In 1983, when I was a budding young student heading off to Princeton to do a physics Ph.D. I had the privilege of spending some time with an elderly John Edsall, an eminent Harvard biochemist, who was friend and collaborator of my father. I asked him if he had any advice for me. I expected him to say something profound and give me a long list of suggestions. He thought for a while and said, "Well I am not sure but I guess it is important to know the literature on what you are working on."
I am not sure to what extent I took on board this advice over the next decade.
But, now I think this was very good advice. The reason is knowing the literature can save you a lot of time. If you are trying to do something someone else has already done then
-perhaps there is no point in trying it yourself
-you may be able to use what they have done to do something even better.
It is amazing what a discerning Google Scholar search can pick up. On the other hand, you need to be careful you don't spend all your time downloading and reading papers. Also, don't assume your supervisor knows the literature.