Disclaimer: we are talking about reasonable end game gear here. At end game, what is most effective? It should be known that enhancing an accuracy offhand is a much easier way to achieve optimum accuracy than enhancing accessories.
Traditionally in BDO most classes go for accuracy offhand, liverto mainhand, and all the AP you can squeeze out of your accessories. Why?
I would like to suggest that it is because the bdo community is lazy in regards to understanding game mechanics/stats. And now you can be too! See the short answer to skip all the mathz.
Short answer, even with the recent buffs to accuracy offhands, you should be rolling straight AP offhand in patch whatever valencia is. Roll accuracy accessories. You will be much happier. Get all your accessories to tri for much cheaper as well, since they are not in high demand. I will be rolling III red coral earrings X 2, IV ring of good deeds X2, and perhaps ancient weapon core neck and belt. Maybe. Maybe tree belt. and ogre ring. Might go straight AP there. god willing. In the future I cant wait to see how manos emerald rings will play into all this.
Long answer: The following examples are for the best class in the game, ranger (I am totally NOT biased i don't know what you are talking about.) After reading many other tests from fellow nerds, we can estimate that a single point of accuracy grants about a 1% increase in hit chance. Accuracy is countered by evasion. Evasion is gained not through DP like previously believed, but now we believe evasion is gained through armor enhancement level. I will be testing this further (with much needed help) And DP is more damage reduction level. SO. Using basic bow skill which gives +6.25% accuracy, the following chart was used to conclude the ACTUAL amount of accuracy an accuracy offhand gives. Testing was done with a bronze dagger equipped, firing at a target 300 times, recording miss to hit ratio, and then enhancing the offhand weapon. We then take the difference and subtract, round for sake of RNG. for instance, bd 0 hits 57% of the time. bd 1 hits 58% of the time. We conclude that enhancing bd from 0 to 1 gives 1 accuracy. Note: main weapon was a not enhanced Styd longbow, which should give base 0 accuracy. Guild buff gave an abnormal amount of starting accuracy which is +5 as I recall. As that stat is static and not dynamic, the accuracy difference between tests should be unaffected. These results do seem to match a korean chart I found here: http://www.inven.co.kr/board/powerbbs.php?come_idx=3584&l=12110
Enchant level | Accuracy Increase | Total Accuracy |
0 | 0 |0
1 | 1 |1
2 | .75 |1.75
3 | .75 |2.5
4 | .5 |3
5 | .5 |3.5
6 | .5 |4
7 | .5 |4.5
8 | .75 |5.25
9 | .75 |6
10 | 1 |7
11 | 1 |8
12 | 1 |9
13 | 1 |10
14 | 1 |11
15 | 1 |12
+16 and beyond can be assumed to be +1 accuracy?
With bronze dagger ult you can add another +7 accuracy totalling +22 for a ult tri offhand. 25 accuracy if we assume +2 accuracy per level beyond 15.
So lets compare traditional build to a build I propose.
Witches earring tri
11 AP
Mark of shadows tri
11 AP
{Crescant tri would grant 14 AP, however be HELLA expensive }
Bronze dagger tri
17 AP 3 DP 22 Accuracy
assuming 100% enhance success you will spend 136 mil on the accessories alone.
9 mil per witch earring X 8 for 72mil mark of shadow at 8 mil a piece for 64mil
VS
Red Coral Earring tri
8 AP 6 Accuracy
Ring of good deeds +4
6 AP 5 Accuracy
Ult Steel dagger tri
34 AP
Assuming 100% success rate you will spend 57.2 mil on the accessories alone
7.4mil per red coral earring X 8 for 56 mil 120k per ring of good deeds X 10 for 1.2 mil
| Traditional | Andrew's way
AP | 39 | 62
Accuracy | 22 | 22
DP | 3 | 0
Stamina | 0 | +100
Cost | 136 mil | 57.2 mil
Now all this of course is if you can get your gear to tri. Cron stones can certainly help, and should be included towards the ring of good deeds because they are so cheap. Only one cron stone is needed when making dou deeds ring into tri. They are the reason I decided to go for +4 on the good deed rings. Also, your necklace and belt have been left bare for this comparison as you probably noticed. If you need more accuracy go for it in belt or neck. Need more AP? Go for it. Note: if bronze dagger gains more then 1 accuracy after +15, you will effectively be missing whatever that amount is. The true accuracy after +15 has yet to be tested and I cannot afford it.
End results: You gain 23 AP my way. You will lose no accuracy. For far cheaper to boot.
So. Part 2 of these tests. How much accuracy does one need?
Well, thanks to reddit users we have a whole jumble of potentially inaccurate information. We know that in terms of accuracy Yuria
WARNING THESE CHARTS WERE NOT TESTED BY ME. NOR DO I TRUST THEIR ACCURACY <-- no pun intended.
They assume a yuria gives the same accuracy as a bronze when enhancing
This chart demonstrates your hit rate without using a main weapon IE just your rings and earrings
DP | Hit rate
174 | 68
155 | 73
124 | 93
111 | 91
after equipping a +15 yuria with the above accessories, these numbers of course still making the massive assumption of the yuria bow providing equal accuracy as it is enhanced (+12 accuracy)
DP | Hit rate
174 | 80
155 | 83
124 | 100
111 | 100
Liverto increases accuracy more then yuria by about 13% (and the liverto VS yuria test results seem extremely accurate based off a sample size of 400) against ult grunil +15 set. So I will assume with this information in the above charts that I will be near optimum accuracy for fighting opponents with 200+ DP. But we will see. It is entirely possible that against opponents with crazy TET ult gear, more accuracy will be required depending on how evasion scales. In this case, I propose my build is STILL better because you can sacrifice AP neck and belt for more accuracy without going over threshold. Or even better, go for ancient weapon belt and neck. + 5 accuracy right there while maintaining great AP.
Liverto + 2 accuracy has a 83% hit rate (using ranger basic bow attack +6.25% accuracy) against a defender with ult grunil armor +15. Up from 70% with yuria +2 accuracy. Couple that with 20 extra accuracy from accessories (22-2 from the test) against a +18 grunil set instead of +15 and you should be good by my calculations. Especially with +5 accuracy from a guild buff and +2 accuracy with the lvl 55 perk.
(Liverto hit chance) + (gear accuracy + guild accuracy bonus + Skill bonus) = total hit chance
83 + 27 = 110% chance to hit a target with ulti grunil +15. At the proposed rate in which enhancement level raises evasion (1% miss chance per piece), assuming it doesn't rise exponentially after +15 and instead continues in a linear fashion, you should have around 98% hit chance. Once again, those hit chance charts still looking wacky. I would like to test them myself to see evasion scaling better.
Anyway. Questions comments and concerns are appreciated.
July 15th 2016, 7:00 pm