General class balance observations

"

Snorkle_uk wrote:

there is a problem with the shadow, its a shit start, you are right but you dont know why you are right. The reason is because the first 5 nodes are all damage no matter what way you chose to go. In highly efficient builds you usually only take damage when it leads to a notable in the majority of cases, where with life you take small life nodes regardless of a notable being there, you take every life node you can reach. That means if you start shadow you probably take all the life nodes at the ranger start, but if you start ranger you will rarely take all the damage nodes in the shadow start because you will get better damage from a cluster containing a notable elsewhere on the tree. This makes the ranger more efficient life start in 95%+ of cases.

Nailed.

Let me quote a reply from before 2.0 to someone asking about a build I'd made as a witch (for 2.0), and my advice on playing it as shadow.

"

TheAnuhart wrote:

You can start as Scion or Shadow but you'd still want to go to witch start.

The Shadow 5 start points are not useless, as are not the 5 (including jewel soc) Scion start points, but they are not points I'd take as a witch (at least not the shadow 5), ever, which basically means they are suboptimal. You can look at the Shadow's 5 starters as replacing the 3 15% wand nodes in the fusilade circle and the 2 sniper nodes in shadow.

This isn't too bad a loss in numbers, losing 75% dmg and 10% proj speed, gaining 56% dmg and 10 ES, basically 19% increased dmg loss which you could justify simply by wanting to play/respec a shadow you have levelled already.

The problem is, those 5 damage nodes for the witch are 'floaters', nodes that you take and drop, take and drop, as you level. When 15% damage may be worth taking on level up, but next level 2 points gets you life, or crit, or aura reserve. Pathing, aura reserve, life, crit, accuracy, come way before pure % damage nodes. As a shadow you lose that flexibility and are locked into those 5 'floaters' from the start.

Now, a caveat, GG high rolls of mana and mana regen everywhere you can on gear (without losing out on crit, life, accuracy, resists, flat/WED) could well mean you can completely lose the witch start nodes. This would then mean your shadow starter 5 are all free, extra, OpieOPness.

"

Snorkle_uk wrote:

there is a problem with the shadow, its a shit start, you are right but you dont know why you are right.

You're right, I don't know why. But I do know how. I trust the results of top players, collected in aggregate, rather than thinking I'm the end-all and be-all of PoE knowledge. So naturally, I find your attitude of (essentially) "fuck statistics, I know everything I need to know" to be, essentially, hubris.

Your feedback on the shadow area (and, to a lesser extent, Scion) is helpful. I still maintain, however, that the Templar and Scion's issues are more pressing. Because numbers.

I hope this thread doesn't become monopolized by any one voice, but instead a group conversation of experienced players handling class balance issues. Which is actually where it seems to be going, so that's good.

"

ScrotieMcB wrote:

"

Snorkle_uk wrote:

there is a problem with the shadow, its a shit start, you are right but you dont know why you are right.

You're right, I don't know why. But I do know how. I trust the results of top players, collected in aggregate, rather than thinking I'm the end-all and be-all of PoE knowledge. So naturally, I find your attitude of (essentially) "fuck statistics, I know everything I need to know" to be, essentially, hubris.

Your feedback on the shadow area (and, to a lesser extent, Scion) is helpful. I still maintain, however, that the Templar and Scion's issues are more pressing. Because numbers.

I hope this thread doesn't become monopolized by any one voice, but instead a group conversation of experienced players handling class balance issues. Which is actually where it seems to be going, so that's good.

Scion's is pretty simple due to the fact jewels give you massive variety in your build and the best build to play rainbownuke (anything Mjolner), low life builds, wanders, Incinerator/RF hyrbids, CI high end physical damage build, Poison Arrow, shatterchuck/coc buids.

Which ironically above are extremely popular builds for either budget (Poison arrow, incinerator) or very high end item value builds. (Basically anything that use T1 items and most of T2 scion uses)

Thus the most played more compared to other classes. (In high ranking wise). The highest ranking characters are more likely to be geared out and played with high end builds, thus the class that plays the highest end build the best is picked more often if it is not a budget build.

Shadow was the most popular last due because everyone played high end crit builds with shavs.

Marauder is most popular in HC because the most popular build is pure-life ultra tanky 2h cyclone. No other class and do this well as marauder, thus everyone that wants to go 2nd cyclone will be play marauder.

Attempting to give you prospective outside of just raw numbers and more reasoning behind things. Understanding the stockmarket and how the number works is one thing, but predicting the numbers and knowing why is how you get rich.

"

ScrotieMcB wrote:

2. Nerf Scion and/or jewels. Scion is pretty much the strongest class at the moment, and the class's copious access to jewel sockets is why. In perm leagues, jewel inflation will ensure the class just keeps getting more and more powerful. I don't think being a "jewel specialist" class is a bad concept, and I like that jewels can be pretty darn powerful, but it's just a tad much at the moment. If jewels need nerfing and the whole "legacy version" garbage, better now than later.

I would rather do the very opposite.

Increase the number of jewel sockets for the other classes.

Jewels are a nice mechanic. If GGG really adds unique jewels that change specific skills, as they mentioned a while ago, jewels would be a great mechancic. Giving players more access to them would be fun.

If Scion currently has more access to jewel sockets than everyone else, increase a bit the number of sockets for the other classes. Not as much as the Scion has, to keep the "jewel specialist" theme, but enough to they can be used more often, and so the ones with a radius can be used in more places.

"

ScrotieMcB wrote:

"

Snorkle_uk wrote:

there is a problem with the shadow, its a shit start, you are right but you dont know why you are right.

You're right, I don't know why. But I do know how. I trust the results of top players, collected in aggregate, rather than thinking I'm the end-all and be-all of PoE knowledge. So naturally, I find your attitude of (essentially) "fuck statistics, I know everything I need to know" to be, essentially, hubris.

but you are evaluating classes base on use of classes, when use of classes is something effected by what skills are currently good, what defenses are currently good, what items are easy to get, what popular streamers are doing... many factors that have nothing to do with the actual class and if its nodes need changed or not. So yes you know the data, you dont know how to read the data because you dont know enough about the game, so u presume, and presume wrongly.

For example, very few people played armour before, so armour needed changed based on stats? Well everyones playing armour now, and its worse than it was compared to the content people are facing than when no one was playing it, it hasnt really changed at all in any positive way. So to draw a conclusion about how armour works from people either playing it or not playing it would be an incorrect way to evaluate those statistics. People playing or not playing armour has to do with whats going on with skill gems, with items, with immortal call, with light coil, with virtually everything other than armour itself.

Templar was one of the most popular classes in the game, Duelist was regarded as the worst class in the game, based on server stats. Neither characters starts have changed since then, and now you are saying the opposite is true and we need to change the starts to solve it.


So actually, your presumption you know how to read the data you are looking at is hubris, and the reality is that I am actually right. So be it hubris or not on my part what I said is actually true. Fine, Im an egotistical arsehole, yep, hands up, thats me no denying it. But what I said is true.

"

CharanJaydemyr wrote:

"

Ladderjack wrote:

This thread really illustrates how the community has tunnel-vision on the meta and cannot back off from that "urge to win" enough to view the game system holistically.

Yeah but who taught them to be that way, Jack?

Cluck, cluck, crack, crack.

Then, explain me. Even if you don't remember me ghosting around these forums a few years ago, you can see my titles. . .and I know you, sire. ;) Yet, long-term citizen of Wraeclast that I am, I have never even looked at the list of uniques. I eschew net builds. I piss on the meta. I choose to experience the game organically. . .and I love this approach. I've been here since almost the very beginning. . .and I appear to be neither chicken nor egg.

I think it really comes down to motives. Most people use video games as a means to an end: a vehicle for the thrill of victory. And with those players, the game is ancillary. . .only victory is important. So they prostitute these amazing and rich game systems and turn them into a dray horse to carry them to their goal. Then, three weeks later, the alpha nerd in their group finds Hearthstone or LOL and it starts all over again: "how can I become godly the fastest in my gaming group and become top nerd?"

The fact that GGG can please those guys and freaks like me is a true testament to their ability.

"

ScrotieMcB wrote:

I still maintain, however, that the Templar and Scion's issues are more pressing. Because numbers.

The problem with the Templar appears very early, in Act 2 Normal. His core area starts off strong, with life regen, +life, ele damage, AoE, and Endu Charge nodes. But once you fill out those clusters around level-20, you find yourself marooned with few good prospects nearby:

http://www.pathofexile.com/passive-skill-tree/AAAAAwUATwRDyFugp1V8Dmjy8B-iAASz4XPAZu0gPAWCm-987Dg62CcvAZGQVS3SGjg=

The routes out of the Templar area are all pretty long-legged, with little in the way of low-hanging fruit. The Templar's outer perimeter is very situational, geared for aura builds (now deprecated), totem, and specific melee types. Marauder core area is six nodes down. Scion wheel of life is seven nodes away. The middle of the Witch area is eight nodes distant. These long spokes slow down your development right around the time you need to power through Act 2 and the first half of Act 3 in Normal. As a consequence, your build is a half-dozen points behind when you take on Dominus in the Sceptre of God. No problem if you're fully twinked out with leveling gear, but for races and rebuilding from scratch the Templar can be a slow-paced start.

"

Qarl and Rory, and Chris because although he isnt involved in balance he is lead dev and has a big say, these 3 guys need a time out. They need to be banned from the office for 4 months, game development taken down to maintenance levels. Their new job is playing the game 10 hours a day 6 days a week in a live league, pref softcore so they dont rip and end up repeating 1-80 again and again and learning fuck all, they need to grind to the mid 90s on many different chars, get a few 100 ex from trade, craft some items, map a shitload for an entire league. Thats how you fix this game, if those guys did this then they wouldnt need us to tell them whats up, they would see it all for themselves and then fix the problems.

This comment amused me, but it's also has a certain appeal. Between 1.1, Vaal, 1.2, Masters, 1.3, Did this have a name?, and 2.0, Awakening. The game has seen a massive amount of change. I'm sure GGG wants to maintain "momentum", but I don't think a season of stability where the devs and designers play their own game on live servers would be a bad thing.

probably they cannot afford (literally) to do a 'dull' season.

poe - probably, i only project other f2p moo's situation - relies on NEW players. big whales with long list of titles are important but in the grand scheme they probably are responsible for less than 20% of total income.

new players wont come to a game attracted by 'we are starting Dull Season, that will last 4 months and nothing exciting is around the corner'. that wont happen. how do you do 'hype/press/streamers' marketing campaign for season like that?

so - for good or bad - expect changes, expect new stuff, expect each new patch to have something 'groundbreaking' (even if it only means big words, noone verifies pre-release press articles)


on topic:
templar situation is a) slow windup (described below. that 5+ stretch of 'boring' nodes is just that - boring) b) result of EXTREME templar over-population in previous leagues (firetrap/prolif/flameblast etc). people that played these builds are probably burnt out (pun intended) and would rather play cow simulator than yet another fire/templar build

duelist got a lot of good nodes in 2.0, same with area between marauder/duelist. good incentive to start duelist vs starting mara in 1.3

shadow is bit dps-heavy. that wasnt a problem in 1.3, it is in 2.0. well. maybe this is by design


but the most important part - numbers are there to be read. and read carefully. ignoring fotm builds (people aiming at the top ladder spots will not play Elemental Hit marauders), fotm uniques and fotm skills in general means that conclusions you draw from the data are an example of going too far

"

Ulrichvonbeck wrote:

"

Qarl and Rory, and Chris because although he isnt involved in balance he is lead dev and has a big say, these 3 guys need a time out. They need to be banned from the office for 4 months, game development taken down to maintenance levels. Their new job is playing the game 10 hours a day 6 days a week in a live league, pref softcore so they dont rip and end up repeating 1-80 again and again and learning fuck all, they need to grind to the mid 90s on many different chars, get a few 100 ex from trade, craft some items, map a shitload for an entire league. Thats how you fix this game, if those guys did this then they wouldnt need us to tell them whats up, they would see it all for themselves and then fix the problems.

This comment amused me, but it's also has a certain appeal. Between 1.1, Vaal, 1.2, Masters, 1.3, Did this have a name?, and 2.0, Awakening. The game has seen a massive amount of change. I'm sure GGG wants to maintain "momentum", but I don't think a season of stability where the devs and designers play their own game on live servers would be a bad thing.

2.0.4 The Learning? The Reality? The Rude Awakening?

-_-