Weapon Tierlists 2025

Nice try, but I'm not falling for that

Don't do it to yourself.

Show me the sim.

I need a simulation to prove that Aeonic scythe is better than calad or prime.

The irony here is that the highly-regarded "all jobs sets" on BG wiki that many people use for their gearing ideas comes from Simulations from someone who probably doesn't have all of the pieces of gear the tool spits out. Even in practical terms, looking at R25/30 Nyame vs JSE+3/4 and the WSD pieces, it is nearly impossible to estimate which pieces are better for WS under specific scenarios, and only a thorough test over a long period of time or a sim can calculate the performance of either piece properly.

I understand the trolltastic level of trolling you are doing to Simon, but a lot of FFXI is literally math...Lol

Real life is a simulation in god's brain, ffxi is a simulation within a simulation

tbh for damage I would put Apoc at C/D, it is only B because of utility which if you consider you could own any weapon in this ranking makes Apoc actually D rank.

This I will answer. I don't prefer Sim, I simply don't have access to all the items in game and don't have time to properly test it for 24h or through 200 Sortie runs. I still play and test stuff in game. I sacrificed 6 Sortie runs early on to test Ruthless WS and ftp for example. If I could close myself in time chember like Goku.. but we are in a real world, not anime :)

What happened to the test dummys??

At the same time, a sim, any sim, is not the appropriate tool to compare things like "how much damage does weapon X on this job do compared to weapon Y on this other job in this content", simply due to the number of variables the sim has to fix in place to get results that make sense, and how much the sim simply ignores because they are hard to model.

The sim is a tool for comparing one set to another set and seeing if/how much of an improvement it makes, not jobs or weapons.

The last thing anyone should be doing in a thread about opinion tier lists for weapons, a troll list from its inception, is bringing maths into it. Opinion, obviously.

Let's ramp up the tier list troll dial

DD jobs
S tier: WAR SAM DNC
A tier: DRK BLU COR RDM
B tier: DRG MNK RNG BRD
C tier: THF
D tier: PUP RUN

And people wonder why I don't do tier lists more often even though they are my most successful videos.

Tier lists are for fun but aren't really the best way to think about things in such a situational game. They do provide a fun format to talk about a set of weapons or whatever though.

Job tiers would be fun, for me
SS Cor
S BRD GEO RDM WHM SCH
A all the DDs
D pet jobs :(

This is true, but nobody uses sims in this way afaik. We don't have the exact stats of Sortie, Omen, or Odyssey bosses, so it can't determine that.

What it can do is calculate average damage over a common enemy that we do know the stats for, with a set of variables (buffs, job traits and abilities, and gear sets) specific to the test. Then, by changing out only the weapon (or a few items in a set), you can see differences in the average damage output of each by tweaking the tool slightly (adding this buff, absorb stat, etc, removing this piece in favor of another). This gives you a very close estimation of comparing singular items like weapons against each other because the environment is extremely controlled and the variables you are changing are only a few, so it is easy to see where the differences occur. This kind of control is actually preferred, because if you go out into the field and whack a mob 100 times, even with near-perfect tracking, you can still be fooled by the eyeball test, or a mistake can happen, a buff can fall off you get enfeebled, or some other factor, which can throw the entire test off. It would also take entirely too long to actually know for sure if something was a better option than another.


I can understand what you mean that a sim is not specifically a job-comparison tool (but it can even be used for that since the tool doesn't know the difference between a job and gear, its all inputs/numbers), but I disagree with the weapon. A weapon is just a piece of equipment, no different than a body armor. You can simulate damage output on weapons just fine and compare them to one another, using the controlled approach I mentioned above. Again, this is way more reliable than the eyeball test.



Really? I get that the entire premise of a tier list is a person's individual opinion, but there is also something considered an "agreed upon" fact from peers as to where items are placed. An example is Naegling, it's so obvious that nobody could reasonably have the opinion that it is not worthy of its placement (in an "opinion" ranking nonetheless). So if 100% of the people have the Opinion that Naegling is S tier, it becomes basically a fact among everyone. Where it gets dicey is when people don't agree on placements at the extreme tiers (A, D). Nobody really fusses over tier placements in the middle rankings, because those can flip flop based on the wind or the time, but the extreme "this is OP"/"this sucks" is more disputed than anywhere else, because people have individual experiences and uses cases they can bring to the discussion that they feel heavily weigh into the placement. From that point, the tier lists just become a huge conversation about people's input about the weapons, which is fine. But that doesn't means someone can't contribute evidence, whether its from a sim or a personal use-case, to help others understand where they might be mistaken about a particular weapon placement. Being ignorant about a weapon is common; not everyone is going to have every single weapon on every single job, so it is perfectly reasonable for someone to share their perspective to educate others who are misinformed.

Now where I will agree is that there are a number of factors involved that comes into play when it comes to using Sims. I believe humans should be smart enough to take the information from the sim and deduce what is the best choice to make based on the information provided on their own; you wouldn't let the tool tell you exactly how you should gear for every particular scenario. That would be just as silly as you asking an AI a question and assume it is 100% correct, and blindly taking it's word on the matter.

Personal job tier list in terms of how much I enjoy playing them
S: WAR SAM
A: THF DRG NIN MNK DRK
B: BRD RDM BLU DNC COR RNG
C: WHM BLM SCH RUN PLD
D: SMN PUP BST GEO

How much I actually play the jobs:
S: WAR SAM BRD RDM
A: SCH COR DRK
B: MNK THF DRG NIN
C: WHM BLU DNC RNG
D: SMN PUP BST GEO RUN PLD BLM

They aren't the best way to look at anything tbh, because they're just entirely subjective after a certain point. You will never be able to convince someone who grew up in the 80s that Michael Jordan isn't the greatest basketball player or Arnold the greatest bodybuilder from the 60/70s, no matter how good anyone in the future ever becomes.

Here’s the blunt version:

You can’t sim exact Sortie/Omen/Odyssey bosses—we don’t know their stats.

What sims can do: model average damage vs a known, common enemy under tightly controlled variables (buffs, traits, gear).

Swap only the weapon (or a few items) and you get clean, comparable deltas in output.

This beats field testing, which is noisy (buffs drop, enfeebles, mistakes) and slow.

A weapon is just another gear slot; simming weapons for comparisons is valid.

Tier lists are opinions, but strong community consensus (e.g., Naegling = S-tier) functions as de facto fact; most arguments happen at the extremes.

Sims and personal use-cases both provide evidence that can correct bad takes—reasonable since not everyone owns every weapon.

Bottom line: sims are tools, not commandments. Use the data, then use your judgment (same as with AI).

Everytime anyone says X does less damage than Y in some content, someone will come along and post sim data as "evidence" of that not being true. Which is insane. The sim cannot simulate content, it can give you an *idea* how they compare everything else being consistent (not equal).

No sim can correctly model how well dds will do in a seg farm. Doing well there has less to do with gear and more to do with how quickly you can move from target to target and one WS kill them.

Same with sortie, these are 30sec-1min fights. The human pressing buttons is the biggest variable in a fight that short, not the jobs.

I do agree it can be used to compare weapons, on the same job given the same buffs.

I disagree it can be used to compare jobs, simply because then you need a standard unchanging way to play those jobs, and that is not realistic.

Goes back to comparing gearsets, not gameplay as in how to play a job.

You stating obvious facts, but this is exactly the reason why sim is better to compare anything. You don't have human factor, it simulate person with the same timing to give you idea what you can change in your gear or buffs to get higher dps. It's strictly gear comparison, or buff comparison, while parses in game are comparing person, gear, ping, skill, etc. which often eliminates any definitive conclusions about only one of those elements. Idk how you see this as flaw of sim instead of advantage. None is telling you that with this gear you will make better dps if you sleep on your keyboard, while other DD with ambuscade gear is actually killing mobs.

Also you can sim everything, it's only a matter of how advanced sim you want to create.

Whats the point of discussing a weapon tierlist if you add in player skill as a factor?

It becomes a pointless discussion.


As soon as "what weapon you are using" is no longer the biggest variable, you have *** up your axioms of the discussion surrounding the topic of "what weapons are the best"

For the discussion to have any purpose, you have to remove any variables that could be a bigger variable than "what weapon you are using" in order to make it become the bigger variable, via axioms of the discussion...

Otherwise you aren't having a good faith discussion and everyone is pretty much getting their time wasted.

So, yes, you have to assume the scenarios and player skill are fixed variables and hand waved away if you wanna discuss "what weapons are the best" and wish to "rank" them in a "tier list"

Thats just sort of a necessary pre-req to the convo being productive and able to get off the ground.

It's literally a shitpost tier list, not calculating how much burn you need to get from earth orbit to moon orbit.

Not that serious. Can't we just let shitposts be shitposts?

will get flack for this, but you could always parse a bot setup with existing tools to compare different jobs. that removes human inefficiency in response and decision making times. same potential problem as sims in that everything needs to be set up ahead of time correctly to get meaningful results, but it's not a huge leap from people whose gearswap files do autows and swap to appropriate sets if they get debuffed, etc. can even program reactions to certain tp moves or actions by other people in the fight. relatively easy to set up if you know the fight well, too.

of course that takes most of the fun of playing the game away, but the information would be pretty accurate. existing sim can't account for variables like changing mobs or reacting to new conditions, but there are still many fights where you don't worry about anything and just mash ja/ws buttons so it still has a place.

The point isn't to be accurate.

The point is to set an expectation you can't meet and be perpetually disappointed.

The difference is in presentation.

The author of the jobs sets simulation said they weren't always going to be optimal due to FFXI being situational. They are showed as guides and examples of what to work with, not "This is absolutely best, fight me bro!".

also many people do not look at the stipulations sim used to get those sets at all, just the gear. not the author nor the sim fault it is misinterpreted.

Skill is a part of power balance, presenting Extra Credits.



Naegling is a really good example of a First Order Optimal Strategy that ran away. Something that is 5~10% "better" but requires 200~300% skill only creates an Awesome But Impractical situation.



https://tvtropes.org/pmwiki/pmwiki.php/Main/AwesomeButImpractical

I mean if you wanna play the simulation card every time I'm going to have to start asking for correlation percentages and levels of confidence.

In other words, bring it if you gonna bring it.

Are they 4 sigma or merely 2 sigma.