Dev Diary: MCOC Release Process and Bug Fixes
KabamDORK
Administrator Administrator › Posts: 219
Greetings Summoners,
We on the Community Team had a moment to speak to the Game Team regarding a sentiment we often hear around why certain issues haven’t been fixed, or what is taking so long around issuing acknowledgement and solutions, and wanted to share with you their thoughts!
This is a complicated and multi-faceted question, and the answers vary depending on the specific issue we’re talking about, but we’d like to spend a little time today talking about one of the many variables we juggle here - our build release pipeline.
One of the challenges of running a live service game, that releases a new build every single month, is that the next release is always right around the corner. To be able to ship builds predictably, we go through a number of build finaling milestones for each release.
When it comes to bugs, there’s two categorizations that are helpful to understand:
Client Bug - This type of bug requires a fix to the game’s client code. We have to release a new build to get the fix in. This requires going through the first parties’ approval processes all over again; when we do this to address critical live bugs, this is what we call a Hotfix Build.
E.g. Issues with animation, VFX, etc.
Server/Data Bug - This type of bug requires a fix to the game’s server code, or to the data that lives on the server (we can do this without having to go through first party approval processes again. While this typically requires significantly less time than a client fix, it still pulls time and resources away from other things)
E.g. Issues with damage values, effect durations, etc.
As a point of improvement, we’ve been making efforts to move more of the game over to server/data so we have more flexibility with fixes, but this isn’t feasible for some systems.
Now let’s take a quick look at an example of how some build finaling milestones overlap each other…
So hypothetically let’s say we identify a live client bug on Mar 14th (new build required to address). We’re 4 days out from when we need to submit the 50.0 build to first parties for approval so at this point, we need to ask a few key questions regarding risk and feasibility with regards to addressing an issue.
From the developer side of things, waiting a few months to implement a bug fix that fits within the build release cycles feels like a rapid and organic turnaround. However, from a player’s perspective, we understand this can not only feel like an eternity, but also a frustrating experience when playing our game every day. Game development on a live service game is often not a sprint, but a marathon, and we need to make sure we properly pace ourselves while figuring out how best to efficiently deliver solutions to issues and pain points.
Now, while the examples used here were pretty straightforward, not all of them are. With 10 years under our belt and over 300 Champions/interactions, there are many compounding issues and challenges that regularly present themselves with each new build. To name a few, older legacy tech that can be difficult to work with, obligations to our partners, impacts on content plans, all the cool new things we’re working on, and the multiple devices, platforms, & versions we support, etc. all need to be weighed out as part of the prioritization equation.
Our goal in this entry is to provide some transparency into how we work but, all that to say, we hear you and we’re continually making improvements to help make your experiences with Marvel Contest of Champions better. Many of us actively play the game every day, and we feel the pain from bugs and issues just like you do. Some of these problems are a lot more challenging than others, and will take longer to address - but we are working on them and to further improve visibility on how we’re addressing them. We look forward to sharing more with you soon in a future entry!
Until then, we’ll see you, In The Battlerealm!
We on the Community Team had a moment to speak to the Game Team regarding a sentiment we often hear around why certain issues haven’t been fixed, or what is taking so long around issuing acknowledgement and solutions, and wanted to share with you their thoughts!
This is a complicated and multi-faceted question, and the answers vary depending on the specific issue we’re talking about, but we’d like to spend a little time today talking about one of the many variables we juggle here - our build release pipeline.
One of the challenges of running a live service game, that releases a new build every single month, is that the next release is always right around the corner. To be able to ship builds predictably, we go through a number of build finaling milestones for each release.
When it comes to bugs, there’s two categorizations that are helpful to understand:
Client Bug - This type of bug requires a fix to the game’s client code. We have to release a new build to get the fix in. This requires going through the first parties’ approval processes all over again; when we do this to address critical live bugs, this is what we call a Hotfix Build.
E.g. Issues with animation, VFX, etc.
Server/Data Bug - This type of bug requires a fix to the game’s server code, or to the data that lives on the server (we can do this without having to go through first party approval processes again. While this typically requires significantly less time than a client fix, it still pulls time and resources away from other things)
E.g. Issues with damage values, effect durations, etc.
As a point of improvement, we’ve been making efforts to move more of the game over to server/data so we have more flexibility with fixes, but this isn’t feasible for some systems.
Now let’s take a quick look at an example of how some build finaling milestones overlap each other…
So hypothetically let’s say we identify a live client bug on Mar 14th (new build required to address). We’re 4 days out from when we need to submit the 50.0 build to first parties for approval so at this point, we need to ask a few key questions regarding risk and feasibility with regards to addressing an issue.
- Is this severe enough that it’s worth risking missing our 50.0 release date?
- Is this severe enough to warrant a 50.0.1 hotfix build? (also stacks risk on 50.1 release)
- If not, then our next available window (assuming that bug can be reasonably diagnosed, fixed, and tested in time) is 50.1, which releases May 5th (52 days away)
From the developer side of things, waiting a few months to implement a bug fix that fits within the build release cycles feels like a rapid and organic turnaround. However, from a player’s perspective, we understand this can not only feel like an eternity, but also a frustrating experience when playing our game every day. Game development on a live service game is often not a sprint, but a marathon, and we need to make sure we properly pace ourselves while figuring out how best to efficiently deliver solutions to issues and pain points.
Now, while the examples used here were pretty straightforward, not all of them are. With 10 years under our belt and over 300 Champions/interactions, there are many compounding issues and challenges that regularly present themselves with each new build. To name a few, older legacy tech that can be difficult to work with, obligations to our partners, impacts on content plans, all the cool new things we’re working on, and the multiple devices, platforms, & versions we support, etc. all need to be weighed out as part of the prioritization equation.
Our goal in this entry is to provide some transparency into how we work but, all that to say, we hear you and we’re continually making improvements to help make your experiences with Marvel Contest of Champions better. Many of us actively play the game every day, and we feel the pain from bugs and issues just like you do. Some of these problems are a lot more challenging than others, and will take longer to address - but we are working on them and to further improve visibility on how we’re addressing them. We look forward to sharing more with you soon in a future entry!
Until then, we’ll see you, In The Battlerealm!
38