Sigurd's Descendants: The Art Of Dragon Slaying.

did you make that account just for that post?

For MNK discussion go here.

I swear, MNK is like the Harvey Weinstein of FFXI jobs, trying to dip its *** in every thread. >_>

You can feel free to change my dev tracker post to fit anywhere if you want since it's applicable :O

I had to do more Sims to include a normal DA set for Rho so that all comparisons can be traced along all possible conditions.

I also cross-referenced all Sim's findings with Spreadsheet and the discrepancy is still the same [around 1.2% in favor of Rho AM] Meaning once you get your Sim findings you add 1.2% more in favor of Rho.

//Frailty,Torpor,Sam,Chaos,Magma Steak HQ//

Are you certain that you used Torpor in your Sims? I think that raising hit rate % contributes to Rho's advancement due to the nature of its TP set built/AM% when you sample out all the hits. (Trish will always be safer)

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

Trish Kat STP: (Can safely be obtained by all players)
Simulated Average: trishula 5324.13 (WS : 22205.5 )
AA Percent: 0.15 WS Percent: 0.85

Trish Kat QA:(DM Event Spam/Vagary)
Simulated Average: trishula 5460.25 (WS : 21949.87 )
AA Percent: 0.15 WS Percent: 0.85

Trish Perfect STP:(Can be obtained after awhile)
Simulated Average: trishula 5394.3 (WS : 22161.57 )
AA Percent: 0.15 WS Percent: 0.85

Trish Perfect QA:(Almost impossible to obtain)
Simulated Average: trishula 5486.11 (WS : 21984.74 )
AA Percent: 0.16 WS Percent: 0.84

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

Rho Kat DA:(Can safely be obtained by all players)
Simulated Average: rhongomiant 5401.21 (WS : 17943.93 )
AA Percent: 0.36 WS Percent: 0.64

Rho Kat QA:(DM Event Spam/Vagary)
Simulated Average: rhongomiant 5483.91 (WS : 18232.77 )
AA Percent: 0.35 WS Percent: 0.65

Rho Perfect DA:(Can be obtained after awhile)
Simulated Average: rhongomiant 5451.9 (WS : 18201.15 )
AA Percent: 0.36 WS Percent: 0.64

Rho Perfect QA:(Almost impossible to obtain)
Simulated Average: rhongomiant 5534.87 (WS : 18368.55 )
AA Percent: 0.36 WS Percent: 0.64

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

ItemSet 354261	ItemSet 354270
(Legs,Body DA5)	(Legs,Body STP8)

Trish Kat STP Vs. Rho Kat DA: Rho wins by 1.4%
-------------------------------------------------------------
-------------------------------------------------------------

ItemSet 354261	ItemSet 354270
(Legs QA2,Body QA3)	(Legs,Body STP8)

Trish Kat STP Vs. Rho Kat QA: Rho wins by 3.0%
-------------------------------------------------------------
-------------------------------------------------------------

ItemSet 354261	ItemSet 354270
(LegsQA2,Body QA3)	(LegsQA2,Body QA3)

Trish Kat QA Vs. Rho Kat QA: Rho wins by 0.4%
-------------------------------------------------------------
-------------------------------------------------------------

ItemSet 354261	ItemSet 354270
(Legs,Body DA5)	(Leg,Body STP8)

Trish Perfect STP Vs. Rho Perfect DA: Rho wins by 1.0%
-------------------------------------------------------------
-------------------------------------------------------------

ItemSet 354261	ItemSet 354270
(Legs,Body QA3)	(Leg,Body STP8)

Trish Perfect STP Vs. Rho Perfect QA: Rho wins by 2.6%
-------------------------------------------------------------
-------------------------------------------------------------

ItemSet 354261	ItemSet 354270
(Legs,Body QA3)	(Legs,Body QA3)

Trish Perfect QA Vs. Rho Perfect QA: Rho wins by 1.0%

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

• The Margin is minuscule between the two, yet in my personal conditions Rho pushes further.

• QA is a game changer for Rho but not so much for Trish.

Something irrelevant, Flamma +2 beats QA3 and no point in swapping it if you got it, Also Zanshin in Sim when?

you’ll get larger than that just comparing a set against itself. and skillchain will make a larger difference than that as well. for example if every other star diver closes darkness, you’ve already made up the difference in damage. rho also assumes no mistakes as well, and unless you’re maintaining aftermath perfectly it’ll fall off more than 1-2%

Yeah, the probability of that to happen on Trish is higher if you went with another DRG/DNC/DRK/THF in DD party or Solo.

We already established the fact about Skillchains being a major factor, and I wish there was a way to factor this in all the research (numerically).

The closest I can share is this:

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

I needed 30k Beads and went to test the factor [Skilchains] on "Quetzalcoatl" solo during /2 Days/:

//Venom Potions,Dragon Breaker,Ancient Circle,Angon//
//TP Set Rho Kat QA, TP set Trish Kat STP//

[Arciela 1, Joach, Cornela, Yagans, Ulima]

Number of Fights: 15 each weapon
Dragon Breaker/Ancient Circle used: 2 times each fight.
Angon used: 3 times.

[Stardiver>Camm>Drake>Camm>Camm]

Trish Kill Time AVG: 12:23 min
Rho Kill Time AVG: 11:55 min

28s between them in favor of Rho

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

I would repeat the same test with DA Rho Set and see but Im sure Trish will win this.

Edit: I went on and divided the AVG clear times between the two weapons to come up with a percentage and:

100%-([715s for Rho / 743 Trish] X100%) = 3.8% in favor of Rho which is really close to Sim's findings.

That Skillchain is whack. Stardiver is what you want to be using when you are rocking Trish as much as possible, and you want to be closing your Radiance/Umbra with Stardiver when you can.

Stardiver > Sonic Thurst > Stardiver > Stardiver. (Trish)

This is a much more brutal skillchain.

Unless Quetz has some darkness defense I am unaware of, that will drop your clear times considerably.

I don't know that we need a test for this TBH, If the DRG is solo skillchaining that added SC damage and Umbra/Radiance is going to blow the lid off the dps gains Rhon had over it.

I did this on purpose Sidra, to have more samples.
I would’ve definitely gone with your skill chain order if it was about “just kill speed”

Needed more samples that’s all.

Well the problem is it's a pointed test. You have 3x Camlann's in your chain, which gets way more benefit from using Rhon. And you don't have more Stardivers (which Trish does better), and you aren't closing your ultimate skillchain (which is the advantage of Trish) with the more potent ws, leading to less ultimate skillchain damage in your test.

It's like trying to determine who is smarter, Albert Einstein or Steve Jobs. And to determine it, you sit them down and give them a computer test.

But rewind here. Are you really trying to see Trish's skillchain damage will outdo the 1-3% Rhon lead without skillchains? I am all for testing things, and A for effort...that just seems like a crazy waste of time. If I had both, I'd be rocking Rhon in HELM burns with 3 dps spamming, and Trish anytime I was solo or in situations with planned skillchains. No testing needed, we are allowed to use logic and understanding when the answer is obvious. This does not need tested.

It was closing Radience Sidra, and I don’t mind doing the test all over again utilizing stardiver Umbra.

Also note that Camm was only 3k when AM needed to be activated and it’s damge is capped, it doesn’t scale.


And I agree with your conclusion about when to use them.

I understand it was closing Radiance. But I think we can both agree that Stardiver does more damage then Camlann's, especially with Trish. I think we can also agree that that ultimate skillchain damage is based on the damage of the closing weaponskill, thus you are doing less ultimate skillchain damage by closing it with Camlann's instead of Stardiver. Lastly, I think we can agree that you can't make an ultimate skillchain with Rhon.

Given we can agree on those 4 points, I am not quite sure how this is a discussion or debate as opposed to "Good point, Sid"

As I mentioned before I only wanted to isolate the skill chain factor and I don’t mind using Stardiver umbra with both.

Yes to all the points you mentioned and should be the go to scenario yet we are comparing two weapons here based on several factors and I hate to break it to you but many people don’t utalize skills chains and prefer Zerg fest.

We all felt this firsthand and TBH with the current trend I don’t see them changing these habits.
Remember how frustrating going DNC to WAR fest?

I know that, man. My only point is that your test was poorly conceived and thus reveals nothing. I say this from a place of love and friendship: wasted your time and forum space. I was not making a statement about the weapons.

My mind about the weapons has been made up for 2 months, I am checked out of that discussion.

Thanks mate, I did not waste my time here I learned many things about this forum in the past month and my only issue was justifications not to build Rho because of its “grind/cost”

I only wish that foR whatever reason someone wanted to build it, won’t get discouraged to do so based on these excuses.

Couldn't post this, phone died and was outside!

Considering the nature of Stardiver, I don't mind doing the same test with it to show its outcomes yet you need to consider the following:

Cam test will be considered the "stable" base that will be AVGed with Stardiver because of:

Cam is a single hit with capped damage and its TP gets 50% ignoring target defense with 3K (which is only going to happen when AM is re applied and still 50% chance) its the exact reason you choose said WS to build your XHIT.

When you want to study/isolate an erratic element (Skillchains) you need to introduce stable elements for it to work and not have many anomalies or what we call in game here "Spikes" and whats even better is that [TP return] can factor the overall curve more accurately, and thats what I had in mind thinking of "isolation".

This Series of skill chains contain two erratic we's (Drake and Stardiver) and two Stable Cam so this methodology will yield isolation eventually overtime since the overall curve is uniform.

Stardiver Test will contain various spikes with in it since 66% of all cycle contains erratic elements and will push the skill chain damage to act that way making pure isolation harder.

This methodology follows specific rules and I hope its understood.

You cannot test which weapon will kill a mob faster by intentionally killing it slower.

You do what's best for each weapon, and let the quantity of the tests smooth out the rough edges and randomness.

But I still think this test is silly. I don't know what you are proving or to whom. For anyone determining if they want to make a Rho or not, there is plenty of info here in this thread already. We know a Trish using skillchains is tops. And we know in 3k tp starts spam non skillchain situations Rhon has a small lead. I fee like we have known that for 20 pages. Or at least I did months ago when I decided to make a Rhon.

So the question at hand for anyone is simply if all of the time and expense associated with making an empy is worth it to have a slightly better performance in a limited, albeit important situation, and lesser performance in a lot of other situation via the free Trish.

That has already been established. What are you trying to establish, now?

The test wasn't designed to "kill faster" it was to isolate SC, I said that before but I digress.

For whatever reason, somehow the BG-Guide tries to shaft Rho and this needs to be addressed.
Unless they want to make it a fact that Empy AG for all weapons are crap.
They added so much BS on guide to shaft it and its ridiculous.
They even added the cost as a valid reason not to go after it haha.

Misinforming everyone based on Cost is ridiculous and serious DRGs will have both anyway so good luck with changing their minds.

And 50 Vit counts ffs in survivability and taking less criticals from enemies... are they high?

Even with current prices, Rhong costs more than a Trish before touching on the time spent in abyssea or camping ***NMs. Your best case simulations are unable to provide any situation where Rhong wins by a meaningful %. Those 1-2% differences become irrelevant when you consider going a single extra attack round without AM3, a single skillchain, or spending more time disengaged will cost significantly more than 1-2%.

The conclusion last time I was in this topic, which I feel was at least a month ago, was that any difference in the two is more than negated by skillchain damage or player imperfections both favoring Trish. Aren't you the one always saying that real players aren't the same as bots and mistakes happen..?

If both weapons were handed out freely, the only time it would be practical to use Rhongomiant is if you were confident you were going to play perfectly AND confident no skillchains would be made. That is such a slight niche, especially when considering how heavily skillchains play into current gameplay, that making Rhongomiant over Trishula is ill-advised.

The gameplay testing is worthless. As Sidra said, you used a stacked skillchain to achieve the result you wanted. Furthermore, given you had the winner in mind before trying it, I would not be shocked if you did things like intentionally allowing extra attack rounds before WSing, or using suboptimal sets. There's no peer review, the tests weren't recorded, when working with numbers this small it carries zero weight. Given how many logging plugins are available for every platform, did you record all of these fights? Do you have the length of all 30? You only posted the averages, there's very little data for anyone to analyze or verify.

I would also be concerned with you posting a single DPS number for each simulation, did you only run them once? Did you keep running them to cherry pick numbers that showed what you wanted? Did you do a huge sample so it didn't matter? I'm not saying any of these mistakes have been made, but as Austar said, the difference between the two weapons is less than the difference you'll get running two consecutive sims on the same weapon.

I will concede that if you are completely into DRG and you feel you want to cover every situation, no matter how slight, it could have some minor use as a supplement to Trishula. This is, of course, pending verification of the simulation numbers. As is, the differences are still unlikely to outweigh even unintentional skillchains when maintaining AM perfectly.

Ok I will take every single statement of yours with a grain of salt and try to replay "logically":


Same goes for every other Empy that players make everyday and its not a factor to use in comparing two weapons performance, Cost doesn't affect performance at all and I'm done discussing this extremely lame excuse.
P.S you can literally farm Abyss drops in few hours if you buy pops from players or ask help from your LS.
So This whole argument of yours fails and mostdef shouldn't be on BG Wiki.




Its more than 1-2% and this was simulated several times across the thread, help yourself reading them.

Regarding Skillchains I offered a test that "tries" to isolate them and maintain a uniform curve free from spikes, and I don't mind doing it again with Umbra route that is extremely erratic to shed more light on the matter.

Spending more time disengaged can be applied on both weapons so its irrelevant.



100% true, and thats why I resorted to isolate skill chains to check this factor. did you even offer a way to test it? No


Simply put, its wrong because as the test shows on Quet the margin was still minuscule and there are many instances in game where you literally can't WS/SC [Schah,Mandy some Omen to name a few if Amnesia hits or buffer wears or BRD wasn't fast enough with replying carols or bubbles wear/die] these circumstances do happen more often than not and thats the exact reason why you should consider an EMPY.


THAT STATEMENT NEVER TOOK PLACE EVER, so please don't put things that weren't there to push a certain narrative.


Provide an alternative, and no I tried to worke methodologically like i stated.
Also join me to one of these tests if you like.




A. Prove it.
B. I will make sure to wear my tinfoil hat next time from all the conspiracy theory spam.


Please be the one to review it I don't mind at all we are on the same server.


If you go back just one page, you will see my spam of Sim, the only reason I started posting singles was because Austar did it.

With that logic you can also dismiss Austar Sims, but we both know you won't do that :)




I started off with 6 samples [clicks] and went to 12 and I don't mind going for a 100 to prove my point.


My intent is to dismiss the horrid description about the weapon in BG-Wiki Guide because it really discourages even making it.

And Yes I am completely into everything DRG DNC and you know it.
I would do all I can to help shed more light on the matter with the time I allocate of course.

BTW, Regardless of everything that went on, many findings were discovered [QA for example, Austar made an incredible Sim and I know it will get better with time]
Yet I don't want people to be misinformed and the guide is excellent except for the Weapons Section for the reasons I presented.

Thanks for you feedback.

Spicyryan is the author of the BG wiki guide and not Comeatmebro

I appreciate Kat's work with Rho. There are those of us who have Trish and Rho or are planning on using both. Her research and simulations are helpful.

That being said, the attempt to isolate SC damage by using a SC five step that contains three Camlann's Torments is going to favor Rhon. In contrast, using Stardiver < Sonic Thrust < Stardiver < Stardiver will clearly favor Trish. You would need to do the 4 step stardiver above to give Trishula a fair shake (as Sidra was saying).

Still, it's really unneeded to bash Kat for the work she's done. It's not perfect but at least she clearly cares about the job and advancing the guide in a positive direction.

Time has value. Gil has value. If you're buying the pops, enlisting help, you're adding to the cost of the weapon. Pretty simple. Reasonably, depending on the value of your time, rhongomiant costs 30-200% more than Trishula. That is a simple fact, and it's always relevant.

Your last post says 1.4%, 3%, 0.4%, 1.0%, 2.6%, 1.0%. It may be technically more than 1-2%, but given the percent of damage composed of skillchains and the bonus offered to them by Trishula.. it's not significant.

You made a stacked test using a skillchain that doesn't utilize Trishula's strengths, and provided no details to the test, nor logs. Nobody has any reason to believe your test, and even if they did, you were using best case for Rhong and a horrible case for Trishula.

Wrong. Trishula's aftermath is always present, since throwing stardiver into your skillchain will guarantee you have AM when you need it, and costs you nothing in terms of DPS. Needing AM3 on Cam means that the more time you spend disengaged, the more of your total TP gain is spent on 3k Camlann's instead of relevant WS. Being engaged less favors Trishula.

I did not dismiss your sim, and I even stated in my post that those mistakes may not have been made. I simply asked for clarification on them.

I'm not going to individually address the rest of your post as it's largely lacking clear points and mostly defending a gameplay test that was ill-conceived from the start.

If you are not just set on your result and you intend to prove which is better, a meaningful gameplay test has to consider the following:

-Optimal usage for each weapon(define all sets you're using for both weapons, define skillchain, lay it out to the reader).

-Human error. A gameplay test should be done using a bot, plain and simple. If you're doing it by hand, you're adding factors that may be uneven between weapons even when they shouldn't be, intentionally or not.

-Target choice. Fighting Quetzacoatl is about as bad as you can get. A single extra terror in 10 minutes will have 5x more impact on your speed than the difference between weapons. A valid target must be free of debuffs that heavily influence damage, and isolate factors that benefit/hurt each weapon. In this case, you would just need something with enough HP to last more than one AM3 cycle, so that's pretty easy. If you were comparing weapons with a large difference in attack, you would want to decide if that attack has value or not and select a target that represents it.

-Buff choice and uptime. Using trusts for buffs is not viable. Any time spent in the wrong buffs will skew your comparison. In most real situations, your songs will not be dropping. Again, the goal is to isolate all external factors and just compare the weapons in conditions representative of what you're arguing. Trusts cycling buffs is very counterproductive to this. Get a BRD to N/T, it should last your whole sample. Ditch the trusts. If using debuffs, strictly control their uptime.

-Accountability. If you are using a bot, this is pretty simple. Describe the bot's AI, if it's within parameters of what should be expected, there's no argument to be made about gameplay. If you are not able/willing to do this, you should at the very least have logs of the fights available. Videos would be better.

With all of this in mind, it's easy to see why so many of the educated posters would disregard gameplay tests as meaningless. It's not easy to make an unbiased test that would hold up to scrutiny.

I finally made you to speak!!! I like this :)

Also I know that, and many people can edit that guide FYI!

First of all, Thanks.

And yes I don't mind to do that or if anyone willing to for that matter.

No one edits it but him. His DRG thread is here

i never posted a result based on one sim run.

Not even once during the 30 fights I was terrored, Im not saying it won't happen Im simply pointing out that Dragon Breaker/Ancient Circle with the rate of intimidation it gets its very seldom.

100% agree on this one from an abstract point of view yet you keep forgetting the fact that we are after all people that play this game and field test of a larger sample can predict human error quite effectively.
I expressed several times on the limitations that Sims/bots try to convey, they disregard human error entirely.

What Im trying to say, a human induced test its more realistic to measure the performance of the weapons.
From an engineering standpoint tests that are constructed without a human error in mind are usually discredited.


I have to point out that I can't do any of that since I don't dual box or use other ways to test this than my own hands.

I think you can shed more light in that department, I simply advocate the personal use above all with minimum buffs and Yes with trusts.

For the reasons I stated before.


In science/engineering the methodology of field tests hold the highest weight.

He meant posting a single result here not doing 1 sim.

Neither of these effect the rate of terror's use and breaker only minimally effects chance of it landing. Are you trying to say you fought quetzalcoatl for a total of 6 hours, 4 minutes, and 30 seconds solo without it landing terror once?

I can agree that human error has value. However, a discrepancy between the amount of error for each weapon can easily invalidate the tests. Let's say you personally are a bit off on the timing for 5 stepping and miss 30% of your Umbra skillchains due to going too early or late(not saying you are, but it's a possibility). Suddenly, Trishula is looking like ***while Rhongomiant is looking great. If the average player only has a 10% rate of missing their skillchain, or lower, your results don't say much of use. I am sure we can agree that the type of player to follow a ridiculous forum argument over a difference of 2% is likely to play much better than the average player.

When using the scientific method, you need to *** all variables and control as many as possible. Your test would be thrown out immediately by any serious scientist, for the reasons outlined in my last post. You also do a full write up including an introspection on your own possible mistakes and all data points, you don't just give the averages.

Yes not even once, and it was done across two days.
Intimidation is enhanced with Ja's
Also, fight on tail?


That is true, but you proved it the moment you pointed out how I care about 2-3%
And again come join on it.


Thats exactly what I did:

Cam was used to control TP return and to balance the erratic outcomes from Stardiver/Drake since 66% can counter any anomaly "Spike", think of it with the same concept of building you XHIT.

Minimum buffs were used (capped Haste/angon only) to remove any
possible alterations with Skillchains from buffs.
Conserve TP is 23% for both and this anomaly can also be negated with a steady curve of skill chains.


Thats the only thing that wasn't provided, and can easily be fixed if there was someone during the test since there can be another argument over "tampering with logs" to prove a narrative.