I haven't seen that video, but as I said the iGPU should be able to access more RAM as needed. Basically, assigning RAM to the iGPU in the BIOS means that memory can only be used by the iGPU, while the rest of the RAM is free to be used by either the GPU or CPU as needed. As long as the memory management of the shared memory is done well, there should be little to no impact from reducing the size of the fixed amount allocated from iGPU (and might even help in some cases, as the RAM is free to be used wherever it's needed most). See here:
https://www.techspot.com/article/1578-amd-raven-ridge-reserved-memory-explainer/