41 replies to this topic

[BlueFlames]

tl;dr:

Spoiler

Nope! Those big percentages just make it easy to overestimate the effect of all of those armor and structure nodes.

On my early examination of the durability tree, I realized that it might be a trap. My initial inclination, upon seeing the current iteration of the durability tree was to call a twenty-six node investment a heavy but worthwhile investment to make mechs extra-tanky. For a sixty-five ton mech--and I was initially theorycrafting skill tree builds for my Linebacker--the bonuses for that investment are 13.6% armor and 25.6% structure, which sounds like a lot.

There's a couple of important questions that need to be asked about those percentages, though: How many absolute hitpoints do those percentages translate into, and what do those bonus HP mean in practical terms?

The first question is simple math, provided in the table below:

Mass    CT Ar/St    26 Ar/St Nodes       CT Bonus HP
 20t      24/12       20.8%/32.8%        4.992/3.936
 30t      40/20       19.2%/31.2%        7.680/6.240
 40t      48/24       17.6%/29.6%        8.448/7.104
 50t      64/32       16.0%/28.0%        10.24/8.960
 55t      72/36       15.2%/27.2%        10.94/9.792     20.73 bonus HP
 60t      80/40       14.4%/26.4%        11.52/10.56     22.08 bonus HP    <- Peak armor bonus
 65t      84/42       13.6%/25.6%        11.42/10.75     22.17 bonus HP
 70t      88/44       12.8%/24.8%        11.26/10.91     22.17 bonus HP
 80t     100/50       11.2%/23.2%        11.20/11.60     22.80 bonus HP
 85t     108/54       10.4%/22.4%        11.23/12.10     23.33 bonus HP | 185.33 total HP
 90t     116/58        9.6%/21.6%        11.14/12.53     23.67 bonus HP | 197.67 total HP    <- Peak structure bonus & peak bonus HP
 95t     120/60        8.8%/20.8%        10.56/12.48     23.04 bonus HP | 203.04 total HP
100t     124/62        8.0%/20.0%         9.92/12.40     22.32 bonus HP | 208.32 total HP    <- Peak total HP

As a function of tonnage, the bonus HP explodes from the nine HP that an unquirked twenty-ton mech gets to its center torso, until it nearly levels off at about twenty-two bonus HP around sixty tons. At that point, the armor bonus actually starts to get smaller, but the structure bonus continues to grow fast enough to make up the difference, until the total bonus HP peaks at about twenty-four for a ninety-ton mech, after which point the structure bonus begins to decline alongside the armor bonus.

In practical terms, a dual-Gauss hit eats up those bonus HP. A dual-(ER)PPC hit steals almost all of those bonus HP. Four IS medium pulse lasers burn through those bonus HP in their 0.6-second burn time. In other words, twenty-four bonus HP basically buys you a single extra hit from weapons that don't badly spread their damage.

Is a single hit worth spending a quarter of your skill nodes? Ultimately, that's for you to decide, but I'm inclined to say, "No." Saving the twenty-six nodes that would otherwise be put into the durability tree allows you to hunt more heat gen nodes in the weapon tree and more torso speed nodes in the mobility tree. Faster twist speed will allow you to better place incoming damage where you want it to hit, and lower heat generation will help you fire the extra shot you need to cut through your target's bonus HP.

Caveat: My interest was in the general case. The effect of quirk HP and the bonus HP that the skill tree provides to locations besides the center torso I leave as an exercise for the reader. Most kills are CT kills, and most mechs don't have (large) durability quirks, so doing the math for an unquirked CT felt like the most productive use of my time. You'll have to do the math yourself, if you want to know if dozens of durability nodes are worthwhile for the special case of an XL Atlas.

[mushis]

excellent content from BlueFlames once again, thanks for gathering all that information man.

[Bud Crue]

BlueFlames, on 19 May 2017 - 08:50 PM, said:

In practical terms, a dual-Gauss hit eats up those bonus HP. A dual-(ER)PPC hit steals almost all of those bonus HP. Four IS medium pulse lasers burn through those bonus HP in their 0.6-second burn time. In other words, twenty-four bonus HP basically buys you a single extra hit from weapons that don't badly spread their damage.

Is a single hit worth spending a quarter of your skill nodes? Ultimately, that's for you to decide, but I'm inclined to say, "No."

To me, 26 points to survive an additional dual-gauss hit is certainly worth 26 points. I mean that is to my way of thinking the core of the ISXL v Clan XL debate. By the time you lose an ST you are generally 1-2 hits away from death. Does that clan XL make a big difference? Hell yes. The ability to stay functional for those 1-2 hits can make all the difference in a match. Thus, looking at the survivability tree, I feel that at least some investment there is nearly always worth it.

[An Innocent Urbie]

damn I need to buy SPs again and remove those that I have put on my linebacker

thanks BlueFlames!

[Gaussfather]

Nice post! Keep in mind that the main IS 100 tonners & mechs like Victor, Highlander, etc have big armor quirks now and this only adds to it... so the increase is perhaps more than 24% max for these... I leave it to you to do the math!

BUT I find that certain mechs feel tankier when you combine this extra/armor structure, and of course adding a light engine helps alot. At some point the cooldown quirks (around 12% max if you grab all of them) have diminishing return vs the amount of SPs you need to spend, because for most builds your limiting firepower issue during a match is heat buildup. And that's the trade off between space and number of DHSs you can put in your mech. Most people get killed during a shutdown!

So I think its better to take few SPs from the weapon tree and put them into an extra coolshot and the heat tree.

At first I was a big critic of the new skill tree but I'm starting to like it more...

[Lostdragon]

It really depends on the opportunity cost, imo. I wanted to play my SDR-5K with HMGs last night and I skilled it into survivability but I probably wouldn't skill an ACH that way. An MG based loadout gets very little from the Firepower tree, many nodes there are completely useless to it and MGs have such short range that the small increase from range nodes is not worth it, imo. The mech doesn't have heat issues so it doesn't need to invest heavily into the Operations tree for heat capacity/dissipation. I am not trying to spend 120k/match in consumables, so no need to go too deep into Auxiliary. It only seemed to make sense to spec heavy into Survivability and Mobility so I can get in close and out manoeuvre opponents while being able to survive a bit longer, but on a laser or missilebased light I would skip survivability for Firepower and Operations probably.

[ScrapIron Prime]

Most mechs I run ignore this part of the skill tree, it's very situational IMO. I dumped points into it on my Pirate's Bane, however, because an extra 6 points of armor and extra structure on each leg keeps that locust in the fight much longer.

[Angel of Annihilation]

BlueFlames, on 19 May 2017 - 08:50 PM, said:

tl;dr:

Spoiler

Nope! Those big percentages just make it easy to overestimate the effect of all of those armor and structure nodes.

On my early examination of the durability tree, I realized that it might be a trap. My initial inclination, upon seeing the current iteration of the durability tree was to call a twenty-six node investment a heavy but worthwhile investment to make mechs extra-tanky. For a sixty-five ton mech--and I was initially theorycrafting skill tree builds for my Linebacker--the bonuses for that investment are 13.6% armor and 25.6% structure, which sounds like a lot.

There's a couple of important questions that need to be asked about those percentages, though: How many absolute hitpoints do those percentages translate into, and what do those bonus HP mean in practical terms?

The first question is simple math, provided in the table below:

Mass	CT Ar/St	26 Ar/St Nodes	   CT Bonus HP
20t	  24/12	   20.8%/32.8%		4.992/3.936
30t	  40/20	   19.2%/31.2%		7.680/6.240
40t	  48/24	   17.6%/29.6%		8.448/7.104
50t	  64/32	   16.0%/28.0%		10.24/8.960
55t	  72/36	   15.2%/27.2%		10.94/9.792	 20.73 bonus HP
60t	  80/40	   14.4%/26.4%		11.52/10.56	 22.08 bonus HP	<- Peak armor bonus
65t	  84/42	   13.6%/25.6%		11.42/10.75	 22.17 bonus HP
70t	  88/44	   12.8%/24.8%		11.26/10.91	 22.17 bonus HP
80t	 100/50	   11.2%/23.2%		11.20/11.60	 22.80 bonus HP
85t	 108/54	   10.4%/22.4%		11.23/12.10	 23.33 bonus HP | 185.33 total HP
90t	 116/58		9.6%/21.6%		11.14/12.53	 23.67 bonus HP | 197.67 total HP	<- Peak structure bonus & peak bonus HP
95t	 120/60		8.8%/20.8%		10.56/12.48	 23.04 bonus HP | 203.04 total HP
100t	 124/62		8.0%/20.0%		 9.92/12.40	 22.32 bonus HP | 208.32 total HP	<- Peak total HP

As a function of tonnage, the bonus HP explodes from the nine HP that an unquirked twenty-ton mech gets to its center torso, until it nearly levels off at about twenty-two bonus HP around sixty tons. At that point, the armor bonus actually starts to get smaller, but the structure bonus continues to grow fast enough to make up the difference, until the total bonus HP peaks at about twenty-four for a ninety-ton mech, after which point the structure bonus begins to decline alongside the armor bonus.

In practical terms, a dual-Gauss hit eats up those bonus HP. A dual-(ER)PPC hit steals almost all of those bonus HP. Four IS medium pulse lasers burn through those bonus HP in their 0.6-second burn time. In other words, twenty-four bonus HP basically buys you a single extra hit from weapons that don't badly spread their damage.

Is a single hit worth spending a quarter of your skill nodes? Ultimately, that's for you to decide, but I'm inclined to say, "No." Saving the twenty-six nodes that would otherwise be put into the durability tree allows you to hunt more heat gen nodes in the weapon tree and more torso speed nodes in the mobility tree. Faster twist speed will allow you to better place incoming damage where you want it to hit, and lower heat generation will help you fire the extra shot you need to cut through your target's bonus HP.

Caveat: My interest was in the general case. The effect of quirk HP and the bonus HP that the skill tree provides to locations besides the center torso I leave as an exercise for the reader. Most kills are CT kills, and most mechs don't have (large) durability quirks, so doing the math for an unquirked CT felt like the most productive use of my time. You'll have to do the math yourself, if you want to know if dozens of durability nodes are worthwhile for the special case of an XL Atlas.

This is a really tough one to decide to be honest. When you look at it from the standpoint a single hit negating the bonus you get, it generally doesn't appear to be worth it but I can't count the number of times I have barely survived a match with my CT hanging on by 1 or 2 points of structure.

The thing of it is, one hit could mean the difference between you dying early in the match having accomplished little to nothing, all the way up to you being the MVP of a match. The problem is how do you determine which situation is more predominant, dying only a few seconds later or lasting out the entire match because you scraped by with 1 point of structure you wouldn't have had without the skill?

Also on top off that you have to also evaluate what your giving up to gain that durability. I mean 26 points is alot of points. 26 points can give you 5-10% better heat management, 5-10% more range, 5-10% more cooldown, better sensors, more agility and speed or some combo of several of these things.

So there is no really easy answer here but for me, I am personally starting to feel that I gain more advantage from spending the points I save on not purchasing points in the survival tree and instead spending it on the mobility or firepower tree more often than not. Depends on the mech of course.

[Methanoid]

the problem is its percentage based, so on lighter mechs that may actually want some better protection, a tiny % of increase to tiny amounts of armor = a tiny insignificant increase thats not worth it.

Tbh the whole skill tree is flawed in that you are forced to pick useless junk you dont want just to take a few skill nodes which you do want that are typically so tiny and insignificant that they are barely noticable at all, in many cases you notice exactly no difference, with the exception of say radar derp, 1 of very few that has a high % bump per node.

The skill tree in no way feels rewarding or useful, all the flawed system does is buff things that are already decent and give un-noticable increases to the things we lack but actually wanted buffed in the first place, its a really terrible system again likely hampered by wanting to have some form of silly attachment to the tabletop system.

For example, my recently aquired Adder Cinder, i wanted a bit of a speed increase to it and thats all in the mobility tree as its default mobility stats are all superb out of the box, yet to just get 5 tiny 1.5% nodes (all located at the far bottom of the mobility tree) i have to unlock a ton of nodes that are of utterly no use to me whatsoever just to gain a pathetic 7.5% speed increase, thats 21 points spend just to get the only 5 nodes of actual use to me and its a tiny buff at that, and only of slight use to mechs that are already quite fast in the first place.

The same applies to many other mechs which have a specific deficiency you would like to correct only to find that because that deficiency is typically due to a low numbered stat it cant be increased in any meaningful way because of the skill trees silly nodes all being low % based and you have to aquire a ton of stuff you dont want or need, and even if you did need them on a ton of mechs those increases barely register.

For example, range nodes on weapon tree.

You have for example c-sml pulse lasers and want a bit of a range buff, you take all the nodes and you aquire an insignificant range buff, those same nodes on say a C-ER-PPC bags you a metric cr*pton of extra range on a weapon you dont want or even need that huge range buff on.

The skill tree is completely backward and silly.

[Reza Malin]

I usually only take durability nodes for larger mechs that are already durable, or, for mechs that have a lot of bonus armour points already.

That way the bonus armour and structure from the skill tree only adds to what is already a large pool and you notice it more.

EG. My Heavy Metal has the highest armour i have seen yet, at 151 CT armour when skilled into durability.

My Catphract can get it up to a really high level too, can't remember off the top of my head but it is higher than a lot of assaults.

So yeah i use it as a complement to an already significantly durable mech to focus its strengths more. Which i thought is what the skill tree was all about?

[Methanoid]

Reza Malin, on 21 July 2017 - 08:28 AM, said:

So yeah i use it as a complement to an already significantly durable mech to focus its strengths more. Which i thought is what the skill tree was all about?

thats just how it ends up because the skill tree ends up useless for the other stuff you might have wanted buffed but dont because its not worth it, given that, why even have the skill tree when its 100% predictable what ppl will skill for in each mech, may as well have just stuck to the stupid quirks given that theres no real choices to be made.

[HGAK47]

For face time mechs like my shadowhawk with the rotary 5`s, I absolutly do think that extra hit or two is worth it but I can easily see why others wouldnt want to take it.

I think building your mech around a role or style is the best way to go. Arbitarily sticking points into survivability is probably not a good idea. Much like sticking points into jumpjets is really not a good idea unless your role depends on it or you just like it. (Save the JJ heat gen, thats really useful on hot jj mechs)

[Reza Malin]

Methanoid, on 21 July 2017 - 08:38 AM, said:

thats just how it ends up because the skill tree ends up useless for the other stuff you might have wanted buffed but dont because its not worth it, given that, why even have the skill tree when its 100% predictable what ppl will skill for in each mech, may as well have just stuck to the stupid quirks given that theres no real choices to be made.

There are choices, but not every mech has every choice.

Which is fine, you still need mech classes with a rough group of roles they fit.

[Methanoid]

Reza Malin, on 21 July 2017 - 08:45 AM, said:

There are choices, but not every mech has every choice.

Which is fine, you still need mech classes with a rough group of roles they fit.

for the few skills needed in those tree's to define their roles we are however typically taking a ton of throwaway skills just to get the few we want, such as radar derp, maybe narc/ecm/other niche skills, the rest is nearly fully disposable or not required making the skill tree a bit of a waste.

[Reza Malin]

Methanoid, on 21 July 2017 - 08:51 AM, said:

for the few skills needed in those tree's to define their roles we are however typically taking a ton of throwaway skills just to get the few we want, such as radar derp, maybe narc/ecm/other niche skills, the rest is nearly fully disposable or not required making the skill tree a bit of a waste.

The alternative would be being able to pick all the most desirable skills, then there would be even less variation between people's builds.

It forces you to spread your points so you cant cherry pick, which is fine.

[LT. HARDCASE]

I have found that it's almost always worth it, to take Survival.

Was just in a match where my Uziel had red CT internals, meaning I would've been dead long ago, without those 25 points.

Taking the 9 points on the right side of the tree is a waste though.

[Snazzy Dragon]

Certainly more worth it than most of what the other trees offer. 3 extra degrees of torso pitch? Psh, get out of here with that crap.

Especially on mechs with lower base agility, where spending 30 points in mobility gets you from 7 to 9 m/s acceleration. Plus, with the threat of more DPS (is UACs, light PPCs, RACs, etc) and higher "prolonged" burst (heavy lasers) weapons, that extra armor/structure to torso twist with is significant.

[Antares102]

If the survival tree had no filler nodes and I could pick armor/structure bonus freely I would use them.
As it stands right now I prefer mobility and RADAR derp/Seismic.
Especially seismic saves you TONS of armor and outright your live if you know what comes your way through a wall.

Edited by Antares102, 21 July 2017 - 09:29 AM.

[Y E O N N E]

Snazzy Dragon, on 21 July 2017 - 09:27 AM, said:

Certainly more worth it than most of what the other trees offer. 3 extra degrees of torso pitch? Psh, get out of here with that crap.

Especially on mechs with lower base agility, where spending 30 points in mobility gets you from 7 to 9 m/s acceleration. Plus, with the threat of more DPS (is UACs, light PPCs, RACs, etc) and higher "prolonged" burst (heavy lasers) weapons, that extra armor/structure to torso twist with is significant.

Speed Tweak back to 10% pls.

[Vesper11]

Armor is really good for faster med/lights that can spread the damage, it also protects against machine guns and other crits as well as clan stuff blowing up. And it's only 16-18 points.