Background:
So parsing makes sense in a game like WoW because you can actually see damage over time inside a raid setting. In FFXIV the devs went out of their way to mask individual damage over time effects in raids. Meaning that parses inherently became fugazi. You always see one tick on a target no matter the number or potency of damage over time effects. To put the cherry on top; fflogs is incapable of seeing people's gear. So not only is parsing in FFXIV a simulator at best; it is a very bad simulator, because they cannot see your gear at all. People seem to dislike this fact. And I feel like this is bad for the game's health. People are using data and performance to judge jobs and combinations when the data is very very wrong.
People then use this data to pick on jobs calling them memes. When the data itself is just worthless because many of the meme jobs are actually used in world first prog teams. People then make the argument that "It is the best we have" and that "Math is not wrong". First of all; the best we have is very bad because devs want it that way. Second, math is wrong if it is imagining data that is not given to it. What do you guys think? and I understand there is much money involved in calling simulators parses and so it threatens alot of people when you say that their parsers are actually just very bad simulators.
Edit: Some rough estimated values. Let me know if they need to be corrected.
- The difference between world's highest dps and world's 99th highest dps is around 3.5% to 4.5% for most jobs
- Around 9% of bard's damage is DOT damage
- Around 5% of black mage's damage is DOT damage
- Around 5% of dragoon's damage is DOT damage
- Around 5% of Samurai's damage is DOT damage
SMN finds out that his dots are completely inaccurate and inflated in ACT. The lead developer (Or PR person? not sure) responds by saying that DOTs are simulated by ACT.
Screenshot of the app:
Any updates to the app. Reading game memory and so on does not make the dot damage accessible as it is never sent to the clients. So unless SE changes their stance on sharing data nothing will change the inaccuracy of ACT and FFlogs.
https://www.reddit.com/r/ffxiv/comme...damage_in_act/
Another response by the same person
https://www.reddit.com/r/ffxiv/comme...uide_wplugins/
There are hundreds of posts on the subject
*This does not include the RDPS which is raid-dps. Raid dps is even more laughable because how do you calculate the damage gained from reduced healing required via a properly timed debuff? when dots themselves are estimated and you start estimating their raid contribution and contribution of raid buffs; the whole thing becomes even more misleading. They did this because basically they had no other way to explain how jobs that were doing much less damage were actually clearing faster. Dancer for example.
Parsing is incapable of seeing actual DOT damage. Incapable of seeing critical strike, direct hit or a regular hit. It therefore has no way to tell what actually happened to a dot ability and thus assumes that it always behaves a certain way. This inherently creates a bias in favor of non-consistent jobs that have crit or direct hit builds over determination. Not only can it not tell what happened to a DOT ability, it also assumes the player's actual crit, direct hit and potency which is also wrong because there is no capability to see gear.
My personal problem with parsing in FFXIV:
I use Details! in wow and I love it. I want more data not less. I want to know who messed up and where so that I can help them. In FFXIV it is illegal to talk about ACT or parsing. You can google it and will find hundreds of people getting banned for mentioning someone's dps. There are also videos on youtube for directors discussing parsers.
it is bad because it is simming data that is not what actually happened. And this simmed data is then used to silently kick, harass and bully people without even telling them why. Because of the TOS people can't even discuss why someone's sim values were lower than others. This makes parsing even more toxic than it is in games where parsing is legal. But the fact that our parser in FF is a SIM makes it so that you will see someone doing more damage than others yet being kicked because they simmed lower.
Edit: I do not believe fflogs developers intentionally scamming people. I believe their app is being used to mislead players by other players. And it is used inaccurately as a replacement for real 100% accurate data which it is not. And it seems to work in their favor financially. The inaccuracy of fflogs however is hidden by the community because many people stand to lose from that idea.
EDIT:
Going to explain what I am saying in simpler terms. Imagine a shopkeeper promises you that every day you visit him he will either give you a glass of juice or not. He guarantees you that you will get a glasses of juice 150 days out of 300. You either get a 0 glass of juice or 1 glass of juice. Now imagine that your doctor says you have an ultra rare case of hyperactive scurvy. And there is a juice shortage. To survive this disease and live to see tomorrow you must drink atleast half a glass of juice. Now your friend has this scurvy too. So the only place you can get juice is at that very shop.
An FFXIV version of a parser would say that you would get half of a glass of juice daily. It tells you that because it is guessing your average from past events. So fflogs is basically saying you never die from scurvy and your friend always dies from scurvy. Now if you go to the shop keeper you may get 0 glasses of juice and thus you both die. Or you can go to the shop and get one glass of juice; saving you both. This means FFlogs was always wrong because you either get a glass of juice or you do not, it is never a percentage or 50%.
If 9% of your damage is dot damage. Crit enhances it by 100% making you do 9% more damage. A direct hit enhances it making you do 25% more damage. If you crit and dhit all uses of this ability then you will do 9% more damage from the crit and 2.25% more damage from the DHIT. If your crit rate and direct hit rate are both 50% then fflogs will sim your values and tell you that you always do 5.62% more damage. In reality however you can also not crit or dhit any of those abilities. Causing a raid wipe, griefing and lost time. The variance between nocrit nodhit vs crit and dhit is 11.25%. FFlogs is off by 5.62%.
Raid dps makes this even more complicated. Now you have a neighbor hand you a card which says the shop keeper will give you 25% of a glass half the time, based on the color of the shirt the shopkeeper is wearing
Now fflogs takes your values and processes them for a while before posting them up. This reminds me that during college research labs there would be people who messed up measurements and took down incorrect values. Either that or they were lazy and did not pay attention. They would go home and do formulae and calculations until they came up with measurement values which not only made sense but also matched their peers. But were their values correct? did they actually learn how to do the measurements properly? answer to that is no.
People who are claiming that fflogs uses good models to sim dot data are actually supporting my argument. A model is a SIM, not a parser. Parsers are not suppose to give sim values. That's what sims are for.
People who are saying the app has been updated over the years and is better at giving different people same values via formulae that are being updated: parsers don't use formulae that's what sims are for. It just means the app is getting better at giving out consistent but incorrect values to different people in a raid setting.
Edit: To the app developers repeatedly stating FFXIV is not a hostile environment for 3rd party apps:
https://na.finalfantasyxiv.com/lodes...0c5d56cb2e4547