June 7th, 2012, by alex

Why Diablo 3 is less addictive than Diablo 2 - Blizzard Responds

Wow.  Looks like my post about Diablo 3 struck a nerve - over 100,000 views in less than 2 days, a great discussion in the comments and over on Hacker News, and a couple replies from the Diablo 3 team! (more on that below)

First, to clear up a few things:

- I realize after talking to people that my post probably came across much more negative than I intended.  I focused on one aspect of the game I find fascinating, but I very much enjoy the game overall.  In fact I’d be playing right now if I weren’t writing this post… Act 2 Inferno won’t beat itself!

- I really thought the “scientific” part was obviously tongue in cheek, but apparently not.  I wasn’t expecting so many people to get sidetracked by that one word in the title (in quotes, no less!)

- Yes, I played Diablo 2 and LOD.  Perhaps not obsessively, my highest character was around level 86, and I didn’t play much in the last few years.  I felt my observations apply best to the “first playthrough” part of the game, which is what most players experience, and where the “addiction” has to start (otherwise they’re less likely to make it as far as endgame).

Anyway, the Diablo 3 team just did a really great AMA on Reddit - I thought the answers were clear, honest, and didn’t shy away from tough questions.

Two replies in particular were very interesting.  Wyatt Cheng talking about the Auction House:

The auction house has absolutely no effect on drop rates. There are conspiracy theories and misunderstandings but I do want to re-iterate, the is NO interaction whatsoever. Bashiok mentioned earlier that we took the AH into account, so let me expand a little bit on that.

The drop rates were tuned for a player who would never use the Auction House. For the majority of internal development we didn’t have an Auction House, we all played using our own drops only. I’ve personally leveled multiple characters from 1 to 60 internally before the game came out using only drops that I found - we all did.
When we say we “took the AH into account” that means it’s one of many factors. ie. some players will choose to play without trading, some players would play in a group of 4 where they share drops among each other, and some (as it turns out, many) players would use the AH.

Three weeks after launch player’s gear is much higher than what we were expecting. When I killed the Butcher on Inferno for the first time I was using a weapon with 492 DPS. There are also certain passives which are much more powerful than they were during internal development. One With Everything, for example, was basically never used internally because we didn’t have an auction House. With the auction house, it feels like a mandatory passive. In retrospect we should have seen it coming. In the game’s current state though, it’s a powerful Monk ability that gives Monks a big survivability boost and has some interesting (some would argue fun, others would argue negative) effects on gearing.

I consider playing without the Auction House to be a very fun way to play the game. I’m personally planning on rolling some new characters that I’ll set aside to be “no-AH/no-twink” characters. Much like in D2 when I would make a new character with a friend and we’d agree with each other not to twink our characters out.

(Reddit post)

I find that fascinating. I thought it was almost commonly accepted wisdom that the Auction House was one of the main reasons for Diablo 3’s itemization system, but that doesn’t seem to be the case.  I even said as much towards the end of my blog post, it looks like I was wrong.

There has been a lot of speculation about the Auction House and Blizzard’s motivations.  The more extreme fringe seems convinced that the AH is nothing more than a cynical cash grab that has influenced all aspects of the game and more or less ruined it.  A much more reasonable explanation is that the Auction House simply started out as a feature of convenience for players, and got somewhat out of hand / had its effects on the game vastly underestimated during development.  A few of the AMA answers really strongly suggest the second option.  This probably also explains why so many people feel the game was too easy!

Next, someone asked if Blizzard read my blog post (I swear that wasn’t me):

Have you seen the article at: http://www.alexc.me/a-scientific-explanation-why-diablo-3-is-less-addictive-than-diablo-2/417/
If so, what do you think of it?

(Reddit post)

Then:

Wyatt is working on a longer answer, but I will say, while I’m not dismissing his conclusions, if you want to prove something with science you need data, not just a theory. The graphs created are based off memory and perception, and so this isn’t very ’scientific’. /pushesnerdglassesupnose

OMGOMGOMG Jay Wilson read my blog! *squee*

More to the point is, again, Wyatt’s answer:

Alright so I’m going to take a stab at this question.

As mentioned in a different thread, the drop rates were carefully tuned for a single player playing through from 1 to 60 without ever using the AH.

All of our items are randomly generated, and so follow a distribution curve in power. Let’s say for the sake of argument that you were to somehow distill an item down to it’s “power level” and created a distribution graph of drop rate vs. power level. This graph would probably be normally distributed with outliers at high power levels dropping at a lower rate.

Looking at this graph, an average item drops every 5 minutes, a higher power item drops every 15 minutes, even higher power drops every hour. etc. As you move up the curve to ever more powerful items, the amount of time it takes to find such an item increases. This is what makes certain items more desirable, this is how things worked in D2.

What happens for a standard player who is playing solo when they first hit level 60 is they see an item upgrade every 30 minutes or so. Pretty quickly it becomes every hour, then every 2 hours. The higher the power level of your gear, the longer it takes to find your next upgrade, that’s just the underlying math of this distribution. It’s not really anything we set either. If we magically made all drops rates 10x higher, all it would do is shift the power curve left or right, it would not change the fundamental property that the higher up in power you go, the longer (statistically) it is going to take until you find your next drop.

So then let’s say you visit the Auction House and get infusion of power that hurls you forward on that power curve. So whereas at one point your gear may be at a point that you are statistically speaking probably going to get an upgrade every 2 hours. After visiting the Auction House you hurl yourself forward on the power curve so far that now you are statistically going to get a drop every 8 hours.

To further illustrate the point, let’s talk about the coming changes in 1.0.3. In 1.0.3 we’re going to start dropping level 63 items in Act I of Inferno. We’re also reducing incoming damage. What do I expect to happen? I expect that there will be a rapid increase in power across the entire community as all of these items become more widely accessible. It’s like we took the distribution curve of items and made everything drop more. That item that used to take 10 hours to find is now a 2 hour item. An item that used to be a 2 day item is now an 8 hour item. After the initial frenzy of power increase, things are just going to settle again. People who think drop rates are too low now will probably still think drop rates are too low a week later when they move to the new point on the curve. I’ve spent a long time on this question so I’m going to move on but hopefully somebody who gets what I’m saying will be able to expand on it more, maybe draw some graphs to better illustrate the point.

tl;dr we could make drops 100x what they are now and it would just cause everybody to settle at a new equilibrium point. Anything you can farm in a few hours you’ll already have, anything that takes longer you’ll wish you could get faster.

Thank you, Wyatt, for a very thoughtful response.  I hadn’t really considered this:  the effect of buying powerful gear from the Auction House is really to skip a lot of the “reward spikes” from Diablo 2, which makes it feel like they’re not really there at all.  I think it provides a great explanation for why the reward pacing feels so different from Diablo 2, even if the mechanism behind it may not have changed.

I will stick to my guns on one point, which I now realize I didn’t really make originally: I maintain that, Auction House aside, the items in Diablo 3 “feel” less rewarding than in Diablo 2, particularly throughout the levelling process.  This partly comes down to uniqueness and variety, as opposed to just raw “power”.  Magefist? Chanceguards? Goldwrap? Sigon’s? Tarnhelm?  I’m sure those ring a bell - sure, they don’t matter at end-game, but I still remember being so excited to find one.  That’s a pretty powerful reward long before end-game, and there is currently no equivalent in Diablo 3, where all items just blend together until level 60.

I’ll get off my soapbox now.  As I said I still love the game, and the great AMA on Reddit combined with the Patch 1.0.3 Design Preview give me a lot of faith that the game is in good hands.  Time to play!

June 5th, 2012, by alex

Why Diablo 3 is less addictive than Diablo 2: a “scientific” explanation

UPDATE: Got a response from Blizzard! Click here to read it.

Lately I’ve been amusing myself by reading Charles Duhigg’s “The Power of Habit” and playing Diablo 3 (not simultaneously). Diablo 3’s launch was a clear success, with 6.3 million sales in the first week, but also came with the predictable chorus of “the old game [...]

December 30th, 2011, by alex

“Go Daddy opposes SOPA.”

Heh.

Sorry GoDaddy, too little, too late. But I’ve been meaning to switch for a while now, thanks for finally prompting me to do it.

September 13th, 2011, by alex

Getting nginx to play nicely with Rails and the static content in /public

I’m far from a good sysadmin. I don’t have the patience to RTFM, I never remember syntax, and I usually don’t have the discipline to document what I do, meaning any sysadmin work usually ends up a frustrating sequence of trial and error.
I’m moving all my stuff from Slicehost to Linode, which I’ve put [...]

August 30th, 2011, by alex

LyricFind PlayBook app is out!

It’s really only a coincidence that my last two posts are about the PlayBook - I haven’t worked with it that much, although it actually is a great little tablet.
Today I pushed out the LyricFind app for the Playbook, and so far so good! Joe Styer of PlayBookDaily.com was kind enough to post a [...]

April 5th, 2011, by alex

MobileMonday Toronto - RIM PlayBook

Just got back from a pretty good presentation for Mobile Monday Toronto by RIM’s Tyler Lessard. The first half of the presentation was about RIM and BlackBerry phones, the second was a demo of the PlayBook. The Q&A session was much longer than it usually is at these events, and quite enjoyable.
I jotted [...]

January 7th, 2011, by alex

Releasing an iPhone game: Whack-a-Kitty three weeks later

I released my first iPhone game, Whack-a-Kitty, on Dec. 20, almost 3 weeks ago. People have been asking how it went, so here are some numbers and first impressions.
First, if you missed it the first time, here’s a neat timelapse video of the game being made:

And a shameless plug for a friend: if [...]

December 20th, 2010, by alex

Whack-a-Kitty is finally out!

It feels like I’ve been working on this game forever, but it’s actually only been 3 months - a lot of it evenings and weekends. The game itself has been done for a while, but it’s been stuck in the approval queue for about 3 weeks. I’ve also been struggling with making [...]

April 22nd, 2010, by alex

Android running on iPhone

This is impressive, to say the least. A Canadian hacker has managed to get Android running on the iPhone. The important stuff seems to work - wifi, the browser, playing music, SMS, making calls… oh, and it dual-boots. Wow.
Check out the video, or download the code from the original post.

February 9th, 2010, by alex

40 Android Business Models

While trying to figure out how to get access to paid apps on the Android Market in Canada (come on, Google… you know you want to, the Canadian dollar is strong nowadays), I stumbled on a blog series listing possible Android business models. There is probably nothing revolutionary, but the ideas are short, sweet, [...]

Twitter