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

[HowYouSoGudd]

Honestly drops seem to predictable to drop on an RNG system. Someone on reddit said that each DPK item in CBT was set with a DPK Min and Max which would explain some of the randomness of results.

There could also be an RNG system along with DPK. With gems as my example there could be a 1/10,00DPK and then a .001%(1/100,000) for it to drop. Ofc thats pure theory with 0 basis but a possibility non the least.

Anyways from the data I’ve seen in the other thread(50k kills) the drops are too consistent for RNG to be more likely. If it was RNG then i’d expect results more similar to it dropping on 2k/10k/20k/35k/40k/70k/73k/99k would be more inline with RNG. But most people said that the rods/maces dropped at around 200kills every time in a particular order and gems around 10k kills. They’re just isnt enough variance to assume a pure RNG system.

Now there is also people who claim to have had multiple DPK items drop nearly right after each other(or atleast far too close to be DPK) or getting none after way too many kills. In the above scenarios its possible that something happened to affect the result although i strongly feel DPK is a real thing and we’re better off trying to work out the logistics rather then arguing over whether it exists or not.

[Takemi]

It can be any of those in that scenario. You can’t just scratch off one, especially not because you think it’s not the best implementation of the system. We are talking about IMC here, they don’t code like that.

The programmers that coded the game were likely hired because they had the education and qualifications necessary for the job. You can’t assume that programmers think on the same level as the people who handle the management and business aspects of the game.

Programmers have to know which data structures are necessary for the task at hand. Yes, the game was not properly optimized, but I imagine that business was the higher priority in this case, and the upper management likely made the call to push the game out and adjust afterwards.

It was probably more like: Originally they wanted to do it on kill. Then they added Oracle, so they had to rethink their system and calculate drops on spawn. But then there was too much memory used so they changed the system that each mob can only drop one item, so only one item needs to remembered per mob. Then they added shiny mobs so there are at least some mobs that drop multiple items.

That’s honestly the most likely scenario in my opinion.

This is not a possible scenario in my opinion. Let’s think about this: because they added a skill that identifies which items a mob drops, they had to entirely scrap the way they wanted items to drop? If that were the case, they’d simply not put the skill in at all. When programming, there are many ways to make something appear a certain way when it actually works in an entirely different way.

The only thing we can safely say is that it can’t be a pure DPK system because sometimes a rare item drops twice within a short time. Everything else is just guessing. All of the findings could very easily be explained with a simple RNG system too.

This is what I’ve been thinking:

DPK does play a major role; however, this system isn’t entirely DPK based. An item will likely not drop at the same amount of kills each time (there is a chance of it happening, though).

From what I’ve observed, drops (the ones I observed on the infrorocktor map, anyway) are likely ~95% DPK (+/- a few percent) with the rest being an added RNG variance.

[rya.reisender]

This is not a possible scenario in my opinion. Let’s think about this: because they added a skill that identifies which items a mob drops, they had to entirely scrap the way they wanted items to drop? If that were the case, they’d simply not put the skill in at all.

That’s not so unlikely. Oracle was already added before closed beta. And there are many other known things they completely changed last minute. Party size, interface, removal of classes …

[Takemi]

The gameplay mechanics take priority over one class’s skill, and party size, GUI, and class inclusion/exclusion are easily altered on the fly.

[chiday]

Maybe system similar to PoE evasion based on “entropy”:
http://pathofexile.gamepedia.com/Evasion

[dahdoczock]

This is something I thought of possible too, yet this does not explain the OPs statement that he is able to pinpoint exactly which mob will drop.
More testing out of curiosity later today on my side.

[chiday]

Every time you enter a map 1st time after maintenance, game generates a number of mobs you need to kill to get item.
For example gem from Lizards:
3.33% chance to drop, so its 30 mobs for sure.
Game generates number from 1 to 30. Lets say it’s 5
You kill 5 mobs, you get the item.
Game generates number from 1 to 30 again.
etc

[lordshredder]

For example gem from Lizards:3.33% chance to drop, so its 30 mobs for sure.

Huh? But Green Gem from Lizardman is NOT a DPK item, just normal RNG.
DPK items are listed as ?%.

[chiday]

okay not a gem, but you got the point

Or probably they make it like:
If the basic chance to drop is 0.01 so the need 10000 kills.
And lets say if the n=10000 so the actual mobs to kill to 100% Montser gem drop is: From n/2 to n (5000-10000)
But its just thougts nothing more

[HowYouSoGudd]

I dont think a system like that makes much sense, items drop consistently +/- a few kills. Someone on reddit said that when the tables were visible that there was DPK min and DPK max columns.
So a Gem is probably listed something DPK min 9,900 DPK max 10,100

I’m still not sure about people claiming to get 2 gems within a short period of time and other things that deter from DPK but as i said in my previous post if the system was RNG you’d expect alot more variance in the amount of kills for a drop.

[Takemi]

Yeah, after I made that statement about the “~95% DPK (+/- a few percent) with the rest being an added RNG variance,” I was surprised to see the link with the min and max columns because this aligned with how I assumed it worked–the min and max being a random added variance to a baseline DPK value.

It would be nice to also see some screenshots of two gems dropping in a short time span. From now on, I’ll be taking a screenshot every time I obtain a gem.

[chiday]

But at least i know that once i got my m.gem from 1.000 kills. Never killed these particular mobs before.
Also i killed 350.000 different mobs in total before, and didn’t get a single gem.

[Takemi]

Were these kills consecutive and on the same channel, or was it over the span of a week? It is very possible to obtain a gem in a thousand kills or even one kill, but it is the timing that matters since the DPK value applies globally applies to everyone on a channel.

The primary rules to DPK are this:

Each channel has a set kill count for DPK items (there is also a random variance value added). If an item has a 10,000 kill DPK value, one person on a channel can kill 9999 mobs, while another person on the same channel can kill the remaining one to obtain the gem. After the gem is obtained, the kill count resets to 10,000. After maintenance, this number also resets.

[chiday]

I made an archer for oneshotting particular mobs that are only 10 on one channel. And i switched channels after i kill all 10 of them. 1 switch per 30 second. 1-6 channels. ~1000 mobs in total.

[HowYouSoGudd]

TBH I’m going to head to the infro map and just log the rod/mace drops over an hour or so. I think that should be able to give enough evidence to support DPK or not. I highly doubt there would be much difference in how DPK works for items and gems and either way this would atleast prove DPK as a theory.

I’ve played RO on/off since beta days, i know how cruel/kind RNG can be, but the thing with RNG is the range is 1 - infinity. ?? items in this game seem to drop at regular intervals but i havent seen anyone provide a decently sized sample that logs every drop and the exact number of kills.

But at least i know that once i got my m.gem from 1.000 kills. Never killed these particular mobs before.Also i killed 350.000 different mobs in total before, and didn’t get a single gem.

Assuming gems drop at a rate of .01% this argument actually supports DPK. Assuming a good porpotion of those kills actually dropped gems, that would mean you didnt roll that .01% X times. A DPK system is more about being at the right place at the right time.

[Takemi]

Yes, under these conditions, you can obtain a 10,000 kill DPK gem in 1,000 kills. The main factor would be how many mobs were already killed before you got there. After you obtained that gem, the kill count reset on that channel.

What we would really like to see is two of the same gems dropping within a short period of each other on the same channel (assuming they were known to be DPK gems). If this was recorded through screenshots and appropriately time stamped, it would dismiss DPK, but all we’ve heard are claims that this has happened and no evidence has been presented.

[lordshredder]

I really hope we can get IMC to change DPK to be per player instead of channel or have every DPK item have an additional normal drop rate…

The way the system works right now is just discouraging.

They listened to the community and implemented a debuff for World Bosses(I wasn’t for it but I don’t really mind) so it’s not impossible to achieve a change but we have to raise awareness and show that this system is flawed.

[Takemi]

I definitely understand your concerns with the way the system is currently. I also feel that many deny that DPK exists due to feeling threatened by it in the same way. I’m sure that the system could be changed if enough people pushed for it.

Knowing that you obtained a rare item through luck is a nice feeling, and the idea of the DPK system takes away that magic. However, after I obtained my boater hat drop, it suddenly became worth much more to me after knowing how difficult it was to obtain.

[dahdoczock]

The way the system works right now is just discouraging.

As long as one doesnt know, he will accept it as RNG. Thats okay[quote=“lordshredder, post:235, topic:304868”]
but we have to raise awareness and show that this system is flawed.
[/quote]

But I dont see why you would call it flawed, could you enlighten me?

[FlanFlan]

But I dont see why you would call it flawed, could you enlighten me?

Would you rather farm for 12+ hours at the chance that every kill you deal could drop what you are farming for? Or the DPK count only apply for you personally instead of being a global variable?

#or

Would you rather farm for 12+ hours at the chance that you’re the player who deals the final kill when the DPK count reaches maximum?