Warning, I’m going to go all over the map in this entry because this is really one of those stream-of-conscious style posts where I just typed out what I was thinking.
There’s often times a lot of speculation about whether or not the next Diablo game will be a full blown MMO or not. Myself, I still believe that there are a few old school executives left at Blizzard who understand the games roots where its core design is based on instanced single player content (with limited multiplayer co-op in order to fulfill the always online component) as opposed to potentially moving towards an MMO-ARPG which have mostly linear gear progression, lack of item/build diversity).
I’ve mentioned the hybrid approach before which is utilized by Path of Exile and David Brevik’s Battle.net Town concept that they were never able to implement (a larger town hub/lobby that is based on the typical persistent world design except just for the hubs. The “challenge” is creating a much more seamless transition when moving between the open world hubs and the instanced zones. It’s been done; good examples are Blade & Soul and Riders of Icarus (when moving from the open world to solo quest or dungeon instances), though there is still stutter during the actual transition (the challenge is getting rid of that stutter completely).
Going on a tangent for a short bit, I have an actual design concept for a large scale game world which blurs the lines between a persistent open world MMO and an instanced ARPG. The easiest way to think about it is to look at your typical MMORPG where the persistent open world you roam around in (with many other players) is one part. Then when you enter a dungeon, that is instanced where the number of players that can enter it is limited. There might also be larger instances like a battleground or siege for example. Now imagine if you substitute a full blown ARPG as an instance that is connected to a larger open world MMO. A player who started their adventures on the MMO side could enter one of these instances and take part in a smaller scale adventure with complete ARPG designs (fog of war, randomized tile sets. non-linear gear progression) either solo or with a party. Similarly, a player could start their adventures on the instanced ARPG and never had to know there is a larger world they could move their character into. Or, they could take their character into that larger persistent open world experience (moving back and forth between the two subjected to certain contraints and rulesets). There could also be multiple such instances (games in their own right) and those instances themselves could have their own expansion content associated with them which could also be connected to the other instances.
It is an extremely lofty idea with a lot of technical design challenges and design compromises. One such challenge in this design is the game server foundation in terms of its social, networking, and security functionality since all of that is critical glue that supports the features used at the game client level for everything from chat, friends lists, data validation. All of that would need to be designed first and be very robust in its ability to scale. The other is an actual system for abstracting character and gear stats because you don’t want gear obtained while playing on the MMO side, to become unbalanced when re-entering an ARPG instance. It’s a design challenge given the dichotomy which exists between the usual linear gear progression in your typical MMO, versus one that has its itemization heavily influenced by RNG. It’s a potential balancing nightmare without a regimented approach to character and gear design (and that balance is where design compromises come into play; it can have negative side effects when it comes to some of the core objectives that are important in an ARPG). The more I thought about it, the more I realized it’s almost impossible to put together a design and development team that would ever be on the same page.
Digressing, the point of the above is to look at how in your typical MMORPG, developers can shrink or expand the persistent world part and the instanced parts. I firmly believe the next Diablo will retain its instanced heritage but expand on the social/networking portions to give those parts a more immersive open world experience similar to the Battle.net Town concept that Brevik had where you can see, chat, and interact with a lot of other players. But once you (seamlessly) transition out of that hub, you are in an instance with much more limited number of players (if you chose to not play it single player).
Which leads to the actual subject of this entry; the camera. The fixed camera 2.5D isometric view is pretty much a huge part of Diablo’s branding and identification. As much as it would be interesting to see a view like the following, deviating from that quarter view game angle would alienate a good portion of the franchises core demographic.
Some folks may then ask why not have options for either. It’s may seem simple but it is actually more complex because of the game mechanics involved with each type. The above utilized a machinima tool that allowed manipulating the camera (usually used to make flyovers of the game environment). Diablo 3 PC however was designed with point and click movement (where your character is either dragged along with the mouse or follows a path to where the mouse cursor is when a button is clicked. The camera being fixed wasn’t designed that way and needs to be manually moved. In the above, if the character is moved, the camera would stay but the character would move out of view. That is unless you add a different control scheme like WASD movement where you are now directly moving that character (and the camera can be made to move along with the character). That would also mean changing the targeting system when it comes to actual combat.
There was a time when Bluehole Ginno was looking into that possibility with Devilian (which is why there was WASD movement, a full zoomed in mode, and assets that were completed more than they needed to be). There were buildings for example in both the Kafkas Canyon and Wintergale Tundra zones where you could walk behind them and there was one house structure in Misteye that you could walk into where the inside was complete (walls and flooring). And while the camera allowed a full third person camera view, the ability to rotate it was never implemented.
You essentially end up with the exploding parts problem trying to deal with implementing two different cameras (when one is fixed and the other movable) and control schemes since it affects things like monster AI as well (nothing that a lot of time and resources can eventually solve but when talking about production budgets, priorities need to take over). With Devilian, the WASD movement was left in (but with a side effect for point-to-click movement when it came to animation lock where your character always completed its last queued instruction; thus force standstill never worked as expected by traditional ARPG players) and zoomed in mode left as a feature to view your character up close.Blizzard has the resources BUT it still comes to down to production costs and coming up with solutions that possibly only a small percentage of the games demographic may end up using. So the simplest solution is to stick with the camera view (and thus combat and control scheme) that the franchise is synonymous with.
