Werner

I think that's a perfectly reasonable plan. Some people seem to want other people to figure everything out and make their own build, but I see no problem with grabbing someone else's good build if all you want to do is to PLAY something great.

The main problem I see with that build cost-wise is the inclusion of a Gladiator's Armor +3%. Those are 2 billion influence on the market, and you have to be lucky to get one, because everyone else wants one too and is willing to pay it. The only fast way to get one is to buy it off market, and last time I paid any attention, that means 2.5 billion. I doubt the whole rest of the build will run you that much. Probably more like 1 billion for the rest of the build (as long as you have the Hamios in it - I didn't check), then 2 billion for that one IO.

So here's what I'd suggest, and this applies even if you find you have the influence - buy everything but that one IO, and leave the slot empty. Then run tip missions. Endless tip missions. I believe the Gladiator's Armor costs 30 alignment merits, which is I think 330 missions across a minimum of 60 active days of playing. You MIGHT have enough influence for it just from selling your Hamios, but you'll seriously be very, very solid without it. You'll probably notice the difference when you add it, but it's not going to be night and day. Oh, and that'll give you time to see if you like the build before slotting the final, most expensive piece. I'll be very surprised if you don't like it, though.

Well, I suppose there is the accuracy issue. No Tactics, no Kismet unique - it's just not built for fighting significantly uplevel enemies. If you like fighting enemies three or four levels higher than you, that could be an issue.

I should let Desmodos address this, but here's a quick partial fix. Since we're over the cap on 1.25% energy/negative defense bonuses, we can afford to lose a Cleaving Blow set with no effect on defense. Let's say that we drop them from Death Shroud and replace them with Multi-Strike acc/dam/end and dam/end. Take the extra slot, add it to Combat Jumping, and put in a Kismet +6% accuracy. It won't make a huge difference, but it'll help. It gets you over 95% accuracy fighting +2s, and 87%+ fighting +3s. Once you get your alpha slot to tier 3, you'll never face more than +3s. So that'll probably have to do. I'm not sure how to add more accuracy without more significant build changes.

Yeah, that looks like my latest and the one I'm probably happiest with. I'm not entirely satisfied with it, but I'm not sure exactly what's bugging me, and even less sure what to do about it. I'd like to get another slot in Divine Avalanche, but not sure I see anything I want to give up to get there. I should note that the Hamios are level 53, which doesn't seem to show in the short version. That's part of what makes it so expensive, and difficult to put together even if you have the influence (that stuff sells maybe once per month).

Even if I personally love the challenge of soft-capping anything and everything, yeah, it's not always the best approach. I started a thread recently involving a soft-capped Katana/Regeneration build, looking for improvements. As the community started poking at it, defense quickly dropped to the lower 30s, and calculated survivability went up the whole way. Defense isn't everything.

For a long time, conventional wisdom was "layer your defenses". You want everything - some defense, some resistance, some hit points, some healing, and so on. Focus on getting more layers, better layers. During this time, I think I could say that I was an advocate for a more extreme approach, the approach of soft-capping defense, even at the cost of other layers (but ideally with other layers left intact). There were solid numerical reasons for this suggestion in many cases. But things now have shifted perhaps too far in the other direction. It seems like everyone now "knows" that for best survivability, you soft cap defense no matter what the cost. That's also the wrong answer in many cases. The cost DOES matter. I think the forum regulars realize that and have probably always realized that, and there are a lot of people advocating a more balanced approach these days, at least on sets that don't provide a lot of defense (such as Dark Armor). I think some of that is a result of the incarnate content and the new soft cap of 59% simply being unattainable in many cases.

Anyway, we're still adapting to the new incarnate content and abilities. Our advice may shift in some unexpected direction moving foward, and it'll probaby take a while before we're confident in our recommendations again. That said, I'm very happy with the performance of my Katana/Dark with the incarnate abilities against the incarnate content, and feel no pressure at all to respec to something better. I'm not even sure what "better" would be.

Hopefully me blathering on is useful instead of tiresome. My Katana/Dark is my favorite character, and defense and survivability are a favorite topic of mine.

What you're hearing might be an abbreviation of this, "It is better to have smashing/lethal defense soft-capped than to have melee defense soft-capped." I think that IS the forum consensus, and one I agree with. But anyone who abbreviates that to "typed defense is better than positional defense" probably doesn't understand how defense works in the game. Typed defense is neither better nor worse than positional defense. The game doesn't care which you have - it just applies whichever one is higher.

In practice, whether typed or positional defense is better tends to depend on what your secondary comes with. Dark Armor supplies a tiny amount of both, so that doesn't decide anything. Parry supplies large amounts of lethal and melee defense. The trouble with the lethal defense is that you can only get set bonuses for smashing AND lethal. So if you want to boost your smashing defense up to high levels, you end up with lethal defense much higher than is normally needed. But hey, that gives you the option to use Parry or not depending on the situation, and some enemies have higher than normal to hit, and so require higher than normal defense. That's when Parry will be very useful. Basically, Broad Sword/Dark Armor doesn't really demand one approach or the other, and either can be viable.

Demodos's build is better for defense than you're seeing. Fire, cold and psionic damage aren't going to go against typed defense because positional defense is better or equal. So you're really looking at 33.3% minimum defense (melee or ranged) against anything but fire, cold or psionic AoEs. That's a pretty small portion of the damage you'll typically face. And the vast majority of damage is going to be going up against either 43.6% or 44.2% defense.

45% defense is referred to as the soft cap. Most enemies in the game have 50% to hit, and 5% to hit is the minimum the game allows. So defense over 45% defense doesn't help against most enemies in the game. Also, as you approach 45% defense, survivability increases drastically. So you might hear things that sounds like "soft cap or go home". That's an exaggeration to be sure, but there's a kernel of truth to it. These days, you might hear that called the old soft cap, though. In the new incarnate content, all (?) of the enemies have 64% to hit (and sometimes higher) instead. So now it requires 59% defense or more to make yourself as unhittable as possible. So we might refer to 59% defense as the new soft cap, or the incarnate soft cap. I'm not sure we've settled on the terminology yet. As you might imagine, this is MUCH more difficult to achieve, so it's important that your build have something to fall back on. In Desmodos's build, you hit the new soft cap to lethal damage with a single Parry, and to melee with two Parries. Then as Dark Armor, you have good resistance and a huge heal to fall back on. Mine does fine in incarnate content. I'm sure Desmodos's does too.

Resistances tend to be simpler than defense. Damage is damage, so there's none of that worry about what the to hit of the enemy is. It is also difficult to boost resistance meaningfully with set bonuses. Therefore, the rule of thumb tends to be simpler - slot your resistance powers decently and call it good. Oh, and pick up Tough. I'm pretty much of the opinion that nearly every build should have Tough (and Weave, but that's not resistance).

Yeah, much as I'd personally love a taunt aura, I agree it shouldn't be added. Some poeple hate them, and I bet a lot of people took Super Reflexes with a ninja concept, for instance, and having a taunt aura would run completely counter to that. I rolled mine knowing it didn't have one. Clicking on "Super Reflexes" meant I was willing to accept that drawback.

Mid's can be wrong, but the bugs are usually more subtle. For instance, I noticed recently that Mids' was including a 6% healing bonus in its calculation of how much Reconstruction heals. Reconstruction, though, has a resistance component, so the game itself doesn't allow the healing buff since it isn't a pure healing power. It makes a difference of I think 20 hit points on my Katana/Regen.

I think what you may have meant was "people who look at a Mids' build and think they know how it will play in the game can be wrong." That's certainly true. The more experience we have, probably the better we are at predicting in-game performance by looking at a Mids' build, but we can indeed be wrong.

And it also leads into your point about the player making a big difference. I have no idea how Iggy Kamakaze does what he does with some of his builds. I'm not saying his builds suck - they're great builds. But he still does more with them than I think I could do with them. But an awesome player with a bad build is still just an awesome player with a bad build. Being an awesome player doesn't make your build good.

Heck, I would even describe some objectively bad builds as good builds. For instance, my Fire/Shield is squishy and has poor DPS (objectively bad), but I still consider it a good build. Why? Because it's a concept build, and to match concept, it has only sword attacks in melee, and no heal. But within those serious constraints, it's been highly optimized with the help of the community.

Some things are matters of player preference and opinion, and a player with a certain play style may do better with some things that are, on paper, suboptimal. For instance, maybe I'm not as good as some at timing my Regeneration clicks, so perhaps I'm better off with more defense and passive regeneration and less recharge than an optimal build. It's still valid for the community to say "that's suboptimal and this would be better" just as it is valid to reply "I tried that, and this suboptimal approach actually works better for my personal play style for this reason." Other things, though, are simple basic facts about the game, like how defense is applied when the game calculates the chance of an enemy to hit you.

How many slots are needed in powers is often more a funtion of what set bonuses you are after than it is about the power itself. You do typically six-slot attacks, though occasionally you get away with five-slotting. For most other things, though, it's highly variable depending on what you're slotting.

You can only have ten enhancements on yourself at the same time. So to pull all those Hamios out of your build, you'd need to do multiple respects, reslotting all but ten every time. You might still use some Hamios, but it's usually not many any more. They definitely have their uses, though. I have 14 in my own build, but I think that's way more than typical.

How times have changed! There are more popular combinations, but yeah, you're no longer a red-headed stepchild.

The approach I took with my Katana/Dark is at least popular on the forum, which is soft-capping your positional defense (45%, a magic number). That makes you very survivable. It also, without an unlimited budget, involves making some compromises that make you less of a Dark Armor Scrapper and more something a bit different. If you're taking that approach, most of your IOs are going to be picked for their defense bonuses. Things like Gaussian Synchronized Fire Control, Eradication, Mako's Bite, Aegis, Blessing of the Zephyr, Red Fortune and Numina's Convalescence. Oh, and definitely a Steadfast Protection unique. I could post example builds (well, for Katana/Dark), but I'm not sure if you're looking for that specifically, or just the kind of thing to be looking for.

Soft-capping, though, isn't necessary. You can be very survivable without it. Pick up some decent defense, but don't go overboard. Don't compromise on your strengths. Make sure you have and use either Cloak of Fear or Oppressive Gloom. Get a bunch of recharge so that Dark Regeneration recharges quickly.

Oh, definitely put a Theft of Essence proc in Dark Regeneration. Desmodos already mentioned it, but I want to repeat it for emphasis.

Umbral was into shock therapy and tough love. "Your build sucks, here's why, and here's how to fix it." And if you disagreed or got upset, he was perfectly happy to jump straight to mockery. Thing is, if you could get past all that, he was a very well-informed and helpful guy. A lot of people just couldn't get past HOW the information was presented, and he'd have probably done better with a softer tone, but he just didn't have the patience for it. I understood where he was coming from. He was my anti-hero, I guess, the forum supervillain I couldn't be.

And yeah, you'd only be Umbral if you now turned to outright contemptuous mockery.

You can, and a Katana/Electric can actually put out some very impressive DPS if you really go for it. Not quite Dark/Shield or Fire/Shield DPS, but without all the set up time to get there, so probably more in many situations. Unfortunately, also without the survivability. But hey, you can still have Divine Avalanche for when things turn south.

But in that case, it's not "built for survivability". It's built for damage with a survivability back up plan. Should be good enough for most content, though. I've been tempted to make one myself.

Katana is capable of great single-target damage. However, you're not likely to see that with a resistance set built for survivability, because you're going to be using Divine Avalanche to cover your melee and lethal defense needs.

I suppose the best explanation is the first part of the earlier post where I try to explain the basic idea. In terms of the flow in the spreadsheet, I should probably highlight the inputs in some way, but they're the numbers to the top and left of the various matrices, and then random bits and pieces here and there. You enter enemy to hit at the very top of the thing (B1). The matrix above the freeze line feeds all the matrices below it.

Oh, and there's a little bit on tab 2 to plug in some recharge numbers for the regen builds to figure out how much down time you have without Shadow Meld or Moment of Glory. I don't think I have it quite right, as doesn't Moment of Glory take effect very early in the activation time? How early? Is Shadow Meld the same way? That could significantly improve the numbers on these builds.

I know it's not user friendly. It grew over time. It was never something I sat down and designed for use, and now I'm so used to it that I have no motivation to make it easier.

I swear that at work, my code is tight and elegant, my GUI is clean and beautiful, and the users have all the information they need at their fingertips, understood at a glance. I swear that my professional output looks nothing like this.

You won't be doing much of that, don't worry. You have solid defense, passable hit points outside of Dull Pain, passable resistance for a Regen, and some pretty decent passive regeneration. Most of the time you'll be cruising along, barely touching your clicks, even against content that most people would describe as nasty. It'll only become an insane clickfest if, say, you find yourself suddenly soloing something really nasty that won't die fast. But doing the dance of the Regen clicks is probably more fun than a visit to the hospital in that situation.

That's certainly my plan. I haven't included that in the survivability calculations because I'm not sure how we'd say what the average enemy resistance to the debuff would be. I'd probably apply AV resistance to it, and it wouldn't have much effect on my numbers. But it can have more effect in the game as a whole.

Yes, I suppose in the same way that I underweight hit points, I underweight resistance. Both are functionally equivalent in many ways. So my Katana/Dark functionally has more hit points most of the time. And then I could argue that I also underweight defense in a different way, which is in its ability to prevent debuffs and other secondary effects that challenge your survivability. And I underweight regeneration, because it helps keep your hit points topped up, which gives you more hit points when you need them, so is functionally kind of like hit points. And I underweight healing, because the ability to heal allows you to respond to spikes of damage in ways that average regeneration does not.

So if I'm underweighting all of the important factors, perhaps I should stop worrying about all the underweighting going on. I'm sure they're not all entirely in balance, but I've never thought of a good way to account for all the issues. I just keep them in mind when looking at things. They're all "yes but".

Ooh! Cool! That'll help my Katana/Dark even more since I'm running Cardiac. I'll need to recalculate.

As another data point along those lines, spikes of damage aren't the only problem - my Katana/Dark's resistance and at least decent hit points tend to keep him from getting "one shotted". Waves of damage are also a problem. I can be cruising along fine without even touching my heal, and then there will just be a big, reasonably-long-lived wave of damage. Get pounded down, hit my heal, and get pounded down again faster than it recharges. That which can't kill me in fifteen seconds can't kill me, but there are things out there that can kill me in fifteen seconds, leaving me popping inspirations. And that seems to be common for me in general across a number of toons. I'm fine for a long time, and then I'm not just popping a small green, say, I'm chugging 1/4 of my inspiration tray to stay alive. And then I'm fine for a long time again.

The Regeneration secondary seems made for these waves of damage. During those forever times when everything is going OK, you aren't even touching your heals. Then when one of those massive waves of damage hits, you're sitting on a huge list of powers that will all help keep you in the game and ride it out. On a Regen, you just about never find yourself just clicking when things recharge, over and over, getting the highest average level of mitigation and damage recovery. You're matching your mitigation and damage recovery up to the spikes and waves. Played well, that makes Regen much stronger than measures of average performance over time suggest.

Now, with Barrier on the incarnate trials, I don't pop it the moment it's recharged, but I don't reserve it for when I'm in trouble either. I just look at my defense, figure it's similarly buffed to the people around me, and I buff it when it I think the group could use a buff. In solo play, it would probably be best to hold it in reserve, as having both Barrier and Dark Regeneration available should get me through one decent wave of damage. Then with any luck, Barrier will be recharged by the time the next hits.

I don't think I'll have much time to do much tonight, but I'll clean up and post the spreadsheet when I can. In the mean time, I'll try to explain the basic idea. And it's a long post since I'm short on time. Hopefully that makes sense to some people.

Well, the BASIC idea is the old immortality line (the name of which Arcanaville I believe regrets to this day).

I divide incoming damage into a matrix by position and type. I divide that across types by blue side percentages given by Besserwisser a long time ago. I split apart into melee, ranged and AoE by 70%, 20% and 10%. Multiply to get the matrix. In theory you can enter every cell in the matrix individually, but I never have because I'm not sure what I'd change.

Now, your resistance and defense combine to form a matrix as well. Each cell in this matrix is a specific damage type from a specific position. Do the opposite of a mitigation calculation in each cell, a "% damage allowed" calculation based on your typed defense, positional defense and resistance. Multiply that by the percent of damage in each cell from the earlier matrix.

Now, sum up the second matrix, and you have an overall percent of damage allowed. That's one fundamental number, basically 100% - mitigation, though "mitigation" here is not the forum consensus definition. It's not a problem, though, as it just means my numbers will be twice as high as if I calculated using the forum consensus. Since I give no scale for the number, and use them only for comparison with other similarly-calculated numbers, this isn't an issue.

Now you need to look at your average damage recovery per second. No matrix here, since you recover from all types of damage equally. Just time average your healing and regeneration.

Divide this total by the damage allowed, and you get survivability. And survivability here is basically the amount of enemy damage output per second that you can survive. More precisely, it's the amount of enemy damage output (split according to that first matrix) that would produce a random walk of your hit points instead of an upward or downward trend.

What it DOESN'T recognize is that a random walk that takes you to zero hit points WILL KILL YOU. That's a big limitation when you're talking about these survivability levels, which on paper are enough to survive crowds of AVs pounding on you.

Hit points, for instance. The more hit points you have, the less likely the random walk will kill you in any reasonable time frame. Resistance. The more resistance you have, the less likely the random walk will kill you. Ability to match healing to the spikes of damage. The better you can do this, the less likely the random walk will kill you. And so on.

It also ignores other things, like debuffs. That's kind of accounted for by being able to plug in a to-hit number, though. It won't give you one number, but you can get an idea how tough you are when things get nasty.

So, tons of problems. The way around them is to write a program that simulates the entire game. That's beyond my interest, and possibly beyond my capability. Certainly beyond my capability in the time I'd be willing to devote to such a project.

Or I could just say that I already HAVE such a simulator. The devs have been kind enough to provide it. I can take my build, build it, and go test it out in their simulator against an amazingly good recreation of the actual game. Why, I'd almost think I was playing the game when I was using their simulator. Unfortunatly, it only does the calculations in real time, so it takes a very long time fiddling with their simulator to get any good numbers out of it. Ah, well. At least they made it fun to use.

Hmmm, it doesn't explain how I account for Barrier or for the defense and resistance buffs. I basically create a matrix for each, and weight them by the average time each is up. I don't calculate survivability for each, because let's say that once per minute, you had a five second power that gave you 100% resistance. Hopefully it's obvious that wouldn't overall give you infinite survivability, even if it did for those five seconds.