if you are deciding whether you want to use the survival tree at all ... on a light mech where the base values are small there is less payoff than with agility; the opposite for lumbering large mechs.)

Thanks again -- and as said before if you are just tired of this . . . I understand completely.

[Koniving]

We dont know the real reason beyond "assault techs should feel tanky", light techs should be agile... The only time I noticed a difference was on a mech that often fell below the Jedi curve so to speak.


There was another case where a mech known in lore for a greater than average sensor suite happened to have all of its sensor related quirks at slightly higher values than within its weight class.

[Kubernetes]

Delores-- ymmv when it comes to armor and other skills. Two equally good pilots may skill out their mechs differently in accord with their individual styles.

I put my mobility points into the acceleration and braking nodes. Why? Because I love hill humping with my high mounts and it's more important to me to be able to move forward and back quickly. Torso twist is important if you do a lot brawling or corner trading. If you can afford both, do it, but otherwise select the nodes that complement your style


I skip armor nodes because I'd rather put the points elsewhere. It still takes 200+ damage to CT me, so that's good. I also only run 2-3 back armor. Survive light backstabs with awareness, team movement, and boatloads of structure. If you're getting singled out by marauding lights, your back armor is not the issue.

Oh and I would recommend you spare the points to get two upgraded Coolshots. They're incredibly powerful for only five SP.

Edited by Kubernetes, 29 May 2017 - 12:16 AM.

[Dace Darkhunter]

I used to win alot of games with the battlemaster-2c variant from which I run 3 LPL's and 2 MPL's STD 325 Engine. It was one of my favorite mechs but now it's no longer viable. The heat is too much for the mech to bear. So sad.

Edited by Dace Darkhunter, 19 July 2017 - 09:50 AM.

[AnswerIsOne]

Gwahlur, on 24 May 2017 - 05:03 PM, said:

I'm still having great success with the blr-2c.
Here's my spec, since why not. Feel free to try, modify, comment or theorycraft on it:

ditch the XL and put a LFE in it,