If 2% Resistance equals 1% Defense, How much Regen is equal to 1% Defense ?

I think most people would rather have a simulator with programmable options that they could set-up, run, and view results. Allow such a program to be given input for how many iterations you want to run, which mob group you want to simulate and how large a spawn to start with. Plug in your character data and punch a button.

Hmmm, I guess my question for you Arcana is whether my numbers (previous page) match what you expect to see on a standard model.
I may not be the only one that wants to hear your input on the regen values presented thus far in the thread.

One approach that I have been working on (not with any real pressing need to present because of how harshly these things are viewed), is similar to your "sawtooth" models, where you have the X-axis as Time, the Y-axis as Incoming damage, and as you defeat your spawn group over time the graphical line result eventually reaches zero, meaning your spawn is defeated. Character effects which increase DPS against the spawn will bring the X-intersect closer to zero, while lowered character DPS will stretch out the Time (X) axis and make the X-intersect move farther away from zero. Effects which increase survival by lowering incoming DPS will pull the graph down in the Y axis and the Y-Intersect would be the start of the fight and is a representation of the alpha strike. The models I have been looking at so far are: 3-minion, 6-minion, and 9-minion. Once the spreadsheet is fleshed out, I would expect that you could program any number of starting combinations and view the graphic results.

I realize that this still amounts to statistical averaging, but two things make it different.

A) You get a graphical representation of a spawn battle and can easily compare two "builds" overlapping each other to see the diferences. Basically, your survival is not a single flat number anymore, but is instead a descending line from alpha to defeat.

B) So far I am seeing some of the iterative events that you have discussed (for regen in particular) where at certain time intervals, you can see the graph rise and fall. Heck even the regen pulses create small ripples in the graph.

Any thoughts ?

Yeah. Think one boss, two Lts, and five minions, and you're in the ballpark of the idea I'm thinking of (the spawn composition used by Circeus' Ice tanker comparisons). Then add in a two-phase computation: an average one for minion damage, and a more stochastic one for boss (higher) damage. And combine the two with a methodology that essentially uses minion damage to estimate health trends and boss damage to calculate the chance of being killed per attack superimposed on the varying health due to minion damage. Sort of a "constant damage slowly whittles you down, and at some critically low level one big hit kills you." That's sort of how scrappers tend to die now, actually. But to make it work with the simplest possible calculations is the challenging part.

Also:

I wrote one of those, but it was in python and text-only, with some ugly graphs tacked onto the side with duct tape. I'm too lazy these days to write a full-blown windows app for that kind of stuff. Also, to generate histograms you had to run tens of thousands if not hundreds of thousands of simulation runs to generate reasonably smooth curves. The results of that sort of thing are in my Scrapper analysis. Some involving millions of simulation runs.

A real windows app that did that, but updated and with more complex conditions; that's more the sort of thing that my European counterpart would have been up to. Dr. Rock, who I once heard described as "the Arcanaville of the Euro boards" (presumably in terms of damage mitigation, not insanity) actually wrote a windows app that did, way way back in I6, more or less what my powerset proliferation spreadsheet did in I10 in Excel, and my scrapper comparisons in I2, I3, I4, I5, and I7, but in a Windows app with graphs. But they too used average equations. In fact, Dr. Rock's equations are essentially identical to the ones I first generated for my scrapper comparison analyses from I2 to I7, but I understand he independently generated them for his own purposes.

Also, a quick search on the internet shows that his comparison app is still available for download and was last updated for Issue 11: http://dr.rock.coh.tripod.com/. Its a nice app: sort of a hybrid between Mids and my damage spreadsheet. It still uses average calculations, but a lot of the stuff my damage spreadsheet has is also there: damage type proportions and defense/resistance/regen calculations, and lots of combat assumptions I discuss but do not calculate around in my scrapper comparisons. Its not a well-known tool (and I don't know why, besides the fact that Dr. Rock was a European board member), but it is an interesting one if for no other reason than to read Dr. Rock's calculation methodology and his combat calculator options, all of which are reasonably well founded.

What if you just have the calculated chance of "insta-defeat" below your averaged spawn graph at certain intervals. So you could visually see the graphic of averaged damage from left to right (Alpha to defeat) and the %chance right below it, also running left to right.
For instance, let's say the alpha damage averages 30% of your health. Your first point (0,30) is part of your graph, but right below the graph at zero is a (16%), representing that you have a 1 in 6 statistical chance of being defeated outright(based on bad luck). You would have to program every mob type into the spreadsheet to allow the alpha spawn to be tailored to real in game damage types and frequencies to make it the most relevant, but I think most people would be happy with some form of "Odds of Defeat" to contrast with the averaging method. I could be wrong.

Thanks for the Tip. I will check this out as soon as possible.

It depends on how much damage is incoming. That's the point. You can compare theoretical amounts of Defense and Resistance and see how well they'd do against varying amounts of incoming damage. But as soon as you start discussing Regen, you need to know a set amount of damage that is coming in before you can really see how it will be effective.

I think I can help you out.

I've got a survivability analysis tool that will help you see how much your regen is doing for you. It's also a great tool to compare similar builds and see which gives you the most survivable incoming DPS.

The link is in my signature.

Thanks. Will definitely give it a look. [nice tool; will come in handy]


(I know that there's not a magic bullet when it comes to giving an overall avg comparison but I believe an average DPS by rank can be determined across the board (at least per Enemy Faction) and that can be levied against DR+Base Regen towards figuring out immortality lines (at various DPS points) and that these lines can be compared against the immortality lines for just Regen to come up with an Res vs. Regen approximation... at least for a predetermined DPS range per faction... could be wrong though)

OMG there should totally be a big game hunter badge