Should we stop using the 21mil dummy for measiring DPS

rpa

21mil dummy

I go and kill something which fights back to get a vague idea where I am.
If I ever get around to do prog trials I do what the group requires, even if it involves bashing a stupid dummy.

FeedbackOnly

3mil dummy

GloatingSwine wrote: »

Depends what you're measuring it for.

If you're measuring for trials then you should use the 21M dummy because you are more likely to have most or all of those buffs most of the time.

If you're measuring for dungeons you should probably use the 6M one because you can't guarantee any buffs you don't bring yourself due to it only being 4 man and not an organised group,.

Elemental catalyst isn't with every trial group

Tannus15

21mil dummy

The 21m is by far the best option as long as you understand what is going on.

I use the 3m and the 6m to test sustain or a build that needs to work without support like if i'm going to do vMA or vSS HM Portals

The most obvious example of a buff that you always get in any group situation is major and minor breach. you won't get this on a 3m or 6m dummy without altering your build and it makes a HUGE difference. So for all stam sorcs suddenly crushing weapon looks far superior on a 3m parse compared to any other spammable. Unless you're also running razor caltrops, except the meta is to run anti-cav caltrops in content because you don't need that major breach.

Same for sub assault vs deep fissue on a warden.

This is just one easy to see example of the issues we had parsing on 3m and 6m dummies "back in the day" before the 21m

sbr32

21mil dummy

ccfeeling wrote: »

No buffs the best, i want the real data

More like real ambiguous data.

Is the parser getting Major/Minor Breach? From themselves? From a friend helping?
Is the parser getting Minor Berserk? From themselves? From a friend helping?
Is the parser getting sustain help from a friend?

A 6 million HP dummy parse won't answer any of those questions while they are completely irrelevant on a 21 million dummy parse.

AlterBlika

I don't actually parse but like clicking on things.

I never liked the idea of parsing on a dummy. It neither shows you how you're going to perform in actual content, nor does it show your skill.

Bald_templar

21mil dummy

6m/3m heavily depends on what buff set and debuff you are running and like different class, different spammable provides debuff or not (while many of those debuff such as breach is provided by the tank) ..21m is more likely to provide comparable results between classes and sets

ccfeeling

6mil dummy

sbr32 wrote: »

ccfeeling wrote: »

No buffs the best, i want the real data

More like real ambiguous data.

Is the parser getting Major/Minor Breach? From themselves? From a friend helping?
Is the parser getting Minor Berserk? From themselves? From a friend helping?
Is the parser getting sustain help from a friend?

A 6 million HP dummy parse won't answer any of those questions while they are completely irrelevant on a 21 million dummy parse.

Yeah , I get you .

B/C the lag issue , I don't play trial forever ( 1G fiber network won't help this game )

I usually PUG , so my build , sets option , CP setting tend to good at solo . Sustain , survivability not a really big problem unless the boss can 1 shot or 2/3 hp a hit .

The number I Major breach myself , swap to shield / healing skill in real fight , I don't use minor berserk b/c I move a lot and I feel the game mechanics not easy to stick at the back all time ( non pre-made group ) ... In PUG , everything could happen , like tank and healer go down , I build myself able to finish the rest or rev them at worst moment .

I know I am not a good dps , seems like a avg all around dps .

If the result lower than 25k , I have to modify some of my setting .

TBH , I think 25 - 30k good enough to complete all 4 man contents even hm , just play longer tho .

I didn't get invite the shipwright's regret group yesterday , the group leader was asking for 100k dps , seriously ? They play non-hm

RedTalon

I don't actually parse but like clicking on things.

Start using the best tank you know to test damage.

preevious

other dummy

Wich brings another question, actually.

70k was considered to be the number to get to to tackle non-HM vet trials ...
With the additionals debuffs on the trial dummy .. what would the new number be?

Varana

21mil dummy

The changes to the dummy are mainly there to mask the overall loss in damage. Generally, the number shouldn't change much.

GloatingSwine

I don't actually parse but like clicking on things.

Remember that the big point of a DPS dummy is to let you assess the impact of changes you made to your build in a fixed set of conditions.

FeedbackOnly wrote: »

GloatingSwine wrote: »

Depends what you're measuring it for.

If you're measuring for trials then you should use the 21M dummy because you are more likely to have most or all of those buffs most of the time.

If you're measuring for dungeons you should probably use the 6M one because you can't guarantee any buffs you don't bring yourself due to it only being 4 man and not an organised group,.

Elemental catalyst isn't with every trial group

No, but the 21M dummy isn't intended to perfectly represent what any one given group does, it's intended to be a fixed baseline which you can use to determine the effects of changes in your own activity (build, rotation, etc), and "just turn everything on" is a reasonable amount of time and energy to spend on making that.

Klingenlied

21mil dummy

21mil dummy still makes the most sense. It makes for the easiest comparison and everyone with at least half a brain should know that this dps does not apply to 4man or solo content.

And in regards to "parse monkeys": believe it or not, people that take their time parsing and practicing rotations very likely will outperform you in every content. There is a major difference in players actually "playing" the game and those assuming some theoretical knowledge alone will get them results. Not saying parsing is everything. But I don't know any player that is doing hm achievements that isn't also doing strong parses.

EnerG

3mil dummy

3 to 6 mil is what I think we should move too there's too many buffs on the 21 now for "realistic trial conditions" and if you can do good dps self buffed I know you have a good build and rotation and that it will only go up in an optimized group.

Iron_Warrior

6mil dummy

If i have to choose only one dummy to parse from now on, i'll choose the 6mil. Try your best on it and then optimize your build when you go into the group content based on the buffs and debuffs that are getting provided

SirLeeMinion

other dummy

Unfortunately, the 21M is not realistic for most groups. It may be that the very best high-end groups can get close to the buffs it provides. However, most trials groups are well below the up-times provided. Dungeon groups are even further below it. At one point, the 21M was useful as a standard. But a standard needs to remain fixed or it becomes meaningless for comparison. They have, unfortunately, compromised this as well with U35.

sbr32

21mil dummy

I don't understand why so many people think the dummies are supposed to represent anything like real game play situations.

The buffs on the trial dummy are to standardize common buffs to make parse comparisons easier, not to simulate real combat.

jaws343

SirLeeMinion wrote: »

Unfortunately, the 21M is not realistic for most groups. It may be that the very best high-end groups can get close to the buffs it provides. However, most trials groups are well below the up-times provided. Dungeon groups are even further below it. At one point, the 21M was useful as a standard. But a standard needs to remain fixed or it becomes meaningless for comparison. They have, unfortunately, compromised this as well with U35.

The standard is fixed though. It isn't like the dummy is changing weekly. It changed once, between patches.

All damage you do today against the dummy can be compared to the damage you do tomorrow against the dummy, and every other time against the dummy.

Allowing you to compare damage, within this patch, at ideal settings. IDEAL settings. Not likely settings in a trial, but it removes all the variables that could levy against a parse and puts everyone on the same footing buff wise. It is the base, raw output you can do based on the familiarity of a build and rotation.

So, it is absolutely fixed. If you are trying to compare your parse post patch against your parse last patch, you are already failing. You should be comparing your parse this patch on the 21M against the optimal parse for the class and build on the 21M from this patch.

Gaebriel0410

Wouldn't it be best to measure DPS on a dummy without any dummy added buffs and without other players assisting?

Since I can imagine that it can look really discouraging for a new player who wants to get into the endgame scene, when you hear about players doing 120k easymode all the time (or whatever the high numbers are) while doing a much lower number yourself, not realizing how much DPS is actually inflated by all those buffs on the largest dummy.

Like, I have played for years myself and I do vet content semi-regularly, but this is so badly explained in the game that I only realized some months ago how much of a difference those dummies can make numbers wise.

I mainly play as a tank but go as DPS offspec now and then as well, so I am not someone who regularly uses dummies really. I have never had complaints about my deeps, and I always measure up fairly in group damage contribution. I even get told I do good damage now and then, so I didn't really worry much about it, and it's my offspec after all.

But often seeing those super high numbers thrown around, I always just assumed the majority of players were a million times better at DPS than I, while the difference may be quite a bit smaller than I thought.

So I think unbuffed dummies would be a lot more "community friendly" in measuring damage, if that makes sense, since I was probably not the only one who didn't know.

jaws343

Gaebriel0410 wrote: »

Wouldn't it be best to measure DPS on a dummy without any dummy added buffs and without other players assisting?

Since I can imagine that it can look really discouraging for a new player who wants to get into the endgame scene, when you hear about players doing 120k easymode all the time (or whatever the high numbers are) while doing a much lower number yourself, not realizing how much DPS is actually inflated by all those buffs on the largest dummy.

Like, I have played for years myself and I do vet content semi-regularly, but this is so badly explained in the game that I only realized some months ago how much of a difference those dummies can make numbers wise.

I mainly play as a tank but go as DPS offspec now and then as well, so I am not someone who regularly uses dummies really. I have never had complaints about my deeps, and I always measure up fairly in group damage contribution. I even get told I do good damage now and then, so I didn't really worry much about it, and it's my offspec after all.

But often seeing those super high numbers thrown around, I always just assumed the majority of players were a million times better at DPS than I, while the difference may be quite a bit smaller than I thought.

So I think unbuffed dummies would be a lot more "community friendly" in measuring damage, if that makes sense, since I was probably not the only one who didn't know.

Not really.

As it has been explained by many people in this thread, different classes bring different buffs, naturally to the table. However, some of those buffs will be completely irrelevant/redundant within a raid environment. Like penetration buffs.

So, say your parse against a unbuffed dummy is on a class/build that has extra penetration, your parse will not be accurate and you will be doing less damage in a situation where the tank and other factors are fully penetrating the enemy. Likewise, a build or class with less penetration will see a smaller parse number, but in a full penetration environment, their parse will be higher.

Amottica

21mil dummy

ccfeeling wrote: »

No buffs the best, i want the real data

@ccfeeliong

Since the dummy in question relates to trials where buff would be provided, it is real data. A dummy without the buffs, especially the sustain assistance, would be fake data as it forces a player to do something different than they would do in a trial fight to help with the sustain.

Beyond that, the only real data comes from a real fight.

Oreyn_Bearclaw

21mil dummy

I would not be opposed to say a 10 mil dummy that tosses non damage synergies, with no armor rating, and gives Hircine's and Worm for sustain. I think that could be useful for a lot of players. In other words, no damage buffs, but you should be able to reasonably sustain and you wouldn't need to worry about penetration. Rest is up to you. (but understand, this would over and underinflate certain classes/specs)

FeedbackOnly wrote: »

GloatingSwine wrote: »

Depends what you're measuring it for.

If you're measuring for trials then you should use the 21M dummy because you are more likely to have most or all of those buffs most of the time.

If you're measuring for dungeons you should probably use the 6M one because you can't guarantee any buffs you don't bring yourself due to it only being 4 man and not an organised group,.

Elemental catalyst isn't with every trial group

No but its in more than you think. Most formal groups have someone dedicated to EC and someone dedicated to Alkosh. Those are really the two main set buffs you have to know whether or not are present in your group. Both are on the Dummy.

There are really only two stats that you need to try to "cap". *Penetration (Alkosh) and **Crit damage (EC). Admittedly the dummy makes it very easy to do that with these sets up full time, but I believe that is a good thing. Pen needs to be at 18.2k, crit damage needs to be at 125%.

After the recent update, you need about 1200 pen, 1170 to be exact, (basically your passive CP and one piece light armor gets you there), and 30% to your crit damage. Lots of ways to do this, but Minor Force (10), fighting finesse (8), one piece Medium (2). The last 10% can come from class passives for Warden, Temp, Nightblade, Kilt, Backstabber, Khajiit (12%, no need for one medium), Orders Wrath, Axes, Shadow, etc., or any combination of the above.

*For pen, you have Major Breach 5948, +Minor Breach 2974, +Infused Crusher 2108, +Alkosh 6000 (any tank worth anything is giving the first 3), for a total of 17,030, leaving 1170.

**For crit damage, you have base 50, +major force 20, +EC 15, +Brittle 10 for a total of 95, leaving 30.

Sure you wont have all that 100% in every raid. Who cares. And sure, you may find that you practically want to build above 125% or above 18.2k in real fights to allow for non optimal uptimes. In any content where your DPS is really being scrutinized, you are going to know what your group is giving you and make adjustments to fill in the gap. Same with Penetrations. Especially on PC, first thing I do in a new trial group is look at some CMX reports. If either of these values are low, I make adjustments. Not hard to make a gear or CP adjustment to bump either of those numbers a bit, but admittedly, you do have to have a basic understanding of the caps and the various sources of pen and crit damage (google is your friend).

When objectively trying to compare DPS between players or between classes, standardizing this stuff makes it far easier for comparisons.

Again, if you get caught up in the raw damage numbers, you are completely missing the point of the Raid Dummy.

veznan5

6mil dummy

For any content that isn't a trial, ie arenas and dungeons, the 6 mil dummy is the most reliable measure since it gives you a more accurate idea of your performance over an extended fight that isn't propped up by buffs you might not have.
Though even then its good for a baseline measure since the buffs you get in a trial are dependent on how coordinated your group is.
If you take your 6mil performance and 21mil performance, your actual trial performance will lie somewhere in between.

jaws343

veznan5 wrote: »

For any content that isn't a trial, ie arenas and dungeons, the 6 mil dummy is the most reliable measure since it gives you a more accurate idea of your performance over an extended fight that isn't propped up by buffs you might not have.
Though even then its good for a baseline measure since the buffs you get in a trial are dependent on how coordinated your group is.
If you take your 6mil performance and 21mil performance, your actual trial performance will lie somewhere in between.

The 6M doesn't actually give you a proper measure for either solo or group play.

For solo play, often, you can run setups that increase sustain based on mob targets in a boss fight. Something that cannot be replicated against a dummy. When I solo dungeons, I will position adds during boss fights to specifically die when I need mag return for sustain. So, I can hit a dungeon boss with millions of health far longer than I can parse on a 3M or 6M dummy without needing a heavy attack.

And for group play, it is highly unlikely that your healer isn't running at least 1 form of resource return. Orbs, shards, set pieces, something. All of which are unavailable on the 3M or 6M dummy.

Vevvev

321.8K HP Precursor or Thrall dummy!!!

Oreyn_Bearclaw

21mil dummy

veznan5 wrote: »

For any content that isn't a trial, ie arenas and dungeons, the 6 mil dummy is the most reliable measure since it gives you a more accurate idea of your performance over an extended fight that isn't propped up by buffs you might not have.
Though even then its good for a baseline measure since the buffs you get in a trial are dependent on how coordinated your group is.
If you take your 6mil performance and 21mil performance, your actual trial performance will lie somewhere in between.

Nope, Nope, Nope and Nope.

1. Arena Pen is all over the place, so not consistent with any of the dummies.

2. Any 4 man dungeon should have a tank who if nothing else should be giving major/minor breach, again inconsistent with the 3/6 mil.

3. Any group, even one with out a dedicated healer, should have some synergies available for sustain (any synergy gives sustain via undaunted passives), which again arent there on 3/6 mil.

4. Dont compare any Raw dummy numbers (3/6/21) to performance in a trial. Too many variables between fights, mechanics, immunity phases, etc. Give me a good raid group on an easy stack and burn like Valariel, and I will beat my trial dummy parse. Give me a fight with a lot of immune phases and I might be below my 6 mil parse, but it might still be a very good parse.

Oreyn_Bearclaw

21mil dummy

Vevvev wrote: »

321.8K HP Precursor or Thrall dummy!!!

They are very useful for practicing my PVP burst rotation on my ganker, I will certainly give you that.

For PVE, they are completely useless. (I also just googled, "you are as useless as". Some funny ones, but not sure they are getting passed the mods )

Dojohoda

other dummy

We need more that talks back. The Precursor is a real confidence booster when he says "oww".

Actually the 6 mil is a good measure. In the old days, one would get a Templar friend to offer shards.

The trial dummy is useful if your group is structured to include the same buffs with an eye on buff mechanics and such, but trial is less useful for pugging. Pug groups are likely to have overlap.

jaws343

Dojohoda wrote: »

We need more that talks back. The Precursor is a real confidence booster when he says "oww".

Actually the 6 mil is a good measure. In the old days, one would get a Templar friend to offer shards.

The trial dummy is useful if your group is structured to include the same buffs with an eye on buff mechanics and such, but trial is less useful for pugging. Pug groups are likely to have overlap.

I think the question should be asked: What is the actual purpose of a dummy parse?

It isn't to actually gauge your DPS for an actual fight.

It is to gauge your competence with the class/build you are using.

Which makes the 21M far more suitable for that, since it removes all of the variables that are entirely out of your control to set an absolute baseline for a parse. To then gauge competence based on your DPS parse against the maximum parse.

None of that is possible on the 3M or 6M as class or build factors then come into play and skew that benchmark. The 21M puts everyone and every class on mostly the same level starting point. Removing crit fluctuations and pen fluctuations, and sustain issues.

Another way to look at it is with a comparison to a race:

2 runners on a track.

In the Trial dummy scenario, those 2 runners are both wearing cleats (full penetration), both wearing light weight clothes (full crit damage cap), both running on the same surface conditions (sustain from shards). These 2 runners are set in an even matchup at the best possible conditions. The winner here is going to come down to skill and not the conditions of the race.

In the 6M scenario, 1 runner is wearing cleats (their class/race has natural penetration or some other buff that the other doesn't have naturally), 1 is wearing jeans, 1 is running uphill. These 2 runners may have identical skill, but the conditions of the race are stacked against them to provide a very uneven starting point.

Neither of these scenarios are going to determine which of these runners is going to be better at running through the woods in the rain. But in the Trial scenario, you can better weed out the skilled player over the unskilled player being supported by gear/class.

SirLeeMinion

other dummy

jaws343 wrote: »

SirLeeMinion wrote: »

[edited for brevity] At one point, the 21M was useful as a standard. But a standard needs to remain fixed or it becomes meaningless for comparison. They have, unfortunately, compromised this as well with U35.

[edited for brevity] The standard is fixed though. It isn't like the dummy is changing weekly. It changed once, between patches.

I seem to remember a change in early 2021 as well. Granted, that is not a weekly change, but it would qualify as an annual change. One regularly sees people discuss DPS creep over the years; a changing standard makes that difficult to evaluate. In contrast I can compare my damage on the 3M / 6M today to my damage two or three years ago on a 3M / 6M and know how it has changed.

The 21M has a place in evaluating damage for people trying to complete organized trials. For those PUGing, for those running 4-person dungeons, for those comparing historical trends, there are better options. Changing the trial dummy alongside the U35 DPS nerfs was, at best, a poor choice. While I appreciate your perspective, I'm going to stick with "other."

Oreyn_Bearclaw

21mil dummy

jaws343 wrote: »

Dojohoda wrote: »

We need more that talks back. The Precursor is a real confidence booster when he says "oww".

Actually the 6 mil is a good measure. In the old days, one would get a Templar friend to offer shards.

The trial dummy is useful if your group is structured to include the same buffs with an eye on buff mechanics and such, but trial is less useful for pugging. Pug groups are likely to have overlap.

I think the question should be asked: What is the actual purpose of a dummy parse?

It isn't to actually gauge your DPS for an actual fight.

It is to gauge your competence with the class/build you are using.

This is everything you need to know right here. Also, solid metaphor as always. Haha