Should we stop using the 21mil dummy for measiring DPS

Zezin

With the constant changes to buffs available to this dummy it is unreliable at best and straight up deceiving at worst. Parsing on the 3/6mil both requires better knowledge of the game and gives a more accurate view of an individual players capabilities without the risk of getting ZoSed. Another big point is that the 21mil dummies are locked behind the crown store while it's counterparts are not.

GloatingSwine

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

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,.

jaws343

It's only unreliable if you are trying to compare DPS between this patch and last patch, it is still reliable to tell you your DPS in the current patch while under optimal buff scenarios.

Zezin

6mil dummy

jaws343 wrote: »

It's only unreliable if you are trying to compare DPS between this patch and last patch, it is still reliable to tell you your DPS in the current patch while under optimal buff scenarios.

I don't think 100% uptime on major slayer is possible no matter how optimal your group is and up until before they added catalyst this patch(the only buff that should have been added IMO) it was incorrectly portraying the strength o light armor builds in groups that forgo Alkosh in favor of EC.

preevious

other dummy

It's hard to answer, really.

Sure, the 21M can be (is) misleading (especially now)
But when I parse on the 3/6M, I get some number .. but I get it without breach .. nor orbs or jabs for sustain.

Wich one is valid? My parse at xxk DPS without breach, or the next guy's parse at xx+5k DPS with breach?
It's the same numbers, essentially, but mine will be considered low.

There's no agreed upon standard for non trial dummy, and so, it's not really usable, except for self-comparison.


Right now, I 'd say there's no good dummy to parse on.

Zezin

6mil dummy

preevious wrote: »

It's hard to answer, really.

Sure, the 21M can be (is) misleading (especially now)
But when I parse on the 3/6M, I get some number .. but I get it without breach .. nor orbs or jabs for sustain.

Wich one is valid? My parse at xxk DPS without breach, or the next guy's parse at xx+5k DPS with breach?
It's the same numbers, essentially, but mine will be considered low.

There's no agreed upon standard for non trial dummy, and so, it's not really usable, except for self-comparison.


Right now, I 'd say there's no good dummy to parse on.

That's why I said parsing on the 3/6mil requires better knowledge of the game and it's mechanics being able to change and adapt your build to a scenario and encounter are a big part of being a good DPS and not just a parse monkey.

Dalsinthus

6mil dummy

I mostly run dungeons and therefore prefer to test on the 6 mil dummy. 3 mil is also fine but 6 does a better job testing your sustain.

I do wish we could customize buffs a bit to match what is in our group. You can pretty reliably depend on having major and minor breach on the boss, for example. However, not every class has easy access to this in their tool kit and one bar builds probably don't have the bar space to include a source of breach. Still you can compare builds and rotations within a particular character to get a sense of how you character is performing.

For trials, I agree the 21 mil dummy makes good sense.

redlink1979

6mil dummy

Yes. U35's trial dummy inflates your real dps by 20%.

ZOS_GinaBruno wrote:

(...)
Adjusted Target Dummies with raid buffs with the following changes:
- Adjusted the Crusher Enchant on Target Dummies to mirror an Infused value, rather than an Infused + Torug's Pact value.
- Engulfing Flames now matches the proper 6% maximum value, down from the old 10%.
- Alkosh's Roar is now the maximum 6000 Armor reduction, up from the old 3010 value.
- Attackers now gain Minor Courage and Major Slayer when hitting the targets.
- Added Elemental Catalyst’s full stack value.

Developer comment:
Note that this may create a situation where your damage seems to have gone up in comparison to other patches, when in reality it is merely closer to a coordinated group environment where these powerful bonuses were being brought. We have tried to focus on important buffs and debuffs that we see a large majority of groups build around, rather than every possible buff and debuff, as many of them are not able to be ran in tandem with one another or require more specific builds to be brought to a composition.(...)

Bat

other dummy

It would be neat if we could

1) Get some appropriate buff foods that make sense for the new hybridised general character building (Like, tristat foods that would also give regeneration of stamina and magicka resources, a couple of similar such that either gave more resource regeneration on magicka or stamina depending on which food is crafted), as well as probably a few more alternatives in terms of parse foods; and;
2) Have a few this-or-that buffs added to dummies below the trial/21M target dummy, not as many as the 21M target dummy (which, really, are best-case-scenarios in terms of buffs and buff uptimes, and likely not fully achieved in most trial groups, perhaps not even in many experienced raid groups) but a few that might make sense in, say, a dungeon designed for 4 people where 2 will provide real group support.

Edit: A/B listing format changed to 1/2 listing format because it made no sense to have in that text.
Edit 2: Added it would be neat if also we were given some more alternatives in terms of parse foods to compliment the overall changes from U33 forwards.

jaws343

Zezin wrote: »

jaws343 wrote: »

It's only unreliable if you are trying to compare DPS between this patch and last patch, it is still reliable to tell you your DPS in the current patch while under optimal buff scenarios.

I don't think 100% uptime on major slayer is possible no matter how optimal your group is and up until before they added catalyst this patch(the only buff that should have been added IMO) it was incorrectly portraying the strength o light armor builds in groups that forgo Alkosh in favor of EC.

None of that really matters though. The point is, the buffs offered are static for the patch they are in. So whatever you or others do on the dummy this patch should be compared against other parses from this patch on the same dummy.

Plus, the 21M dummy offers group sustain that the other dummies do not offer. Which means it will always be closer to a trial environment than the other dummies.

tonyblack

6mil dummy

I just wish people stop throwing parses on raid dummy for anything other than 12 man group content. You would never see so many buffs in dungeons or arenas and parses on 6 mil dummy would be more representative of your actual performance. Never mind that i saw people arguing about their dps changes on trial dummy while admitting they only interested in overland and occasional normal dungeons. Performance on 21 mil dummy only relevant for trials, for anything else it might be rather misleading.

Styxius

6mil dummy

6mil is/should be used to help show minimum damage outputs. Zos has messed with the 21mil at the worse times too frequently for it to be reliable. I was/am planning to move my end game guild back to 6mils because of these awful changes over time.

TheGreatBlackBear

A 21 mil parse is not about wholly about what you hit in a vacuum. The idea is to standerdise access to buffs and debuffs across all classes to the only variable when parsing is class strength and player skill. That helps raid leads make informed decisions about what class and what player to bring. Some classes like warden have native access to fracture and minor berserk so their non 21 mil parses could appear inflated when compared to other classes. It was not uncommon to have a tank or healer to puncture or ele a parse for non wardens and stamblades back in the day. Depending on the your raid lead you were allowed to get combat prayer to simulate minor berserk when the berserk became passively granted from just having Bird of Prey slotted on a warden.

MudcrabAttack

6mil dummy

Honest question:

When 80% of players who would strive to supply raid buffs vanish overnight, leaving you with not enough people to even join a raid, how valid is a 20mil dummy parse? The percentage of players just doing small group and solo will be skyrocketing

mocap

3-6 mil dummy to test your pure dps and more importantly - sustain. Especially if it's not HA build.
21 mil dummy mostly for premade/static vet trials or if you just like high dps numbers for no reasons.

TheGreatBlackBear

Zezin wrote: »

preevious wrote: »

It's hard to answer, really.

Sure, the 21M can be (is) misleading (especially now)
But when I parse on the 3/6M, I get some number .. but I get it without breach .. nor orbs or jabs for sustain.

Wich one is valid? My parse at xxk DPS without breach, or the next guy's parse at xx+5k DPS with breach?
It's the same numbers, essentially, but mine will be considered low.

There's no agreed upon standard for non trial dummy, and so, it's not really usable, except for self-comparison.


Right now, I 'd say there's no good dummy to parse on.

That's why I said parsing on the 3/6mil requires better knowledge of the game and it's mechanics being able to change and adapt your build to a scenario and encounter are a big part of being a good DPS and not just a parse monkey.

What you're saying makes zero sense. Does the 3/6mil move or attack you or otherwise make you struggle for time on target?No. How on earth does that dummy require better skill and knowledge? This entire thread reeks of casual or lower midtier players giving opinions on things they don't understand.

mocap

TheGreatBlackBear wrote: »

Does the 3/6mil move or attack you or otherwise make you struggle for time on target?

Sustain and other self parts for PUG dungeons, vet DLC in particular.

Maybe you're used to play in a static group where everything is perfect, but many players play with randoms and there can be anything - fake heal or bad sustain from the heal (no Orbs, no EleDrain). You have to be ready for anything, including to even heal yourself. In that case it makes sense to test your dps on 3-6.

sbr32

21mil dummy

TheGreatBlackBear wrote: »

A 21 mil parse is not about wholly about what you hit in a vacuum. The idea is to standerdise access to buffs and debuffs across all classes to the only variable when parsing is class strength and player skill. That helps raid leads make informed decisions about what class and what player to bring. Some classes like warden have native access to fracture and minor berserk so their non 21 mil parses could appear inflated when compared to other classes. It was not uncommon to have a tank or healer to puncture or ele a parse for non wardens and stamblades back in the day. Depending on the your raid lead you were allowed to get combat prayer to simulate minor berserk when the berserk became passively granted from just having Bird of Prey slotted on a warden.

This is the correct answer.

Varana

21mil dummy

The main point of the 21million dummy is not to have a realistic scenario in actual content. Sometimes it's a good approximation, but that's not the point.
The main point of that dummy is to have a standardised, class-agnostic testing environment in a trial group. Their changes to the trial dummy are unfortunate (like providing buffs that have nothing to do with classes, like EC, and esp. after a major change like now), but they're not invalidating the dummy within one patch cycle.

The non-trial dummies have the inherent issue that classes are not equal, and you're forced to build for unrealistic circumstances. In actual content, I have a tank to provide Breach, and others raising sustain, and so on. Some classes have easy access to some of those buffs in their main toolkit and don't need to change anything. Others have to specifically build for them, sacrificing damage in the process.

As such, the non-trial dummies are not able to be used as a general benchmark. Before the trial dummy, we had things like "assisted parses", i.e. having a second player standing around to Breach the dummy and throw orbs for sustain, and stuff like that. Players did that precisely to even the playing field and create a more useful metric to compare performance.

So while the non-trial dummies may be useful for build testing under certain conditions, they're not helpful for what the trial dummy is designed to do.

TheGreatBlackBear

mocap wrote: »

TheGreatBlackBear wrote: »

Does the 3/6mil move or attack you or otherwise make you struggle for time on target?

Sustain and other self parts for PUG dungeons, vet DLC in particular.

Maybe you're used to play in a static group where everything is perfect, but many players play with randoms and there can be anything - fake heal or bad sustain from the heal (no Orbs, no EleDrain). You have to be ready for anything, including to even heal yourself. In that case it makes sense to test your dps on 3-6.

What you're saying still makes no sense. Are you parsing on the 3/6 mil while using a heal or shield every 10 or so skills? Are you recasting ground dots early because adds and bosses can and will move out of them? How are you simulating sustain passives on weapons like 2h or destro or certain CP nodes that restore resources or gives increased recovery when you kill something? Are you factoring in the undaunted passive that restores resources for activating any synergy? Are you using more expensive cleave oriented skills when parsing instead of a single target dot or spammable? Parsing on on a 3/6 mil dummy is in no way shape or form a worthy test your output in anything other than being an insufferable hipster

Kirawolfe

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

I've always parsed on the big guys with my main, but I work on the little ones to get to a certain number before bothering on the trial dummies.

I only clicked this option because it was hilarious, but, I think the trial dummies are still important. They help you see your long term dps (in ideal situations, I realize) but also see where you might need to work on sustain, which I appreciate.

I think what ZOS seems to have done - trying to minimize how this update looks to have affected people by buffing the trial dummy - as really sus(pect), but I would rather they return it to it's original format than have to stick with parsing the little guys.

Zezin

6mil dummy

Varana wrote: »

The main point of the 21million dummy is not to have a realistic scenario in actual content. Sometimes it's a good approximation, but that's not the point.
The main point of that dummy is to have a standardised, class-agnostic testing environment in a trial group. Their changes to the trial dummy are unfortunate (like providing buffs that have nothing to do with classes, like EC, and esp. after a major change like now), but they're not invalidating the dummy within one patch cycle.

The non-trial dummies have the inherent issue that classes are not equal, and you're forced to build for unrealistic circumstances. In actual content, I have a tank to provide Breach, and others raising sustain, and so on. Some classes have easy access to some of those buffs in their main toolkit and don't need to change anything. Others have to specifically build for them, sacrificing damage in the process.

As such, the non-trial dummies are not able to be used as a general benchmark. Before the trial dummy, we had things like "assisted parses", i.e. having a second player standing around to Breach the dummy and throw orbs for sustain, and stuff like that. Players did that precisely to even the playing field and create a more useful metric to compare performance.

So while the non-trial dummies may be useful for build testing under certain conditions, they're not helpful for what the trial dummy is designed to do.

The best possible solution would probably be a dummy that only apply readily available class buffs and debuffs to the character and dummy I.E. minor and major breach, minor berserk etc. I find this option very unlikely to ever be implemented however, and even then with how ZoS changed the dummy to mask the damage reduction I don't trust them to not do the same again in the future with such a dummy.

Parsing on the 21mil dummy right now leaves a very sour taste in my mouth and it's why I don't want it to continue being the standard.

CleymenZero

other dummy

The full parse dummy should be more a tuned to what a top tier optimized raid team runs.

If even top tier groups can't maintain full buffs all the time.

The dummy has EC but doesn't have z'en.

It has 100% major slayer and force. I'd have to look at logs but any fight longer than 4 minutes or smth won't see 100% major slayer.

Alkosh is debatable because I've looked at a bunch of logs from groups that DEMANDED Alkosh on a DK last patch and if I saw a 70% uptime, I was impressed. A lot of times it was in the 50s and 60s. 50% uptime on alkosh might as well slap Crimson on tank...

Would it just be possible to toggle buffs and debuffs or sell different versions? Instead of having multiple skin for the same parse dummy, why not have the iron one have x, y, z buffs and debuffs and another with w, x, y buffs and debuffs.

To add Major Slayer was a mistake imo. In my groups we get max RO/Jo uptimes which is around 70%. Slapping it at 100% vs not having it, especially at a time when they're nerfing dps so hard seemed intentionally deceptive...

Troodon80

As others have said, it's unreliable when changes are made to the dummy between patches. I don't think we should stop using the raid dummy as a baseline, but in order for that to remain true changes should not be made between primary updates. Changes should be made in incremental patches so that changes between one PTS cycle can be compared to whatever the current Live version is.

3m / 6m used to be the staple for DPS parsing, but there's a reason why people use the raid dummy now. It's because there are too many ways to cheese the 3/6m dummies, generally using penetration; something you won't be doing in a live environment (i.e. a raid with supports providing resistance debuffs to a boss). You'll want to focus different things, like penetration or weapon/spell damage over crit when you're under pen-cap (which I have referenced before in another topic), but you'll want to prioriitise critical damage when looking at the raid dummy and in a proper raid.

Troodon80

Varana wrote: »

The main point of the 21million dummy is not to have a realistic scenario in actual content. Sometimes it's a good approximation, but that's not the point.

The main point of that dummy is to have a standardised, class-agnostic testing environment in a trial group.

While I do generally agree with the sentiment, it's also now unreliable because of how the debuffs work. For example, Warden is currently performing as one of the worst classes in raids and on the target dummy partially because it's overpenning on the raid dummy compared to other classes which get weapon/spell damage or critical strike/damage. Unfortunately, it's not class agnostic because of this. One could, in theory, run a mostly warden group and optimise around the penetration bonus you get from your class, not using Alkosh or Crimson, for example, and results would possibly be similar to those groups optimising around needing that penetration.

Casul

6mil dummy

3 mil for convenience. 6 mil for actual measure. Maybe I am just old fashioned.

Suna_Ye_Sunnabe

3mil dummy

As the leader of a trials teaching guild, I will no longer be accepting 21 mil parses when this patch goes live on console. Only 3 and 6 mil parses. The 21 is a joke now (like the entire patch, but I digress...)

Necrotech_Master

other dummy

i use the dummies as tests and it depends on what kind of test im doing

burst dmg - single 300k dummy

boss + adds to test cleave dmg - 1x 3 mil dummy + 2x 300k dummies

boss without adds for single target dungeon/WB dmg - 6 mil dummy

theoretical dmg in a trial group - 21 mil dummy (i just got one of those like a month ago cause i was missing that test case)

i tend to use the dummies for a lot of things other than parsing, such as testing what a proc does

Oreyn_Bearclaw

21mil dummy

If we are talking PVE DPS, this is not even a debate. The 21 mil dummy checks all the boxes. Other dummies certainly have their uses, but the raid dummy is the gold standard everyone should be using for End Game PVE DPS.

-It is enough health to adequately measure sustain, player fatigue/endurance, and remove weird cheesy burst from the equation that is typically found in the opening or execute. 3 mil allows for way too much cheese, and 6 mil is not realistic to sustain without help.

-The buffs remove the need for support players when practicing or parsing. DPS really matters in the context of trial prog groups. It makes sense to not have people wildly altering their rotations for things like breach, magic steel, etc. They wouldn't do it in a raid, they shouldn't have to do it on the dummy. Also, smashing synergies is good practice.

-Most importantly, they are the best and most consistent measure of a players potential, and the best tool for comparing classes for balance purposes. By standardizing buffs, it levels the playing field for both players and classes. Who cares if it over inflates the number, that is completely missing the point of their value. Any given patch, metas emerge and people generally now what the upper limits are. Raw DPS numbers in any given patch are only useful in how they compare to the meta.

If I have 2 sorcs for one raid spot, and one pulls 95 and the other pulls 70, there just isn't a lot of debate as to who pulls more DPS. It doesnt mean I expect the first guy to pull 95k in every fight, that would be insane. (And please, no nonsense about how people that pull high damage are automatically bad at mechanics. It's a fallacy put forth by people with bad DPS as far as I can tell. It is simply not supported by any of my experience in Tamriel).

Furthermore, if the meta sorcs are pulling 115 K, the player knows how he compares to the field as well. I am generally about 95% of the youtubers. If I am below that, I know I have practice to do for that particular spec. If I am there or even above, I know its about as good as I am going to get.

Now dont get me wrong, they might have over buffed it recently, but that is no reason to abandon it.

Amottica

21mil dummy

The sustain is why the trial dummy is relevant for a dummy parse. Anything else requires a player to do something they would not normally do in a raid to adjust for the sustain, which is a bad practice.

It is irrelevant if some of the buffs are different since it is still a constant and still serve the intended purpose. What a player can do in a real raid fight where mechanics need to be dealt with is what matters most.

ccfeeling

6mil dummy

No buffs the best, i want the real data