Favourite Arthur C. Clarke

[AE35Unit]

Great book.... her's my 1956 edition...

wow that looks well loved!

 

[SpanishMill]

I've been trying to find an "affordable"..."The Nine Billion Names of God" hardcover book. Why is that one sooo expensive?

Yay.... awaiting arrival of my 1st edition

 

[AE35Unit]

Very nice ! Thats a real find!

 

[SpanishMill]

Finally got it in... ex library but still good enough...

 

[Fried Egg]

Over the last five years, I read his five inclusions in the SF Masterworks series and while none of them have blown me away, none have been that bad either.

There's not much between the best and the worst but I would rank them in the following order:

1) Childhood's End
2) Fall of Moondust
3) Rendevous with Rama
4) The City and the Stars
5) The Fountains of Paradise

I don't know whether it's true that these are a representative sample of his best work but it leaves me feeling glad that I've read them but not really inclined to read any more. Perhaps I will try some of his short stories, but as for novels...

 

[J-Sun]

I'd like to "revise and extend my remarks". I've since read or re-read a couple and realized that I've covered more of his works than I'd thought in my first post in this thread.

Clarke's first books (Prelude to Space (1951), The Sands of Mars (1951), Islands in the Sky (1952 juvenile)) are readable and okay but are largely travelogues that are deficient in plot/conflict and other things.

He hit his novelistic stride with Childhood's End (1953) and Against the Fall of Night (1953) but not so much in stride that he didn't basically rewrite the latter into the much better The City and the Stars (1956). Meanwhile, he produced the neglected sub-classic-but-still-neat Earthlight (1955).

After the revision, he went on to produce the superb The Deep Range (1957) and A Fall of Moondust (1961). Then he wrote a second juvenile and a non-SF novel (Dolphin Island and Glide Path, both 1963, both unread) and finally hit the Big Time with 2001 (1968).

He actually announced his retirement from novel writing but was perhaps lured back by a massive contract with a record-level advance and his "return" was greeted with much rejoicing and Rendezvous with Rama (1973) swept the major awards. He followed this up with the extremely anemic and uninspired (IMO) Imperial Earth (1975) before producing one of his very best books with The Fountains of Paradise (1979).

Then came the recycling period, with 2010 (1982) which I actually liked, and a rather placid The Songs of Distant Earth (1986, inspired by an earlier tale), and the pointless and virtually unrelated (IMO) third Odyssey, 2061 (1988).

I quit paying attention after that. All that remained was the collaborative/recycling period from 1988-2008 with "collaborations" and sequels with Gentry Lee, Greg Benford, Mike McQuay, Michael P. Kube-McDowell, Stephen Baxter, and Fred Pohl (running the gamut there). The only titles with just Clarke's name on them were The Ghost from the Grand Banks (1990), The Hammer of God (1993), and the final Odyssey, 3001 (1997). The only one I read was 3001 - it was much much better than 2061 but not great or essential.

So my Clarke Masterworks would be

Childhood's End (1953)
The City and the Stars (1956)
The Deep Range (1957)
A Fall of Moondust (1961)
2001 (1968)
Rendezvous with Rama (1973)
The Fountains of Paradise (1979)

(Fried Egg: I'd agree that, if you had to limit it to five, the SF Masterworks series seems to have nailed it. (Though it really has to be at least seven.) But if that didn't blow you away and make you want to try more, I doubt you'll really love the stories either. I'd recommend The Collected Stories of Arthur C. Clarke (2000), generally, but many of his stories are only adequate - there are numerous all-time classics and I don't know of a "best of" that really captures them and it sounds like you'd want a concentrated dose of the very best if Clarke's best novels don't light you up. Maybe The Nine Billion Names of God (1967) or The Wind from the Sun (1972) but I suspect they have too much (relative) filler for you.)

 

[Fried Egg]

Thanks for that overview of Clarke's career J-sun, it was most interesting.

 

[J-Sun]

Thanks - I'm glad if it was of any use.

 

[psikeyhackr]

A Fall of Moondust

I am not sure if it is the first Clarke novel I read. I believe I read Childhood's End near the same time. But I think Moondust more than any other book caused me to go for engineering in college.

psik

 

[Rafellin]

The only Clarke I have and one of my all-time favourite books: The Deep Range.

 

[Jesse412]

A short story of his I found particularly moving is The Star. Earth explorers led by a astrophysicist/priest, travel to a distant star system that was destroyed by a supernova. They discover a time capsule sealed in a vault that was left by the dead civilization that once lived there. I can't really explain anymore than that without spoiling the ending but it was shocking and I was very moved.

 

[gdoc]

I loved Rendezvous with Rama and The City and the Stars.

 

[scidata]

Rendezvous with Rama is an interesting book. It has the technological depth of a Clarke novel, the epic social implications of an Asimov novel (the long sweep of history), and the descriptive texture of a Jules Verne novel. It will probably remain a classic long after its original readers are all dead and gone.

 

[AE35Unit]

Got this recently from bookmooch.com
Its been a while since I read it. And I dont think I've ever read Lion.

 

[Frost Giant]

While J-Sun's comment is certainly true (about the book being for juveniles) , I still have to say Islands in The Sky is my sentimental favorite. It was the first Clarke story I ever read and while it is a travelogue I enjoyed it very much as a young person. It was what sucked me in to the rest of his work.

 

[Edward M. Grant]

I read Islands in the Sky recently and liked it. What really dated the book was not the space travel, but the way they use humans for so many things that we'd do with a computer; it's funny that Golden Age SF writers massively overestimated progress in spaceflight and massively underestimated progress in computers.

Edit: actually, I think one of the problems with his books is that they were usually so hard SF based on the scientific and engineering beliefs at the time that they often don't stand up as well as softer SF of that era. I re-read Fall Of Moondust recently, and it's a good book, but we now know that it wouldn't happen because the Moon doesn't have seas of dust. Had it been set on the third moon of Jaglon Beta, it would have survived the decades much better.

 

[Frost Giant]

Edward M. Grant - "it's funny that Golden Age SF writers massively overestimated progress in spaceflight and massively underestimated progress in computers."
You're right about that, they were victims of the age they lived in. Saturn rockets were being launched and we were reaching the moon, so it didn't seem too improbable that humans would spread across the solar system. The thing was, the computer the Apollo astronauts had was less powerful than my wristwatch and true interplanetary travel was expensive and dangerous. Once the cold war sputtered out, that was pretty much it for the notion of humans living on other worlds. Many writers didn't foresee that our culture would turn inwards and develop computers into what they are today.

 

[Ray McCarthy]

Computers though are just smaller and cheaper. They also over estimated progress in what class of tasks/ problems can tackle. There is about zero progress in real AI since 1946. The inherent kinds of things that Turning, Shannon, Von Neuman and others 100% knew COULD be done with enough storage hasn't changed. The power (i.e. speed) of a computer doesn't allow more complex things, just faster completion.

They only underestimated the cost and size. The IC was what made that possible:

Newly employed by Texas Instruments, Kilby recorded his initial ideas concerning the integrated circuit in July 1958, successfully demonstrating the first working integrated example on 12 September 1958.

Commercially viable ICs from about 1960
ANY current computer design can be made with 1930s mechanical relays or electronic valves (tubes). It would just be VERY large, expensive and slow. Tommy Flowers proved a valve computer could be reliable during WWII. Conrad Zuse had a reliable relay based computer by 1939. The German Military were not interested. The ARM based CPU core in phones and tablets is lower power than Intel / AMD x86-64 because it uses very much fewer transistors.
A discrete transistor is about 1mm x 1mm x 0.1mm in a 4mm diameter tube. IC transistors are about 25nm, an nanometer is a million times smaller than a mm.
Acorn Archimedes - Wikipedia, the free encyclopedia About 4.5 million instructions per second in 1987. A 1945 computer was about 500,000 times slower, though at code breaking only about 1000 times slower as that is what it was used to help with.
Turning proved what class of problems a computer could solve. That hasn't changed. The so called Turing test was just an unproven idea, almost a parlour game, not a proof of AI or any suggestion AI was possible. There is still none. "AI research" today uses its own narrow meanings for language to create the illusion of progress.

No one in 1940s or 1950s imagined that in 1969 we would have men on the moon and then before the end of the 1970s simply lose interest.

Golden Age SF writers massively overestimated progress in spaceflight and massively underestimated progress in computers.

So I would argue the reverse is true for up to 1973 (The first Intel 4004 microprocessor shipped in 1971, Affordable "pre IBM" PCs by 1976)
Intel 4004 - Wikipedia, the free encyclopedia
Today's Graphical User Interface, mouse and networking / Internet concept all working by 1976 on mini-computers.
Computer just have got cheap and small. A decent display now costs more than the rest of the laptop!
Entire computer including storage, RAM, graphics and I/O can be one chip, though typically three in a phone / tablet and many more in Intel type Laptop. It can't do any kind of thing the 1939 or 1944 computer in theory could do very slowly.

 

[Edward M. Grant]

ANY current computer design can be made with 1930s mechanical relays or electronic valves (tubes).

It could. But it would be lucky to execute the first instruction before you had to replace some of those billions of valves.

I suspect one reason Clarke believed there'd be men on his geosynchronous communication satellites was because they'd need someone up there to replace the valves when they blew!

I agree on the AI: that's one area where they overestimated computing progress.

 

[Brian G Turner]

He followed this up with the extremely anemic and uninspired (IMO) Imperial Earth (1975)

I've picked this up from a pile of old books I have. The first chapter was lovely, but it's gone downhill since then. 60 pages in and I'm wondering whether to risk continuing or just quit while I'm ahead. I figure better to read a good Clarke, than have Imperial Earth cloud my opinion of his writings. Unless, of course, anyone thinks it's worth continuing with?