As a Roblox developer, it is currently too hard to save a visual of user generated content from within an experience. For an example, my game Clip It uses ViewportFrames, tons of compression, optimization, culling, and other systems to allow us to display previews of users creations (clips) on profiles, in our in-game messaging system, etc… you can see examples here.
If Roblox is able to address this issue, it would improve my development experience because we would not have to load tons of assets into memory just to display a preview of a user-generated creation.
In my ideal scenario I have an API where I feed a model, a camera cframe/fieldofview and aspect ratio/resolution and Roblox takes an image of that model from the camera’s perspective, then prompts the user to upload it to Roblox. The user would be prompted via a CoreGui that explains how they are responsible for moderation on this asset. This would now be a regular roblox image asset that can be loaded into an ImageLabel and save us tons of resources.
I love the idea, but I can understand why this would be nearly impossible to implement in a way that is not extremely limited. This is largely a technical challenge when it comes to data storage, not only would it require massive amounts of storage space, but it would potentially be consuming massive amount of IO which over time = large amounts of money when you have to replace storage devices because devices are failing (both SSD and HDD have finite amount of Read/Write before they start failing), now this can be handled by cache but cache is not cheap either when you consider thousands of games, potentially each having thousands of images cached, which is a whole technical challenge on its own.
Now if we could use a external service this would be different, but this would be unsafe in so many ways entirely bypassing Roblox moderation.
This would be amazing for in-game report systems which could allow users to take snapshots of rule-breaking behavior, upload it as an image and send it straight to Moderators via something like Discord, which, mind you, is already how the in-built Roblox report feature works.
What’s mentioned here is something I was thinking of when I read this post, but seems to be only half of the solution. CaptureService already has all the technology needed to take a screenshot of the user’s surroundings built into it, and can output it as an image just fine. Using EditableImages is possible but it’s relatively costly (I think?) to send these off somewhere external, and I’m not even sure if it’s possible to translate them into an image that can be used externally.
