Asimov wrote almost nothing but science fiction for his entire early career, say from 1939 to 1958 or so. His strength was not in ideas per se but in the working out of the implications of ideas. This always worked better for him at shorter lengths. Most of his early “novels” are actually collections of novelettes. His first true novel is the almost forgotten The Currents of Space.
His weaknesses were in the ordinary literary virtues: plot, dialog, characterization, depth. He couldn’t write human-sounding dialog ever: his characters simply lectured the plot mechanics at one another. Even worse, he had a bizarre weakness. He was incapable of writing a conversation with more than two people speaking. Even when he depicts a group of people in a meeting, first he has two people talk at one another, then two more people talk at one another, then two more, never a true conversation.
The less said about his female characters the better. Obviously, there is no sex in his works either.
He wrote that when Sputnik happened he had almost an epiphany. He lost interest in writing fiction and began writing about the sciences, basic knowledge that seemingly the entire country badly needed.
Suddenly all his idiosyncrasies became strengths. Simple clear explanatory prose at short length poured out of his typewriters. Even his books, when analyzed, prove to be individual topic chapters put together rather than a developing argument. He was probably the best ever at explaining the basics of science.
His books on other topics are not up to the same standard. You can present history as a straightforward timeline, but you can’t say anything interesting about history without analysis and this he could not manage.
You can do this with mysteries, though, and he wrote hundreds, almost all pure puzzle shorts with minimal characterization.
After he became famous in several fields, his publisher dangled huge checks in front of him to start writing novels again. He pumped out big fat bestsellers of no merit whatsoever, bloated, static, talky books in which all of his flaws were magnified. Bestsellers are a genre unto themselves and seldom have anything to do with literary merit, as the many discussions of Dan Brown here have shown.
I can’t imagine where that comment about his wife having to lock him into his office came from. Every word out of his mouth while he was alive was just the opposite. His idea of a vacation was to be left alone to write more. By the end of his life he didn’t have to write anything for any reason. He just did.
It’s quite true that he wrote incredible numbers of books, but they do need to be put in some perspective.
Literally dozens of those books are reprints of reprints, repackagings of works previously published in book form. Over one hundred are very short works in instructional series for children. Asimov had no real part in any of the 180 or so anthologies that Marty Greenberg packaged under his name. Asimov would take the already collected stories and write an introduction.
Writing 200 real books should be enough of an accomplishment for any one career. Falsely inflating the numbers to over 500 was simply a manifestation of his ever-increasing ego as he grew older.
I respect Asimov’s career greatly. I can’t believe it will ever be matched. He was a magnificent self-promoter and a champion of literacy. But his science fiction has been superseded in quality and impact by later writers, and the progress of science makes his non-fiction less useful with every year. He was not a hack, merely a representative of his time who could not grow out of his roots.