Elric Galad

Trickster / Beguiler is in my opinion a very fun build. - Melee (on the upper end of DPS), - High Defensive potential from illusion spells, and you might add Borrowed Instinct to stack even more (active) defenses, - A lot of abilities, very active character, - Very open building possibilities (you don't have to select abilities from every Cipher PL to spend your focus pool, and you already have a lots of guile-spender, so feel free to select what you like, and respec if necessary), - Subclasses are Thematically consistent and fit very well a tricky pirate.

Well, that's because I'm implicitly comparing the endgame of Multiclass vs endgame of Single Class. That's why I'm so concerned about a single ability. A lots of powergamers are more concerned about endgame efficiency ^^ I won't pretend it is rational. Single Class characters also gains abilities a lot sooner. So even if Toxic Strike is a secondary argument for the build, getting it a lot sooner would provide a lots of benefits to the Single Class midgame.

Yup, but the whole point of Vanishing Strike is to increase damages for a short time. (while being semi-invincible as a "side effect"). The problem in that case is that this period is SO short that increasing it a bit through INT becomes critical if you don't want the burst duration to become too short. Doing damages "fast" is not specific enough if one doesn't specify a duration. A Vanishing Strike build aims to optimize DPS over : 3 x 10s = 30s (the number of Vanishing Strike which can be cast in 1 encounter, not counting Empower) (estimate duration of invisibility per cast with high INT/PL) With more MIG and less INT, you might be able to do a bit more damages on a shorter duration, but the duration would be far shorter (because INT is basically 5% multiplicative) and the increased DPS won't be that great (because MIG is basically 3% additive with the crapload of bonus damages that Rogues already get). Then as Boeroer said, high INT is also highly beneficial for Toxic Strike which is a "backup mode" of this build (or possibly its alpha strike). This build doesn't aim to maximize the steady DPS of a Rogue, because I think there are better builds than Single Class Rogue for this purpose (Such as Mindstalker).

To be fair, the thread initial question was specifically about Single Class Rogue, this is the reason why I'm so focused on Vanishing Strike. Toxic Strike was referred as a secondary effect. In order to specifically max Toxic Strike effects, it is clear that INT is incredibly important. It is probably even worth to go Monk Multi in order to max INT to maximize toxic strike damages. As a result, IMHO : 1) For passive DPS : DEX=PER>>MIG 2) For usual DoT : MIG=INT>PER (because dealing damage faster is as important as dealing more damages. If only damages are considered INT>MIG>PER. Using AoE makes INT even more interesting.) 3) For Toxic Strike : INT>>PER>MIG (as long as you hit, the exponential nature of Toxic Strike damages will benefit much from INT) 4) For Gouging Strike : MIG>PER, INT doesn't matter 5) For Vanishing Strike burst : INT>DEX>PER>>MIG So, for a Single Class Rogue, given that my priorities would be 5 then 1 or 3, my stat choice would be INT=DEX>PER. Given the protection provided by invisibility, I would be able to dump a bit CON and RES, so I would have enough points for a bit of PER.

Salvation of Time from an ally would be another good way to increase invisibility duration by the way.

Toxic Strike was a secondary reason though. The main reason was the duration of Vanishing Strike invisibility which is a basically a short duration god mode. Assassination + backstab add + 25Per, +4Pen, +100% Damages and +50% critical damages so any way to increase the duration would provide a lot of benefit to the spike DPS which is the reason for a Single Class Rogue build. Maxing Dex is anyway a priority if one wants to max the number of strikes while invisibile. So this is about Per vs Int. Or maybe maxing both while dumping a bit other stats. For a "normal multiclass rogue", I wouldn't prioritarize Int over Per.

But what about building around vanishing strike duration ? Would you consider it a valid strategy ?

Chuck Norris ranger's pet is his beard. Chuck Norris beard is able to bite Eder if he tries to pet it.

Chuck Norris can imbue roundhouse kick with Meteor Swarm.

Pfff No one can beat Ranger. Cause it's Chuck Norris class.

The main advantage for me of Single Class Rogue is Vanishing Strike since it enables backstab and assassination while invisible for its duration. Therefore would you consider Max INT for this build ? Mig is relatively unsignificant for endgame rogue. Con/Res is not the point of such build. Per might not be that important due to assassination already great Acc Bonus. Dex is a very important stat, but this leaves room for at least another maxed stat. So why not Dex=Int>Per for stat priority for such a build ? In addition, Int is highly beneficial for Toxic Strike, which can be optimised too on a Single Class Rogue. So, on paper, it sounds good...

Pure class arcane archer seems also tempting by the way. Double Death Ring per attack with driving flight, with good range, foe-only damages and sometimes insta-death seems a pretty fast and powerful way to deal AoE damages.

Critical acclaim. Sales failure. Obsidian managed to provide a better heir to Planescape : Torment than InXile did This is an excellent news and well deserved !

Multiclassing with casters is the point of the class. Not sure if the regeneration is strong enough though.

Althought I do think that Balance between classes and races is quite good in PoE1, I couldn't prevent myself from thinking that it could have been even better with a couple of changes. It is not very likely that patches would still be released, so Mods are the only way now. The good thing with mod is that the authors are free to make more radical changes since there is no risk to disappoint or frustrate casual gamers or anybody else. To be honnest, I don't plan to create the mod in a near future due to personnal constraints, but this is an idea I would keep on my list of “things to do in life”. General Principles : So let's go for the detailed list of changes. Feel free to make comments if you have other ideas or if you find some stuff weird or OP. Also tell me if something isn't clear or if you have comment about the post readability. I – Race Balance : II – Resolve & Constitution Balance : III – Rogue Balance : IV – Per Rest Suppression : a) Vancian Casting : b) Other Class Changes : V – Various Changes :

Roses are red, Violets are blue, Rogues are rouge. But you're probably right about abilitiy cost (but it isn't going to change on PoE2 anyway).

I think so. We just need to discuss about what will need to be revised or improved, so devs would be able to use this for reference. That is another topic. I was speaking about the time after the final patch. ​But for sure, discussing here about what will need to be rebalanced could help the devs before this.

Such as ? ​Power Strike ? It was buffed in the last true balance patch. They are probably thinking twice before buffing it again. Same story for Hastening Exhortation. Barbaric Blow ? Yup but 1 rage would be too less. So not easy fix either. I don't think there are so many obviously broken abilities with obvious fixes at the moment. Even the title of the thread suggest that they are more "meh" than "bad". By the way, I'm arguing right now, but I'm pretty convinced that a Balance Mod made by fans would be a very good thing at some point. Probably after the "final patch". Really, the main "problem" before making such a patch might be simply that the game is still evolving. ​It would need time and concertation on forums by the way.

Not heard of it, and it is unlikely to exist as long as the game keeps being updated. You seem to think that the imbalanced abilities are intended by Obsidian and that a modder with good will could get a better result. What usually happens is that dissatisfied modders buff the abilities they like, usually beyond measure. ​The result is as imbalanced as the original game, just translated in the direction of power creeping. Balance requires time (so $$) and skill and, to be honnest, has not so much commercial value for single player game. Obsidian simply does not have time for this. At least they are trying to address the most glaring issues. And remember they are still working on it, so there's hope. PoE1 was basically left in a pretty good state (bar rest spamming with Vancian caster which was a problem inherent to the gameplay based on BG2 nostalgia), at least all classes had good abilities. EDIT : sorry if it sounds fanboyish. Maybe Obsidian could do better. But I just wanted to defend the view that balance is not easy to establish, even if it sounds "obvious" for us on forum. Consider that we would probably not agree on the solutions ourselves ^^ By the way, this is another issue with current multiclassing : if abilities have a semi-random level of power, the more abilities you have access to, the more powerful you can be.

Easier yes. Easy probably not. Obsidian games are like beautiful old mansions full of termites. Especially for the ruleset part. Even if you add a single nail in the wall to cling a painting, you're likely to awaken the whole colony. Sometimes you can't even figure from which new hole they have escaped (new dual wield bug is a good example). (Yet, these beautiful old mansions have way more charm than a standardized luxury Golden Tulipe hotel. I mean AAA games ) Also the rules and balances changes are likely to draw criticisism and discontent, especially from casual players who just want to play the game. And new content can be sold. Balance change has little commercial value.

I think the whole point of Boeroer is to suggest a change that would be likely to be implemented. More variety requires developpment cost for Obsidian, so it isn't likely unless really necessary. But well, for PoE3 maybe

Well I still believe that Con is most of the time above 10, even for characters that didn't add points to it. But I can see why you say it is not that much a buff. And you're right, it is supposed to be rewarding for High Con anyway. It won't make the game easier anyway because I suspect foes have usually more Con than party Obsidian games are a bit like Egyptian fields. They are quite likely to become preys of bugs. We must remember which is "Obsidian spirit animal". I admit this is the very best argument.

Hostile effect reduction is weird because of its incredible increasing return. 30% for 10 points is okay. 30% when one already has 30% is incredible. It is close to a permanent Clarity of Agony. I thought 5% reduction would have been better, but in the form of 1 / (1+5%) reduction. 5% in the current form means immunity to hostile effect at 30 Resolve (or lower hard cap). 4% in the current form means immunity to hostile effect at 35 Resolve (or lower hard cap). This would be too strong. By the way Hostile effect reduction from Resolve is the reason why all stats are capped at 35. Increase the magnitude and you'll mess up the whole system. So I think it's currently okay. People like to play all-in, so invincibility is underrated.

So did we come to the conclusion that Resolve is balanced ? It is my opinion too. Healing received is clearly the most logical choice, but it has several problems (given that some of the below are mutually exclusive) : 1) Con Affliction would be redundant with Con < 10. 2) It is annoying for UI. The spells and effects won't be able to show the actual healing value since it would vary between targets. 3) If it is added to the current effects of Might for healing done, it would be a big buff to Healing spells and effects. I beleive that Con is above 10 most of the time, especially in the endgame with all those buffs and items. All of them might require tweaking. I believe this is a problem. 4) If it replaces Might effect to healing done, it slightly decreases the efficiencies of healing spells since it is easier to have 1 char with high Might than several with high Con. 5) If it replaces Might effect to healing done, then Might become worse than Per (similar effects for damages, but Per applies to debuff too). Then another bonus should be added to Might. I had this weird idea about Con applying to healing done instead of healing received. Even for me it sounds strange (not much more than might applied to healing by the way) but I can't prevent from thinking that it would actually solve issues #1, #2 and #4 from the above. Issue #3 would not be applicable. Also adding an active property to Con is appealing IMHO from gameplay perspective. Then another effect should be added to Mig

It sounds good. I don't know if it's balanced (and I admit I don't have the time for a complete review) but as I said above it would have been difficult anyway with the numbers of abilities you added. Anyway, Spell removed while shifting and decreasing defense in human form makes the spellcasting abilities of the druid more dangerous and complicated to use (while still useable). So yeah, now it sounds more like twist than power creeping.