Recent Features

Randomization and level design in the future of Diablo 3
Creating good random map generators is a difficult task in video games, and something we’ve seen done with relative success in all of the Diablo games. Diablo 1 had four dungeon types with quite different maps for each area, though all were just big squares with different arrangements of the pieces within them. (The Hellfire […]
The Diablo 3 Podcast #155: Fave LGems and Achievement Bonuses
The fans have voted their favorite Legendary Gems, but what if LGems could be used on Followers? Why are players suddenly rolling over Greater Rift 50+? Should Achievements grant in-game bonuses as well? And Blizzcon 2014 is coming in just two weeks! Show features HolyKnight3000, Rankil, and Flux Click through for approximate segment starting times […]

Travis Day weighs in on Reaper of Souls Legendary Item drop rates

Posted 23 Jan 2014 by

Travis Day

Travis Day

Travis day has visited the ever popular MVP critique thread and dropped some hefty insight. Travis delves into the disparity of loot, Reaper of Souls legendary item drop rates, current drop rates and the ideal ratio of loot vs time spent.

Nothing is spoken about the color of legendary crafting materials and I’m sure Flux is still dismayed. The posts and more after the bump


Hello folks,

This has been a topic of much discussion within the community so I wanted to take the time to sit down and talk to everyone about drop rates. I’ll start out by saying when F&F went out we set the rates far higher than we knew was appropriate because we wanted to get as much feedback and testing on the new and revised legendary and set items as possible. Over the course of F&F the unanimous feedback both internally and from the community was that the drop rates were too high. We knew this ahead of time but we wanted testers to get their hands on as many cool things as possible while also helping us debug and, ya know, test things.

Travis started the interaction than followed up on questions/comments/critiques.


Hello folks,

This has been a topic of much discussion within the community so I wanted to take the time to sit down and talk to everyone about drop rates. I’ll start out by saying when F&F went out we set the rates far higher than we knew was appropriate because we wanted to get as much feedback and testing on the new and revised legendary and set items as possible. Over the course of F&F the unanimous feedback both internally and from the community was that the drop rates were too high. We knew this ahead of time but we wanted testers to get their hands on as many cool things as possible while also helping us debug and, ya know, test things.

Going into our subsequent beta waves we said we wanted to start narrowing in on what the appropriate drop rates were for legendaries. We had set forth some goals and stated internally what our design intent was so we adjusted these values to reflect a more practical farming environment. As a general rule people won’t speak up if you being too rewarding, usually only a handful of people in a crowd will vocalize that “things may be a bit too high” in terms of drop rate. So when the entirety of our player base and internal team were swarming us with this feedback we knew the drop rates were actually FAR too high. Based on that feedback and also the detailed drop rate data that we track and review on a daily basis, we lowered the drop rates to something more reasonable and closed to what we wanted for release of RoS. For the record, I literally told people internally that I actively wanted to hear the feedback that drop rates were too low. We decisively found where too high was, we needed to find the lower bounds of reasonable so that we could narrow in on a place that would be best for the game.

Now with that said, the data that we have has shown us that drop rates are in the right ball park but that’s not to say they are final. A big piece of the pie that was missing was that we always intended for Kadala to be able to generate legendary items. More specifically she was intended to help give players more direct control over their ability to find specific items they were after. Not to the tune of “I want X item so I buy it” but we wanted players to have a degree of control by saying “Man I really want X helm, so I’ll spend all my Blood Shards buying helms and try to get the one I’m after eventually”. The environment that you have all been playing in has been missing this key piece of the puzzle for quite some time. When we are talking about the overall income of legendary items and finding a good pace for them Kadala has always been intended to factor into this. We are very actively tracking data and adjustments will be made to narrow in more as we proceed. However since obviously you all haven’t had the insight into these conversations I’m here to say that you’ve only been seeing a piece of the pie and with the latest patch should now be seeing the full picture.

A couple of other related notes, Rifts have always been intended to feel highly rewarding and we don’t think they are there yet. We want the obvious answer when someone asks “Whats the most rewarding thing I can kill with the time I have?” to be Rifts, this is part of why access to them is restricted by keys, they are meant to be AWESOME!. We have a plan changed that I’ll be hotfixing in hopefully today or tomorrow to add a reasonably large number of Blood Shards to the loot run boss but more importantly to give him a chance to drop Forgotten Souls. One of the problems that has been expressed here is how legendary and rares compare, barring the obviously build changing ones. A large part of the reason that players sometimes have a hard time replacing a rare with a legendary that should be an upgrade is because, generally, you are comparing rares that have been enchanted many times over to a stock legendary. We want the act of enchanting legendaries to have some degree of investment but what you are seeing in the beta is too much. In addition to being able to use the, soon to be acquired, crafting mats to enchant your legendary items it will also allow everyone a reasonable amount more access to the crafting legendaries.

Lastly on the topic of split farming, this is very clearly not how we want players to play the game. We know of the problems it causes as well as the reasons its occurring and are actively discussing ways to address the problem. While we love coop and we want players to enjoy playing the game together, the right answer should obviously not be “Lets play coop and all go solo in the game!”. I’m walking into a meeting as soon as I post this to discuss the issue and possible solutions.

Hopefully this helps clear the air some and I’ll touch back in later to answer more questions.

Kudos to Travis for weighing into this quagmire. It has to be tough to try to please everyone, and even with the right answer you’ll upset someone else.Not to say I’m defending him but it has to be daunting. Obviously the term “in the ballpark”is a scary thought and Travis quickly took to some responses:


Me again to clarify a couple things.

First off, Kadala is not meant to be the primary source of legendary items, she is meant as a self targeted supplement to the items you find from killing monsters. Diablo is an ARPG and killing monsters should always be the most rewarding thing to do. If Kadala is representing too large a portion of the incoming legendary items than I will address that. She’s meant to be a bonus, not the primary loot acquiring mechanism.

Second, the drop rates being “in the ballpark” means they aren’t off by a factor of 10. The F&F drop rates were at least 5 times higher than what we were aiming for, and the current drop rates are probably likely within the 50-100% range. Again Kadala is meant to cover some of this ground and if she ends up covering more than we want her to we will reduce her chance to generate legendary items and shift the difference into the monsters. It’s a big ballpark but from a top down look, it’s not TOO far from where we want it to be.

I’ve found myself on the end of the spectrum where monsters and Kadala are not giving me much so I may have hesitation to hear the drop rates are 50-100%. However I can also see where the F&F was too high so I can assume game development is hard.


Let me ask you this. What amount of time do you personally think it should take to get a legendary?

After you answer that I’ll give you guys some insight into the numbers we have for contrast.

I’ll say outright that when it comes to rewards what players want and what designers want are generally not going to sync up. We have a game trying to meet the demands of a huge spectrum of players. It’s important to us that everyone have a good experience regardless of how little they play, this is a huge part of why first kill legendary drops were added to bosses. It’s also important to us that people who enjoy playing the game for hundreds of hours to be able to do so without completely exhausting the rewards in a week.

This is something we kind of touched on last podcast. Uniques rained from the sky in d2 but a lot of them were crap. That didn’t diminish the greatness from a good drop however. Right now D3’s legendaries are dropping far too sparsely and are of such a low quality that a great drop seems both far off and far fetched.


Honestly the numbers you guys are throwing out are all reasonable, they are also pretty close to the drop rates that our data is showing. I said I’d show and tell a little so I will.

The drop rates in F&F alpha were as long as 40 minutes per legendary and in the case of MP6 Crushing Blow farming as short as 20 minutes per legendary.

The data that we currently have shows legendary drop rates at roughly 2 hours per, and that’s before Kadala was changed to be able to give legendary items.

The fact that some people claim to go for 10+ hours without finding a legendary is not only not our intent, but should in fact be impossible. We added a system in the expansion that tracks the amount of time you spend fighting creatures without finding a legendary and after a certain period of time will slowly start increasing the legendary drop rate. Once a legendary drops for you, actual item not crafting recipe or material, we reset that timer. This is meant to be a safety net so that the random can never be too extreme to the negative end. If players are legitimately going 18+ hours and not seeing a legendary it’s possible that there are some bugs floating around that need to be identified.

And before anyone responds saying TWO HOURS IS TOO LONG! let me say that we know as time goes on and we patch content or players get better gear and find better specs that will not be the reality. Higher difficulties do in fact increase your chance to find legendary items, coupled with people playing in coop also increasing the rate that everyone overall finds items, etc. Our final target is roughly in the 90 minute range for advanced players. This doesn’t factor in crafting items, trading with friends in coop or any of the other means that we already know players can use to accelerate their legendary acquisition rate.

Betas are incredibly important to the development process, once upon a time we stated we wanted 4 hours per legendary. At the time that we stated that we were living in a land where people on the team had literally never seen a legendary drop so even one every 4 hours sounded like insanity. As the development of RoS has continued and as we have played and tested the game ourselves and also due to the feedback of the community we have moved to a more and more generous environment. We genuinely want D3 RoS to be everything it can be and everything you want, and every change we make when tuning the game is with that goal in mind.

So the question then remains is two hours per legendary a good ratio? I certainly haven’t found them at that rate but maybe I am an outlier. Do the drop rates they have take into account split farming? Travis addresses it as not intended but not if it has impacted their data. One every 40 minutes sounds right by comparison to my F&F experience,90 minutes seems alright but isn’t 60 a nice round number for everyone? A greater question would be: Is Diablo meant to have time played incentives? Is waiting it out an acceptable farming method? One of the things that appealed about d2 was jumping on for a short session, running Meph or Baal, and getting some shiny loot in the process. Not everything was great but you always knew you and a chance. Currently hope seems diminished by comparison. Even in Vanilla you could play for a short duration and gather something that could be sold,which in turn could get you an item you wanted. Now the option is only to wait. The other factor not addressed is the wall players seem to hit where their rare gear is not being replaced by the legendaries that they find. Perhaps if the legendaries are scaled up this wouldn’t be a problem?

As usual answers beget more questions, which beget more questions. What are your thoughts on what Travis Day had to say?

Update: And to prove that, Travis dropped by to clarify more on this subject, based on feedback to the above.