You actually have a lot of control over the C# GC. One common technique is to suspend GC before entering a tight loop and then manually running GC when your pro

You actually have a lot of control over the C# GC. One common technique is to suspend GC before entering a tight loop and then manually running GC when your program has downtime. For instance, you might pause GC during your game's frame render routine, then unpause in the space between frames. Otherwise, GC runs when it feels like it, which always seems to happen in an inconvenient place that causes visible stutter. You also have the concept of "pinning" a variable, which prevents GC from moving it around memory (for instance, if you pass a very long-lived pointer out of C#, you might want to pin it so the pointer address is always constant). So you can suspend and resume GC globally, you can manually run GC at any time, and you have ways of exempting certiain objects from GC. I don't know offhand if you can GC one specific object manually, but I would be very surprised if you couldn't. Also, all of the above is considered very bad practice. As a rule of thumb, you are not smarter than the GC and you will do a worse job unless you are very sure of what you're doing.