|
|
All Jobs Damage Simulator and Gear Sets
By SimonSes 2023-06-19 03:01:59
regular zanshin is pretty pathetic anyways, since you have to have a single hit attack round that misses. can't proc multi hit of any kind.
It's not THAT pathetic. Even at capped accuracy, with Zanhasso build, you will have like 60% rounds when you do single hits with 95% hit rate. In those rounds thx to Zanshin you will have 0.025% chance to miss, instead of 5% (not to mention if you miss a hit and hit with Zanshin you get massive TP boost from Ikishoten too). Makes whole gameplay with "pure" Zanhasso build much more consistent. Ofc with uncapped hit rate (blind, lost buff) Zanshin usefulness only goes up.
 Ramuh.Austar
サーバ: Ramuh
Game: FFXI
Posts: 10481
By Ramuh.Austar 2023-06-19 04:00:44
I know what ikishoten is, I know what it does, and the zanhasso set was something my simulations generated years ago when I brought it up originally. Single hit misses are pathetic. Using taints set since it's probably up to date, I see 3 QA, 6 TA, 8 DA (/DRG assuming,), 77 zanshin rate which would give me a 26% chance at zanhasso. Running 10K attack rounds gives me a distribution of
10,120 normal hits, 12 single hit misses, and a total of 2085 zanshin hit, which are generated from zanhasso and the single miss rounds.
That's 0.12% of your hits that can swing a second time due to a miss alone. 12 * 0.77 of those will attempt a normal zanshin hit, 9.24 and then 0.95 hit rate on that for 8.778 hits out of those 2k+ ones.
I will even be a little generous and say that roughly 0.5% of your zanshin hits are from an initial missed swing and the other 95.5% from actual zanhasso.
By SimonSes 2023-06-19 04:34:14
I see 3 QA, 6 TA, 8 DA (/DRG assuming,), 77 zanshin rate which would give me a 26% chance at zanhasso. Running 10K attack rounds gives me a distribution of
Ok so you are using this set, because it fits your argument.
My set has 101% Zanshin and only 3%QA and 4-5%DA
ItemSet 375956
It also matters when those hits occur not just how many of them on avg. They occur when you would otherwise miss and get 0TP. It makes gameplay much more smooth, because you almost never have a round with missed hit, while without Zanshin 3% of my rounds would miss.
By Dodik 2023-06-19 04:43:12
It doesn't work that way. Zanshin on missed hits is still subject to the same 95% hit rate. Zanshin swings from misses are not guaranteed to hit, they just gain more accuracy. Even at accuracy cap they, the zanshin miss swings, will hit up to 95% of the time.
Your total accuracy is 95% regardless. You just have more swings.
By SimonSes 2023-06-19 04:47:29
It doesn't work that way. Zanshin on missed hits is still subject to the same 95% hit rate. Zanshin swings from misses are not guaranteed to hit, they just gain more accuracy. Even at accuracy cap they, the zanshin miss swings, will hit up to 95% of the time.
Your total accuracy is 95% regardless. You just have more swings.
If you have 100% chance to make another hit with 95% hit rate after a missed hit with 95% hit rate, then your chance to get a 0TP round is only 0.025% instead of 5%. That's why I wrote "almost never".
サーバ: Carbuncle
Game: FFXI
Posts: 167
By Carbuncle.Samuraiking 2023-06-19 10:50:10
I see 3 QA, 6 TA, 8 DA (/DRG assuming,), 77 zanshin rate which would give me a 26% chance at zanhasso. Running 10K attack rounds gives me a distribution of
Ok so you are using this set, because it fits your argument.
My set has 101% Zanshin and only 3%QA and 4-5%DA
ItemSet 375956
It also matters when those hits occur not just how many of them on avg. They occur when you would otherwise miss and get 0TP. It makes gameplay much more smooth, because you almost never have a round with missed hit, while without Zanshin 3% of my rounds would miss.
Did you plug in Nyame hands into a calculator and it came out over Tatena hands somewhow, or do you just prefer the full cap on PDT? I generally run 42% PDT and don't die due to physical dmg, but it's also way less MEVA when using Tatena as well.
By SimonSes 2023-06-19 11:53:20
Carbuncle.Samuraiking said: »Did you plug in Nyame hands into a calculator and it came out over Tatena hands somewhow, or do you just prefer the full cap on PDT? I generally run 42% PDT and don't die due to physical dmg, but it's also way less MEVA when using Tatena as well.
Its for sure less dps than Tatenashi, but I prefer survivability. Drop on meva on feet is hard enough already and since I have 100% Zanshin anyway, Tatenashi advantage is mostly that 6~8sTP and I can live (literally ;P) without it.
サーバ: Carbuncle
Game: FFXI
Posts: 167
By Carbuncle.Samuraiking 2023-06-19 13:53:01
Nah, that's perfectly reasonable, was just making sure it was purely for survivability. I usually think the same way, but have been playing with 42% PDT and Tatena hands for quite a while now, and the only place I personally use SAM on and need the MEVA is Dynamis wave 1 boss where I am opening for the COR, but I just swap sets for that.
 Ragnarok.Primex
サーバ: Ragnarok
Game: FFXI
Posts: 229
By Ragnarok.Primex 2023-06-19 14:06:00
is there a way to forecast dps attributed to counters and retaliations from MNK and WAR, starting the baseline assumption of 100% uptime on agro/tanking? or, alternatively has anyone done any long parses as mnk or war tank with max or at least measured counter/retaliation rates? I know it would be pretty hard because there's other factors to consider such as weapon, mob def, atk etc. but has anyone ever taken a shot at this?
サーバ: Fenrir
Game: FFXI
Posts: 1339
By Fenrir.Richybear 2023-06-19 15:19:35
[+]
By Izanami 2023-06-19 21:33:08
Started playing with the damage program and still learning it. With Jinpu it says to use Fotia Gorget over SAM+2 neck. Most the sets i see show SAM+2 neck. I guess question is two fold, is Fotia Gorget better for hybrid WS
Hybrid WSs are tricky. The total damage dealt by hybrids is the sum of a physical and magical component, where the magical component is based on the physical component's damage and both components gain Weapon Skill Damage+ on their first hit. I'll approximate things below to explain why the code sees Fotia > JSE+2 for Jinpu in most situations.
Short Answer: Fotia > JSE+2
Fotia's FTP bonus applies to the physical and magical portions of the weapon skill, which are also both boosted by weapon skill damage. This results in Fotia adding roughly 5% to 15% damage to Tachi Jinpu, with a realistic boost of +10% (over having no neck equipped).
If you aren't attack capped, then the SAM JSE+2 neck only provides STR+25 to Jinpu's damage, which is only worth about +3% damage against enemies with relatively low VIT. If you are attack capped, then the JSE+2 neck should beat Fotia by about 1%, but you lose Fotia's "occasionally do not consume TP" bonus.
Recall that hybrid WSs have three FTP values. Fotia Gorget/Belt provides +100/1024 (~0.098) FTP to the first physical FTP value and the magical FTP value:
Physical FTP1 applies only to the first hit (always 1.0 for hybrids)
Physical FTP2 applies for extra hits after the first (1.0 unless FTP transfer, then copy FTP1)
Magical FTP applies to the total physical damage to create the magical damage (use the BG Wiki values; scales with TP)
First, we look at the general damage equation for Hybrid WSs:
Damage = (Physical*(1+wsd) + Magical*(1+wsd)) = (physical*(1+wsd) + physical*(1+wsd)*magical_ftp*(1+wsd)) = physical*(1+wsd)*(1 + magical_ftp*(1+wsd))
where only the first physical hit gains the (1+wsd) multiplier.
For a 2-hit Hybrid like Jinpu, the physical portion has 2.0 total FTP (1.0 for each hit). Adding 0.098 FTP to the first hit from Fotia Gorget is a direct gain of about ((1.0 + 0.098)*(1+wsd) + 1.0) / (1.0*(1+wsd)+1.0). If you consider multi-attack procs, then this equation is generalized to (wsd*(1+fotia) + fotia + N)/(N+wsd), where N is the number of hits the WS has after multi-attacks, fotia is the Fotia bonus (~0.098), and wsd is the total weapon skill damage stat (50% = 0.50). For a triple attack (4 hits total) and 50% WSD, Fotia Gorget directly increases physical damage dealt by (0.5*(1.098)+0.098 + 4) / (4+0.5) = 3.2%.
For the magical damage: we first multiply the physical damage by the magical FTP, which is the value reported on BG Wiki modified by Fotia bonuses. For Jinpu, the magical FTP ranges from 0.5 to 2.5 based on TP. Fotia will therefore add 19.6% to 3.9% bonus FTP, depending on your effective TP. At 2000 TP with 0% WSD, this is a 6.5% increase in magical_ftp. With 50% WSD that becomes 9.75%.
Now putting these two terms together, we see that Fotia adds (wsd*(1+fotia)+fotia+N)/(N+wsd) from the physical term and (1+(magical_ftp+fotia)*(1+wsd))/(1+magical_ftp*(1+wsd)) from the magical term. We can simply multiply these together to get the full fotia bonus.
Fotia Results
For the best-case scenario for Jinpu with no multi-attacks (N=2), 111% WSD (best gearset with traits), and 1000 TP, fotia increases damage dealt by [(1.11*(1.098) + 0.098 + 2) / (2+1.11)] * [(1 + 0.598*2.11) / (1 + 0.5*2.11)] = 1.066 * 1.1006 = 1.174, or +17.4% damage at best.
In the worst case scenario for Jinpu with double quad-attack procs (8 hits without missing), 111% WSD, and 3000 TP, fotia increases damage dealt by 1.0226 * 1.0329 = 1.056, or +5.6% damage at worst
Realistically, you'll WS at 2000 effective TP and have ~45% DA (subjob WAR) and 3% QA for an average of 2.90 hits per WS, including a 95% hit rate cap. Fotia adds +10.3% damage in realistic situations over no neck option.
At 2500 effective TP, the bonus is 9.3% in the realistic case.
Edit: Adding magic multipliers to the magical damage term only increases these values slightly. But since this is a rough estimate, I exclude them here.
In comparison, estimating the damage contribution to Jinpu from the SAM JSE+2 neck is easy. If you aren't attack capped, then JSE+2 neck only provides +25 STR. Jinpu uses +30% STR (+10% DEX:Utu) mods, and my code shows 344 STR + 252 DEX in the best Jinpu set with Fotia, resulting in 128 WSC, plus 330 weapon DMG from Dojikiri and 14 fSTR (against a low VIT Apex Bat). The base damage is then 128 + 330 + 14 = 472. SAM JSE+2's +25 STR is worth 25*0.3 = 7.5 WSC and ~7 fSTR for a total base damage increase of ~15, worth (472+15)/472 = 3% bonus damage. On a respectable enemy with much higher VIT, or with ML>20, this value is lower. Fotia Gorget is 1.5x better at its worst, and roughly 3x better in reasonable settings, than the JSE+2 neck when not attack capped.
If you are attack capped, then the PDL+10% on SAM JSE+2 neck is immediately +10% damage on top of the +3% from the +25 STR. In this case, the JSE+2 neck should pull ahead by about 1% or so, depending on the enemy's VIT. For what it's worth: you also lose the 1% chance to not consume TP from Fotia.
Note that the other Great Katana hybrids have the same magical_ftp values, but are all 1-hit WSs, so fotia is worth even more on them (~12% vs ~10%).
how closely should i kinda follow the outputs?
Think of the code as a spreadsheet. The sets you typically see in guides are from players manually checking outputs on a spreadsheet one at a time until they find a reasonable-looking best set based on their intuition. My code does the same thing automatically, but checks tens of thousands of sets, typically improving upon the sets that players manually build by 1~3%. This is usually only 1 or 2 minor swaps to their sets, though.
The sets that the code outputs, like the sets you see from spreadsheets, are all situational for the situation that they were created in. You should make reasonable swaps based on your situation, but I recommend using the quicklook buttons to see how much damage you'll lose with each swap.
[+]
サーバ: Carbuncle
Game: FFXI
Posts: 167
By Carbuncle.Samuraiking 2023-06-20 05:30:53
Does anyone know the proper stats for the Sortie bosses, just the basement ones, by chance? I'd love to plug in proper sets for E, F, G and H but have no clue what their stats are.
Similar question with Wave 3 Dynamis-D Fomors. Obviously their eva/def vary WILDLY based on which job you are pulling and if they are one of the three small ones, or a mini-boss, but a somewhat high-tilted average would be nice to ballpark on them if anyone knows. I have no idea how to go about testing it myself and haven't seen anyone post the exact stats outside of a 1600-ish average for their evasion. Nothing about def or stats though.
By SimonSes 2023-06-20 07:02:38
Carbuncle.Samuraiking said: »Does anyone know the proper stats for the Sortie bosses, just the basement ones, by chance? I'd love to plug in proper sets for E, F, G and H but have no clue what their stats are.
Similar question with Wave 3 Dynamis-D Fomors. Obviously their eva/def vary WILDLY based on which job you are pulling and if they are one of the three small ones, or a mini-boss, but a somewhat high-tilted average would be nice to ballpark on them if anyone knows. I have no idea how to go about testing it myself and haven't seen anyone post the exact stats outside of a 1600-ish average for their evasion. Nothing about def or stats though.
I think they were only tested by JP group using mages strategy, so they only tested stats relevant for magic damage. Afaik none knows exact VIT and DEF values.
サーバ: Carbuncle
Game: FFXI
Posts: 167
By Carbuncle.Samuraiking 2023-06-20 11:46:17
Carbuncle.Samuraiking said: »Does anyone know the proper stats for the Sortie bosses, just the basement ones, by chance? I'd love to plug in proper sets for E, F, G and H but have no clue what their stats are.
Similar question with Wave 3 Dynamis-D Fomors. Obviously their eva/def vary WILDLY based on which job you are pulling and if they are one of the three small ones, or a mini-boss, but a somewhat high-tilted average would be nice to ballpark on them if anyone knows. I have no idea how to go about testing it myself and haven't seen anyone post the exact stats outside of a 1600-ish average for their evasion. Nothing about def or stats though.
I think they were only tested by JP group using mages strategy, so they only tested stats relevant for magic damage. Afaik none knows exact VIT and DEF values.
Good for hybrids, thanks.
Quetzalcoatl.Tomasello
サーバ: Quetzalcoatl
Game: FFXI
Posts: 313
By Quetzalcoatl.Tomasello 2023-06-20 12:02:14
Think of the code as a spreadsheet. The sets you typically see in guides are from players manually checking outputs on a spreadsheet one at a time until they find a reasonable-looking best set based on their intuition. My code does the same thing automatically, but checks tens of thousands of sets, typically improving upon the sets that players manually build by 1~3%. This is usually only 1 or 2 minor swaps to their sets, though.
The sets that the code outputs, like the sets you see from spreadsheets, are all situational for the situation that they were created in. You should make reasonable swaps based on your situation, but I recommend using the quicklook buttons to see how much damage you'll lose with each swap.
Totally agree 100% and thank you for all your hard work on this. Amazing program, I've been using it extensively to tighten up all my sets. It's really appreciated!
By Izanami 2023-06-20 12:41:13
Been using this to work out nuke sets on BLM SCH RDM GEO in Sortie, is there any way to get it to account for player base INT and Mlvl INT?
The current version of the code does not allow you to adjust master levels. For now, the code assumes ML20. If you want to add more INT to increase dINT, then I suggest removing some INT from the enemy instead.
In a few weeks, I'll release an updated version of the code that includes a drop down for master levels, which will increase base parameters (STR, DEX, etc) and combat skills, and unlock different tiers of subjob traits and abilities (such BLM/SCH Ebullience and more WSD+ from /55DRG).
サーバ: Carbuncle
Game: FFXI
Posts: 167
By Carbuncle.Samuraiking 2023-06-20 13:58:26
Not to add to your workload and make you feel like we are piling up requests, but that's exactly what I am doing... any chance you can add a save feature for the mobs at the top? I'd like to save mobs/stats presets for different things instead of trying to remember and reset them every time if that isn't too much trouble for you.
I absolutely hate the spreadsheets everyone was previously using and love that you are making user-friendly UI systems.
[+]
By Izanami 2023-06-21 07:11:29
Carbuncle.Samuraiking said: »any chance you can add a save feature for the mobs at the top? I'd like to save mobs/stats presets for different things instead of trying to remember and reset them every time if that isn't too much trouble for you. This is already a feature in the updated GUI since the entire app is (poorly) written as a Python class object. I tested this a while ago with changing the default gearset, job, and ML, but pretty much everything that is selectable in the GUI can have its default value changed from outside of the app object itself.
I plan to add an input file to read default values from and a menu option to output the current values to the input file in the proper format.
Could you add an option for Subtle Blow set if possible? The code only swaps two items at once, which usually only leaves room for two conditions at once, which we currently use on PDT+MDT. I think it should be possible to allow up to three conditions with two swaps, but I'll have to be clever with how I write the automated set finder. Either way, this won't be available for a long time if it is possible.
[+]
サーバ: Asura
Game: FFXI
Posts: 25
By Asura.Kaelann 2023-06-21 18:12:04
Found one small typo: Heathen's Earring +1 is listed as 89 PLD instead of 8.
[+]
サーバ: Carbuncle
Game: FFXI
Posts: 167
By Carbuncle.Samuraiking 2023-06-22 04:14:15
This is already a feature in the updated GUI since the entire app is (poorly) written as a Python class object. I tested this a while ago with changing the default gearset, job, and ML, but pretty much everything that is selectable in the GUI can have its default value changed from outside of the app object itself.
I plan to add an input file to read default values from and a menu option to output the current values to the input file in the proper format.
I'm stupid. I didn't think it was possible to open Python files with notepad++ and edit them, I thought it had to be done with a special program or something. I'll dig around in the tab_inputs.py file and rework some defaults, thank you very much for the info. This will make it a hell of a lot easier. I'll dig around the rest and find the one with job/gear etc. in it, too.
Also, confirmed the SAM stuff is working like you said. When I put the proper stats and capped haste etc. in it started printing out what people thought was the BiS already, Ryou feet, Moonlight Nodawa, Aug Orb +1 etc. for a proper Zanhasso set.
I saw someone mentioned earlier that you only had two variables, PDT and MDT, but were thinking about adding a third if you can make it work, like SB. Is it possible to edit the PDT/MDT (temporarily, for now) to edit out those words in the file to other stats like "crit rate" or whatever it's labeled as and have it prioritize that?
By Izanami 2023-06-22 06:48:04
Found one small typo: Heathen's Earring +1 is listed as 89 PLD instead of 8.
That's a lot of Paladin in my Dark Knight. I'll fix this issue later, thanks. For now, you should be able to directly edit the gear.py file to fix it in your own version of the code.
Carbuncle.Samuraiking said: »This is already a feature in the updated GUI
I'm stupid. I didn't think it was possible to open Python files with notepad++ and edit them, I thought it had to be done with a special program or something.
I should be more specific. The current version of the GUI (on GitHub) is written using procedural Python and can't easily be updated as is (unless you're running the .py version of the code and not the .exe). The updated GUI (still in progress) is object-oriented and will allow for easily changing the defaults within a single input file.
For now, there is no easy way to update the defaults with Notepad++.
Carbuncle.Samuraiking said: »I saw someone mentioned earlier that you only had two variables, PDT and MDT, but were thinking about adding a third if you can make it work, like SB. Is it possible to edit the PDT/MDT (temporarily, for now) to edit out those words in the file to other stats like "crit rate" or whatever it's labeled as and have it prioritize that?
I could add a third conditional and have the code zero-out the PDT and MDT conditions if the Subtle Blow condition is non-zero. This is probably the easiest way to handle it. I'm out of town for a while so I won't have a chance to work on this for a few days at least.
サーバ: Asura
Game: FFXI
Posts: 595
By Asura.Melliny 2023-06-22 11:37:49
Quote: That's a lot of Paladin in my Dark Knight.
That you Cecil Harvey?
サーバ: Asura
Game: FFXI
Posts: 135
By Asura.Chendar 2023-06-22 11:38:13
Wondering why does it change out my weapons sometimes? example would be RDM i use kraken a lot but it never wants to keep it. Thanks.
If you have other weapons selected in the gear selection tab it'll swap them out for whatever gives the most ws damage.
Just uncheck all the offhand weapon options and it should stop doing that.
[+]
By Masaru 2023-06-29 03:37:05
DNC Sub shows Dual Wield: 25. I think it have to be 15. Version: 2023 June 18
[+]
By Izanami 2023-06-29 08:41:56
The app add MDT from Shell V when creating any Sets if WHM is ticked, however, Shell V is not chosen in the drop menu. This was intentional at the time it was added since I assumed if you have Dia/Haste/Boost, then you have Shell V by default. But it does make sense to leave Shell 5 as optional to help build sets for fights where you need to remove Shell to avoid having it absorbed by the enemy.
I've added a Shell V (MDT-29%) checkbox. WHM must be enabled to use this checkbox, and selecting/deselecting WHM will automatically select/deselect Shell V, which can be manually deselected if you do not want it on. This is the same behavior as before, but with the option to disable it.
DNC Sub shows Dual Wield: 25. I think it have to be 15. Version: 2023 June 18 Thanks. This was a typo and has been fixed.
|
|
