100% agreed!

tbh Hyperion Cantos also had a great premise but an unsatisfying conclusion to me too

Simmons comes up with great beginnings but can't quite deliver when it comes to the ending

Yeah. I read the third in the Cantos and couldn't be bothered to finish. I think the second book would have been a better place to end.

The Hyperion Cantos is probably one of the greatest literary disappointments of my life. Couldn’t put the first book down from start to finish. Second was a downhill slide culminating in one of the most anticlimactic endings I’ve ever read. Couldn’t bring myself to pick up the third 😞

I do remember reading a very funny takedown of it (amongst other books) by David Langford in a New Worlds revival anthology the library somehow had.

I think he was also snarking about the climax to Stephen Baxter's Xeelee sequence because there's a whole thing the problem with trying to do Olaf Stapledon when you're not Olaf Stapledon that's seared into my memory and kept on coming to mind when I was reading Liu Cixin's Deaths End.

(Olaf Stapledon, now there's an sf author with impeccable progressive credentials. Unfortunately they're pre-WW2 progressive credentials which means the whole eugenics thing is a bit jarring.)

BTW I translated Star Maker to Polish (first full Polish translation, although there were fragmentary ones before and another full one since)

I think more SF writers should stick to short stories for a longer while before they even try to tackle a multi-volume epic saga

Yes. I think this is one way in which the dearth of good markets for short fiction has caused problems for the genre.

short stories are where SF excels more than novels, they're the bloodstream of the genre, but are sadly neglected by readers and publishers these days

(GRRM will always be "the author of Sandkings" to me first and foremost)

Some of Phillip K Dick's best work is his stories; less room for him to go off on one. (I think this is also true of horror; a good horror novel is just more difficult to write the longer it gets.)

WRT horror, he's a terrible example to cite now, but Neil Gaiman has always been a master of shortform horror who got lost whenever he wrote something longer than 30 pages. The strength of Sandman was that it gave him space to write shortform horror and pretend it was a series.

Never much liked it.

It's funny how I'm mentally going "Is it too much of a basic bitch opinion to say I really like A Song for Lya" when even knowing that GRRM had a whole career pre-ASOIAF is semi-rare.

That story definitely shows GRRM at his best.

mindflayers and githyanki being a thing in D&D at all are because GRRM wrote something similar in a book in the 70s and a teenaged Charles Stross got inspired by that to write up statblocks and other information about them and submit it to Dragon magazine

I mean, it’s good too, Sandkings just made the biggest impression on me as kid

The best market is Clarksworld which is a pretty religiously hard sci fi magazine