Combat Feedback for Update 5.

[Eladrin]

One of the major issues that we’ve been working on is the dps lag problem in high level content (especially raid content). We’re attacking this problem from many angles, and since some of the changes we’re considering will have significant impact on many characters, we’re interested in your feedback before it goes to Lamannia or live. We're trying to aggressively attack the issues that cause these lag problems and are interested in feedback regarding how aggressively we should be attempting to tackle them.

My post is a bit more technical than I normally write for public consumption, but I want to make the issues and the reasons we’re considering these changes extremely clear.

Some people may have noticed that problems occur most often when there are many monks or two weapon fighting characters that are loaded with many on-hit effects attacking en masse. These characters currently perform many attacks very quickly, and each attack is pretty heavy performance-wise. Rather than attack the problem by changing common heavy-load equipment, we’ve already taken some steps to optimize the way we perform attacks, and are considering the following changes to reduce overall performance loads:

Currently a single two weapon fighting attack makes a physics detection check with your main hand, followed by a second detection check for your off hand (roughly 0.15 seconds after the first one). Instead of making multiple physics checks, all two weapon fighting attacks (weapons or unarmed) would now make a single check for your main hand attack, and would “piggyback” on that detection check and have a chance to proc (trigger) an off-hand attack based on the number of two weapon fighting feats (or related enhancements) you possess. Off-hand attacks would have a chance to proc on any main hand attack now, instead of being predetermined on certain attacks. Having more TWF feats increases the % chance of proccing an off-hand attack.

We would now assume that if your opponent was within range for your first attack, it will still be in range 0.15 seconds later. We also go about determining whether or not a particular swing gets an off-hand attack in a much cleaner manner than before.

We’re also considering introducing a new mechanic to replace some speed bonus effects, called double strike:

A character with a double strike chance has a chance to make an additional attack roll with their main hand weapon any time they make a main hand attack, on the target of the first attack. (Note that two handed weapons count as “in the main hand” for these purposes.)

The Fighter Alacrity capstone, and the Paladin spell Zeal would be changed to a +10% bonuses to double strike.
A TWF Paladin that Smites something under the effect of Zeal would attack with the main hand, have a 10% chance to attack with the main hand again, and would have a chance (based on TWF feats) to attack with the off hand in one sequence. Level 20 Fighters with the capstone would simply always hit twice 10% of the time with their main hands. For a future update, we’re considering revisiting some old items (such as Jorgundal’s Collar, which currently doesn’t stack with Haste) and replacing the effects with a double strike effect.

For those that are interested in the “how does that help? You’re still making 11 attacks in 10 swings” – double strikes, like the proposed off-hand attacks, would bypass the additional physics detects and secondary characteristics of normal attacks such as glancing blows, while still providing an increase to overall damage over time. It would have the advantage over speed boosts of sometimes proccing on special attacks, and opens up a new itemization and enhancement pathway. (Warchanters, for instance, are likely to end up with a double strike song in a future update.)

Proposed Numbers:
A character with no two weapon fighting feats has a base 20% chance to proc off-hand attacks.
The Two Weapon Fighting, Improved Two Weapon Fighting, and Greater Two Weapon Fighting feats grants a +20% bonus to proc off-hand attacks.
Monk Air stances now grant a +2.5%/+5.0%/+7.5%/+10% insight bonus to double strike chance instead of an insight bonus to attack speed. (Enhancement bonus to attack speed remains.)
Ranger Tempest I and II now grant a +10%/+20% bonus to the chance to proc off-hand attacks when dual wielding instead of a bonus to attack speed. Ranger Tempest III now grants a +5% double strike chance while wielding two weapons.
Monk Air Stance and Ranger Tempest I and II now stack with each other for these purposes.
Attacks with two handed weapons while moving no longer perform Glancing Blows.

Fighter Alacrity (capstone) now grants a +10% bonus to double attack with the main hand instead of a 10% speed increase.
The Paladin spell Zeal now grants a +10% Sacred bonus to double attack with the main hand instead of a 10% speed increase.

Things like the Shield Bonus and to-hit penalty reductions that Tempest grants when wielding two weapons, or the enhancement bonuses to speed from Wind Stance remain unchanged.

For example, a Ranger 18 (Tempest III) / Monk 2 in Wind Stance I would have a 100% chance to proc off-hand attacks per swing and a 2.5 to 7.5% chance to double strike (depending on whether they are fighting unarmed or with a pair of kamas), while a completely unskilled character would only generate off-hand attacks 20% of the time. A Monk 20 in Wind Stance IV would have a 80% chance to trigger off-hand attack hooks, and a 10% double strike chance.

I won’t hide that these proposed changes do reduce the effectiveness of off-hand attacks, which reduces the two weapon fighting style’s extreme dominance over two handed fighting. Skipping the extreme outlier of the Epic Sword of Shadows (we got a little carried away there, didn’t we?), estimated damage output over time for the two styles should be extremely close to each other if we go this route. (Which style is superior ends up being heavily dependent on your character's abilities.)

It’s likely that we’ll change Glancing Blows to operate in a similar manner in the future as well (chance to proc based on feats/enhancements rather than guaranteed on swing X), but right now I have to stay pretty focused on TWF and monk unarmed combat.

These are not the only things we’re currently investigating, and will not solve all of the dps lag issues - it’s one of many steps we’re taking to pursue them. I also highly recommend reading this post and thread - especially the part about intra-DAT fragmentation[step 3] if your install is a year or more old - if you’re suffering serious problems, it may help with some client-side issues.

We wanted to get your reactions and comments to this now though, before (and after) it hits Lamannia.

---

Some sample attack percentages with this system (assuming you took all possible available feats and enhancements) would be:

INVALID CHART FOLLOWS:
Build				Main-hand	Off-hand
20 Ranger (Tempest 3)		100%		85%
20 Monk (Wind Stance 4)		100%		75%
20 Fighter (Alacrity)		110%		55%
20 Paladin (Zeal)		110%		55%
20 [Other]			100%		55%
12 Ftr/6 Rng/2 Mnk		100%		70%
15 Pal/3 Mnk/2 [Any] (unarmed)	110%		60%
14 Pal/6 Rng			110%		65%
18 Rgr/2 Mnk			100%		90%

---

Edit:
Thanks for the massive amount of feedback all.

We're currently thinking of adjusting the numbers to:

	Doublestrike	Bonus	Main hand	Off hand
No feats	0	20%	100%		20%
TWF		0	+20%	100%		40%
ITWF		0	+20%	100%		60%
GTWF		0	+20%	100%		80%
Tempest I	0	+10%	100%		90%
Tempest II	0	+10%	100%		100% 
Tempest III	+5%*	0	105%		100%
Wind IV		+10%	0	110%		80%
Zeal		+10%	0	110%		80%
Alacrity	+10%	0	110%		80%

* Only when wielding two weapons.

This set switches Wind Stance and Tempest III to doublestrike bonuses, increases the benefits of the TWF feat chain, and adds an additional feat for high BAB characters.

All of the bottom rows assume that the person has GTWF.

Edit 2:
Updated information in the main post to reflect the most recent changes.

[GoldyGopher]

I'm not sure I am reading this right, but in general what I am seeing is you are going to Nerf TWF by reducing the output of the offhand while increasing the procs in the mainhand?

Don't get me wrong, its not like I would totally disagree with what you are doing, but ...

[le_goat]

How about you nerf the greensteel and not the monks.

[Mockduck]

Interesting! If this largely solves DPS lag I know I'd be willing to accept significant changes to the mechanics of TWF. I'll be very interested in testing this. Say, on Lammania. In the new Update 5 quests. hint hint...

[Gercho]

Let me see if i understand correctly, before a paladin (or fighter) with full twf line would have 100% of off hand attacks and now 55%? so, you are nerfing off hand attacks 45% for the paladin (or fighter)
Besides, the 10% double attack now will only affect the main hand, before a 10% attack speed increased main hand and off hand attack speed... i m ok with consolidating check for lag controlling, but why the nerf?

Full twf should have 100% off hand attacks, and double attack chance should procs for main hand and off hand (depending on how many twf feats the char has, so for a full twf char it will proc for both attacks 100%)

Keep ranger tempest bonus a 10% to double attack as paladin zeal and fighter capstone, if you want you can give extra 5% for tempest II and III...

[Sirea]

I don't know if I like the part about the attack speed bonus being taken away from Tempests and especially Monks.

I really feel like I'd have to actually test it to see the true impact, though.

[Daehawk]

It seems interesting and I would be interested to see how all this plays out. With you revealing this pretty heavy change before the weekend I'm going to assume that you want to hear the voices over the weekend to decide on if it will hit Lam when U5 hits Lam next week?

One question though about the Double Strike. So you're saying that basically with that for every strike we make with our Main Hand there would be a chance to make a second strike with the main hand as well?

Double strike will be a bonus given to Paladin Zeal, Figher Capstone, and Jorgundal’s Collar(possibly and other items that give Melee attack speed?)

[Eladrin]

I'm not sure I am reading this right, but in general what I am seeing is you are going to Nerf TWF by reducing the output of the offhand while increasing the procs in the mainhand?

It reduces the number of permanent speed increases in the game that are doing nasty things behind the scenes, significantly reduces the overhead in a bunch of the calculations, and makes the off-hand more unpredictable than before. It is also (on average) reducing the frequency of off-hand attacks.

How about you nerf the greensteel and not the monks.

I am actively seeking to avoid making broad changes to existing treasure systems.

[Irinis]

Wow, I love the double strike thing, will be especially cool for Warchanter. *squeee*

Interesting thoughts on offhand attacks, however tempest and wind stance should still have faster attacks somehow. Just more offhand attacks in the chain doesn't seem quite right for the concept of a whirling fast class enhancement. I'm sure you'll be testing how much efficiency you get from the reduced computing requirements, but please don't take away the speediness of the attacks! The reduction should already help a lot.

Plus, some people get DPS lag and others don't, so part of it may be the way the combat log is handled going from the server to the client. If the information was split before being sent to the client (I read somewhere that ALL combat info gets sent out then is processed on the client side for what to display in the log) that may also reduce the problem significantly.

[Zenako]

Interesting approach, which accomplishes a number of goals at once. Reducing the shear number of calcs to process when fighting, but at the same time tweaking the fighting styles to become closer to each other in the good ole dps metric that drives many build decisions.

The downside from a PR standpoint is that most players will probably not really get the benefit (since the lag was most pronounced in top end raids with top end builds and gear all getting in the same group and in sync) but will get a reduction.

From a deisgn standpoint it seems like a prudent choice. (Be it known I have characters of all styles, including many TWF'er so it will affect me.) I also have a few classic S&B tank types as well. While details of other factors or proposed changes would be nice but probably out of scope, at least some broad indicators of other things being worked or looked at would help put it into context.

Thanks for asking for feedback.... I am sure you will get some.

[moorewr]

Interesting! If this largely solves DPS lag I know I'd be willing to accept significant changes to the mechanics of TWF. I'll be very interested in testing this. Say, on Lammania. In the new Update 5 quests. hint hint...

One of the things I see in Eladrin's post is that they are focused on keeping the result (in terms of off-hands procs) in the correct flavor for D&D. If this comes close to that goal the technical methods used to get there are not important to us players.

Devil, meet details, but still, I certainly support where they are trying to take this.

[Omega2K]

I like the part about cutting the physical contact check in half but the rest of it I do not agree with... I don't even play many two-weapon fighting characters but you will upset a good portion of your player base by gimping them in your proposed manner. What I would recommend is reducing the number of dice rolls and references to your random number generator table. In raid situations where lag is prevalent, you should instead duplicate previous rolls for additional swings. The number of times you duplicate it will increase performance (decrease lag). Example:

First swing:

physical contact check, to hit roll: 15, 1d10 damage roll: 6, 1d6 pure good roll: 3, 1d6 acid roll: 5, and so on

The outcomes of the above rolls are applied to your next two, three, four, etc. weapon swings producing the same to hit and damage output

New rolls are performed after the last duplication for the next sequence of to hit and damage results

I feel that this would be a better solution as it can quickly cut the number of rolls in half, down to a third, down to a fourth, etc. while leaving the rest of the game basically untouched.

Try it out and see what the Devs think.

[Eladrin]

It seems interesting and I would be interested to see how all this plays out. With you revealing this pretty heavy change before the weekend I'm going to assume that you want to hear the voices over the weekend to decide on if it will hit Lam when U5 hits Lam next week?

I want to see both "instinctive feedback" as well as actual gameplay feedback once it his Lamannia. I can't say when Lamannia will be going live with these changes (that's beyond my control), but I figured better to get this out early.

One question though about the Double Strike. So you're saying that basically with that for every strike we make with our Main Hand there would be a chance to make a second strike with the main hand as well?

Correct.

[MrkGrismer]

I like it although it would seem to me that Tempest Ranger should have 100%/100%, if it is to keep its place as the example of 'best/perfect twf'.

I actually got the DPS lag thing one night in Devil's Assault (normal, believe it or not), and suffered Lag-death as a result.

[Coldin]

Eh....I really have reservations about this. I do like the efforts made to reduce DPS lag though. Just anything with random chance tends to be more against me than for me. How it is currently, you get that guaranteed attack roll with an off-hand.

I'm wondering why TWF grants a larger bonus than ITWF and GTWF. To me those should be more equal.

Monk air stance is interesting. Right now it gives a bonus to attack speed, but this change seeks to make air stance more of a TWF only doesn't it? Not that there's probably many monks that are using air stance and only single wielding kamas or using quarterstaves.

I suppose what I really don't like is overall attack speed being decreased. Attacking slowly is just...more boring. I like being a whirling blur as a monk in air stance with haste.

A couple questions for you Eladrin. You mention Zeal adding 10% chance for an extra hit with the main hand. Would this extra hit also deal smite damage (if that hit was a smite). Would it roll a separate d20 to determine if it's a hit?

And just to make sure about something else. These off-hand chances. When they do proc, do they automatically hit then?

[maddmatt70]

Why do rangers get such a higher offhand attack and fighters and paladins do not? That is not the way DDO works currently. It would probably be a good idea to post the current numbers next to the proposed numbers so people got a better visualization of the changes.

[rest]

Helping DPS lag by reducing the amount of DPS?

That sucks.

[Hutoth]

If you're also trying to equalise the playing field re TWF (which is secondary to fixing DPS lag), I'd prefer a boost to other styles, rather than a nerf on TWF.

[Vhlad]

would a double striked main hand attack have a chance to proc an off hand attack as well?

would a double striked main hand attack have a chance to proc another main hand attack?

[Gercho]

So, translating the actual attacks to the new system, you have now

Build				Main-hand	Off-hand
20 Ranger (Tempest 3)		110%		110%
20 Monk (Wind Stance 4)	dont really know
20 Fighter (Alacrity)		110%		110%
20 Paladin (Zeal)		110%		110%
20 [Other]			100%		100%

[/QUOTE]

And you will have in the future with your proposed changes

Build				Main-hand	Off-hand
20 Ranger (Tempest 3)		100%		85%
20 Monk (Wind Stance 4)		100%		75%
20 Fighter (Alacrity)		110%		55%
20 Paladin (Zeal)		110%		55%
20 [Other]			100%		55%

So you are effectively nerfing all the twf, and paladins and fighters get the biggest hits... (well, and tempest 1 too)
You really nerf paladins even more cause for paladins smites and sacrifices allways went twice, now they will got half the time...