I find in fundamental sciences, sometime newer editions of book are not better, especially if you have a book where the first edition was like 100 years ago, then it build up a reputation (because it was good), next edition were getting better and better. Then the person who made these new editions retires/dies and somebody incompetent takes over, tries to give the book a 'new' organisation, and it turns into a big mess.

I remember when I asked a prof for a book recommendation, and he warmly recommended a title and author. Got the book from the library and it was a mess. Years later I saw an older edition (which the prof studied as a student) and it was beautiful. If you just flip through the new edition one the flaws are not obvious immediately.