40 replies to this topic

[Jman5]

Great post and I agree with most of it. It was nice to see that bit about the heat skills. I was just playing around with it yesterday and came to similar conclusions. Though really, it hardly seems noticeable either way. I haven't messed with your spreadsheet yet, but I look forward to checking it out.

On the subject of Sensor Skill Class I disagree that it's a waste or neglible. I get so much mileage out of Seismic sensor that, it alone almost completely justifies the investment.

Then there is Radar Deprivation. No matter how good you are, you will take LRM damage semi-regularly. Some maps it's worse than others and some games there are more or fewer LRM-boats. However, Radar Deprivation has a practical effect of reducing average damage taken.

ECM skill for ECM mechs is like night and day. When the skill tree first dropped I didn't understand these skills and played a bunch of games without it. It honestly just felt like I didn't even have an ECM. These two skills make a huge difference.

Sensor Range, Target info gather, Target Decay, Target Retention: I can live without these, but they are decent filler skills that bring me down to the skills I really want. Being able to hold a lock long enough to see the paper doll can turn a slug match into a trivial fight. Don't forget they significantly boosted the target info gather times last patch.

My thoughts on the sensor Class is incredibly useful for:

• ECM mechs
  
• Short Range Mechs
  
• Lights
  
• Solo queue

If you're building for competitive, or play with a group that basically slaughters 90% of the teams out there, then it's probably less useful. Long range mechs, it's not a terrible investment, but many skills start to lose their value the further back you fight.

I also think there is an argument for ditching it in Faction Play because of the way Siege Mode is structured.

Edited by Jman5, 09 June 2017 - 02:00 PM.

[Appogee]

Fantastic work, Oni, thank you.

[Edward Scissorhands]

Honest Question: Why is there no mention of investment into radar deprivation? I feel that this is an imporant skill. is there a reason others arent investing here?

Also i invest 9 points into mobility into some mechs. i feel the acceleration and deceleration bonus is very good for faster positioning changes

Edited by Edward Scissorhands, 09 June 2017 - 02:54 PM.

[Ghogiel]

Edward Scissorhands, on 09 June 2017 - 02:54 PM, said:

Honest Question: Why is there no mention of investment into radar deprivation? I feel that this is an imporant skill. is there a reason others arent investing here?

Also i invest 9 points into mobility into some mechs. i feel the acceleration and deceleration bonus is very good for faster positioning changes

It's ok for 100% solo puggers because you can still get a bit of good use out of it while only committing to one side of the tree.
I expect most people, even pugs, to just get 2-3 derp nodes + 1 seismic (which is about 90% less effective as having both but NOT worth the point investment to do both) or completely skip the tree unless they are a lurmer or require ECM. Everyone else is min/maxing elsewhere because lurming, derping, or ECM isn't any of the metas right now.

Basically you want firepower and consumables as 100% mandetory you are doing it wrong if you don't, then what ever else on surv and cooling, and maybe comprimise for mobility if you are a light, brawler or mid poke mech that isn't a slow heavy or assault.

[Ultimax]

Ghogiel, on 09 June 2017 - 03:23 PM, said:

It's ok for 100% solo puggers because you can still get a bit of good use out of it while only committing to one side of the tree.
I expect most people, even pugs, to just get 2-3 derp nodes + 1 seismic (which is about 90% less effective as having both but NOT worth the point investment to do both) or completely skip the tree unless they are a lurmer or require ECM. Everyone else is min/maxing elsewhere because lurming, derping, or ECM isn't any of the metas right now.

Basically you want firepower and consumables as 100% mandetory you are doing it wrong if you don't, then what ever else on surv and cooling, and maybe comprimise for mobility if you are a light, brawler or mid poke mech that isn't a slow heavy or assault.

Exactly.

My builds are: (generally speaking)

1) Go deep into firepower. (typically 30 points minimum, average is probably around 35 nodes)
2) Pick up at least a 3rd consumable slot, and double up on the most important/useful consumable for the build.
3) Go heavy into Survivability.
4) If hot build 15 points into ops for max heat containment and 3/5 Cool Run Nodes.
5) If pure solo build, consider Radar dep - otherwise leftover points go into mobility prioritizing right side of tree for ridge humpers or poke mechs, left side of tree for mechs that need fast torso twist.
6) If for some reason the build can skip some of the stuff in points 3 - 5, unlock another consumable.

Edited by Ultimax, 09 June 2017 - 03:45 PM.

[Errinovar]

Jman5, on 09 June 2017 - 01:56 PM, said:

Great post and I agree with most of it. It was nice to see that bit about the heat skills. I was just playing around with it yesterday and came to similar conclusions. Though really, it hardly seems noticeable either way. I haven't messed with your spreadsheet yet, but I look forward to checking it out.

On the subject of Sensor Skill Class I disagree that it's a waste or neglible. I get so much mileage out of Seismic sensor that, it alone almost completely justifies the investment.

Then there is Radar Deprivation. No matter how good you are, you will take LRM damage semi-regularly. Some maps it's worse than others and some games there are more or fewer LRM-boats. However, Radar Deprivation has a practical effect of reducing average damage taken.

ECM skill for ECM mechs is like night and day. When the skill tree first dropped I didn't understand these skills and played a bunch of games without it. It honestly just felt like I didn't even have an ECM. These two skills make a huge difference.

Sensor Range, Target info gather, Target Decay, Target Retention: I can live without these, but they are decent filler skills that bring me down to the skills I really want. Being able to hold a lock long enough to see the paper doll can turn a slug match into a trivial fight. Don't forget they significantly boosted the target info gather times last patch.

My thoughts on the sensor Class is incredibly useful for:

• ECM mechs
  
• Short Range Mechs
  
• Lights
  
• Solo queue

If you're building for competitive, or play with a group that basically slaughters 90% of the teams out there, then it's probably less useful. Long range mechs, it's not a terrible investment, but many skills start to lose their value the further back you fight.


I also think there is an argument for ditching it in Faction Play because of the way Siege Mode is structured.

Pretty much exactly how I feel, except I always wonder why people so undervalue Radar Derp as a flanking device. I mean Radar Derp + ECM is basically invisibility from the mini map unless you are are in LOS. If you know the maps its so easy to flank in a mech that has both, and often you can rout an enemy position or perform a pincher maneuver if your timing is right and you have the fire power. I would argue that the OP is fantastic when you are talking about maxing performance in a face to face brawl, but it ignores other less quantifiable benefits.

[Angel of Annihilation]

I disagree about the importance of agility, in fact aside from heat management skills, it is about the most important trait for your mech to have. I mean why do think most people now think the KDK's are junk mechs when before the Engine Desync they were supposedly godlike and OP? It was because of their speed and agility. That is prehaps a extreme example but I think it underscores just how important speed and agility is for a mech. I am not going to go into details because if your an experienced player I think it goes without saying how important things like being able to stop and reverse back into cover when confronted with overwhelming firepower or race behind a tall building when the LRM storm is on its way truly is. Positioning is only going to get you so far.

[Shifty McSwift]

Well firstly I would point out you can get 80% of the armor specific buffs for 17 points in structure.

And while you may feel that way about mobility, increasing your top speed is strong in all mechs. And some mechs mobility has suffered to the point that ignoring the mobility tree effects your survivability in twisting.

The notion positioning beats mobility is working under the assumption that you will ALWAYS have the best positioning, which just isn't the case.

[Onimusha shin]

ScrubLord1, on 09 June 2017 - 12:18 PM, said:

Some preliminary thoughts here, damage output is a combination of max dps and sustained dps. Your calculations is based on max alphas, which only take into account max dps, and heat containment will always win out here. Cool Run affects sustained dps, and time to return to 0 heat, which Heat Cont. doesn't. While heat containment extends the time you can stay at max dps, once you reach heat cap, it basically does nothing. Aside from that, cool run also has a greater contribution in fights where you don't reach heat cap.

Also in the spreadsheet, the max # of alphas has the same value as time to overheat, think there might be an error in the equation there.

Agreed. Sustained DPS is just as important. But CoolRun is best used in a build with 20DHS or more. And Coolshot consumables are available for flushing out your heatcap. Hence the 5 SPs needed for another 4% CoolRun are better spent on getting Coolshoot skills. And you're heat capped for more than 5 alphas, despite using a Coolshot, either your build is WAY too hot or you're in a bad spot of being bumrushed.

As for the spreadsheet, try downloading it and key in your build heat profile values in the "Calculations" tab. It certainly could use some refinement but I'm busy preparing for a family road trip next week. Sorry about the sloppiness.

Joe Decker, on 09 June 2017 - 12:56 PM, said:

The Goal of the Skilltree is to give you the Ability to play whatever you want. Armor, even Radar Deprivation or Speed Tweak might not be that important if you build a Mech with a different Goal. Use what enhances the Role that your Mech should play. In Teamplay pick the Nodes that improve your Role in the Team.

It should not be possible to generalize the Skill Tree. If that is possible the Skill Tree might need a Rebalance.

Bad news, UAC Jam Chance skills are useless except when stacking on existing Jam Chance reduction quirks. JJ skills are only for flavouring and the SP investment is not for performance but rather a bit more fun. Speed Retention gives you the illusion of choice but doesn't do anything drastic to improve battlefield performance. Sensor skills would be WAY more useful if they weren't so spread out and segregated. As it is, InfoTech can be a huge boon to the battlefield with changes to spotting/lock rewards/bonuses to light 'Mechs but the 13-18 SPs investment is a heavy price to pay for non-existent scouting bonuses.

Jman5, on 09 June 2017 - 01:56 PM, said:

My thoughts on the sensor Class is incredibly useful for:

• ECM mechs
  
• Short Range Mechs
  
• Lights
  
• Solo queue

If you're building for competitive, or play with a group that basically slaughters 90% of the teams out there, then it's probably less useful. Long range mechs, it's not a terrible investment, but many skills start to lose their value the further back you fight.

I also think there is an argument for ditching it in Faction Play because of the way Siege Mode is structured.

You're absolutely right on these. But because PGI has structured the CBill/XP reward system around the concept of dealing the highest damage and kill scores on the battlefield, any SP investment outside of the offensive/defensive skills meta is merely a sacrifice to your performance on the battlefield.

Like Tarogato, I've started taking out the ECM on my ECM-equippable 'Mechs and don't see a major performance hit as long as I play tactical. Furthermore, the base game is ultimately a team PvP so aside from the early scouting, Sensor skills don't play a huge part in the main engagement, only the early-game for setting up and late-game for any hunting of strays. So if I'm sticking to the team most of the time, it's unlikely that the Sensor skills will be in play often.

Edward Scissorhands, on 09 June 2017 - 02:54 PM, said:

Honest Question: Why is there no mention of investment into radar deprivation? I feel that this is an imporant skill. is there a reason others arent investing here?

Also i invest 9 points into mobility into some mechs. i feel the acceleration and deceleration bonus is very good for faster positioning changes

As I've stated in my Skill Tree tier guide here, Radar Deprivation is really only good for awareness (that radar blip thingy). Yes you do break LRM locks faster if you have it. But IMO, it's better to stick to hard cover and make it a habit.

As for mobility, you won't be using that much of it unless sniping from a distance and repositioning or for NASCAR'ing. So those accel/decel skills help a bit but not 80% of your engagement with the enemies, especially if you stick to deathballing with your team.

Viktor Drake, on 09 June 2017 - 05:16 PM, said:

I disagree about the importance of agility, in fact aside from heat management skills, it is about the most important trait for your mech to have. I mean why do think most people now think the KDK's are junk mechs when before the Engine Desync they were supposedly godlike and OP? It was because of their speed and agility. That is prehaps a extreme example but I think it underscores just how important speed and agility is for a mech. I am not going to go into details because if your an experienced player I think it goes without saying how important things like being able to stop and reverse back into cover when confronted with overwhelming firepower or race behind a tall building when the LRM storm is on its way truly is. Positioning is only going to get you so far.

Sorry, positioning is EVERYTHING. Comp teams will tell you that firing lines with great sight lines are way more important than mobility. Movement synergy among team is more important than a few extra 4.5-7.5% kph on top of your max speed. Accel/decel skills ARE useful but on highly agile mechs like the VPR or LCT, do you really need it? If your positioning is good, it DOES reduce your exposure time but superior positioning means you have far less risk when shooting at someone.

Shifty McSwift, on 09 June 2017 - 09:41 PM, said:

Well firstly I would point out you can get 80% of the armor specific buffs for 17 points in structure.

And while you may feel that way about mobility, increasing your top speed is strong in all mechs. And some mechs mobility has suffered to the point that ignoring the mobility tree effects your survivability in twisting.

The notion positioning beats mobility is working under the assumption that you will ALWAYS have the best positioning, which just isn't the case.

I'm not going to beat a dead horse. Mobility is a personal choice but from my experience, positioning is a better strategic tool to use than your mobility. The mobility spreadsheet I shared in the first post will have all your stats based on datamined game files. I've yet to do all the math but often, you won't be able to catch up with the top agile mech of your class if you were already near bottom.

PGI seems to believe that the bottom scrappers of each class are the best performing mechs DPS/durability wise and hence the nerf to mobility post engine desync. They're not entirely wrong though, but my poor KGC cries in its sleep for a mobility buff every night.

[Bluttrunken]

Hmm, no.

[QuantumButler]

Viktor Drake, on 09 June 2017 - 05:16 PM, said:

I disagree about the importance of agility, in fact aside from heat management skills, it is about the most important trait for your mech to have. I mean why do think most people now think the KDK's are junk mechs when before the Engine Desync they were supposedly godlike and OP? It was because of their speed and agility. That is prehaps a extreme example but I think it underscores just how important speed and agility is for a mech. I am not going to go into details because if your an experienced player I think it goes without saying how important things like being able to stop and reverse back into cover when confronted with overwhelming firepower or race behind a tall building when the LRM storm is on its way truly is. Positioning is only going to get you so far.

It's because mechs with atrocious agility, that really need a boost, get negligible returns from investing into it.

+24% to a 7.2kps accel/deccel is pretty much worthless, so investing into agility is pretty much just flushing your points down the toilet.

Now, on mechs that already have decent mobility it can be worth it to take some agility nodes depending on build, but sluggish lumbering beasts like Atlas, Kodiak, and King Krab will never really get enough bonus agility to make those points really worth it.

[Palfatreos]

How you (and Onimusha shin) calculate total heat cap confuses me as far i understand you using this formula : (30 + (trudubs+poordubs) * 1.5) *(1+%heat node)

- trudubs = *2 not *1,5 you can test test against a mech withh 10 trudubs and one with 5 trudubs 5 poordubs your heat % will be high on the latter one.
- heat node works only on the base 30 heat cap not on the heatsink despite what pgi statement.

Source : http://steamcommunit...searchtext=heat
although the values are outdated the formules still accurate. (testing the formula on forest wit erppc and calculating the heat %)

[J0anna]

Onimusha shin, on 10 June 2017 - 12:58 AM, said:

Agreed. Sustained DPS is just as important. But CoolRun is best used in a build with 20DHS or more. And Coolshot consumables are available for flushing out your heatcap.

That makes more sense, I was using a 22 DHS Hellbringer (getting ready for Virago) as my test heat platform, and cool run did seem to make a notable difference.

Overall I tend to agree with your and Tarogato's findings related to Survivability, and have been moving some SP into that tree on most of my QP mechs (Group que mechs may do better with more Firepower - need more data). It's really just building skills like I build mechs - start with max armor, and work down from there.

However, I tend to value mobility a bit more than you. While it's true most assaults can't gain their former agility. A few points in mobility makes a significant change in some heavies and mediums. In lights, I will only put a few points here, to boost acc/deccell - they are agile enough.

What I don't like about the ST, is the amount of time one needs to invest when PGI does their inevitable "balance moves" - change the heat/range on lasers...I'm probably making 20+ ST moves on a few dozen mechs to try to find the proper balance again. I'm dreading June's patch - and haven't even touched my IS mechs as July's new weapons will probably throw most ST's into a drastic change.

What I am finding is that the JJ trees do make a noticeable difference in mechs with 5+ JJ's. I still think it's only for niche builds, but the effect is fun.

Thanks for the work.

[meteorol]

Jman5, on 09 June 2017 - 01:56 PM, said:

Then there is Radar Deprivation. No matter how good you are, you will take LRM damage semi-regularly. Some maps it's worse than others and some games there are more or fewer LRM-boats. However, Radar Deprivation has a practical effect of reducing average damage taken.

While this is true, you will take damage from everything semi-regulary. While radar derp also somewhat helps to hide your movement once you go into cover, as a damage reducer it only works against LRMS, which are among the games worst weapon systems before radar derp even comes into consideration. I have 0 points in sensor on all my mechs and i'm playing 99% solo queue, if group queue in a group of 2. I have not died to LRMS once since the patch.

Putting points into survival increases survability against all kinds of damage sources, while radar derp only helps against lrms. Imo, LRMs are simply too bad as weapons to justify investing into sensors just to reduce their damage.

Jman5, on 09 June 2017 - 01:56 PM, said:

ECM skill for ECM mechs is like night and day. When the skill tree first dropped I didn't understand these skills and played a bunch of games without it. It honestly just felt like I didn't even have an ECM. These two skills make a huge difference.

Also true. Which is why i simply swap ECM for an aditional DHS, even in soloqueue.

QuantumButler, on 10 June 2017 - 04:55 AM, said:

It's because mechs with atrocious agility, that really need a boost, get negligible returns from investing into it.

+24% to a 7.2kps accel/deccel is pretty much worthless, so investing into agility is pretty much just flushing your points down the toilet.

Now, on mechs that already have decent mobility it can be worth it to take some agility nodes depending on build, but sluggish lumbering beasts like Atlas, Kodiak, and King Krab will never really get enough bonus agility to make those points really worth it.

While this is true, people simply seem to forgot what was preached about torso twist speed for years. Whenever someone said "i downgraded the engine, because i don't need the speed", everyone jumped on him, stating that the higher twist speed of huge XL engines are even more important for survability than speed.

Now, all of a sudden after ST and engine changes, people happily ignore like 20% twistspeed for whatever reason, even though the the twist values are considerably higher than the accel values even for sluggish mechs and 20% twistspeed often gives a considerable bonus.

Lets just take the HBK IIC for example. It has ~ 20 base accel, which means it will end up hat like 24 with skill tree, but it has 80 twist speed, which means skill tree can get it up to 100. Even the NTG, which has an atrocious 13 accel, which puts it at 16 with skill tree has 66 torso speed, which ends up at 82.

It seems odd how happily people are skipping torso speed even on mechs that get a good return of it, after the whole forum and reddit preached how important torsotwisting and twistspeed is for surviving like an echochamber. For years.

[Alienized]

good work oni but....
why would anyone actually go with meta when everything else literally works as well?!? lol.
you can do just as fine whatever you do as long as you put the 91 nodes somewhere.

well, propably not if you max out the consumables.....

but then, a bad player just wont be good even with the meta-est tree of all meta's.

and there are far too many people out there that require much more than just a good skill tree.

(GET THIS DAMN MOVEMENT AND MAP CONTROL SORTED -.- !)

[MustrumRidcully]

Viktor Drake, on 09 June 2017 - 05:16 PM, said:

I disagree about the importance of agility, in fact aside from heat management skills, it is about the most important trait for your mech to have. I mean why do think most people now think the KDK's are junk mechs when before the Engine Desync they were supposedly godlike and OP? It was because of their speed and agility. That is prehaps a extreme example but I think it underscores just how important speed and agility is for a mech. I am not going to go into details because if your an experienced player I think it goes without saying how important things like being able to stop and reverse back into cover when confronted with overwhelming firepower or race behind a tall building when the LRM storm is on its way truly is. Positioning is only going to get you so far.

The problem is: IF you buff a a low value by a few percentage points, you get very little out of it. PGI understood this for the armor tree, but they didn't apply the same logic to the mobility tree. (or the range nodes...)

If your mobility is **** before the skills, it will still be **** after.
If your mobility is great, it might be fantastic after spending points on mobility.

The idea that a slow mech with poor twist rates should boost his mobility is sound, but the mobility tree fails to achieve the objective.

Quote

Yes you do break LRM locks faster if you have it. But IMO, it's better to stick to hard cover and make it a habit.

You can't stay perpetually in cover.
You must be able to reposition yourself, or the enemy knows you're pinned down and where and can move his units into position.

And Radar Deprevation helps you there - you can risk leaving cover briefly and move to a new position. Most of the LRms that people might lob your way will probably hit dirt afterwards.

Edited by MustrumRidcully, 11 June 2017 - 01:45 AM.

[LORD ORION]

Onimusha shin, on 08 June 2017 - 09:21 AM, said:

In summary, across multiple build types with constant -9% Heat from Firepower tree, max Cool Run (10%) has the lowest effect of raising your DPS compared to running max Heat Containment (15%) in conjunction with moderate Cool Run (6%) with the figures shown below as reference.

Ummm.... so you showed the best RoI (do my unused skill points earn c-bills for me? )

All at the cost of a Fubar weapon skill tree that you now have points to put into that you argue are pointless to spend elsewhere?

Would it not be better than to just advocate going full heat gen, heat containment and cool run? (52 points)

Also, you can take one reinforced casing out (by the two nodes on the right side) of the tree and save a point. (32 points instead of 33)

That would be 84 for full heat management, armor and structure on hot mechs

Still leaving some spares for whatever you want in Miscellaneous.

Here, I made the image

Edited by LORD ORION, 11 June 2017 - 01:58 AM.

[Onimusha shin]

LORD ORION, on 11 June 2017 - 01:54 AM, said:

Ummm.... so you showed the best RoI (do my unused skill points earn c-bills for me? )

Well, earnings in this game is mostly decided by your dmg dealt so, just measuring the DPS improvement given by the SPs invested

LORD ORION, on 11 June 2017 - 01:54 AM, said:

Would it not be better than to just advocate going full heat gen, heat containment and cool run? (52 points)

Also, you can take one reinforced casing out (by the two nodes on the right side) of the tree and save a point. (32 points instead of 33)

Yeah, you could actually. Any balance leftover you have would be better used for enhancing your Firepower tree first (build specific stuff) then maybe Mobility if you feel REALLY sluggish (caveat: using Mobility skills on slow assaults like DWF, MAL, AS7, KGC, KDK rarely helps).

And tks for pointing out the excess SP spent. Now I have to edit again!!! =___=|||

[GenghisJr]

ThoseWhoFearTomorrow, on 08 June 2017 - 11:35 AM, said:

Have never used coolshot and never will.

**** up and drink the cool-aid

[Onimusha shin]

Well, with Coolshot looking like it'll be nerfed soon, might be better to divert the SPs for it over to strikes instead.

Can anyone say..... StrikeWarrior:Online?