Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 25, 2014
arrowPress Releases
October 25, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Torchlight 2 Vs. Diablo 3: ARPG Fight 2012
by Josh Bycer on 01/25/13 02:22:00 pm   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


Last year we saw a variety of ARPGs hit the market. With Diablo 3 and Torchlight 2, we had a situation where two different development teams set out to create different takes on the same design.

Reprinted from my site: Game-Wisdom

2012 for PC gamers was an ARPG smorgasbord, with titles like Drox Operative, Path of Exile beta and Borderlands 2 for instance. But the two that I find the most interesting were Diablo 3 and Torchlight 2.

With these two games we had an almost David vs. Goliath scenario. With Diablo 3 coming from Blizzard, which is arguably one of the biggest developers in the world. And Torchlight 2 from Runic games, which is helmed by the co founders of Blizzard North and the creator of Fate.

Both developers set out to create an ARPG of a similar mold. While on the outside both teams created a similar game, if we look closer there are very subtle differences that make it that much harder to pick a winner.

Diablo 3

When it comes to ARPG design, there are three areas that we can judge them on: Character Design, Loot and Combat. When I broke both titles down into these categories, the winner wasn't as black and white as I hoped.

The Man/Woman Behind the Armor:

Class design is a major point in ARPG design, as this is what defines your abilities and search for loot. With both Torchlight 2 and Diablo 3, the designers had a different philosophy that wasn't quite perfect.

In Diablo 3, each class was designed as a defined character, with a back-story, unique abilities that fit and preferred gear and attributes.  When you played as a Barbarian, you were never going to use bow and arrows and equip quivers; likewise a Monk would never need a Witch Doctor's sacrificial totem.

Abilities were set in stone in terms of their use, potential and utility and would never change. The rune system that allowed players to attach a skill modifying affect, while having a major impact, still only had set effects on the specific skills.

Meanwhile, Torchlight 2's set of classes were more open and less defined. Each class had three skill trees unique to them, but was not restricted by gear. Meaning, if I wanted to take the engineer (heavy melee class) and use a ranged weapon that was fine.

One of the reasons for the openness was that there was no such thing as a primary attribute. Instead each one had an important purpose that could fit into the player's style which I'll be coming back to in the loot section.

The skills available to each class could be improved by investing skill points into them. Unlike Diablo 3, the skill's basic functions are not set in stone and can be improved over time. Each skill has three tiers that the player can level them up to, which will increase the functionality of the respective skill.

For example one skill that weakens all enemies around the player for a few seconds will have its range increased with each tier. But at the same time, the duration and effect will go up with each skill point added. While this does allow the player more control on skill growth, the skills themselves lack the variety and depth that Diablo 3's classes had. And since you had to pump additional points into a skill before you saw any changes, meant that skill progression was slower compared to Diablo 3.

Overall, to have to decide on a favorite of the two games, I'm split. On one hand, the better defined classes of Diablo 3 made them more unique and diverse, with a more varied selection of skills. But Torchlight 2's ability to improve skills beyond their original functionality offered more customization and freedom in defining your character.

Looting Around:

Loot is a cornerstone of any ARPG: both as a form of progression and as a motivator for the player. If there is anything that can sink an ARPG faster, it's not having good loot design. Once again looking at Torchlight 2 and Diablo 3, both titles have good ideas, but also some problems.

Let's start with Diablo 3, here the rarity scale is: normal, magic, rare, set, and legendary. Going up the scale, items become more powerful, have greater stats and at set and higher may have special abilities.

The most important factor when determining what is considered good loot for a character is if it provides a bonus to the class's primary attribute. As the primary attribute not only affects how much damage the character can do, but also is a factor of their defense.

Diablo 3
The primary attribute’s effect on loot makes upgrades very noticeable, but occurs rarely.

When the player reaches the end game of Diablo 3 with the current cap of level 60, loot continues to level up three more times to a total of level 63.

Now, level 60 and up items don't appear until the player is in the final act of Hell difficulty (second hardest difficulty.)

The increase in stats from 60 to 63 is huge, as weapons can start providing thousands of extra points of damage with a major defense increase to boot.

Torchlight 2 handles things abit differently. First is that the rarity scale is condensed to just: normal, magic, rare and legendary with set items appearing in rare. Because there is no such thing as a primary attribute, a character can equip any weapon that they want. The strength attribute does affect how much physical damage a player can do, while focus affects magical. But the main determination of a weapon's damage is the actual damage stat on the weapon.

Because of that, the player has more freedom in finding loot as they aren't restricted by one attribute. Good loot can be found anywhere, and the world of Torchlight 2 is randomized with each game. The player can also reshuffle cleared areas to make them new again. Throughout the game, the player may find enemies called phase beasts, that when killed will open up a portal to another area where the player can find more loot.

The problems with loot design between the two games are a simple matter of missing the target of keeping loot varied and improving. Diablo 3 falls short of the target while Torchlight 2 overshot it.

In Diablo 3, the chance to find loot that can provide any kind of upgrade to your character becomes more and more limited the further you play. The reason is due to the reliance of the primary attribute. It doesn't matter at all what the other stats are for a weapon (and in some cases equipment,) as without the primary attribute the item won't be anywhere as effective as those with it.

On the flip side, Torchlight 2's loot problem is that the window of finding upgrades is too wide and makes the feeling of finding new equipment lessen. The reason is that since there isn't a focus on what attributes are needed for gear (with exception to damage and armor on weapons and equipment respectively), you're free to use whatever you want.

However, while it gives the player more freedom in finding items, it also makes it hard to find meaningful upgrades - or pure upgrades. In Diablo 3 it's very easy to see what gear would be better or worse for your character. But in Torchlight 2, it's very easy to be overwhelmed by different gear for the same slot with minute differences.

Another problem is that loot doesn't scale up as quickly in Torchlight 2 compared to other action rpgs. If I'm in a level 28-30 area, I may still find loot for level 24 and below dropped by enemies. Due to how varied the loot table is, that level 24 item  may be as good as something around level 29, but the chance of that is less, compared to finding something of a higher level.

In terms of both games I would give the nod to Torchlight 2. As having more choices is always better than less, and the random world design mixes things up better then what is currently in Diablo 3.

For the last category: combat, this can be a tricky one to look at. This is hard to examine critically while you're being surrounded by dozens of monsters.

Calculating Combat:

Combat seems like something so simple in an action rpg, but there is a lot of subtlety hidden under the surface. We could actually have a full post related just to talking about what goes into making combat interesting.

There is no point in explaining combat in Diablo 3 and Torchlight 2 before talking about them in depth, as both use similar systems. The key difference is how combat flows in either game.

In Diablo 3, combat is just a tad slower pace than Torchlight 2. Granted on the higher difficulty levels, one mistake can leave you dead or close to it. However enemies move and react slow enough that the player has a chance to use a skill or actively avoid a high damaging attack.

Diablo 3
Torchlight 2's special effects can easily get in the way and lead to death.

The use of the skill system allows the player to define their move-set for any given fight: Focusing on area of effect skills, direct damage, slowdown and many more.

And the combat is slow enough that you can set up attack strategies like putting down a slow trap as a Demon Hunter, and following it up grenades when the enemies are grouped up.

Torchlight 2 plays at a faster pace, with more enemies attacking per group compared to Diablo 3. Due to the pace, it is harder to set up skills in a strategy as enemies will quickly close the gap and do damage. Ranged projectiles fly a lot faster and in more numbers, making it harder to actively dodge them while attacking.

Because the general size of the enemy on screen is smaller in Torchlight 2 compared to D3, it makes it easier to lose enemies in a crowd of special effects. During a boss fight for instance, I went from full health to dead in one second and I had no idea what actually happened.

Granted I've had the same thing happen to me in Diablo 3 inferno mode, but I was at least able to see it coming before dying horribly. That's why I give the nod to Diablo 3 in terms of combat, as it delivers on both fast pace, and allowing the player to react to what's going on.

The Final Tally:

When I first started writing this post  I was leaning towards Torchlight 2 being the better ARPG of the two. However, after spending time playing both games back to back, I had to give a slight nod to Diablo 3.

The reason is that while both games have design problems, the issues that I have with Diablo 3 (loot, more randomness to world) are easier to fix then Torchlight 2's problems (combat pacing, class refinement.)

With that said however, it's not exactly a clear win by Diablo 3, as both titles are good for the genre and are neck in neck in terms of edging out the other one for the top position. With a Diablo 3 expansion planned and more patches, it will be interesting to see how the game will change from launch in a year's time. And with Path of Exile now in open beta, more people can see another variation on the ARPG genre.

Related Jobs

Digital Extremes
Digital Extremes — LONDON, Ontario, Canada

University of Central Florida, School of Visual Arts and Design
University of Central Florida, School of Visual Arts and Design — Orlando, Florida, United States

Assistant Professor in Digital Media (Game Design)
The College of New Jersey
The College of New Jersey — Ewing, New Jersey, United States

Assistant Professor - Interactive Multi Media - Tenure Track
Bohemia Interactive Simulations
Bohemia Interactive Simulations — Prague, Czech Republic

Game Designer


John Flush
profile image
So the analysis puts them neck and neck, which means one then has to find additional data points that make a difference to them. The fact Torchlight 2 can be played offline, in a LAN environment and for only $20 should easily sway many a person to the fold. Alas, Blizzard has massive marketing appeal and the ability to buy your way to success. It is no wonder so many people are cheering on the little guy.

Jacob Pederson
profile image
Another point that a lot of people wouldn't think about. Runic managed to make a game very very close to the quality of Diablo III at what percentage of the development cost? 10% 1%?

Adam Bishop
profile image
I have to disagree with your analysis that the games are close. I think Torchlight 2 is better by a very wide margin. (For the record, I'm comparing the base game, which is what the vast majority of players will experience, not the end game which is aimed primarily at an exceptionally hardcore crowd).

You say that TL2 is faster paced than Diablo, but I would say it's the other way around, and to Diablo's detriment. Diablo often feels basically like a button masher. You hit your primary attack over and over again to build up your primary resource then you spam one or two big attacks (or if you're playing a class that starts with a full resource bar, you spam your big attack first and then hammer on the primary attack button to refill your resource). Every fight in D3 plays out virtually identically; there's virtually no variation in enemy variety or the tactics necessary to defeat them. TL2 is a much more strategically demanding game. Part of this is because resource management is more important in TL2 (I virtually never used health potions in D3 and there are no mana potions), and to me resource management is one of the most important aspects of good RPG combat design.

TL2 also supports much more novel strategies. For example, I played a character who had one ability that put a hex on the ground which reduced enemies' armor and another ability that threw out a projectile which bounced off walls and did rebound damage. Combining those two abilities meant I came up with a strategy whereby it was to my benefit to try and get enemies trapped in corners where my abilities could stack for maximum damage. There's nothing like that in D3.

I also found myself have to approach different enemy types with different tactics in TL2, whereas in D3 almost every enemy can be beaten by spamming the same attacks in the same order every time. This is especially frustrating when facing D3's bosses, which are basically just very large versions of the standard enemies, having slightly more hit points. In TL2, by contrast, the bosses employed widely varying behaviours which created novel challenges not seen in regular combat. They require the player to think. D3 never asked me to do that.

Josh Bycer
profile image
"You say that TL2 is faster paced than Diablo, but I would say it's the other way around, and to Diablo's detriment. Diablo often feels basically like a button masher. You hit your primary attack over and over again to build up your primary resource then you spam one or two big attacks (or if you're playing a class that starts with a full resource bar, you spam your big attack first and then hammer on the primary attack button to refill your resource). Every fight in D3 plays out virtually identically; there's virtually no variation in enemy variety or the tactics necessary to defeat them. TL2 is a much more strategically demanding game. Part of this is because resource management is more important in TL2 (I virtually never used health potions in D3 and there are no mana potions), and to me resource management is one of the most important aspects of good RPG combat design."

I think this is where we're not going to agree . As I had more of a button mashing experience playing TL 2 then I did D3.

As an outlander every fight had the same result: Start shooting, get one shoot off and the enemy group is on me. Every time I used a skill to slow or stun, by the time I got far enough way to start attacking again, the skill worn off and the enemy was right next to me within a second.

With D3, I was able to pull off skill combos a lot easier. Such as using entangling shot to slow down a group, trap to do more damage while my sentry gun hammered them from a far.

Regarding the boss fights: I though that D3 was a lot more interesting compared to TL 2 as each fight was with a different enemy. In TL 2, the fact that every fight had infinitely spawning enemies just turned them into a slog and lacked the epic sense of the one on one fights such as the butcher.

Robert Boyd
profile image
From what I've played of each (I included Titan Quest as well since that was my favorite hack & slash RPG before the recent batch of new ones)...

Character Design = Path of Exile > Titan Quest > Diablo 3 > Torchlight 2
Loot = Path of Exile > Titan Quest > Torchlight 2 > Diablo 3
Combat = Diablo 3 > Torchlight 2 > Titan Quest > Path of Exile

Of the 4, I think Torchlight 2 has the best variety in enemy types & AI. Also, Torchlight 2 & Titan Quest get bonus points for having strong mod capabilities (Titan Quest in particular has some very impressive mods since it's been out for a while).

Josh Bycer
profile image
I'm curious as to why you would rank POE at the bottom when it comes to combat? I recently started playing it again and I like how action skills are tied to drops and can be improved and combined with modifier gens.

Ricky Bankemper
profile image
Overall, solid analysis. I agree with most everything you said.

Only point I slightly disagree on is which game I think is overall better. I give the slight edge to TL2 only because I enjoy the superior customization it offers, which I find to be the true heart of an ARPG. The fact that my choice of weapon affected the gameplay shouldn't be overlooked. The one thing I think really holds TL2 back from being much better are the problems you described in the loot system.

One thing I would consider adding is Plot/Story. While I would agree that you mainly judge an ARPG on Character Design, Loot and Combat. The story delivery influenced my reception quite a bit in these two games.

Diablo 3 bombarded you with cutscenes, while TL2 was a little more subtle. Seeing as both games are designed to be played over and over, it was a curious move on Diablo 3's part. To constantly interrupt the player with cutscenes when they probably don't care past the first time was rather frustrating in my opinion, even if you can skip them... you still had to skip them, unlike D2 or TL2.

Ron Dippold
profile image
'the issues that I have with Diablo 3 (loot, more randomness to world) are easier to fix then Torchlight 2's problems'

I don't think you can easily fix the Diablo 3 loot problem since it's caused by game being balanced for the Auction House, which is one of the primary design pillars of the game. They certainly haven't fixed it yet.