Hi everyone ! Sorry this was a bit late; I thought it'd be a quick topic to cover, it took me almost twice as long to edit than my previous videos! Also, I put together a calculator sheet on my DPS calculations spreadsheet, where you can input the stats on two pieces of gear and compare how good they are, based on your stats: docs.google.com/spreadsheets/d/1ZrJokNh_0AF_9welc7Etz6K8jlpzi5bXpiWz-mQZa78/edit#gid=2055222127 Also, to clarify a few things / address a few oversights: Flat ATK actually has slightly higher value because it'll additionally benefit from any ATK% buffs your comp has. As such, the breakpoints for how much base ATK you need before ATK > Flat goes up by a variable amount depending on how many buffs you have - check my spreadsheet for accurate calculations! There are a few missing points I didn't make in the Crit section - first off, every 1% crit added is 0.5% dmg on average - but this is additive to a singular modifier! Refer to this table: prnt.sc/gpeQu1aFOWrJ Additionally, several high-level enemies in Vera have high critical resist, making the value of crit even lower. I've also heard some concerns about people worrying that some may take "ATK % is better than DMG %" at face value - remember, at a 1:1 scale, it's not! It's better (in general) to roll on for gear, but when comparing it from other sources, the outcome is likely different based on how much % of each you're comparing!
How much Crit % should I have for Crow's matrices to be better than Shiro's? I read somewhere that I shouldn't be using Crow's until I have 80% Crit :/
Firstly, Thanks for you hard work =) I got some question... Isn't Frigg only has 25% increase frost attack? I'm not sure if A6 is adding another 25% on top of 15%? or just increase to 25% also if you add metrics on frost team, it would easily reach the limit...
Maybe I am just blind, but where can I find the accurate calculation for the threshold of when ATK % > Flat in your spreadsheet? I was looking around, but couldn't find it.
Your topics and production value are incredible. You really know how to respect your viewer's time as well by getting straight to the points! Thank you so much!!! :)
Great explanation. Getting the gear with great sub stats is the most difficult part after the stat locker arrived in the game. Good luck to everyone on their gear hunt!
The production/editing quality of your content is astonishing giving the size of your channel. In the community I've seen that there are mainly two ppl who are "highly appraised", Gateoo, and you Maygi. And because of that, if you keep it up I'm sure you'll grow quite good :D Sorry I never comment (on YT I do like 1 comment every two or three months). Just know you have a loyal fan from me and a faithful community, at least on reddit.
I agree with you.. the reason why I prefere to use elemental atk % instead of elemental damage % is because elemental damage % doesn't work on Lin, while if you use elemental atk % it will give Lin more damage, I tested both a 2.02% elemental damage and a 4.14% elemental atk and the elemental damage % does slightly more damage than the elemental atk % but if you have Lin it is better to use the elemental atk % imo
We don't deserve Maygi. Not only are you dedicated to putting the right info out there, you're well versed in both math and programming? Gonna check out the open source since I'm curious although I'm hesitant as to what ToF thinks about it. Is it potentially against ToS?
This is so informative and a big help cause i dont have time to do the math on my own(really just s2pid when it comes to math lol). Also your voice is so damn cute so now you've got your self a new subscriber ❤️
With Lyra upcoming I think something you should mention is also the fact that some characters have seperate scalings Like healing for example on lyra : you heal yourself for 15% of your atk on each punch that hits an enemy If you get a lot of damage% yes you get a little more damage but you are missing out on a considerable amount of healing That's why I personally like Lin's realm giving atk% instead of damage boost because that means it affects the healing lyra provides Also why I would argue Tsubasa trait may be overall a better trait to use for lyra dps as it's easier to keep up in comparison to Samir's trait due to lyra being a face tanker and gives atk% instead of damage boost
shiro's trait is superior in most cases for lyra. 26% attack up, 50% of the time. Some would say that averages to a 13% attack up, but in reality, your gonna be using that period its up to burst damage more than not.
AS everytime, very valuable video. Good luck for your overlay plugin, I’m sure we’ll all rush on it (at least the ones not yet disgusted by the game ;))
Sadly there is a small mathematical error at 3:40. The damage % increase you get from additional crit chance is dependant on your current crit chance and has "diminishing returns". Going from 0% to 1% crit chance gives this 0.5% DMG increase, but going from 50% to 51% crit chance only gives 0.4% DMG increase. The correct formula for DMG% by critrate should be (((1-old_crit%-delta_crit%)+(old_crit%+delta_crit%)*crit_dmg) / ((1-old_crit%)+old_crit%*crit_dmg)) -1
additionally mobs at lvl 80 have 30% crit resist. So the first 30% crit rate are 0% dmg increase. it might be worth looking into gloves and boots again. Sadly for volt players crits are essential for crow/samir and provide more than the simple 1.5 crit dmg multiplier.
Could you make a video for the best simulacra traits to use? All ppl recommend always samir but it would be nice if someone like You could make a video about it comparing samir, tsubasa, Lin, shiro, frigg, etc...
I wonder if damage% is actually multiplicative or if it's additive within certain buckets and the buckets are multiplicative in the formula. Hard to tell because there's some variance between individual attacks, so you can't really test it easily. For instance, formula might be (elemental atk) * (1+ sum of atk% buffs) * (skills scaling) * (1+ sum of damage % buffs) * (1+ sum of elemental damage % buffs) * (1+sum of resistance reduction % debuffs)), where stacking a bunch of damage buffs isn't great, and you would rather have some atk% buffs, some damage % buffs and some resistance reduction debuffs for optimal damage. Or it could be (elemental atk) * (sum of atk% buffs) * (skills scaling) * (1+damage % buff A) * (1+ damage % buff B) ...etc * (1+elemental damage % buff A) * (1+elemental damage % buff B)...etc * (1+resistance reduction % debuff A)...etc, where you just want as many damage % buffs as possible and the single strongest atk% buff you can get. Genshin for instance used the first formula, and you were getting diminishing returns if you stacked just a bunch of elemental damage % buffs compared to spreading out between Damage %, elemental damage % and atk %. E.g. 250% elemental dmg would be less damage than 200% elemental damage and 50% attack.
Attack % increase boosts are more future proofed I feel, even if early on the the flat ones are better. Don't the visors also have flat crit too btw? Flat crit vs. flat attack, flat crit should win out. Crit % is probably future proofed as well though since there seems to be an offset in your level to the flat crit you have that crit % would supercede. I'm not sure how quick we will power creep to such points though, especially on global. In CN, maybe it will occur a lot faster though.
@@merem6972 This in no way shape or form makes crit less worthwhile. Crit is easily the most important stat to get from gear as it's so limited. Flat attack has diminishing returns, while crit doesn't. Like, 7000 crit is far more impressive than like 2500 attack. Which, is what you'd be looking at comparatively for crit rolls vs. attack rolls on gloves and boots.
ah yes true i forgot to add the real tldr: "just roll whatever gets any dmg sub because beggars cant be choosers and we are all beggars in the context of gear"
You're perhaps the most informative content creator on ToF. I have a question about crit. Let's say I have 12K crit and 20K frost attack, and using Saki, Lin, Frigg. Would a gear with 2K crit and 800+ frost attack be better or worse than a gear having 2.8K crit? Thanks in advance.
Without even doing the math, I can tell you with confidence that the 800+ Frost attack is worth way more than the 800 Crit. Crit on boots/gloves is nice, but it's not THAT good! (I did the math anyway. The pure crit one is around 2/3 as good as the one with attack)
@@Maygi Thank you so much! I'm glad I saved the gloves and boots. I'm using the 2.8K crit version as everyone keeps telling me that crit is everything in this game, that always go crit gloves/boots. But with the crit resist in Vera mobs, I've been rethinking this.
this is calculation on papper on field battle Atk% had more advantage on most situation cause usually average player only build 1 elements you ask why? Dmg Boost shine better to Atk% when you hit enemy with weakness elements other than that atk% shine more be with enemies with resist nor normal one you can't see much using cactus on Hota studio you need to compared it on live battle to see it by your own in conclution if you want to be Op on any situation using dmg boost% make sure you fight those with your elemental weakness
From my testing with my ATK% / Bonus% [ without ATK% / Buff ] i got 21k ATK [ 24k with ATK% ] & i got try the value of +2,09% or +4,14% [ Bonus Value was better than ATK% ] depend en ATK Base & if you have some buff or Final Damage Bonus (i think) u already want some Bonus Damage% 😆
Hey Maygi, I was wondering, could you make an updated guide about gear stat preferences comparing wether dmg%, atk% or flat atk rolls are better, especially at endgame/maxed stats/gear, since this guide is over 5 months old by now? I just got a penta roll on a reactor and just noticed how even that over 8% atk reactor is actually worse than the 800 combined flat attack, 2.XX atk% reactor I had (34,8XX atk on the 8% flame atk% reactor, but 34,9XX flame attack on the 800 flat attack reactor, and that before any additional buffs and matrice effects id gain in combat) and noticed that with multiple other pieces as well. Especially once buffed and matrices etc in effect, it just seems like atk% is really really bad compared to flat attack. However with damage rolls its not that easy to compare but I feel like it may be worth revisiting with the stats we can achieve. Also we have many new chars, matrices, gear and especially now titan gear. Also on the augmented gear I noticed, that the max upgrade you can achieve (the green/blueish number under the stat on the augment window) vary a lot, but I simply dont want to throw a lot of mats at the system to test it out. On the 8% reactor for example I only get about +1.3% as maximum upgrade, which is less than 1/4th of the total stat, on the other reactor its 539 +max 297 flat attack and 322 +max 359 flame attack. So its more than the base flame attack as a maximum upgrade and about 60% of the flat attack value as an upgrade. That just seems like once that thing would be maxed out, flat attack rolls value would sky rocket compared to atk%. Is there something I'm overlooking? Maybe the answer is even to aim for a good mix of both?
Flat ATK is buffed by ATK%, as mentioned in the pinned comment. If you check the gear calculator sheet in the spreadsheet, there are checkboxes to add certain attack buffs, and then you can evaluate the comparison between ATK roll value vs ATK% there by inputting your stats! On the topic of an updated guide, though, I'll probably look into making something for titan gear optimization at some point...
Thankyou, I was wondering what exactly made the change from lins frost damage% to atk% so much worse. So it's going from a multiplicative buff into an additive buff, good to know this. Jeez and to think with all these insane nerfs from CN for balancing, they won't even give us a reasonable amount of advanced modules for gear. I am still 10 away from a full level 20 set (with my level 40 gloves) and the other modules aren't much better, and yet we're supposed to somehow be caught up with cn in 2 months who had pretty much maxed out gear by the time vera rolled around, and we're missing a bunch of characters. They are just shafting global by not giving us any reasonable amounts of materials, I mean seriously we still haven't gotten a re-run of nemisis yet.
you missed only one point in your Video Maygi which is Lin gives 15% Atk increase with the Skill globaly and with Frost Realm it's 10% more making Lin at A6 in Frost Comp giving 48% Atk % Buff combined with A6 Frigg and the Frost Resonance a Frost main A6 player gonna have 103% Atk % Buff without Coco and her matrieces with that being said it's only possiable for Whales at this point but great Video as Always
Just because I was curious I went and made a spreadsheet for all of the threshold values for if you roll 5 times into either Flat attack(s) or ATK %... Then made a simple calculator at the top so people could just see what they'd need for a given Flat attack vs ATK % because I realized people would probably not want to read through a huge table. docs.google.com/spreadsheets/d/1KkALOdt5igCa0-TtzQZAXn8vCZ3YiOTZJ0ll766EA-s/edit?usp=sharing Example: If you have a combined Flat attack of 1000 then the thresholds of your Attack (ignoring what is added by the piece(s) you're looking at of course) that ATK% would be: By # of rolls into the stat 0: 79365.08 1: 37037.04 2: 24154.59 3: 17921.15 4: 14245.01 5: 11820.33 Which means that for someone like me with over 15k in my attack stat I would need at least 4 rolls into ATK % to beat a 1000 Flat attack piece.
I have 28k frost dmg... I have 1 item with 1.6 % atk frost , 260 atk frost and 60 atk ... And i have other item with 3.5 % dmg frost what is better? pls help me
The DMG% is better in this case. For future reference, you can use the gear comparison calculator I made, where you can input any stats and see the value: docs.google.com/spreadsheets/d/1ZrJokNh_0AF_9welc7Etz6K8jlpzi5bXpiWz-mQZa78/edit#gid=371662811
How does damage formula work? Is the damage the number you see when you attack a mob (Damage recieved = (Attack x Buffs) - Enemy Resistance)? If this is so, wouldn't damage be a bit better if it's calculated after their resistances?
Any possibility that the tof overlay also calculate damage? played some MMO before... kinda triggered me that the damage meter TOF used only calculates overall dmg instead of damage per second... It would be great if you could add it... also useful to see which weapon dmg breakdown (samir does 65% of dmg etc).
Now we just need to figure out why % Elemental Atk on new gears from vera doesn't give any CS. My Engine went 4 times into % Volt Atk which is great but my 5 star golden picece only gives 478 CS. It's a joke cause I have crew mates with % Element Damage with 6 and 7k CS on 1 item!!
Flat crit on gloves and boots is a different story - it's quite strong in comparison to flat attack at current levels! You can use the gear calculator to pinpoint exactly what's better with your stats: docs.google.com/spreadsheets/d/1ZrJokNh_0AF_9welc7Etz6K8jlpzi5bXpiWz-mQZa78/edit#gid=2055222127
There's no difference between "final damage" and "elemental damage" %. The wording they use across the game is inconsistent, so it can pretty much mean anything (Couant is "final damage" and is additive, and most other sources of final damage/elemental damage are multiplicative).
@@Maygi I noticed that the wording "final damage" started to be used after Lin. Perhaps it was introduced so that players understand that the "final damage" will increase not only the damage of their main element, but also the altered?
@@Maygi but you know Maygi I tested Scylla and Haboela, Haboela is always stronger. Matrix Used, (2pc Crow, 2pc Samir), (4 pc haboela), (4pc Lan) There might be a difference between these enhancements.
Yep, at the end of the day the damage calculation is a ton of multipliers smushed together, resistance being one of them, so relatively speaking, any stat/dmg% buff is worth the same regardless of resistance~
The wording is completely arbitrary. DMG% is just DMG%... but note that some forms of DMG% are additive to each other, such as that from Couant (listed as "final dmg") and shield (which is just "damage"), and even several matrices such as Shiro ("damage"). The wording literally means nothing and it can be confusing...
@@Maygi so what you wanna say, it's basically the same? I ask this because I noticed it at the new Bosses SSR Matrice at 2.4 update. Habra increase Final damage and Scylla increase elemental damage
hi, im using crow as my main dps with 53% crit rn. I have 2 gloves, one would give me addtl 990 atk while the other will give me 11% crit. What would be better?
kinda yes if you have around 16500 attk currently, you need total of 280 flat attack (elem attk + general attk) for each roll to compare with attk%. so if you roll total of 900 flat attk, 3 roll on attk% are only better with 840 flat attk below. But when you reach 18000 attk and you roll total of 900 flat attk, 3 roll on attk% are around the same value as 900 flat attk. 3 roll on attk% also may come with flat attk so you need to minus it first to compare. Getting 900+ and above on flat is harder than to get 3 roll on attk%, so attk% is better since you got 2 roll that may go into flat attk and attk% is a fix value. Also 900+ flat attk to get with only with 3 roll is super rare. So I think a single roll of attk% is still a better roll than a high roll of flat attk.
"Frost ATK is boosted by 2.4% every 3 seconds upon entering combat. Stack up to 10 times. Lasts for 5 seconds. In addition, gain hyperbody and become immune to control effects in the Frost Domain." Since it's an additive you'll get the dimenishing returns, yes, but at full stacks you'd still have the same Frost ATK increase as you would if you'd just gotten 24% straight away. (Floats are the percent increase and then we add 1 for 100% since multipling a float less than 1 would give us less than our initial value.) 10,000 * (1 + (0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024)) = 12,400 10,000 * (1 + (0.24)) = 12,400 If it was DMG% instead you'd get: 10,000 * 1.24 = 12,400 Now hold on, that's the same value! Well that's because we're only looking at one increase source. Lets look at two examples of another set. One will have two 10% ATK buffs, and the other will have one ATK% and one DMG% of the same value: ATK%, ATK%: 10,000 * (1 + (0.1 + 0.1)) = 12,000 ATK%, DMG%: 10,000 * (1 + (0.1)) * 1.1 = 12,100 Heck! Let's double each set! Quad ATK%: 10,000 * (1 + (0.1 + 0.1 + 0.1 + 0.1)) = 14,000 Double Sets: 10,000 * (1 + (0.1 + 0.1)) * 1.1 * 1.1 = 14,520 The key thing to note is this is only in comparision to other equivalent DMG% options. All this diminishing returns stuff means is that if you _could_ get an equal DMG% then you'd value that higher than an equal ATK%. This is why Maygi went into the part about reaching 100% ATK increase being the start of the diminishing returns because only at that point, with our doubled ATK% over DMG% rate, do we actually start to see the diminishing returns.
Samir trait is better than the frigg trait if comparing the damage buff given BUT if you are not good at dodging and always keep dodging, frigg trait might be better since you will be hitting enemies more.
So, thanks to the changes in global DMG % is useless, because with the buffs of cn that goes over 100% would be usefull I didn't understand the part where flat atk is better than atk % :(
There are thresholds of stats where rolls on Flat ATK are better than the 1.44% ATK roll on gear. The lower base ATK you have, the higher the relative value of Flat ATK!
It's a matter of how much actual attack you'd get from 1.44% increase and how much attack you'd need to have before that % increase would be better. You can find out the result by taking the flat attack value and dividing it by 0.0144 which gives you the threshold value, or to put it another way at that threshold value you'd recieve that much effective attack so anything higher than that threshold is better for ATK%. Here's a table of all possible upgrade values (all ATK% upgrades are exactly 1.44% so we only have to factor that): i.imgur.com/SJZHmre.png You can essentially look at that table and find where your Attack stat is at to find out how much Flat attack you can accept. For example: Say I had 15,000 ATK. Anything less than 216 would be a worse roll than rolling ATK%. In the end though, ATK% will stick around longer as you gain base attack stats from other sources.
The flat attack table confused the heck out of me Shouldn't wording be "before flat attack > attack%"? i.e. for when you have 21667 attack, you need 312+ flat attack for it to give more attack then 1.44 attack %
I agree that this table is a bit wordy and unintuitive... but no, this is the intended order! The more total ATK you have, the less flat ATK is worth (relatively) in comparison to ATK %. However, one thing to note that I didn't mention on the video is that Flat ATK is also subject to ATK buffs itself, so the left column of that table should be "Flat ATK * your avg ATK buffs", raising the thresholds even higher for ATK% to be better. (do check out the gear calc spreadsheet for a dynamic calculator where you can select your buffs: docs.google.com/spreadsheets/d/1ZrJokNh_0AF_9welc7Etz6K8jlpzi5bXpiWz-mQZa78/edit#gid=0 )
Those flat attack values are possible values within the upgrade range of both Attack and Elemental Attack. Since the ATK% upgrades are all 1.44% that's the value that was used. You can find it out the threshold for any given value by dividing the Flat attack by 0.0144. This will give you the amount of Total attack you'd reach before that Flat starts having less than the 1.44% in increase. I've screenshotted a table of every value from 93 to 312 rounded to the tenths place here: i.imgur.com/SJZHmre.png
![[Math] Lin Combo Breakdown](http://i.ytimg.com/vi/P4pS4dyd0I4/mqdefault.jpg)
Hi everyone ! Sorry this was a bit late; I thought it'd be a quick topic to cover, it took me almost twice as long to edit than my previous videos!
Also, I put together a calculator sheet on my DPS calculations spreadsheet, where you can input the stats on two pieces of gear and compare how good they are, based on your stats: docs.google.com/spreadsheets/d/1ZrJokNh_0AF_9welc7Etz6K8jlpzi5bXpiWz-mQZa78/edit#gid=2055222127
Also, to clarify a few things / address a few oversights:
Flat ATK actually has slightly higher value because it'll additionally benefit from any ATK% buffs your comp has. As such, the breakpoints for how much base ATK you need before ATK > Flat goes up by a variable amount depending on how many buffs you have - check my spreadsheet for accurate calculations!
There are a few missing points I didn't make in the Crit section - first off, every 1% crit added is 0.5% dmg on average - but this is additive to a singular modifier! Refer to this table: prnt.sc/gpeQu1aFOWrJ
Additionally, several high-level enemies in Vera have high critical resist, making the value of crit even lower.
I've also heard some concerns about people worrying that some may take "ATK % is better than DMG %" at face value - remember, at a 1:1 scale, it's not! It's better (in general) to roll on for gear, but when comparing it from other sources, the outcome is likely different based on how much % of each you're comparing!
please stream more
Don't be Vtwitter D:
How much Crit % should I have for Crow's matrices to be better than Shiro's? I read somewhere that I shouldn't be using Crow's until I have 80% Crit :/
Firstly, Thanks for you hard work =) I got some question...
Isn't Frigg only has 25% increase frost attack? I'm not sure if A6 is adding another 25% on top of 15%? or just increase to 25%
also if you add metrics on frost team, it would easily reach the limit...
Maybe I am just blind, but where can I find the accurate calculation for the threshold of when ATK % > Flat in your spreadsheet? I was looking around, but couldn't find it.
Your topics and production value are incredible. You really know how to respect your viewer's time as well by getting straight to the points! Thank you so much!!! :)
Great explanation.
Getting the gear with great sub stats is the most difficult part after the stat locker arrived in the game. Good luck to everyone on their gear hunt!
The production/editing quality of your content is astonishing giving the size of your channel. In the community I've seen that there are mainly two ppl who are "highly appraised", Gateoo, and you Maygi. And because of that, if you keep it up I'm sure you'll grow quite good :D Sorry I never comment (on YT I do like 1 comment every two or three months). Just know you have a loyal fan from me and a faithful community, at least on reddit.
Awh, this is really wholesome !! Thank you !!
I agree with you.. the reason why I prefere to use elemental atk % instead of elemental damage % is because elemental damage % doesn't work on Lin, while if you use elemental atk % it will give Lin more damage, I tested both a 2.02% elemental damage and a 4.14% elemental atk and the elemental damage % does slightly more damage than the elemental atk % but if you have Lin it is better to use the elemental atk % imo
WHat about Flat atk?
I'm so happy that a girl is representing the best content creator of tower of fantasy
We don't deserve Maygi. Not only are you dedicated to putting the right info out there, you're well versed in both math and programming?
Gonna check out the open source since I'm curious although I'm hesitant as to what ToF thinks about it. Is it potentially against ToS?
Lovely explanation. You earned yourself a sub!!!
we love you, magical math fairy
woah almost ditched my maxed volt atk% for volt dmg%, so it really is the best on my server thanks
we need more people like you in this game, not some self entitled bastards that only care about packages.
It's like you read my mind with this video
The goddess has graced us. 🙇
thank you for explaining this, actually really good video
Thankyou pinky math sensei
Ooh, this channel may just have been gold struct. Great video!
Maygi if you reading this, i love you hshshshshsh your video detail is really helpful
thank you for the wisdom and knowledge, magical math fairy
This is so informative and a big help cause i dont have time to do the math on my own(really just s2pid when it comes to math lol). Also your voice is so damn cute so now you've got your self a new subscriber ❤️
Math archon is back to bless us. Thank you!
Wow very helpful info, thank you miss magical math fairy. Subbed!
With Lyra upcoming I think something you should mention is also the fact that some characters have seperate scalings
Like healing for example on lyra : you heal yourself for 15% of your atk on each punch that hits an enemy
If you get a lot of damage% yes you get a little more damage but you are missing out on a considerable amount of healing
That's why I personally like Lin's realm giving atk% instead of damage boost because that means it affects the healing lyra provides
Also why I would argue Tsubasa trait may be overall a better trait to use for lyra dps as it's easier to keep up in comparison to Samir's trait due to lyra being a face tanker and gives atk% instead of damage boost
shiro's trait is superior in most cases for lyra. 26% attack up, 50% of the time. Some would say that averages to a 13% attack up, but in reality, your gonna be using that period its up to burst damage more than not.
Man o man i loved your character and it suits your voice very well
AS everytime, very valuable video. Good luck for your overlay plugin, I’m sure we’ll all rush on it (at least the ones not yet disgusted by the game ;))
Thanks a maths video was what I was looking for
Great content, waiting for tian lang global analysis with samir and crow replacement comparison
Thank you !! I really have my work cut out for me with Lyra and Tian coming out at the same time but I'll do my best ^^
Just subbed thx!
Great video once again, thank you
Let's say if u have 21500 volt atk, which engine better to use ? 4.4 volt atk % or 1200 flat volt atk or 2.0 volt dmg boost %.
omg how i don't know about you until now
cute and clever
thanks for the video
Sadly there is a small mathematical error at 3:40. The damage % increase you get from additional crit chance is dependant on your current crit chance and has "diminishing returns". Going from 0% to 1% crit chance gives this 0.5% DMG increase, but going from 50% to 51% crit chance only gives 0.4% DMG increase. The correct formula for DMG% by critrate should be (((1-old_crit%-delta_crit%)+(old_crit%+delta_crit%)*crit_dmg) / ((1-old_crit%)+old_crit%*crit_dmg)) -1
additionally mobs at lvl 80 have 30% crit resist. So the first 30% crit rate are 0% dmg increase. it might be worth looking into gloves and boots again. Sadly for volt players crits are essential for crow/samir and provide more than the simple 1.5 crit dmg multiplier.
Yeah, I'm aware of this! I updated the pinned comment a while ago with an updated table and notes about it~
i was very confused on this thank you and your voice is cute!
Yes, exactly the topic i wanted to know more about, thank you, math loli!
My head imploded. I simply ascribe to the school of "big number gud".
Great video :)
Thx for sharing!
i wish i can be good at math like you
Could you make a video for the best simulacra traits to use? All ppl recommend always samir but it would be nice if someone like You could make a video about it comparing samir, tsubasa, Lin, shiro, frigg, etc...
I wonder if damage% is actually multiplicative or if it's additive within certain buckets and the buckets are multiplicative in the formula. Hard to tell because there's some variance between individual attacks, so you can't really test it easily. For instance, formula might be (elemental atk) * (1+ sum of atk% buffs) * (skills scaling) * (1+ sum of damage % buffs) * (1+ sum of elemental damage % buffs) * (1+sum of resistance reduction % debuffs)), where stacking a bunch of damage buffs isn't great, and you would rather have some atk% buffs, some damage % buffs and some resistance reduction debuffs for optimal damage.
Or it could be (elemental atk) * (sum of atk% buffs) * (skills scaling) * (1+damage % buff A) * (1+ damage % buff B) ...etc * (1+elemental damage % buff A) * (1+elemental damage % buff B)...etc * (1+resistance reduction % debuff A)...etc, where you just want as many damage % buffs as possible and the single strongest atk% buff you can get.
Genshin for instance used the first formula, and you were getting diminishing returns if you stacked just a bunch of elemental damage % buffs compared to spreading out between Damage %, elemental damage % and atk %. E.g. 250% elemental dmg would be less damage than 200% elemental damage and 50% attack.
Thanks for another great video.
Why didn't I find this channel months ago.
Attack % increase boosts are more future proofed I feel, even if early on the the flat ones are better. Don't the visors also have flat crit too btw? Flat crit vs. flat attack, flat crit should win out. Crit % is probably future proofed as well though since there seems to be an offset in your level to the flat crit you have that crit % would supercede. I'm not sure how quick we will power creep to such points though, especially on global. In CN, maybe it will occur a lot faster though.
Crit would've nice if there wasn't a crit resistance in mobs (30% in lv 70?)
@@merem6972 This in no way shape or form makes crit less worthwhile. Crit is easily the most important stat to get from gear as it's so limited. Flat attack has diminishing returns, while crit doesn't. Like, 7000 crit is far more impressive than like 2500 attack. Which, is what you'd be looking at comparatively for crit rolls vs. attack rolls on gloves and boots.
the goal is 5k crit on gloves and boots, 1200 elemental attack on helm/body/belt/vambrace, and attack % on engine and visor for best possible outcome.
Great Explanation.. but I just go what ever TOF stats(Full Resistance) gave me.. Yeah Thanks TOF GODS.
ah yes true i forgot to add the real tldr: "just roll whatever gets any dmg sub because beggars cant be choosers and we are all beggars in the context of gear"
You're perhaps the most informative content creator on ToF. I have a question about crit. Let's say I have 12K crit and 20K frost attack, and using Saki, Lin, Frigg. Would a gear with 2K crit and 800+ frost attack be better or worse than a gear having 2.8K crit? Thanks in advance.
Without even doing the math, I can tell you with confidence that the 800+ Frost attack is worth way more than the 800 Crit. Crit on boots/gloves is nice, but it's not THAT good!
(I did the math anyway. The pure crit one is around 2/3 as good as the one with attack)
@@Maygi Thank you so much! I'm glad I saved the gloves and boots. I'm using the 2.8K crit version as everyone keeps telling me that crit is everything in this game, that always go crit gloves/boots. But with the crit resist in Vera mobs, I've been rethinking this.
ty this helps a LOT!
She has a funny diction. Sometimes it's hard for me to understand what she says in principle :)
Maybe its cute, but its bad in general if you want to understand something
Educational cinnamon roll.
Thank you!
Thanks Math Lady, very cool.
You can't choose your stat rolls anyways. The difference between the two stats is minimal so just use whatever has better rolls, as always.
this is calculation on papper
on field battle Atk% had more advantage on most situation cause usually average player only build 1 elements
you ask why?
Dmg Boost shine better to Atk% when you hit enemy with weakness elements
other than that atk% shine more be with enemies with resist nor normal one
you can't see much using cactus on Hota studio you need to compared it on live battle to see it by your own
in conclution if you want to be Op on any situation using dmg boost%
make sure you fight those with your elemental weakness
i will just put a like and leave no wonder i always get 1/20 in math
2:48 Lin also gives 15% attack for all elements with her elemental skill, So she gives in total 48% attack buff
15+23+10
@@gapmama3030 48
From my testing with my ATK% / Bonus% [ without ATK% / Buff ] i got 21k ATK [ 24k with ATK% ] & i got try the value of +2,09% or +4,14% [ Bonus Value was better than ATK% ] depend en ATK Base & if you have some buff or Final Damage Bonus (i think) u already want some Bonus Damage% 😆
Hey Maygi, I was wondering, could you make an updated guide about gear stat preferences comparing wether dmg%, atk% or flat atk rolls are better, especially at endgame/maxed stats/gear, since this guide is over 5 months old by now? I just got a penta roll on a reactor and just noticed how even that over 8% atk reactor is actually worse than the 800 combined flat attack, 2.XX atk% reactor I had (34,8XX atk on the 8% flame atk% reactor, but 34,9XX flame attack on the 800 flat attack reactor, and that before any additional buffs and matrice effects id gain in combat) and noticed that with multiple other pieces as well. Especially once buffed and matrices etc in effect, it just seems like atk% is really really bad compared to flat attack. However with damage rolls its not that easy to compare but I feel like it may be worth revisiting with the stats we can achieve. Also we have many new chars, matrices, gear and especially now titan gear.
Also on the augmented gear I noticed, that the max upgrade you can achieve (the green/blueish number under the stat on the augment window) vary a lot, but I simply dont want to throw a lot of mats at the system to test it out. On the 8% reactor for example I only get about +1.3% as maximum upgrade, which is less than 1/4th of the total stat, on the other reactor its 539 +max 297 flat attack and 322 +max 359 flame attack. So its more than the base flame attack as a maximum upgrade and about 60% of the flat attack value as an upgrade. That just seems like once that thing would be maxed out, flat attack rolls value would sky rocket compared to atk%. Is there something I'm overlooking? Maybe the answer is even to aim for a good mix of both?
Flat ATK is buffed by ATK%, as mentioned in the pinned comment. If you check the gear calculator sheet in the spreadsheet, there are checkboxes to add certain attack buffs, and then you can evaluate the comparison between ATK roll value vs ATK% there by inputting your stats!
On the topic of an updated guide, though, I'll probably look into making something for titan gear optimization at some point...
Thankyou, I was wondering what exactly made the change from lins frost damage% to atk% so much worse. So it's going from a multiplicative buff into an additive buff, good to know this.
Jeez and to think with all these insane nerfs from CN for balancing, they won't even give us a reasonable amount of advanced modules for gear. I am still 10 away from a full level 20 set (with my level 40 gloves) and the other modules aren't much better, and yet we're supposed to somehow be caught up with cn in 2 months who had pretty much maxed out gear by the time vera rolled around, and we're missing a bunch of characters. They are just shafting global by not giving us any reasonable amounts of materials, I mean seriously we still haven't gotten a re-run of nemisis yet.
thank you!
Thank you very 😩 luv you
What would we do without you :D ? Thanks a lot. This rly helps a ton :)
the math archon is real
you missed only one point in your Video Maygi which is Lin gives 15% Atk increase with the Skill globaly and with Frost Realm it's 10% more making Lin at A6 in Frost Comp giving 48% Atk % Buff combined with A6 Frigg and the Frost Resonance a Frost main A6 player gonna have 103% Atk % Buff without Coco and her matrieces
with that being said it's only possiable for Whales at this point but great Video as Always
Sweet, thanks.
Just because I was curious I went and made a spreadsheet for all of the threshold values for if you roll 5 times into either Flat attack(s) or ATK %... Then made a simple calculator at the top so people could just see what they'd need for a given Flat attack vs ATK % because I realized people would probably not want to read through a huge table.
docs.google.com/spreadsheets/d/1KkALOdt5igCa0-TtzQZAXn8vCZ3YiOTZJ0ll766EA-s/edit?usp=sharing
Example: If you have a combined Flat attack of 1000 then the thresholds of your Attack (ignoring what is added by the piece(s) you're looking at of course) that ATK% would be:
By # of rolls into the stat
0: 79365.08
1: 37037.04
2: 24154.59
3: 17921.15
4: 14245.01
5: 11820.33
Which means that for someone like me with over 15k in my attack stat I would need at least 4 rolls into ATK % to beat a 1000 Flat attack piece.
very good explanation! cong
Hi, do you know where I could find the math for lyra 2 pc healing?
You are a miracle
2:22 phys always hoed
I have 28k frost dmg... I have 1 item with 1.6 % atk frost , 260 atk frost and 60 atk ... And i have other item with 3.5 % dmg frost what is better? pls help me
The DMG% is better in this case. For future reference, you can use the gear comparison calculator I made, where you can input any stats and see the value: docs.google.com/spreadsheets/d/1ZrJokNh_0AF_9welc7Etz6K8jlpzi5bXpiWz-mQZa78/edit#gid=371662811
how about final damage
Thank you for this. But even then... my rng is booty q-q...
Thanks magical math pink fairy
How does damage formula work? Is the damage the number you see when you attack a mob (Damage recieved = (Attack x Buffs) - Enemy Resistance)? If this is so, wouldn't damage be a bit better if it's calculated after their resistances?
How much attack is that in flat numbers to get 100% attack?
Any possibility that the tof overlay also calculate damage? played some MMO before... kinda triggered me that the damage meter TOF used only calculates overall dmg instead of damage per second... It would be great if you could add it... also useful to see which weapon dmg breakdown (samir does 65% of dmg etc).
Now we just need to figure out why % Elemental Atk on new gears from vera doesn't give any CS.
My Engine went 4 times into % Volt Atk which is great but my 5 star golden picece only gives 478 CS. It's a joke cause I have crew mates with % Element Damage with 6 and 7k CS on 1 item!!
0:53 yea
So i should ignore crit on boots and gloves and pull for flat attack on them??
I main frost comp. Thanks so much in advance, and great video!
Flat crit on gloves and boots is a different story - it's quite strong in comparison to flat attack at current levels! You can use the gear calculator to pinpoint exactly what's better with your stats: docs.google.com/spreadsheets/d/1ZrJokNh_0AF_9welc7Etz6K8jlpzi5bXpiWz-mQZa78/edit#gid=2055222127
107474 CS 30671 Attack and 112149 CS 30555 Attack which one is good? 🤔
CS doesn't mean anything, just maximize damage stats! Without seeing the rest of the stats, more ATK = better
great video
What about Final Damage? How is it calculated? Which gives more damage, final or elemental?
There's no difference between "final damage" and "elemental damage" %. The wording they use across the game is inconsistent, so it can pretty much mean anything (Couant is "final damage" and is additive, and most other sources of final damage/elemental damage are multiplicative).
@@Maygi I noticed that the wording "final damage" started to be used after Lin. Perhaps it was introduced so that players understand that the "final damage" will increase not only the damage of their main element, but also the altered?
@@Maygi but you know Maygi I tested Scylla and Haboela, Haboela is always stronger. Matrix Used, (2pc Crow, 2pc Samir), (4 pc haboela), (4pc Lan)
There might be a difference between these enhancements.
thanks Teach!
How did you get the formula and stats? where is the source?
I simp maygi 😫
Hi Maygi thanks for the explanation but, what if the enemy has high resistance stat? Is the result still the same or not?
Yep, at the end of the day the damage calculation is a ton of multipliers smushed together, resistance being one of them, so relatively speaking, any stat/dmg% buff is worth the same regardless of resistance~
Which one better DMG% or Final DMG% ?
The wording is completely arbitrary. DMG% is just DMG%... but note that some forms of DMG% are additive to each other, such as that from Couant (listed as "final dmg") and shield (which is just "damage"), and even several matrices such as Shiro ("damage"). The wording literally means nothing and it can be confusing...
@@Maygi so what you wanna say, it's basically the same? I ask this because I noticed it at the new Bosses SSR Matrice at 2.4 update. Habra increase Final damage and Scylla increase elemental damage
I like you accents, smol math fairy
hi, im using crow as my main dps with 53% crit rn. I have 2 gloves, one would give me addtl 990 atk while the other will give me 11% crit. What would be better?
11% crit ofc
Why school didn't teach us this?
sound like fairy elves vs math teacher
Hey Maygi , elemental damage boost% doesn't increase for Lin's damage right? if it doesn't then it's another reason for atk% to be better
Yep, DMG % doesn't affect Lin's damage at all! I did include it in a text note at 3:05 :3
@@Maygi oh thanks , I watched it but too much maths damaged my brain and made me not notice xD
How much DMG will I lose if I use Frigg's trait instead of Samir's ?
You'll lose roughly 5% overall DMG assuming you're using A6 Lin / Frigg ~
Kinda confusing, so it's ATK%>FLAT ATK/ELE ATK>DMG%?
kinda yes
if you have around 16500 attk currently, you need total of 280 flat attack (elem attk + general attk) for each roll to compare with attk%. so if you roll total of 900 flat attk, 3 roll on attk% are only better with 840 flat attk below. But when you reach 18000 attk and you roll total of 900 flat attk, 3 roll on attk% are around the same value as 900 flat attk. 3 roll on attk% also may come with flat attk so you need to minus it first to compare. Getting 900+ and above on flat is harder than to get 3 roll on attk%, so attk% is better since you got 2 roll that may go into flat attk and attk% is a fix value. Also 900+ flat attk to get with only with 3 roll is super rare. So I think a single roll of attk% is still a better roll than a high roll of flat attk.
Does frigg trait have the “diminising return” thing because you only getting 2.4% atk at a time?
"Frost ATK is boosted by 2.4% every 3 seconds upon entering combat. Stack up to 10 times. Lasts for 5 seconds. In addition, gain hyperbody and become immune to control effects in the Frost Domain."
Since it's an additive you'll get the dimenishing returns, yes, but at full stacks you'd still have the same Frost ATK increase as you would if you'd just gotten 24% straight away.
(Floats are the percent increase and then we add 1 for 100% since multipling a float less than 1 would give us less than our initial value.)
10,000 * (1 + (0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024 + 0.024)) = 12,400
10,000 * (1 + (0.24)) = 12,400
If it was DMG% instead you'd get: 10,000 * 1.24 = 12,400
Now hold on, that's the same value! Well that's because we're only looking at one increase source.
Lets look at two examples of another set. One will have two 10% ATK buffs, and the other will have one ATK% and one DMG% of the same value:
ATK%, ATK%: 10,000 * (1 + (0.1 + 0.1)) = 12,000
ATK%, DMG%: 10,000 * (1 + (0.1)) * 1.1 = 12,100
Heck! Let's double each set!
Quad ATK%: 10,000 * (1 + (0.1 + 0.1 + 0.1 + 0.1)) = 14,000
Double Sets: 10,000 * (1 + (0.1 + 0.1)) * 1.1 * 1.1 = 14,520
The key thing to note is this is only in comparision to other equivalent DMG% options. All this diminishing returns stuff means is that if you _could_ get an equal DMG% then you'd value that higher than an equal ATK%. This is why Maygi went into the part about reaching 100% ATK increase being the start of the diminishing returns because only at that point, with our doubled ATK% over DMG% rate, do we actually start to see the diminishing returns.
Visors dont have final crit rate roll maybe only for global lol who knows
Wait--lemme get coffee first 😅
Can you tell me which one is better? Samir trait or frigg trait ? Regardless the immunity of friggs trait .. just in damage output ..
Just in terms of damage, Samir trait >> Frigg
@@Maygi i see .. is it a lot further increase if I change into samir trait ?
Samir trait is better than the frigg trait if comparing the damage buff given BUT if you are not good at dodging and always keep dodging, frigg trait might be better since you will be hitting enemies more.
So, thanks to the changes in global DMG % is useless, because with the buffs of cn that goes over 100% would be usefull
I didn't understand the part where flat atk is better than atk % :(
There are thresholds of stats where rolls on Flat ATK are better than the 1.44% ATK roll on gear. The lower base ATK you have, the higher the relative value of Flat ATK!
It's a matter of how much actual attack you'd get from 1.44% increase and how much attack you'd need to have before that % increase would be better. You can find out the result by taking the flat attack value and dividing it by 0.0144 which gives you the threshold value, or to put it another way at that threshold value you'd recieve that much effective attack so anything higher than that threshold is better for ATK%. Here's a table of all possible upgrade values (all ATK% upgrades are exactly 1.44% so we only have to factor that): i.imgur.com/SJZHmre.png
You can essentially look at that table and find where your Attack stat is at to find out how much Flat attack you can accept. For example: Say I had 15,000 ATK. Anything less than 216 would be a worse roll than rolling ATK%. In the end though, ATK% will stick around longer as you gain base attack stats from other sources.
@@Maygi what's the max overall roll for flat Atk in combat engines? thanks!
@@earthlingx5267 312 attack, same as other gear pieces!
@@Maygi assuming max rolls of flat Atk x 5
vs.
Atk% x 5
the breakpoint would be around 19000+ base atk?
I now like math
The flat attack table confused the heck out of me
Shouldn't wording be "before flat attack > attack%"?
i.e. for when you have 21667 attack, you need 312+ flat attack for it to give more attack then 1.44 attack %
I agree that this table is a bit wordy and unintuitive...
but no, this is the intended order! The more total ATK you have, the less flat ATK is worth (relatively) in comparison to ATK %. However, one thing to note that I didn't mention on the video is that Flat ATK is also subject to ATK buffs itself, so the left column of that table should be "Flat ATK * your avg ATK buffs", raising the thresholds even higher for ATK% to be better. (do check out the gear calc spreadsheet for a dynamic calculator where you can select your buffs: docs.google.com/spreadsheets/d/1ZrJokNh_0AF_9welc7Etz6K8jlpzi5bXpiWz-mQZa78/edit#gid=0 )
Damn man. It's just a game.
just one clarification, when you talk about flat vs % attack, you give values for flat better than %, but how much % did you use as comparison?
I used 1.44%, which is the static value of ATK % rolls! (there's no variance in ATK % rolls on gear)
Those flat attack values are possible values within the upgrade range of both Attack and Elemental Attack. Since the ATK% upgrades are all 1.44% that's the value that was used. You can find it out the threshold for any given value by dividing the Flat attack by 0.0144. This will give you the amount of Total attack you'd reach before that Flat starts having less than the 1.44% in increase. I've screenshotted a table of every value from 93 to 312 rounded to the tenths place here: i.imgur.com/SJZHmre.png
@@Maygi that threshold for flat atk is base atk right?
@@__Newones Yes! Base attack before any modifiers, including resonance.