Changes Focused on Boosting Game Performance

[Azoyhn]

Thanks!
I was hopeful at first, trying to ask what is going on and what will be done to make good on the nerfs as I read and understood the intention of a bow pass to be a buff not a nerf, so I'm looking at it from this perspective.
When I started I was told archers sucked along with bows but I have always found them fascinating (more interesting than a crossbow or throwing dagger - fantasy wise, not DPS in this game)
and always liked them so I wanted to play them here. I understand that I could just switch to dual crossbow and play a much better ranged build but - I like longbows and what is dual crossbows even? lol
Hopefully something positive will come out of this "bow pass" so we can have fun.
I look forward to being able to play an arcane archer.

First of all, welcome to the game - you're going to see this a lot. Second thing, be prepared for the inevitable. I am surprised SoS hasn't hasn't nerfed through all of this. It's kind of ridiculous the sheer power it holds. Either way, they specifically specified in the lammania that was released with Alchemist was that it was 100% balanced and the way they like it - now it's nerfed with a sledgehammer. A few people run Alchemist now, and Gisty was the most popular Alchemist and he's quit the game (not due to switch and bait, but the Alchemist nerf was insanely dumb). But this will be a frequent occurrence just be prepared for the inevitable. Today bow pass, tomorrow SoS nerf to 19-20x2

[iba]

This seems to be an insanely ambitious undertaking. Even if it goes right there will be many negative consequences, the changes are huge. Vorpal, Adrenaline recharge, CC, ki generation, aggro management (maybe some of these have been answered, haven't read it all).

Best of luck devs, I'll be very cautious with my DDO money for about half a year

Given how the change with belts debuffed underperforming archetypes that devs have stated need to be buffed/reworked for years and still haven't had easy fixes to the loss in spellpower there, my trust is near zero on sweeping general changes such as this until there starts to be a better pattern of fixing such issues. Reallly worrying about the epic destiny changes and won't be playing much at all til seeing how HW/ED pans out, let alone spend money on DDO.

[Madja]

Translation : The raid crowd / end game crowd is causing its own lag over which it whines heavily.

This is roughly what i have expected. Those who are the loudest ones about lag are also the source of it.

Yeah, people should just stop raiding or playing endgame if they don't want lag! It's their own bloody fault! Some people have a lot of nerve thinking they can just play the game the way it was designed.. It infuriates me!

[OfElectricMen]

I don't know if this will achieve what the Devs are hoping it will achieve, but mechanically and stylistically, it makes a ton of sense following the history of the game's development. This is just taking us further down the DPS Is The Only Important Thing road that we've been on for a while.

Like someone suggested, I hope you pass out +20 Hearts to all characters and some free XP stones because we shouldn't have to pay actual money to rebuild all the things you're about to trash.

[Tilomere]

Hey how about un-nerfing damage boost so people use that, then deleting haste boost from the game?

[FengXian]

Alright. Let's put it like this then. Why should we leave them alone (which we really aren't because if the range isn't reduced then they get the benefit, which means they are getting affected)? Because they are difficult to acquire maybe? Because some of those items are from raids of a bygone era? Because those bows in particular don't do enough damage? I want to understand your reasoning for being so adamant against this change. You say they could put an augment slot onto them, but is that going to seem like a fair trade to you? Are you going to be placated because of the addition of an augment slot to the weapons? What to you is worth the range that the bows are losing?

You are inverting the burden of proof. We should be given reasons for the nerf, not the other way around.

"Design space" is clearly a lie: not ALL of those bows limit design space.

I would argue that not even very common and/or easy to get bows like Silver, Sinew or Pinion are limiting design space.

But the (very) rare ones definitely don't. Are you gonna dispute this? Are you going to say they can't design cool new bows because everyone will be using Epic Thornlord over them? So why are they getting nerfed?

[Hawklynn]

Other than the bad on the procs change, is there any good that could come from the procs change? Change can be tough but sometimes there are new discoveries.

Originally reading this thread, I also was hit with despair on the % loss of procs. Personally, I’ve been a Proc Lover ever since my first Earthgrab from a GreenSteel. It’s an aspect of the game that allows some unique thinking and hunting to find niches of power.

My post focuses on what good might (or could) come from this change, that are hidden in enhancements. Of course the primary intent to reduce lag will be good if it hits the mark. I’ve not found procs that benefit themselves but with other actions (Adrenaline) they help. Also, some procs could keep the same return, pending developers make a code adjustment.

Stacks is a proc that many have requested to be multiplied. The initial post states this:

“The combat code saves the number of hits and passes it down the chain until it gets to the place where effects actually deal their damage. From there, the effect will understand how many times it needs to be multiplied.”

Devs – Could you (or can the code) multiply stacks by that number (that don’t have a timer)? Procs that go into the queue (I believe) are not increasing. Just the multiplier is being used here.

Examples

- Adrenaline Fury. With DS 100, a Vorpal would return 2 (its 1 recharge * 2). There will be highs and lows of charges coming in.

- Ruby of Endless Night – Vorpal gives 2 Neg levels (its 1 Neg * 2). Again, highs and lows so less even.

Suggestion on DC Roll Procs (bring some procs back to life and some get buffed):

- Earthgrab – 4% chance of Reflex DC 35. Landing that 4% proc on your one hit, and getting Doublestrike, make the DC 35*2, thus DC 70. You’ve got the multiplier of 2, just not sure if the code is difficult to adjust that DC. Weak ones, like Telekinetic DC17 (can become DC34) .. they become new again. Legendary Ice DC100 .. well at DC200, a big buff. It doesn't have to be *2, but there are other equations that can use that 2. (DC *1.2 giving 20% increase)(DC * (1+1/2)) giving a 50% increase).

I’m not sure how to soften the changes on other procs mentioned like Nerve Venom, Vorpal, etc., but I see other suggestions being provided.

[Bluenoser]

From Coco: "An important thing to note when going into this is that when an effects queue happens, it does not only affect your instance, but every other instance on the server that your instance is being processed by. Server in this case does not refer to world, such as Ghallanda or Sarlona or Wayfinder, but the actual machines that process DDO and share the load of each instance between them."

This strikes me as the most important comment of the OP. Lag in one world causes lag in other *worlds*. I assume this is because DDO runs virtual servers on shared hardware, which is cheaper than separate hardware servers for each world. Sounds like the best solution to the lag is hardware upgrades, rather than nerfing all ranged and TWF builds into oblivion. I assume that's not on the table, despite those pretty robust-looking financials we've seen.

Oh, and notice that these changes do nothing to address lag caused by all of those casting-based procs, like EVERY SINGLE CASTING AOE, and things like multiple LGS salt calculations on each Multivial, magic missile, or meteor. Or, as noted above, strikethrough. (Yes, it is a remarkable coincidence that lag got noticeably worse with the introduction of strikethrough and alchemist multivial...)

Overall, what I'm reading here is that the range of viable builds in challenging content is about to be dramatically reduced (it's already far too narrow) to THF, SWF, natural fighting, and casters, and the main reason is that the hardware cannot keep up with what the devs have designed. (On a similar note, the changes to doubleshot that nerf proc effects and Archer's Focus make any bow pass stillborn.)

[Valerianus]

From Coco: "An important thing to note when going into this is that when an effects queue happens, it does not only affect your instance, but every other instance on the server that your instance is being processed by. Server in this case does not refer to world, such as Ghallanda or Sarlona or Wayfinder, but the actual machines that process DDO and share the load of each instance between them."

This strikes me as the most important comment of the OP. Lag in one world causes lag in other *worlds*. I assume this is because DDO runs virtual servers on shared hardware, which is cheaper than separate hardware servers for each world. Sounds like the best solution to the lag is hardware upgrades, rather than nerfing all ranged and TWF builds into oblivion. I assume that's not on the table, despite those pretty robust-looking financials we've seen.

Oh, and notice that these changes do nothing to address lag caused by all of those casting-based procs, like EVERY SINGLE CASTING AOE, and things like multiple LGS salt calculations on each Multivial, magic missile, or meteor. Or, as noted above, strikethrough. (Yes, it is a remarkable coincidence that lag got noticeably worse with the introduction of strikethrough and alchemist multivial...)

Overall, what I'm reading here is that the range of viable builds in challenging content is about to be dramatically reduced (it's already far too narrow) to THF, SWF, natural fighting, and casters, and the main reason is that the hardware cannot keep up with what the devs have designed. (On a similar note, the changes to doubleshot that nerf proc effects and Archer's Focus make any bow pass stillborn.)

oh, they'll change all those things you mentioned too.

and i agree about the hardware not being upgraded vs those robust-looking financials. and agree that in the end it's about what is has been designed with no regard for how it would have worked in potato servers.

i guess the new owner just want to see the cash flow till it flows. oh i'm becoming so pessimistic about the game i love duh, we players are getting so many bad bad bad signals, bad vibes.

[Lynnabel]

the hardware not being upgraded

To be completely clear, the kinds of problems we're trying to solve on the design end are not problems that can be solved by new hardware. We have a robust datacenter, with all of the completely normal industry standard bells and whistles to monitor and observe the metrics of both our baremetal servers and our virtual servers, and all the firewalls and connections and switches and routers that make up DDO as a whole are hooked up to extensive monitoring to prevent and mitigate any potential bottlenecks. If this were as simple as buying more RAM, we'd do it, but it's not, so here we are.

[Tilomere]

That's not exactly how the effects queue works from what I surmised.
The effect would be sent through queue, and then, instead of rolling a save, the boss would say "immune". It wouldn't reduce the amount of times the effect attempts to proc at all.

My understanding is that the filtering of procs to say 10% on-hit from 100% on-hit reduces the proc server load by 90% and what was they were initially looking to do. So the boss ward would simply change it to 0% when the boss is already debuff stacked for that proc type for awhile. The procs would never occur, and never hit the proc server.

So boss goes up to 30 vulnerability stacks (first 20 only used for damage amplification) then boss ward or aura or gaurd or pulse disables vulnerability effects raid-wide for 30 seconds while 10 of those stacks decay. Vulnerability property on gear gets effectively disjuncted or has its proc rate set to 0% for 30 seconds. No more vulnerability procs period in raid for 30 seconds are generated, eliminating server load for that proc type for 30 seconds. After that, say vulnerability stacks up to 30 again in 3 seconds, it again gets turned off for 30 seconds.

Overall would eliminate maybe 90% of the debuff and many damage over time procs on bosses in raids. We simply allow every debuff to overstack an additional 30 seconds and decay one at a time but keep the max effectiveness of those debuffs the same, and deactivate those debuffs by setting their proc rate to 0% for 30 seconds when the overstack is full. Maybe have a boss start with a ward that cuts proc rates in half that updates to the dynamic ward as above to reduce initial debuff spike load.

Then for good measure we nerf warlocks. Because I mean, why not?

-----------------------

They also could just simply release a 15-20x3 Legendary Sword of Shadows.

Or for SWF/TWF remake epic Forrester's Brush Hook to take sentience and have proper damage die for epic versions. Get rid of the Normal/Hard/Elite versions and put 1 version in that works.

[voxson5]

To be completely clear, the kinds of problems we're trying to solve on the design end are not problems that can be solved by new hardware. We have a robust datacenter, with all of the completely normal industry standard bells and whistles to monitor and observe the metrics of both our baremetal servers and our virtual servers, and all the firewalls and connections and switches and routers that make up DDO as a whole are hooked up to extensive monitoring to prevent and mitigate any potential bottlenecks. If this were as simple as buying more RAM, we'd do it, but it's not, so here we are.

Are the server instances/threading bound to the same single core arch design as the client?

[GroundhogDay]

To be completely clear, the kinds of problems we're trying to solve on the design end are not problems that can be solved by new hardware. We have a robust datacenter, with all of the completely normal industry standard bells and whistles to monitor and observe the metrics of both our baremetal servers and our virtual servers, and all the firewalls and connections and switches and routers that make up DDO as a whole are hooked up to extensive monitoring to prevent and mitigate any potential bottlenecks. If this were as simple as buying more RAM, we'd do it, but it's not, so here we are.

Even i were to buy this you're still asking the players to, figuratively, eat the cost. This has all the hallmarks of something hastily put together without much thought put into it, otherwise you would've considered all the ramifications nerfing doublestrike will have on so many builds.

[Lencrennis]

This has all the hallmarks of something hastily put together without much thought put into it.

[IBCrabin]

My understanding is that the filtering of procs to say 10% on-hit from 100% on-hit reduces the proc server load by 90% and what was they were initially looking to do. So the boss ward would simply change it to 0% when the boss is already debuff stacked for that proc type for awhile. The procs would never occur, and never hit the proc server.

So boss goes up to 30 vulnerability stacks (first 20 only used for damage amplification) then boss ward or aura or gaurd or pulse disables vulnerability effects raid-wide for 30 seconds while 10 of those stacks decay. Vulnerability property on gear gets effectively disjuncted or has its proc rate set to 0% for 30 seconds. No more vulnerability procs period in raid for 30 seconds are generated, eliminating server load for that proc type for 30 seconds. After that, say vulnerability stacks up to 30 again in 3 seconds, it again gets turned off for 30 seconds.

Overall would eliminate maybe 90% of the debuff and many damage over time procs on bosses in raids. We simply allow every debuff to overstack an additional 30 seconds and decay one at a time but keep the max effectiveness of those debuffs the same, and deactivate those debuffs by setting their proc rate to 0% for 30 seconds when the overstack is full. Maybe have a boss start with a ward that cuts proc rates in half that updates to the dynamic ward as above to reduce initial debuff spike load.

Then for good measure we nerf warlocks. Because I mean, why not?

-----------------------

They also could just simply release a 15-20x3 Legendary Sword of Shadows.

Or for SWF/TWF remake epic Forrester's Brush Hook to take sentience and have proper damage die for epic versions. Get rid of the Normal/Hard/Elite versions and put 1 version in that works.

They could remove debuff from the game and reduce the scaling of mob hp mrr/prr.

The game has been on a trend that lead to the current situation i.e. -> allow mrr/prr debuff going negative -> mob die too fast -> increase the hp + prr/mrr of newer content. cycle repeat with every effect they add. This started with the introduction of vulnerability.


I am trying to figure out why they added crown of summer as a modifier after Lyn said they don't want to have any melee or range power on item going forward.

They should remove the range / melee power modifiers from items in this update too because they have become special case like the enhancement spell power, threat range.

[Lanadazia]

Doing this would re-introduce the effects queueing problem. We are trying to reduce the number of on-hit procs. Applying them multiple times is not an option.

No offense but stop asking for procs chances to be increased. The WHOLE point of this change is to reduce the the number of procs being generated.

i understood it this way: every single bit of damage (like weapon damage, elemental/alignment procs on hit, etc) are each calculated on their own.
so if a weapon had 2 extra on-hit effects like flaming the damage would calculate in 3 different procs: 1. weapon damge 2. proc1 3. proc2
that'd be 3 calculations on one hit, right? so a multihit would just increase those damage numbers times the multihit number, instead of calculating 6/9/12 things

i thought if all of the current procs that occur on each hit would be condensed to a single number, that'd leave enough room to boost occurance of vorpal/percentage procs.
if thats not the case, i'm sorry, i got it wrong then.

if this is how its going to work, how are you going to alternate the vopral/% procs to compensate for the flat nerf of them?

damage proc x multihit-value -> yeah.. that'll work on raids, but is most likely an overkill everywhere else, making that proc almost useless. but i guess there is no workaround here
debuff proc x multihit-value -> just let them apply more stacks?
DC procs x multihit - value -> what is going to happen to them? like blackrazor's slay living? are we just gonna miss out on ~ half the proc chances? a higher DC maybe?
CC procs x multihit - value -> how is this going to work out? stunning, paralyzing? higher DC?

this still looks unfinished
and i think alot of builds and especially items will lose their special something

are the vorpal/% procs really the problem here? you said you got to ~50% framerate with the changes instead of ~99%
so there would not be room for those specific procs to work the same? or are those procs actually taking roughly 50% of the performance?

[Xgemina]

Will this boosted game performance, or the bow update, fix the phantom missiles issue that I see when using bows or crossbows? What I mean is: you have a mob targeted (soft or hard), you have a clear LOS, you go through the animation to fire your weapon, but nothing happens - there is no hit roll nor any bolts/arrows stuck in the ground or invisible wall. I see it more often on repeaters, but it happens with bows and dual crossbows as well.

[Tilomere]

They could remove debuff from the game and reduce the scaling of mob hp mrr/prr.

Ya, a lot of procs are irrelevant, and could be allowed to be swapped out on items to a DR bypass type or seasonal set bonus. They could also just have a ward that becomes active in all dungeons and reduces all procs by a % if the server starts becoming overloaded.

[OfElectricMen]

What I'm seeing in this thread is that the Devs haven't been clear and concise enough on explaining how things work so people can more fully understand the changes being proposed/taking place.

You can tell from all the "I surmise" comments.

[rabidfox]

What I'm seeing in this thread is that the Devs haven't been clear and concise enough on explaining how things work so people can more fully understand the changes being proposed/taking place.

You can tell from all the "I surmise" comments.

Proprietary information + keeping info about aspect of software/hardware for security/other reasons is pretty typical for any MMO so guessing about certain things is to be expected. There's whole aspect of stuff that players have "worked out" about reaper and from what devs have said, we still haven't figured it all out on the mechanical side.