Hi @ngaywood,
Welcome to the forum ![]()
Yes, that’s right.
How well it works will depend on a number of things, such as the type of GPU, what it’s being used for, and so on. It is possible to share a GPU among multiple users in ThinLinc, however 50 users would probably require very substantial GPU resources.
Remote acceleration is achieved using a third-party piece of software called VirtualGL. We have an article which explains how this can be used with ThinLinc here:
Is it possible to provide server-side graphic acceleration via ThinLinc?
Hope that helps!