Umbral

The manner in which it is misleading is simple: you're suggesting that every toggle costs the exact same amount of endurance. The assumption would be that 4 toggles costs half as much as 8 toggles. This isn't always, nor is is often ever true.

All of those toggles running together cost 2.44 end/sec base. All of your Defender's toggles running together cost 1.56 end/sec base. The defender is only running 56% more end/sec in toggle costs than the Defender, and, interestingly enough, 3 of those 8 toggles operate to increase damage dealing capability (and thereby end efficiency for killing) rather than direct survivability.

The entire point I'm making here is that, even if you didn't intend it, you were trying to make it sound like the Scrapper was running through twice as much endurance per second from toggles as the Defender was (re: "HALF as many as my Scrapper") and still getting better endurance efficiency than the Scrapper. The manner in which you made the comparison (re: "# of toggles : # of toggles" rather than the more appropriate "end/sec : end/sec") generated a flawed argument, which simply demonstrated your poor analytical skills, which virtually everyone already knows about and cites as to why your analyses are fundamentally flawed.

Actually, I mean that no one believes you. Even when you try to put out numbers to validate your personal experience, you get those numbers blatantly wrong (and everyone repeatedly brings forth the numbers to prove you wrong each time). The problem with all of your personal experiences is that every time you've brought them up, you've suffered enormously from perception bias without ever doing anything to mitigate or even mention it while simultaneously using automatically skewed comparisons to further make your point.

Testing has actually been done and people have checked the specific listing of the powers within the confines of the game and found that all of the heal IO procs benefit from the slotting of the power that they are in. This is why you want to slot the +regen procs into a power that is getting more +heal slotting (re: Why top tier builds put their Numina's procs into RttC/Integration rather than Fast Healing). The +recov values are similarly benefited, except that there aren't many power that accept both end mod and healing IOs, much less ones you want to put the procs into.

Unless the devs changed the operation of these specific procs in I16 (and I haven't seen any listing of those changes), it will work exactly as I described.

This is horribly misleading to say, especially since there isn't a single standard toggle cost. Tough and Weave each cost .325 end/sec. Dispersion Bubble costs .52 end.sec. Maneuvers costs .39 end/sec.

Depending on what build Ultimo is specifically using with his Scrapper, every single one of the toggles that he's running is probably going to cost less than any of the toggles that he's running on his Defender. Not to mention that his Scrapper (Claws/Regen, iirc) actually has several other reasons for vastly increased endurance sustainability (Claws/* has 25% lower base end costs, */Regen has only a single toggle that only costs .21 end/sec and Quick Recovery to make endurance even easier).

It's information like this that Ultimo is great at putting out, which is why no one really believes anything he says: he'll put out some numbers that, at first glance, look as if they might be accurate. Sometimes they even are (in only the most general terms). The problem is that they're interpreted completely and totally wrong just to make his point seem more valid. I've yet to see any of his "analysis" or "conclusions" withstand any degree of scrutiny.

It's 12.5% +recov base (.209 end/sec with 100 end), or, in qualitative parlance, half of Stamina. If you fully slot it (3 even level end mod SOs), you'll get ~24.375% +recov, which is, with 100 end, .407 end/sec. With the +end accolades, you'd pull it up to .448 end/sec.

I don't really encourage much in the way of slotting up PP for end mod or heal. The base values are simply too low in my opinion (re: the +regen is similarly unimpressive). Of course, it's a great spot for the Perf Shifter, Numina, Regen Tissue, and Miracle procs (especially the +recov ones because you can slot the end mod to make them bigger), which is what I'd advocate slotting into it.

Yeah... I was talking numbers at the two extremes : SOs and unlimited budget. I'm pretty psyched about Elec Armor, though I'm not particularly enthused about it in the hands of a Scrapper. After I get my Elec/* Scrapper (now undecided on what I want his/her secondary to be because Elec/Shield seems like it'll be a massive FotM), I'm planning on doing an Elec/* Tanker of some kind.

It's not just the fact that Power Sink is stronger. It's also the fact that the end redux isn't all that impressive. For normal gameplay, the +end redux is going to contribute slightly less than an SO over time (47.7% uptime * 59.6% end redux = 28.4% average end redux). At top tier play (250% +rech), it's only going to contribute 50% end redux, which isn't really going to solve many endurance problems. Power Sink is significantly stronger (assuming decent native slotting for end redux thanks to IOs, you'd need to be consuming more than 4 end/sec on average in order to equal Power Sinks' contributions, assuming they're both used at the appropriate intervals to use up the same percentage of animation time).

Either way, because the heal got made significantly bigger, my sincere recommendation would be not to use Energize in a permanent manner. The self heal is the most substantial component of it and should be the center piece of any tactical use of it. If you're having to use it to keep yourself endurance sustainable, then you're doing something very wrong (and wasting a lot of endurance because it costs 10.4 endurance like all of the other similar powers).

Well, considering the 120 second base recharge, 1.32 sec Arcanatime animation, and 30 second duration, it's highly doubtful you'll ever get perma-Energize. You'd need to pull the recharge down to 28.68 seconds (319% +recharge) in order to get it permanent, and, considering that 250-260% is considered to be stretching it even for builds that are packing native +rech (re: */Elec and */SR) and IO'd out to the gills, it's pretty doubtful you'd even get it permanent.

Even so, the +regen component is actually less important than the self heal portion and the +end redux component is largely pointless when you consider Power Sink. A 25% heal does twice as much for you as 30 seconds of 100% +regen does at base hit points. In order for them to be equal, you'd have to have double base hit points (impossible since the hp cap is 180%) and no +heal set bonuses.

Import/Export>Short Forum Export

Pick a decent color scheme (the US color schemes generally work the best) from the color scheme section in the top left, make sure that you're using the "phpBB / Zetaboards" formatting code in the top right, and then hit "Export Now". From there, it's just a matter of pasting it where you want to see it. Make sure you don't preview seeing as the data links don't withstand the changes that they go through when you preview.

As to the build, it's not too bad, but there is simply a lot of slotting that is pretty much pointless. Most of the 6 piece set bonuses you went for (Hecatomb, Doctored Wounds, Aegis, Crushing Impact) don't really do much for you. Your best bet would be to reduce the slotting down to 5 pieces and use those slots elsewhere for other, more useful set bonuses.

Here's a build that you should work better for you. It manages better survivability, better damage, and it's got better endurance sustainability (even assuming you were planning on getting the passive accolades).

Hero Plan by Mids' Hero Designer 1.401
http://www.cohplanner.com/

Click this DataLink to open the build!

Level 50 Mutation Scrapper
Primary Power Set: Martial Arts
Secondary Power Set: Regeneration
Power Pool: Leaping
Power Pool: Fighting
Power Pool: Speed
Power Pool: Fitness
Ancillary Pool: Body Mastery

Hero Profile:
Level 1: Storm Kick -- Hectmb-Dmg/Rchg(A), Hectmb-Acc/Dmg/Rchg(3), Hectmb-Acc/Rchg(3), Hectmb-Dmg/EndRdx(5), Hectmb-Dam%(5), Mako-Acc/EndRdx/Rchg(46)
Level 1: Fast Healing -- Numna-Regen/Rcvry+(A), Mrcl-Rcvry+(29)
Level 2: Reconstruction -- Dct'dW-EndRdx/Rchg(A), Dct'dW-Heal/Rchg(7), Dct'dW-Heal/EndRdx/Rchg(7), Dct'dW-Heal(9), Dct'dW-Rchg(9)
Level 4: Quick Recovery -- P'Shift-End%(A), P'Shift-EndMod(11), EndMod-I(11)
Level 6: Focus Chi -- AdjTgt-Rchg(A), AdjTgt-ToHit/Rchg(13), RechRdx-I(36)
Level 8: Crane Kick -- T'Death-Acc/Dmg(A), T'Death-Dmg/EndRdx(13), T'Death-Dmg/Rchg(15), T'Death-Acc/Dmg/EndRdx(15), T'Death-Dmg/EndRdx/Rchg(17), T'Death-Dam%(17)
Level 10: Dull Pain -- Dct'dW-EndRdx/Rchg(A), Dct'dW-Heal/Rchg(19), Dct'dW-Heal/EndRdx/Rchg(19), Dct'dW-Heal(21), Dct'dW-Rchg(21)
Level 12: Combat Jumping -- LkGmblr-Rchg+(A), LkGmblr-Def(23), Zephyr-Travel(23), Zephyr-Travel/EndRdx(25), Zephyr-ResKB(25)
Level 14: Super Jump -- Zephyr-Travel(A), Zephyr-Travel/EndRdx(27), Zephyr-ResKB(27)
Level 16: Integration -- Numna-Heal/EndRdx(A), Numna-Heal(29), Numna-Heal/Rchg(31)
Level 18: Crippling Axe Kick -- T'Death-Acc/Dmg(A), T'Death-Dmg/EndRdx(31), T'Death-Dmg/Rchg(31), T'Death-Acc/Dmg/EndRdx(33), T'Death-Dmg/EndRdx/Rchg(33), T'Death-Dam%(33)
Level 20: Boxing -- Amaze-Stun/Rchg(A), Amaze-Acc/Stun/Rchg(48), Amaze-Acc/Rchg(48), Amaze-EndRdx/Stun(50), Amaze-ToHitDeb%(50)
Level 22: Tough -- ImpArm-ResDam(A), ImpArm-ResDam/EndRdx(34), ImpArm-ResDam/EndRdx/Rchg(34), ImpArm-ResDam/Rchg(34)
Level 24: Weave -- LkGmblr-Rchg+(A), LkGmblr-Def(36), LkGmblr-Def/EndRdx(36)
Level 26: Dragon's Tail -- Oblit-Dmg(A), Oblit-Acc/Rchg(37), Oblit-Dmg/Rchg(37), Oblit-Acc/Dmg/Rchg(39), Oblit-Acc/Dmg/EndRdx/Rchg(39), Oblit-%Dam(39)
Level 28: Instant Healing -- Dct'dW-Heal(A), Dct'dW-EndRdx/Rchg(40), Dct'dW-Heal/Rchg(40), Dct'dW-Heal/EndRdx/Rchg(40), Dct'dW-Rchg(42)
Level 30: Hasten -- RechRdx-I(A), RechRdx-I(37), RechRdx-I(42)
Level 32: Hurdle -- Jump-I(A)
Level 35: Health -- RgnTis-Regen+(A)
Level 38: Moment of Glory -- LkGmblr-Rchg+(A), LkGmblr-Def/Rchg(43), RechRdx-I(43), RechRdx-I(43)
Level 41: Focused Accuracy -- GSFC-ToHit(A), GSFC-Build%(42), GSFC-ToHit/Rchg(45), GSFC-ToHit/Rchg/EndRdx(45), GSFC-Rchg/EndRdx(45), GSFC-ToHit/EndRdx(46)
Level 44: Stamina -- P'Shift-End%(A), P'Shift-EndMod(46)
Level 47: Super Speed -- Zephyr-Travel(A), Zephyr-Travel/EndRdx(48), Zephyr-ResKB(50)
Level 49: Resilience -- S'fstPrt-ResDam/Def+(A)
------------
Level 1: Brawl -- Empty(A)
Level 1: Sprint -- Empty(A)
Level 2: Rest -- Empty(A)
Level 1: Critical Hit

It's lacking in some of the QoL features that I've become used to, like the enhancement summary window. >.< Figuring out enhancement values now is much more annoying.

The answer is... maybe. It depends on how good you are with those click powers. If you can't really predict incoming damage well or have a hard time managing clickies, you're gonna have a hard time soloing large group content. Even if you're not a great player, you should still have no problem soloing normally. There isn't a Scrapper alive that can't solo normal content.

This is gonna sound bad, especially since I'm reasonably sure that someone has done it, but, no, you're not going to solo AVs without getting some damage mitigation to augment your damage recovery. */Regen's damage recovery is nice, but it's not enough on its own 99.99% of the time for AVs, especially the harder ones.

From what I've been told (I haven't really kept abreast of the PvP changes), MA/Regen is actually one of the better PvP sets. MA is supposed to have decent burst damage in Eagles Claw and */regen is, iirc, considered to be the best PvP Scrapper secondary thanks to the incredible peak survivability.

You can stack regen, but recharge is going to do more for you. Honestly, if you're looking for effective, you're best bet is to simply ignore +regen and focus on recharge first, second, and third, and defense 4th. Everything else (after making sure all of your important powers are getting decent values from their enhancements) should take a back seat.

As far as PvE is concerned, one of the big advantages of */regen is that it actually gets additional survivability from one of the most important offensive attributes around: +rech. Since optimized attack strings play such an integral part in top end DPS (and, I would argue, in optimized DPS even at low levels), enhancing */regen for survivability will work double duty to increase your damage dealing capabilities. */WP doesn't have that benefit, as is generally seen by the substantially lower amounts of theoretical damage output by top end */WP builds as they compare to */regen. Of course, they'll sustain higher theoretical survivability thanks to not having the controllable peaks and valleys of survivability that are the hallmark of */regen, but they don't have as much offensive synergy to play with.

If you want to check to see whether any specific build breaks the rule of 5, go into Mids and hit "Ctrl + B" (or, for the mouse inclined, Window>Sets and Bonuses). It'll give you a list of your totaled set bonuses and a separate list of each of the sets you've got along with a warning next to any set bonus that breaks the rule of 5.

You may want to edit it so that you're using the "phpBB / Zetaboards" formatting rather than the "Official Forums (UBB.threads)" formatting. The Official Forums no longer use that same formatting code. It would make it substantially easier to read.

As to the actual build (from what I can read of it), you're doing a lot of stuff wrong. You're short slotting the recharge on MoG and IH for defense and healing, respectively, when those are the attributes that contribute very little to your actual survival (defense because MoG is already giving you redundant defense without any slotting and IH because only 200% of the baseline 800% +regen is actually enhanceable). You completely shattered the "rule of 5" for the 5% +recharge set bonus with your 10 copies of it (re: the last 5 of those set bonuses don't do anything). You're also stupidly breaking the rule on the 1.12% +hp and 10% +regen set bonuses.

You haven't even got the recharge to run MG>Smite>Siphon>Smite chain. MG, which requires 235% +rech (re: 4.488 second recharge) for that string to run seamless, is nearly a full second off. Unless you plan on actually investing in the expensive purple sets that are actually going to allow you to run that attack string (Hecatomb for MG at the very least), you're not going to have the recharge for it.

As to your build, don't break the rule of 5. Having that many Crushing Impacts, Obliterations, and Doctored Wounds sets is pointless beyond mentioning. The primary benefit of Doctored Wounds and Crushing Impact is the 5% +recharge set bonus and you've rendered most of them entirely pointless by having so many of them. Switch back so that you've only got 5 of those set bonuses.

Stick with the 3 slot slotting for QR and Stamina (it'll open up better slotting options everywhere else and get you more end/sec recovery). If you don't know what it is: Perf Shifter proc, Perf Shifter End Mod, level 50 common End Mod.

Don't use Crushing Impacts in your attacks. Use 6 piece Touch of Death. The +def(melee) is incredible and it'll give you enhancement attributes that are just as good.

Health and Fast Healing shouldn't get loads of slotting. Their actual contribution to your survivable is minimal at best. Don't give them Doctored Wounds at all. Just give them either a heal set proc or a level 50 common heal IO.

Instant Healing needs more slot loving than Integration does. Integration doesn't cost much endurance as is on at all times. It only needs heal enhancing. Instant Healing on the other hand needs recharge more than anything else. Additional +heal slotting just makes it nicer. Switch your slotting of the two powers, switching the Heal/End for the Recharge when you give the Doctored Wounds to Instant Healing.

The 3 piece LotG set bonus is largely redundant unless you plan on doing everything without the 4 passive accolades (which you plan on getting, right?). Because you've got perma-DP, all that +hp is eaten by the hp cap. It's only effective when DP is down and, I assure you, if you're using DP correctly, is barely any time at all.

As to power selection, if you're going budget, don't try for MG>Smite>Siphon>Smite: you can't afford it. Take Shadow Punch and accept that you'll have to use MG>Smite>Shadow>Siphon>Smite>Shadow. It's worse on the damage by a little but it's actually attainable with a budget build.

This build will work much better for you. It's pretty much the same cost for substantially better damage, survivability, and overall playability. It uses the attack string I mentioned, is endurance sustainable with everything running, and has excellent melee defense so that AVs will be dealing substantially less damage.

Hero Plan by Mids' Hero Designer 1.401
http://www.cohplanner.com/

Click this DataLink to open the build!

Level 50 Magic Scrapper
Primary Power Set: Dark Melee
Secondary Power Set: Regeneration
Power Pool: Leaping
Power Pool: Fighting
Power Pool: Speed
Power Pool: Fitness
Ancillary Pool: Body Mastery

Hero Profile:
Level 1: Shadow Punch -- T'Death-Acc/Dmg(A), T'Death-Dmg/EndRdx(3), T'Death-Dmg/Rchg(3), T'Death-Acc/Dmg/EndRdx(5), T'Death-Dmg/EndRdx/Rchg(5), T'Death-Dam%(7)
Level 1: Fast Healing -- Heal-I(A)
Level 2: Smite -- T'Death-Acc/Dmg(A), T'Death-Dmg/EndRdx(7), T'Death-Dmg/Rchg(9), T'Death-Acc/Dmg/EndRdx(9), T'Death-Dmg/EndRdx/Rchg(11), T'Death-Dam%(11)
Level 4: Quick Recovery -- P'Shift-End%(A), P'Shift-EndMod(13), EndMod-I(13)
Level 6: Reconstruction -- Dct'dW-Heal/EndRdx(A), Dct'dW-Heal/Rchg(15), Dct'dW-Heal/EndRdx/Rchg(15), Dct'dW-Heal(17), Dct'dW-Rchg(17)
Level 8: Siphon Life -- T'Death-Acc/Dmg(A), T'Death-Dmg/EndRdx(19), T'Death-Dmg/Rchg(19), T'Death-Acc/Dmg/EndRdx(21), T'Death-Dmg/EndRdx/Rchg(21), T'Death-Dam%(23)
Level 10: Combat Jumping -- LkGmblr-Rchg+(A), LkGmblr-Def(31)
Level 12: Dull Pain -- Dct'dW-EndRdx/Rchg(A), Dct'dW-Heal/Rchg(23), Dct'dW-Heal/EndRdx/Rchg(25), Dct'dW-Heal(25), Dct'dW-Rchg(27)
Level 14: Super Jump -- Zephyr-Travel(A), Zephyr-Travel/EndRdx(29), Zephyr-ResKB(31)
Level 16: Integration -- Numna-Regen/Rcvry+(A), Numna-Heal(27), Mrcl-Heal(29), Mrcl-Rcvry+(40)
Level 18: Dark Consumption -- Oblit-Dmg(A), Oblit-Acc/Rchg(31), Oblit-Dmg/Rchg(33), Oblit-Acc/Dmg/Rchg(33), Oblit-Acc/Dmg/EndRdx/Rchg(33), Oblit-%Dam(34)
Level 20: Boxing -- Acc-I(A)
Level 22: Tough -- ImpArm-ResDam/EndRdx/Rchg(A), ImpArm-ResDam(34), ImpArm-ResDam/EndRdx(34), ImpArm-ResDam/Rchg(36)
Level 24: Weave -- LkGmblr-Rchg+(A), LkGmblr-Def(36), S'dpty-Def(36), S'dpty-Def/EndRdx(37)
Level 26: Soul Drain -- Oblit-Dmg(A), Oblit-Acc/Rchg(37), Oblit-Dmg/Rchg(37), Oblit-Acc/Dmg/Rchg(39), Oblit-Acc/Dmg/EndRdx/Rchg(39), Oblit-%Dam(39)
Level 28: Instant Healing -- Dct'dW-EndRdx/Rchg(A), Dct'dW-Heal/Rchg(40), Dct'dW-Heal/EndRdx/Rchg(40), Dct'dW-Heal(50), Dct'dW-Rchg(50)
Level 30: Hasten -- RechRdx-I(A), RechRdx-I(42), RechRdx-I(42)
Level 32: Midnight Grasp -- Mako-Acc/Dmg(A), Mako-Dmg/EndRdx(42), Mako-Dmg/Rchg(43), Mako-Acc/EndRdx/Rchg(43), Mako-Acc/Dmg/EndRdx/Rchg(43), Mako-Dam%(45)
Level 35: Hurdle -- Jump-I(A)
Level 38: Moment of Glory -- LkGmblr-Rchg+(A), LkGmblr-Def/Rchg(45), RechRdx-I(45), RechRdx-I(46)
Level 41: Resilience -- S'fstPrt-ResDam/Def+(A)
Level 44: Focused Accuracy -- GSFC-ToHit(A), GSFC-ToHit/Rchg(46), GSFC-ToHit/Rchg/EndRdx(46), GSFC-Rchg/EndRdx(48), GSFC-ToHit/EndRdx(48), GSFC-Build%(48)
Level 47: Health -- RgnTis-Regen+(A)
Level 49: Stamina -- P'Shift-End%(A), P'Shift-EndMod(50)
------------
Level 1: Brawl -- Empty(A)
Level 1: Sprint -- Empty(A)
Level 2: Rest -- Empty(A)
Level 1: Critical Hit

Werner has a Kat/Regen, not a BS/Regen.

As to the build, my first question would be what attack string are you using? You may be using too much Parry, which is going to hurt your DPS. Even so, you should be able to do enough damage to whittle down AVs, little by little. Endurance sustainability shouldn't be an issue either since you've got both Stamina and Quick Recovery (3.69 end/sec net passive to burn on your attacks means that you pretty much can't run out of endurance).

The other question I would ask would then be what level AVs you're fighting. Soloing an AV more than 2 levels above you is going to be a stretch, not only from the damage department but also in the survivability department. Soloing highly lethal resistant AVs is also going to be a problem (Infernal is pretty resistant to lethal, iirc).

Your best bet is to simply go with an HS>Parry>Hack>Parry attack string. You'll get decent enough damage to start bringing the AVs down (~140 DPS), but it'll still take a while. I don't really see any reason why you'd be having problems.

Apparently mediocre tohit debuff balances out the offensive disparity. Getting Nightfall rather than either Shadow Meld or Darkest Night really let my jaw dropped. Just... bad...

You're assuming that everyone should be able to overcome a given threat that is designed around having 6 players present. That's... just bad.

It was never intended for anyone to be able to solo the entire game. It has happened, but only for a specific few that are specifically built for it. AVs were never intended to be solo content. Trying to claim that, because your Defender can't solo AVs, you're too weak is simply ignorant.

Of course, I would use the first half of your quote to explain why the second half is wrong. If it's a power that is highly situational and can be skipped with little to no drop in real performance, that's a contemptible power from where I'm standing. Sure, it's not like Resilience (5.635% s/l resist and stun protection on a melee AT that's available 10 levels after you get your mez toggle... are you serious?) or Revive (worst self rez in the game), but it barely warrants the power choice when you consider everything else you could pick up that would contribute more.

I will admit that a lot of this feeling comes from having used it in its pre-nerf state wherein it was actually useful for dealing damage. It's a pale reflection of what it once was. It used to be taken because it was effective, and you were crazy for not taking it. Now it's just kinda of... there.

You're going to want to use the IP>ToF combo right at the start simply because it will allow you to prevent all of that damage. Using it late doesn't really hinder you much, but it's kinda pointless to do damage first and then control when the target is half dead. Your best bet would be to jump in, pop MoG, use ToF on the boss while letting everyone group up around you, and then dropping IP for spawn controlleryness. Keep in mind that fear won't keep from from attacking like a hold will. If you're dealing damage to them, they'll break the fear every 5 seconds or so to throw an attack your way.

Glimpse is cheap as dirt. You can easily get the entire set for under 100k.

Well, first off, you can't actually get that attack string. At the recharge cap, Smite would be recharging in 1.2 seconds and you'd need it to recharge in 1.056 seconds (the Arcanatime normalized animation time of Shadow Punch). Secondly, even assuming you were using an MG>Smite>SP>Wait>Smite attack string (with 250% +rech in Smite, which is more than a bit of a stretch), you'd still be doing worse damage simply because of that wait. You'd have to use MG>Smite>SP>Wait(.66 seconds)>Smite which would generate 429.3 damage in 6.336 seconds for 67.75 DPS. MG>Smite>Siphon>Smite generates 506.4 damage in 6.732 seconds for 75.22 DPS. In order for MG>Smite>SP>Smite to even equal MG>Smite>Siphon>Smite, it would need to be absolutely seamless, which, as I said before, isn't gonna happen.

It's actually closer to a 20% +def difference. You've got 11.1% +def to all. My build is packing 31.57% +def(melee) with Stealth turned on (using half value for combat). I actually made sure that the build is endurance sustainable against a single target with Stealth on so that I could count on the extra 2.17% +def(all).

As to including the Fitness pool (if that's your thing), you could always get rid of the Concealment pool and Super Speed. You'd have the extra slots for Stamina by pulling them out of Super Speed. Of course, you could also replace Invisibility with Conserve Power for another option.

My build is intended to be the absolute most that you can eke out of a DM/Regen. Highest survivability with the highest possible DPS. I've made plenty of builds that achieve greater survivability and also plenty that achieve greater DPS, but I've yet to see a DM/Regen build that can manage to achieve such a high level of survivability with such a high level of damage.

As to where endurance powers fit in to that, it's not really all that important. Once you become endurance sustainable, anything over that is simply a buffer zone to make you happier. My feeling was that I should simply achieve sustainability without going over because it would bring down either my DPS, my survivability, or both.

Brutes always got more mileage out of Buzzsaw simply because damage slotting and damage buffs mean less to them overall because they can get huge amounts of +dam via Fury and, thanks to their low damage scalar, procs add more functional +dam for Brutes than they do for Scrappers.

From what I know of Brutes, it's not as bad an idea as it is for Scrappers, but it's not the best path like it used to be, especially for knocking out large targets.

DM/Fire follows pretty much the same advice that I give to all */regen Scrappers (*/Fire in general does, but DM/Fire moreso than the rest because of DC, Siphon, and Soul Drain):

1. Get more recharge.
2. What do you mean "enough"? I said more recharge.
3. Do I have to repeat myself again? More Recharge!
4. You're sure you can't get any more recharge without gimping your build? Are you sure? Okay, now stack as much defense as possible.