NoChain : an application to maximize your DPS


all_hell

 

Posted

In retrospective to the above, I'm getting at most 150 dps on my StJ scrapper /w damage calculated at 3 combo points with the finishers included. With arcanatimes active. I think you may need to be more clear on where you get your damage numbers from Mids, because I know thats below what I do in-game.

Another example was my PB doing sub 100 dps which isn't true. Something is going on here that I can't figure out.


Also, see if you can work with the guy who made mids and aquire the sourcecode for RAWR, a program for an mmo that analyzes DPS based on fight parameters, gear, ext... and tells you what you can improve to get more dps out of a build. Though I assume the above isn't very time worthy for a game with sub 1mil subs like CoH.


 

Posted

Quote:
Originally Posted by Archmage MC View Post
In retrospective to the above, I'm getting at most 150 dps on my StJ scrapper /w damage calculated at 3 combo points with the finishers included. With arcanatimes active. I think you may need to be more clear on where you get your damage numbers from Mids, because I know thats below what I do in-game.

Another example was my PB doing sub 100 dps which isn't true. Something is going on here that I can't figure out.
Did you count incarnate powers like Musculature/Reactive when you put in the damage numbers? 150 DPS actually sounds about right for most scrappers before you count incarnate powers, Combat Readiness, etc. NoChain also doesn't appear to know how to handle -res debuffs, so you'll have to account for those yourself.


 

Posted

You know, about that whole mids/NoChain combination to make a CoH RAWR, I wouldn't mind helping with the formulas if we get a team to make this program. I can't program, but I'm very good at math and can do that bit and hopefully just have most programming be "This equals this variable, which is used here in this line which is used here in the final product."


 

Posted

I haven't fiddled with it. But could I put in ALL my attack powers available, and it pick the best chain? IE. put all of DM's attacks, Smite, Shadow Punch, Shadow Maul, Siphon Life, Midnight Grasp, Ring of Fire and Fireblast, and it pick out that Smite->MG->Smite->Siphon Life is the best chain. Or will it try to make a chain with all the powers given?



Your character does not have capped defense. Depending on your AT the cap is between 175% - 225%. Your defense is not teal in the combat window, it can go higher. STOP SAYING IT IS CAPPED! The correct term is Soft Cap.
I enjoy playing in Mids. I specialize in Melee Characters, other AT's usually bore me.

 

Posted

Well, when I tried it earlier with my Broadsword character, it (correctly) found the attack chain of Headsplitter-Hack-Disembowel-(short pause)-Hack. It didn't put Slice into the chain at all, even though I had entered it. So it appears smart enough to not include all the attacks, even in a tricky case like Broadsword where you can get better results by spending .2s doing nothing than spending 2.244s doing a poor attack.
I would assume that it takes longer to calculate and compare the possible chains when there are more attacks to choose from, is all.


 

Posted

Quote:
Originally Posted by _Pine_ View Post
I haven't fiddled with it. But could I put in ALL my attack powers available, and it pick the best chain? IE. put all of DM's attacks, Smite, Shadow Punch, Shadow Maul, Siphon Life, Midnight Grasp, Ring of Fire and Fireblast, and it pick out that Smite->MG->Smite->Siphon Life is the best chain. Or will it try to make a chain with all the powers given?
It will leave out attacks. It seems pretty darn bright, all things considered.

Also, after considerable thought (and looking at mah screen while I hit stuff), I have had a notion concerning AOE's.

Namely:

You can pick one of three levels. The basic level is the AOE damage is multiplied by 2.5 for a 'teaming, noodling around' setting.

Or, you can set AOE damage to multiply by 4.5 for 'I aim with a little care and have friends who play really fast.' For intense play, I think 4.5 is a reasonable place.

Or, set AOE damage to multiply by the powers target cap, for 'I Liek Farmz'.

Just a thought.


 

Posted

Now, will it be able to calculate a chain for my Ill/Cold? :P



Your character does not have capped defense. Depending on your AT the cap is between 175% - 225%. Your defense is not teal in the combat window, it can go higher. STOP SAYING IT IS CAPPED! The correct term is Soft Cap.
I enjoy playing in Mids. I specialize in Melee Characters, other AT's usually bore me.

 

Posted

Tested with Wine 1.3.36 - Works as advertised. Finally a tool for us Linux folks.


 

Posted

When you guys are working out AoE and debuffs what do you do?

Is it just increase the damage by 2.5 for AoE?


 

Posted

This application is also great for other games right..

So don't do the 2.5 for AoE, that is a bum figure anyway.

Add an option to add number of targets hit, that multiplies the damage.


He will honor his words; he will definitely carry out his actions. What he promises he will fulfill. He does not care about his bodily self, putting his life and death aside to come forward for another's troubled besiegement. He does not boast about his ability, or shamelessly extol his own virtues. - Sima Qian.

 

Posted

NoChain 0.15 is now available

Changelog

  • Fixed a bug where setting the power chain size to 1 would make NoChain crash (thanks Hyperstrike!).
  • Changed the whole DPS formula for it to handle even non-trivial power chains.

That last point took me some time to figure out, but with the help of Dark Damor, I've been able to implement a formula that can evaluate the DPS of every possible chain. That means results should now be more accurate, even though I still can't guarantee that NoChain will always find you the same "optimal chain" for the same given powers.

Quote:
Originally Posted by mauk2 View Post
OMG, you should speak to the Mid's guys and get this integrated right into Mids as a popup window. THAT would raise the awesome to QUANTUM AWESOME.
Like I said, that's what I wanted to do. I also know that it's something that the "Mid's guys" want to do. I haven't contacted them yet, but now that both my algorithms seems to be working all right, I might give it some thought.

Quote:
Originally Posted by Hopeling View Post
I would assume that it takes longer to calculate and compare the possible chains when there are more attacks to choose from, is all.
Actually, the time it takes to find the best chain only depends on the size of the chain you want, but not on the number of powers you input. If you want a chain with 4 powers, it will take the same amount of time with 2 or 20 powers as input. Well that's not exactly true, but you get the idea.


 

Posted

Observations/Suggestions:
NoChain 0.15 shows NoChain v0.13 in the window titlebar.
A Save/Load function would increase useability.
If the source code were available under a permissive license, it would be nice.

Questions:
Is there a way to find the optimal sustainable attack chain? If I put the number of powers in attack chain to 100 for example, would it be possible to determine when the attack chain begins to repeat? Or are there sets where the attack chain never repeats?


 

Posted

Quote:
Originally Posted by The_Treasurer View Post
NoChain 0.15 shows NoChain v0.13 in the window titlebar.
Indeed. I was sure I changed it this time (I already forgot to change it in v0.14 as you can imagine), but I did after building the release.

Quote:
Originally Posted by The_Treasurer View Post
A Save/Load function would increase useability.
I'm definitely thinking of something, but I want to see if something can be done with the Mids team first.

Quote:
Originally Posted by The_Treasurer View Post
If the source code were available under a permissive license, it would be nice.
Actually it is, it's under GPLv3, the SVN repo is here. It's Python, and even though my code is far from the best, I try to keep it documented.

Quote:
Originally Posted by The_Treasurer View Post
Questions:
Is there a way to find the optimal sustainable attack chain? If I put the number of powers in attack chain to 100 for example, would it be possible to determine when the attack chain begins to repeat? Or are there sets where the attack chain never repeats?
Short answer : no. I can get into the details if you want.


 

Posted

Great tool! My suggestions for handling AOEs:

Add an AOE checkbox to the powers list (maybe optionally a 'max targets' field).
Then, beneath the "How many powers?" box, add an "How many enemies?" field. When set to higher than 1, automatically multiply the damage of the AOE abilities by the number specified (without going over the power's max targets number) to determine the best chain for variable groups of enemies. So we can put '1' to determine the best single-target chain, '3' for the best chain against a group of 3.. etc.

Additional suggestion: A slot or checkbox for an "aura" power that doesn't factor into the chain at all, but whose DPS (based on one tick every 2 seconds, right?) is added to the total count (just for completion's sake). Maybe just have it be any power you set the recharge to '0' to be considered an aura?


 

Posted

There are probably too many things I take in to account when calculating attack chains to use this sort of tool. Here's my general process, and perhaps some of it can be of assistance. I go back and forth, skip around a bit, redo things as needed, so it can take a while to get right.

General Rule: The fastest attack power gets a purple set. You often use that power multiple times in a chain, and the purple proc can add a lot of damage. I usually skip the Dam/Rech because fast animating powers rarely need more Recharge, and that means the power is moderately overslotted for damage. This power is often a very low level one, so it is still available and may be damage capped when exemping. For Blasters it can still be used when mezzed.

Start by listing every attack power available to a char, including all pools (epic and otherwise). Sometimes the use of a pool power can significantly affect a chain, like Gloom as you already mentioned. Calculate the base DPSA (Damage Per Second of Animation) using Arcannatime for the animation and 95% Damage slotting. That will give you a good idea of which powers are the better ones and which are losers. I also see how they look with a purple damage proc slotted.

Next comes the initial recharge calculations. I generally make purpled builds with Hasten, so most attack powers with slotting have around +200% Recharge Reduction. Based on that, pick the best looking chain.

Next I adjust for damage buffers and resistance debuffers. Claws: Follow Up usually gets used as often as possible. Tanker Bruising powers get used as well. Sonic Blasts make this section hard, but worthwhile to get right.

Next comes other secondary effect considerations. Defense debuff? Ignore. Defense buff? Double stack! Mag 3 Hold? Figure out if you want damage or hold duration. Trollers and Doms with a single target Hold normally get it slotted for Hold due to Hami raids and LGTF Green mitos. Immob? Ignore. ToHit debuff? Maybe slot with an Enzyme later when considering END cost.

Next comes redraw considerations. Which powers have redraw? Generally every one where you draw a weapon in some way. Which powers trigger redraw? Ones where you attack without the weapon. That includes things like Boxing and Gloom. :-( How long does that take? Um... I haven't done one of these chains in a while. I'd have to test it or look it up. You can safely Brawl while using many weapons in hand, but you rarely want to.

Long recharge powers like Soul Drain or Accelerate Metabolism don't really fit in to chains. I run some basic numbers to make sure they're worth using at all, then assume they will be used whenever possible. Most are worthwhile, but not all. Widow Teamwork: Placate gives you an extra crit, but that crit does less damage than you could do if you attacked instead.

Next comes the purple proc slotting. You can only slot a single Hecatomb proc. If you're Dark Melee and your attack chain is Smite, MG, Smite, Siphon Life, then obviously the proc goes in Smite. But if Shadow Punch did damage proportional to Smite, then it would be a better choice as it animates faster. Based on which powers get purple procs, those powers normally get full purple sets, while others get Kinetic Combats, Crushing Impacts, Decimations, and/or Thunderstrikes with other available procs.

Next comes final Recharge calculations. Based on the best chain and the rest of the build, find out how much Recharge each power needs. On powers with purple sets I generally skip the Dam/Rech, but sometimes it's worth doing slightly less damage if it means less time paused. This is especially true if you have Alpha Musculature to make up the difference.

Then comes the final END calculations. Calculate EPS (End Per Second) based on the chain. Frown. Run some numbers and realize that higher damage always means higher END, because the END cost is based on the damage done. Go back to Mids, add the Accolades that give extra END, add Miracle and Numina procs, compare damage EPS plus toggle EPS to your END Recovery. Frown again. Consider going with Alpha Cardiac. Remove the minor damage procs and replace with End Reduction IOs. Be glad you went with Alpha Musculature. Add it into the damage now that you rememer it.

Then comes exemping considerations. Do I really want an EPP Gloom to be a major part of the attack chain on a char I plan to exemp all the time? No. That Alpha Spiritual Core Paragon with 45% global Recharge Reduction doesn't work when exemping. Do you want your Broadsword Scrapper to have Hack as his only attack when level 20 or below, like when he runs WWD#1 every week for that yummy Hero Merit? How about a Tanker with massive global recharge and the attack chain Barrage, Energy Transfer, Barrage, Total Focus. But below level 30 he only has Barrage. Not good. Most melee sets have a melee cone that you can take to help fill the chain at low levels, and it may be worthwhile to use at higher levels too.


Goodbye and thanks for all the fish.
I've moved on to Diablo 3, TopDoc-1304