DB & AVs?

DM/SD or DM/SR.
There are others but those two are the best/most efficient.

Are you talking about the specific powers needed?

Ablating Strike > Vengeful Slice > Sweeping Strike

The great thing about it is that, if you toss in Blinding Feint beforehand, you get an excellent attack string with incredibly low +rech requirements.

BF: 12 sec base, 5.28 sec needed, 128% +rech needed
AS: 6 sec base, 5.544 sec needed, 9% +rech needed
VS: 8 sec base, 4.092 sec needed, 96% +rech needed
SS: 11 sec base, 5.28 sec needed, 109% +rech needed

Np for off-topic

Imo you're forgetting the most important point apart from Umbral being extremely helpful, he's extremely entertaining too!

Kudos to you umbs

Yes. The EndMod enhancement enhances the +recov from the Miracle and Numina's procs, allowing PP to grant more +recov than Stamina.

Wow that is some *** kissing.

I think it's fine that some people take a great deal of time to help other people. It's good. There are a lot of, as equally good min/maxxers in game, that wouldn't. For they only worry about themselves and their own. The fact that you guys take the time is something.

I had actually been working on "arcanatime" since I11 beta, off and on, and was in the middle of the final leg of my testing when coincidentally someone (it might have been Shred Monkey) sent me a PM that said "look at this thread, tell me what you think" referring to the pylon testing people were doing. I basically told them "what a coincidence: I'm working on that now."

So it looked like I looked at what the pylon testing people were doing and took a look at it right away, but actually it took far longer than that for me to figure out anything. I just happened to get an answer around the same time other people were starting to ask the question.

Its a funny story actually, and one I probably can't repeat all the details on. But you have pohsyb to thank for arcanatime. Back in I11 I discovered that my timing measurements for Dual Blades wasn't matching the reported cast times. That led me to do a whole analysis of cast time, root time, and in the process work with BaB to figure out how the animation system worked.

*That* caused pohsyb to ask me this question: "how long does it take for a critter to eat a donut?"** The reason: oh, this little project he couldn't talk about that involved letting players write their own missions. Which promptly sent me on a rather epic analysis of how animations work in general, and that led me back to reexamining my cast time measurements done back in I11 beta. Which led me to discrepancies which caused me to suspect I was seeing evidence of large server-side clock ticks (greater than 1/30th of a second). That led to my arcanatime article.

By the way: I presented my findings to pohsyb, and he said "oh yeah, the combat clock is 0.125 seconds and it aligns to the 1/30th second clock, so you should be seeing about 0.132s server clock ticks." I also presented to BaB, who said "yeah, and?" Then he gave me a no-prize.

It was something of an anticlimactic moment, like the American Indians watching Columbus discover America.***

Oh yeah, and it takes about eleven seconds to eat a donut, although that time is variable because the act of eating a donut has a branching randomness to it.


My recollection was that you were the first poster that plugged arcanatime numbers into an attack chain that was then used to compute an estimated time-to-defeat for a pylon, whereupon it matched your previously observed time. There might have been others that posted partial confirmations prior to that, but yours is the one I remember most clearly.

I think you're the first person to call it "arcana-time" too (it never occured to me to give it a name, which is actually unusual for me).



** Because when pohsyb asked his fellow NCSoft employees "can you help me figure out how long it takes to eat a donut" they invariably said "get the *$#^ out of my office, CW."

*** Columbus didn't originally land on the North American continent, but I'm not changing the metaphor.

I know for sure I wasn't the first person to call it Arcanatime. I just liked it when I saw it, so started using it. I can't remember who first used it, unfortunately.

Numerical error detected.

s/24/240

Alert: Backlog has grown beyond desired parameters.

Ahah! Aracaville was wrong about Werner being the one to call it Arcanatime. She can make a mistake!

(although she'll probably write a response to this post demonstrating how she make the correct assumption based on all of the information at her disposal, which included things most of would have never considered before).

The discovery of Arcanatime was really a significant occasion to min-maxers. Prior to that point the way we calculated was to use published cast times. And from that we accounted the difference between actual damage output and reality by saying oh.. well, there some unexplained ineffecency that costs us about 10-15% DPS.

But what we didn't know is that our builds were actually less then optimal because of what that inefficiency was. For one, we had way more recharge then needed to run a chain tight. With longer cast times, the total recharge needed is often much lower and therefore we can add more 'other things' in our builds. Those 'other things' these days are what gets the sets other then /SR up to soft-cap defense numbers. Previously we hadn't started doing that to our AV killer builds because we thought we needed the recharge.

Furthermore, we now look at some attack chains that we would have thrown away before because their DPS was 'too low.' These attack chains are infact the best chains in most sets. Not using arcanatime, empower>sweeping strikes has, as I recall, 250+ DPS. With the same recharge the BF>AS>SS>AS chain only does around 230 DPS. Getting more recharge is next to impossible, so nobody would do BF>AS>SS>AS. However, the empower chain actually only does about 212 DPS whereas the BF>AS>SS>AS can actually run tight and is still at 230 DPS.

When I went back through all the attack chains after learning about Arcanatime, I believe I ended up finding new better attack chains in MA, DB, Claws, DM, Katana, and Broadsword. That was quite a significant change. Granted it also meant we couldn't do as much damage as we thought we could... but we did learn how to do more damage then we were actually doing.

Here's my little part of the story Arcana was telling earlier. I built shred monkey to use the empower>sweeping strikes chain and tried it out. I discovered to my frustration I wasn't doing the DPS I thought I should. By accident I started trying BF>AS>SS>AS because it just seemed like my attacks were up enough to do that (although by my math they shouldn't have been up fast enough to run without gaps). Unable to explain why I was seeing what I saw, I started plugging numbers in my spreadsheet. The only explination I could come up with was that there seemed like there was a lag of about 0.2 seconds after each power. I emailed BaBs to see if there was an animation at the end of each power 'returning to home' that took 0.2 seconds. He didn't respond. But when I sent the same PM to arcanaville, she PM'd some detail about what she'd been studying (as she mentioned above, that became arcanatime).

And for the record, one might expect an average of 1.5 ticks to be added to each attack from arcanatime. (half-way between 1 and 2 ticks). 1.5 ticks is (1.5*0.132= .198). So my 0.2 guess was really pretty decent for doing all my calculating by just tweaking numbers in a spreadsheet and visually watching my powers recharge.

That's why I slotted my Dark Regen the way I did. The only fuel I need for it is the AV itself.

It heals back half of my health off ONE target, so I can kill everything else off with impunity and it doesn't hurt me that much.

Originally Posted by Dechs Kaison See, it's gems like these that make me check Claws' post history every once in a while to make sure I haven't missed anything good lately.

I actually did the same. The heal is enhanced 84.4%. That's probably enough for most AVs without resorting to minions, but I think it'll be stronger with minions. Figured I'd give myself options.