Tree of Savior Forum

[DropPerKill] Drop rate system- DPK- Explained [Updated]2

Exactly this.
There is no concrete proof of DPK in this thread. Every theories and tests/experiments shows the possibilities as the result shows both DPK and RNG are possible.
As for the video, it shows only 1 incident. Would provide better idea if it happened over and over on a circumstance where he is the only person killing them and happening at the exact number of kills. Just being around the number cannot prove anything.
But as you can see with gem drops, it is very obvious the game drop system is NOT SOLELY based on DPK. Which approves there is RNG without any questions asked.
While DPK is just a theory for now, possibilities remain either DPK existing to enhance the drop rate or not. Never DPK being the sole drop system.

That's how you read

https://youtu.be/9U60zfomh8I

Would provide better idea if it happened over and over on a circumstance where he is the only person killing them and happening at the exact number of kills. Just being around the number cannot prove anything.

Can you understand that DPK system may have definite range of kills? not exact number

I did this for almost 1200 kills, but I suppose a video that long might be a bit boring to watch. There is a small difference on drop amount.

Confirmed. Not yet by which margin.

1 Like

Eh if for anything, at this moment, the substantial drop rate increase theory makes the most sense, meaning after a set amount of kills, the drop rate of a certain item increases and thus a lot more likely to drop. This means the item can still drop before the set amount of kills, just with a lower chance like how itā€™s designed.

At least the above theory can fit both the mark tunic / aras recipe testing as well as the hanaming gem drop observations.

Note: it can also be a self tweaking system where if the overall rate is much higher than the designed rate, the chance of such item dropping is lowered to make it overall normalized. This kind of system is also used in a lot of games.

Well give or take, it still doesnā€™t prove anything because ā€œaroundā€ the number only emphasized the RNG rather than DPK. Unless proven in a larger amount confirmed as in hundreds of same item dropped at certain range.

1200 is a very small number to define difference between RNG and DPK, and you do know that already.

Just like what jjjsong and Iā€™ve been saying.
RNG is confirmed drop system seeing gem drops
As for DPK it is not confirmed if this system is even implemented or not. If it is then it only enhances the drop rates, never the core drop system.

Around a certain number emphasizes DPK. You remind me of back in elementary school when the science teacher asked us to stand randomly around the room and everyone in the class stood evenly apart. This is not random, random would have clusters and gaps. All of the data iā€™ve seen for DPK items(not gems) have shown a pattern in drops around a certain threshold.

Assuming that @ingel did not tamper with his data, do you not understand the likely hood of roughly 20 items dropping within 96-109 kills without any drops outside that range? Iā€™m not exactly sure how to do the maths for this one but lets even as far to say that its a 50% for a 1% item to drop in the range of 96-109 kills. By 6 kills the likelihood of that happening is 1.5%. Ofc Iā€™m sure its alot lower then a 50% chance but you get the point.

2 Likes

I canā€™t believe there are still people doubting the existence of DPK for ā€œ?%ā€ chance items just because there is a slight variationā€¦ That is not hard to program at all and highly possible.

Thank you @ingel @dahdoczock @Bald_Dad @FlanFlan for all your efforts.
Iā€™m going to use this knowledge to abuse DPK as much as possible and collect some data myself.

3 Likes

I just want to add that Iā€™d think a DPK system per player would be pretty horrible. I think itā€™s important for MMORPGs to offer situations where you are lucky and get a rare item you didnā€™t even hunt for. Not everyone wants to kill 100000 or 10000 or even 1000 monster of the same time and in a DPK per player system, those players would never ever get a single rare drop. Pretty boring. Mindless grinding shouldnā€™t be rewarded like that imo.

Now a DPK system is generally bad even if it is per-channel. Because it reduces the chances on being lucky tremendously, especially considering that most players leave the map once they got the item they wanted.

RNG system is simply best for open world MMORPGs.

Now it would be a different story for instanced drops. For example I wouldnā€™t mind if exactly every 25th dun90 cube opened would yield an Arde Dagger. Then I could at least get it in 9 days rather than weeks.

At the same time, players putting effort into farming for items and never being rewarded for their time has the adverse effect of significant reduction in player retention. I quit 2 MMOs in the past over spending significant amount of hours and not seeing a single reward for my time invested. Like straight up ragequit, uninstalled, never went back lol. Iā€™m certainly not alone in this feeling.

6 Likes

But the feeling to finally get an item is even greater if you donā€™t know when you get it. But I also understand the frustration it brings.

Some RO private servers started to introduce systems like ā€œ0.1% drop rate, but guaranteed item after 2000 kills if item didnā€™t drop until thenā€ kind of thing. Maybe that would be an option too.

5 Likes

That is the change Iā€™m hoping for.

1 Like

That would actually be a good system. RNG first but guaranteed to be rewarded eventually instead of beating your head against your keyboard.

1 Like

That actually seems like what ToS is doing to be honest.
Unlike lordshredder mentioned about ā€œdoubting the DPK for a slight variationā€, I actually did some research since someone mentioned Korean ToS believes in DPK for months by now. I found out they believe monster gems are occuring every 10k kills give or take. However, you can find in this post the number varies from 3 to 10k that is just way too huge difference in count to become DPK.
Iā€™m not neglecting the existence of DPK in the game system. And I do appreciate all these people testing to make sure.
Just the thing is RNG is proved to exist as you can see from 3 to 10k kill difference, while DPK seems to exist but also at the same time very questionable as there arenā€™t enough data collected just as of now.
Anyways, it would be very nice if RNG+DPK is working for individual without flushing the count via maintenance and channel swapping. If not then just need to hope for a good luck, since either RNG or DPK guarantees you a reward for your hard work.

Just read C4Murderā€™s post about swell body. If that is a case then it seems like DPK is written on top of RNG to enhance the drop rate perhaps? Just like what rya mentioned about private RO server. Or maybe RNG is set to very low on all items, and DPK occurs at the same time because wouldnt make any sense to have high RNG especially for those recipes.
While digging up KToS data, Iā€™ve manage to find a server with old ToS client unpack, and it actually states % chance of every single items that exist back in december last year. Now it is known to be server side save only and you cannot get any drop data from client unpack anymore. Hope IMC can just clarify or provides it in client just like old RO days

Iā€™m wondering if IMC did consider channel crashes or maintenance resetting the DPK values when they implemented it.

This is the main issue and makes some items virtually impossible to obtain, if itā€™s pure DPK.

Assuming they have a DPK system in the first place. =p

from reddit about DPK



unfortunately i canā€™t find his old post-investigation

1 Like

I dont know about this but for example, I.have killed 11k shardstatue and got only 3 bracelets at 900, 1800, 2700ā€¦ coincidence? yes. because I never saw a bracelet again, yet im going for the 12k kills now, bye

pd: my brother has killed 11k and he has received 13 bracelets

That reinforces the DPK theory.

now i understand why OP deleted his post earlier. All these people do not read thread at all.

Dear Morritz, read first post of thread at least



My grandmother killed Hitler with bare hands