Watch: Improbable VP Says There's Little Interest in Massive Single-Shard MMOs in the West -- But Their Platform Can Support 20,000 Concurrent Players in a Continuous, Seamless Virtual World

You'll probably have to turn up your volume and put on headphones, but here's my GDC 2019 interview with Paul Thomas, VP of Product at Improbable, a well-funded startup that launched in 2012 with a mission to enable the creation of "worlds which can be run in real time, simulating the behaviors and interactions of millions of entities." 

In the short term, however, as Thomas explains, the main goal is to support "all kinds of games", as opposed to single-shard MMOs. In fact, the only single-shard MMO that's been announced for Improbable's Spatial OS cloud platform is Seed, the planet-colonizing game.

Why? For the most part, there's just not as much market interest in single-shard MMOs in the EU and North America -- or as he puts it: "In the Eastern regions, massive interest there, but in the Western regions, yeah, it tends to be more session-based type games."

Then again, we are still seeing interesting uses of the platform -- for instance, Thomas mentions Mavericks, a 1000 player battle royale with persistent, virtual world-type elements.

And yes, I asked Paul if Second Life could be deployed on Spatial OS, and his answer was more or less: Maybe, but it would be very difficult. 

At one point, he mentions that they recently tested a single-shard MMO and achieved high simulated user concurrency, but then had to check with his team on the exact numbers. A rep came back with this statement:

"Our current runtime could support 20,000 players in a relatively low-fidelity but seamless world (low fidelity here meaning not requiring the rapid updates of a twitch-based game like an FPS). That's an indicative example - you could optimize in different ways, but that number would allow you to offer a you a recognizable MMORPG experience.

"To provide some context, while a 'Realm' in a traditional MMO might have thousands of players (who would be the players allocated to that shard), the seamless area you could see and interact with on a single server architecture would likely be limited to a few hundred connected users maximum. The comparable unit for New World Notes' purposes would be a 'sim' - the area in Second Life which you can interact with directly."

20,000 is pretty impressive if it's truly seamless -- if as a player, you could travel, in a single unbroken trip, from one side of the world to the other and encounter 19,999 fellow players along the way. (Second Life has player concurrencies in the 40,000-60,000 range, but most of them are located on geographically separated private islands you must teleport to, i.e. not seamless.)

Then again, compare and contrast that 20,000 benchmark with Dual Universe's recent test of 30,000 concurrent (simulated) users. More on that world soon!

Comments

[Willliam Burns]

I think they said everything they needed to say in their answers.

They indicate that MMO isn't popular in the west, but go on to say "games" and indicate that a use case like Second Life would be a "maybe".

This is really all you need to know.

It's not the MMO that's the problem it's the open world sandbox they aren't able to apply to their system that is the problem.

Thats why it isn't popular. It's not the end users being fickle, it's the company unable to apply their technology in a manner that resonates with the end user.

In this manner, Dual Universe has the clearest path to this end goal at the moment.

Especially since they are showing 30,000 agents in a high fidelity test (not low fidelity). Where Improbable hits a brick wall, DU is already climbing well above that.

[MINIMAN10000]

I disagree that the problem is their technology being unable to adapt to the market. The market could very well utilize their technology if they wanted and the only reason he doesn't know if it would work in second life is because it was build for unity and unreal engine in mind, so unless they actually have knowledge of the second life codebase they simply don't know how much work it would be to adapt to their codebase.
I don't know about the east but here in the west MMOs are generally considered to be around 10 years behind in gaming technology, their very basic from a gameplay and quality perspective. On the other hand we have a bunch of high quality session based games. When it comes down to it quality and gameplay is the backbone of the games and MMOs are severely lacking in those areas. If they could pull off that same quality in a single shard in a way where the large number of players doesn't feel out of place I'd expect it to do just as well.
As with all games though, it really is difficult to predict success.
Now I'm thinking back in the day the game EDF was part of a series of shovelware games which were made on a low budget, but it was the most successful. They have since created a franchise off of it. Makes me wonder if created a bunch of low quality games in order to find a target audience is still a viable option these days with such a focus on fidelity.