New Member ASRock GPU Thread

You have now set the minimum allocation to 512mb, this is the lowest amount that
the OS will report to applications. As I said, the memory sharing scales to what
the programs need. it dynamically allocates and reallocates RAM to vRAM and back.

There is absolutely no incentive for manufacturers to trick or persuade users to
buy a dedicated GPU, there is absolutely no guarantee that the user will purchase
a GPU from them over a competitor. iGPU functionality is a balancing act beteen
system resources and GPU performance. That is the only reason shared memory is
dynamic. It allows the system to allocate more memory to the system when there
are no graphically intensive loads and more to the iGPU when there are. It's
that simple. When you factor in that the majority of people using an iGPU are using
more budget focused systems, it also stands to reason that they would not have
huge amounts of RAM to work with. iGPU RAM allocation becomes a non issue with
32gb and beyond.

On my systems that don't need/use a dedicated GPU I leave the share memory allocation
set to auto.