Distribution

Edgegap’s modern, regionless orchestration platform is built from the ground up to provide a multi-tenant environment. Each studio can manage multiple productions within a single, geographically distributed, and highly available environment.

Edgegap prides itself on leveraging its patented orchestrator on the world’s first, and largest, edge network built for multiplayer game server hosting. It includes, as of writing, 615 locations worldwide across 17+ cloud and bare metal providers who are all available to deploy game servers on-demand. 

Edgegap's platform instantly distributes multiplayer games worldwide without the need for selecting regions like in traditional orchestration platforms.

Multiplay advertises access to “over 290 data center locations across the world”. However, it does not offer a list of these locations throughout their documentation for Multiplay hosting (it only does for their relays). Also, as clearly stated through their documentation and pricing, Multiplay is a traditional “per region” orchestration. Meaning that each location must be purchased individually.   

Latency

Edgegap’s platform, using its patented decision-making algorithm and the world’s largest edge network, to deploy game servers closest to users. Which enables game developers to deliver:

• Reduces players’ latency by “58% on average vs. public cloud”;

• Ensure “78% sub-50 milliseconds (i.e., "real time") latency vs. 14% for public cloud,” alongside “91% sub-100 ms latency (i.e. “sweep spot” for non-eSports multiplayer) vs. 67% for public cloud.”

Critically, this ensures a “95% improvement of players' experience” worldwide, which helps game developers ensure a  certain consistent end user experience including traditionally challenging markets such as Oceania and Asia which doesn’t always justify hosting in these markets with traditional orchestration given certain countries’ lower average revenue per user or small  population size.  

Additionally, it helps game developers avoid static, region-locked matchmaking which helps increase match quality for players.

Unity’s Multiplay platform does not advertise on their website, or documentation, any latency improvements.

Unlike Edgegap which has an edge network, Multiplay clearly states it uses a network of “data center locations”.  

Players in both cities and farther geographic locations are likely to have poor latency due to being farther away from the data centers. As stated in this article, Edgegap’s collaboration with a AAA publisher showed that, despite having the AAA studio’s large number of locations (more than what even most studios would be able to afford), by using traffic from 600,000 transactions and comparing the results with a AAA studio’s current architecture, only Edgegap demonstrated an average latency reduction from 116 milliseconds to a drastic 48 milliseconds.

Scalability

Edgegap’s performance benchmark proves its orchestration can consistently scale at 40 deployments per second, sustained for 60 minutes, for a total of 14 million concurrent users (“CCU”) of players worldwide. Thanks to its patented decision-making and rapid-scaling technology. Stacking two of such instances on Edgegap’s platforms allow game developers to manage as much traffic as Fortnite had during their peak launch (100 req. per seconds).

This allows game developers using Edgegap’s orchestration to scale and ensure to succeed the biggest scaling challenge of orchestrators; namely meeting player’s demand over a short period of time such as a midnight launch, a game’s addition to a subscription service, or a “streaming sensation” overnight popularity.

Unity Multiplay mentions on their webpage they offer automated orchestration. Namely that it “scale the number of servers in your fleet up and down automatically in response to player demand” as to ensure “enough capacity available without wasting resources.”

Unity Multiplay offers no performance data or historical data to confirm the level of synchronicity between player requests and their deployment.

Of worry is Multiplay’s orchestration architected being based on a fleet manager. Which means that rapid scaling can be more challenging since servers can take up to 15 minutes to boot when requested. Even if a server is available, deployment of the game server itself has been reported as very long, with a terrible boot time as noted by an ex-Multiplay user:

"[Multiplay has a] cold start of 16 minutes to boot a server when a request is done!” – Multiplayer developer who migrated to Edgegap

The game developers is responsible to cover the entire cost of every server in every location, whether they have players playing or not.

Resilience

Edgegap’s vast network telemetry allows it to detect issues with sites or providers, such as outages, and instantly redirect deployments across its 17+ providers across cloud and bare metal.

Edgegap’s platform has been running live 24/7 for the past six years, maintaining over 99.99% availability.

Unity Multiplay’s website advertises “an uptime SLA of 99.95%” (i.e., about 4.3 hours of downtime every year), which is well below Edgeap’s 99.99% uptime (i.e., 4x less, namely less than an hour of downtime per year).

Unity Multiplay’s service status is available through Unity’s status page, including an incident and maintenance history, for up to 3 months.