65 replies to this topic

[Felicitatem Parco]

LordNothing, on 07 February 2017 - 01:35 AM, said:

as usual they are not interested in feedback, only stats.

while it might be interesting to see what becomes of it, i have some serious doubts that the refunds will be worth the upgrades. geting 125 formetly mastered mechs back to the peak of perfection is going to suck.

Another MechWarrior who supports the old skill tree.

Sorry, if you consider your Mechs to be currently "Mastered" and you do not want to level them again, that means you want to keep the current skill system.

Gas Guzzler, on 07 February 2017 - 09:26 AM, said:

Eh, I didnt think there was a need to overhaul the whole thing, just needed to fix that pinpoint skill. Content for the sake of content should not be encouraged. I could honestly do without it, but since we are getting it I am left concerned with any negative consequences of it.

And another.

Sorry guys - the current tree is crap and just nothing but a way to make your Mechs "better" without any kind of customization or thought required.

Edited by Prosperity Park, 07 February 2017 - 01:49 PM.

[Jman5]

If it's anything like previous PTS expect most of the community to dislike everything about it after briefly skimming part of the patch notes.

[MechaBattler]

Aren't we getting the experience back though with the new skill tree? So you don't really need to re-master them?

The only problem I see is they're assuming we all bought modules for every mech to justify having a c-bill cost for ever skill. I have maybe 4 modules total. Comes to around 12 million c-bills. If each mech skill set is going to cost me as much as the potential modules it would have had. Then anyone who didn't buy modules is going to have to get back to grinding.

[Gas Guzzler]

Prosperity Park, on 07 February 2017 - 01:45 PM, said:

And another.

Sorry guys - the current tree is crap and just nothing but a way to make your Mechs "better" without any kind of customization or thought required.

You're assuming there will actually be meaningful tradeoffs in the skill tree. I expect there to be some clearly superior "trees", with others that only get filled when mechs have a bunch of extra skill points (if they choose to balance things that way, that is).

I'm just concerned the overall state of balance isn't going to be very good.

Jman5, on 07 February 2017 - 02:48 PM, said:

If it's anything like previous PTS expect most of the community to dislike everything about it after briefly skimming part of the patch notes.

I think that is an unfair evaluation of the last PTS, knowing that you were pro-Energy Draw and presumably aren't happy that it got canned. The fact is there were people on both sides of the fence that didn't try it. I know I did...

Edited by Gas Guzzler, 07 February 2017 - 02:54 PM.

[Ultimax]

Gas Guzzler, on 07 February 2017 - 09:26 AM, said:

Eh, I didnt think there was a need to overhaul the whole thing, just needed to fix that pinpoint skill. Content for the sake of content should not be encouraged. I could honestly do without it, but since we are getting it I am left concerned with any negative consequences of it.

Honestly this is one of those things that I wish didn't exist at all.

I doubt many players actually feel a sense of progress or achievement with skilling out a mech - and certainly not 3x. It's just a grind.



I'd rather there were no skills, and it was just baked into the mechs themselves. (This would also improve NPE).

Time spent in a mech could just grant an XP style bar that gives you achievements, unlocks for that mech (decals, one-shot paint, etc. Goodie bag type stuff).


Instead were going to get something to satisfy the immersion crowd that the rest of us will find a way to min max and frustrate the immersion crowd.

They're under the illusion they can "make their mech their way!" - you can do that now with weapons but its very clear when your way is badly designed.

The same will happen with skills, it will be just another layer on mech building that some people will be bad at - except this time it will be expense to fix.

Edited by Ultimax, 07 February 2017 - 03:09 PM.

[Andi Nagasia]

Gas Guzzler, on 07 February 2017 - 02:54 PM, said:

You're assuming there will actually be meaningful tradeoffs in the skill tree. I expect there to be some clearly superior "trees", with others that only get filled when mechs have a bunch of extra skill points (if they choose to balance things that way, that is).

I'm just concerned the overall state of balance isn't going to be very good.

I think that is an unfair evaluation of the last PTS, knowing that you were pro-Energy Draw and presumably aren't happy that it got canned. The fact is there were people on both sides of the fence that didn't try it. I know I did...

we all know the weapon trees will be a heavy Focus of most people,
the Question is if the rest of the Trees will be balanced against them to make hard Choices,

how much Armor or Structure would you have to get to make it a Viable choice over a -20% PPC heat Quirk,
how will this be balanced, if its done right it could be amazing, if not then people will just choose the Best Quirks,

[Bud Crue]

As with the last PTS runs, I'll play the PTS at least daily for as long as it runs. I will give my honest feedback in the forum room that they presumably set up for such feed back. Then, if history is a guide, the skills tree will be pulled and never heard from again. Just like ED. Just like Infotech and ghost damage. That is what I expect to happen.

From the point of view of the scientific method or ANY testing criteria you care to name, two weeks (or so it would seem if Russ's tweets are to be believed) is simply not enough time to get a baseline set of values and then make comparative analysis of multiple variables across all aspects of this game. If the skills tree is as large and complex as what they presented at mechcon that alone has far more possible combinations and potential influences as to make testing it in that limited space of time a mathematical impossibility. In a lab you might test a few variables in that time. PGI is apparently going to test all aspects of the skills tree and all aspects of the requirked "baseline" assisted mechs, simultaneously.

Not only is it impossible to "test" all of those variables simultaneously it is a fact that any data they get from such a test is going to be scientifically or even comparatively useless. That isn't an opinion, that isn't a bias. It is a consequence of the testing schema and there is nothing to be done for it, other than to slow down and test iteratively and that would take WAY, WAY longer than two weeks.

Edited by Bud Crue, 07 February 2017 - 03:17 PM.

[Felicitatem Parco]

Bud Crue, on 07 February 2017 - 03:14 PM, said:

As with the last PTS runs, I'll play the PTS at least daily for as long as it runs. I will give my honest feedback in the forum room that they presumably set up for such feed back. Then, if history is a guide, the skills tree will be pulled and never heard from again. Just like ED. Just like Infotech and ghost damage. That is what I expect to happen.

From the point of view of the scientific method or ANY testing criteria you care to name, two weeks (or so it would seem if Russ's tweets are to be believed) is simply not enough time to get a baseline set of values and then make comparative analysis if multiple variables across all aspects of this game. If the skills tree is as large and complex as what they presented at mechcon that alone has far more possible combinations and potential influences as to make testing it in that limited space of time a mathematical impossibility. In a lab you might test a few variables in that time. PGI is apparently going to test all aspects of the skills tree and all aspects of the requirked "baseline" assisted mechs, simultaneously.

Not only is it impossible to "test" all of those variables simultaneously it is a fact that any data they get from such a test is going to be scientifically or even comparatively useless. That isn't an opinion, that isn't a bias. It is a consequence of the testing schema and there is nothing to be done for it, other than to slow down and test iteratively and that would take WAY, WAY longer than two weeks.

I think the very first "testing" will simply be seeing where people first dump their points into. That will give PGI a good idea of what we consider to be the best or most desirable quirks.

For instance, if everybody goes Heat Reduction or Goes Home, then PGI will likely nerf that in order of make other quirks seem more worth their weight in Nodes. The opposite with Skill branches that nobody invests a single point into.

[Andi Nagasia]

Ultimax, on 07 February 2017 - 03:09 PM, said:

Honestly this is one of those things that I wish didn't exist at all.

I doubt many players actually feel a sense of progress or achievement with skilling out a mech - and certainly not 3x. It's just a grind.

I'd rather there were no skills, and it was just baked into the mechs themselves. (This would also improve NPE).

Time spent in a mech could just grant an XP style bar that gives you achievements, unlocks for that mech (decals, one-shot paint, etc. Goodie bag type stuff).


Instead were going to get something to satisfy the immersion crowd that the rest of us will find a way to min max and frustrate the immersion crowd.

They're under the illusion they can "make their mech their way!" - you can do that now with weapons but its very clear when your way is badly designed.

The same will happen with skills, it will be just another layer on mech building that some people will be bad at - except this time it will be expense to fix.

all this would really mean is instead of everyone taking a AS7 for Amazing Tanky ness(Armor & Structure)
you can now make your BNC just as Tanky, or your AWS, ect ect, people will be able to make their favorates amazing,

personally i love the LCT, and im looking forward to making it Amazingly Tanky,
or even perhaps a Joke Build with a COM with a LRM5 with -50% cooldown,

Edited by Andi Nagasia, 07 February 2017 - 03:22 PM.

[Sjorpha]

BLOOD WOLF, on 07 February 2017 - 06:48 AM, said:

lets be certain we are getting things right. They hold feedback subservient to Objective stats. How does anyone separate the feedback that is actually accurate from the obvious B.S that I know some people are going to try to pull via various fallacies? You compare the statements to the Statistics, like any rational person would.

Feedback might be 100% inaccurate but the stats are the raw numbers from the data that shows how things performed. The only drawback to looking at stats only is you might not get the context for why. An example would be players going into the PTS and sabotaging the numbers to prove a point, those extreme ends of the data get mixed in and might be misconstrued to be a real scenario

The most important thing about a game feature is that people enjoy it, that doesn't show up in any stats.

Subjective feedback about how people experience the feature is the actual relevant data for knowing if the feature is good, because good means as many players as possible liking it and nothing else.

The relevant questions are: Is the skilltree fun to play with? Does the skilltree feel well done and intuitive to use? Does the skilltree seem reasonably fair in terms of balance?

Balance is where stats can be used to some degree, if you can get competitive players to play enough matches, but that's very unlikely given the short time and likely miniscule population with too much skill variation. Therefore even for the balance question the impression of competent players are likely to be more accurate than any stats generated in this kind of test.

[Bud Crue]

Prosperity Park, on 07 February 2017 - 03:19 PM, said:

I think the very first "testing" will simply be seeing where people first dump their points into. That will give PGI a good idea of what we consider to be the best or most desirable quirks.

For instance, if everybody goes Heat Reduction or Goes Home, then PGI will likely nerf that in order of make other quirks seem more worth their weight in Nodes. The opposite with Skill branches that nobody invests a single point into.

No doubt, and if we were just looking at filling up a row or not filling up a row (in this case heat reduction quirks) versus one other row, and then compare population behavior across all mechs, and how those two comparative values alone affected the game, two weeks might be doable. Maybe. But the skills tree appears to have far more than just two choices (obviously) and we have several hundred distinct mech variants that are going to be influenced by those choices. Two weeks sound reasonable to get any useful data in that environment?

I just don't see it. Nor can I see how this PTS is going to provide much useful data for them.

That is why I keep coming back to the conclusion that this PTS is just being done for forms sake and the conclusion that they want is predetermined.

[Andi Nagasia]

Prosperity Park, on 07 February 2017 - 03:19 PM, said:

I think the very first "testing" will simply be seeing where people first dump their points into. That will give PGI a good idea of what we consider to be the best or most desirable quirks.

For instance, if everybody goes Heat Reduction or Goes Home, then PGI will likely nerf that in order of make other quirks seem more worth their weight in Nodes. The opposite with Skill branches that nobody invests a single point into.
\

im hoping that all quirks will have equal use, but i dont want to see any weapon quirks under 10%,
i would rather make all the other Quirks usefull, in relation to those weapon Quirks, that what im hoping for,

we also dont know how points will be spent,
will it be 1 to 1, 1 point for ever 1 skill or skill advancement?
or will it increase per point spent in a tree, 1st skill is 1 point 2nd is 2 ect,
also will it be the same per skill tree, some trees may cost more to invest into,

[Gas Guzzler]

Andi Nagasia, on 07 February 2017 - 03:22 PM, said:

all this would really mean is instead of everyone taking a AS7 for Amazing Tanky ness(Armor & Structure)
you can now make your BNC just as Tanky, or your AWS, ect ect, people will be able to make their favorates amazing,

personally i love the LCT, and im looking forward to making it Amazingly Tanky,
or even perhaps a Joke Build with a COM with a LRM5 with -50% cooldown,

If a BNC is allowed to take as much structure as an Atlas, then balance is doomed and my greatest fears will be realized.

Prosperity Park, on 07 February 2017 - 03:19 PM, said:

I think the very first "testing" will simply be seeing where people first dump their points into. That will give PGI a good idea of what we consider to be the best or most desirable quirks.

For instance, if everybody goes Heat Reduction or Goes Home, then PGI will likely nerf that in order of make other quirks seem more worth their weight in Nodes. The opposite with Skill branches that nobody invests a single point into.

Based on early testing? That would be very foolish. What people think is good does not necessarily mean that is what IS good.

[Felicitatem Parco]

Gas Guzzler, on 07 February 2017 - 03:31 PM, said:

Based on early testing? That would be very foolish. What people think is good does not necessarily mean that is what IS good.

I guess my emphasis would be on "desirable"

It would take a month or two for us to really know what's best across 200 Mechs.

Edited by Prosperity Park, 07 February 2017 - 03:44 PM.

[Felicitatem Parco]

Actually.. I think the best data will be based on how often we ReSpec and determine that a skill is not worth it...

The Skills that are dumped the most might get some love. We all know what we're gonna try first, but we don't know if it will be actually good enough to keep. If we find a skill sucks and we dump it, then there's a good indicator of value.

[AnTi90d]

Bud Crue, on 07 February 2017 - 03:14 PM, said:

As with the last PTS runs, I'll play the PTS at least daily for as long as it runs. I will give my honest feedback in the forum room that they presumably set up for such feed back. Then, if history is a guide, the skills tree will be pulled and never heard from again. Just like ED. Just like Infotech and ghost damage. That is what I expect to happen.

I sincerely hope that PGI's developers can shove their egos where the sun doesn't shine and take constructive feedback for once in their lives. I truly do want to see some form of the new skilltree make it to the live servers.

What I really want out of all of this is to not have to spend 10+ minutes juggling modules in PGI's crappy interface that shoves you back to the Home screen nearly every time you hit Save instead of leaving you in the Mechlab.. just to try out a new combination of mechs for a dropdeck or to fill tonnage in a group QP match.. all while maintaining some resemblance of balance.

My current apprehension lies with PGI's track record of constant failure throughout 2016.. from the Minimap and its icons getting nerfed.. to the rescale totally screwing over some mechs like the Firestarter.. to all of CW/FPv3.. Frozen City turned from a fun brawling map into some abominable snipe fest.. any of numerous iterations of Alpine Domination.. et cetera.

I'm really holding my breath on how many base level quirks crap-mechs like the Victor/Awesome/Vindicator will get to keep.. how many weapon quirks the Clanners, especially the Kodiak 3, will be able to have.. how much structure/armor/speed they're going to strap on top a damned Locust.. and how they're going to maintain FP balance with two different tech bases.

(..and that's not even mentioning the tech update that's coming with the timeline advance.. and all the respeccing fees and releveling that will be required of people who just want to try out the new weapons.)

[Bud Crue]

AnTi90d, on 07 February 2017 - 04:01 PM, said:

Hope...

Yep. Me too. There are just so many variables (skills, underlying quirks, need for 3 being eliminated, respecing as a presumed revenue stream, etc.) at play with this PTS to give me any confidence that PGI has thought ALL of their impacts on one another through, or that a 2 week test of all this stuff together...at the same time...will be remotely sufficeint to give them any useful data on how all those variables are interacting an affecting one another; let alone how they might be affecting underlying mechanics of the game or player motivations/performance. Add in the historical conduct of PGI when they change merely one or two things at once (flammers, gauss, Minimap, ECM, etc.) and it is enough to convince me that what they are attempting here (with a PTS or not) is simply beyond their abilities.

I sincerely hope I am wrong, cuz the thought of making all my Quickdraws truly unique appeals to me greatly, but as you say, just looking past over the last year and my fears feel justified.

Think about it this way: How long did they have to analyze their "data" to finally come up with a way to bring the Kodiak-3 to heel (a bit)? How much ancillary damage was done to other mechs as part of that effort? Now consider what they are attempting to do with the skills tree...in two weeks time.

What could possibly go wrong?

[LordNothing]

Prosperity Park, on 07 February 2017 - 01:45 PM, said:

Another MechWarrior who supports the old skill tree.

Sorry, if you consider your Mechs to be currently "Mastered" and you do not want to level them again, that means you want to keep the current skill system.



And another.

Sorry guys - the current tree is crap and just nothing but a way to make your Mechs "better" without any kind of customization or thought required.

im not totally opposed to the new trees, the benefits may yet outweigh the inconvenience. and thousands of wasted man hours is quite an inconvenience (and i very much doubt the refunds will be representative of time invested as their value has changed). but it may be worth it on those mechs that can only have one iconic loadout, though i have a feeling its going to hit mixed builds hard.

no the thing that really bothers me is that mastering all the things was kind of an exit point for me. new system you can spec and respec as many times as you want, and so there is no point of completion. so its just another nail in the lid of the skinner box.

Edited by LordNothing, 07 February 2017 - 04:21 PM.

[FupDup]

For the whole topic of Banshees getting the same structure buffs as the Atlas or whatever else, my idea for that would be a "skill multiplier" set per mech. Basically, each of the skills in the tree comes with a low default value, and then this value is increased based on your specific chassis/variant multiplier.

So, that Banshee might only get 5% structure at the lowest level of the structure skill tree, while an Atlas might get 10% to 15% for the same skill point cost.

This would allow for a mostly global skill tree while still giving the poppy gundams more benefit than Kodiaks.

Edited by FupDup, 07 February 2017 - 04:23 PM.

[Jesusguy]

Personally I am hoping for $50k range for the skills or maybe even tiered, the better the skill the higher the cost.

One question i haven't seen answered (although i haven't looked to hard) my timberwolf has close to 500k XP it gets to keep that right? it's not just lost to the cosmos because i didn't spend $$ on making it general XP.