AlphaGo and the future of game AI

If you aren't familiar with the game of go (also called baduk, weiqi, and a few other names) the recent news of a computer beating a human grandmaster may seem kind of boring and irrelevant.  After all, haven't computers been routinely beating human grandmasters at chess since 2005?

Chess, ironically, has more complex rules than go.  You can learn how to play go in, literally, about five minutes.  Yet making a computer program that can play go at an even semi-professional level has proven to be impossible until just now.

Why?

Because you can't predict your way to victory in go and that's how the chess playing programs win.

Deep Blue won against Kasparov not by actually playing chess, but by predicting what would happen from each move.  There's only 64 squares on a chess board, each piece only moves in a limited number of ways, therefore there's only so many moves that can be made at any given moment.  Deep Blue could predict hundreds of moves into the future, pick the move that would give it the best outcome, and win.

That approach doesn't work with go simply becuase the play space is too big.  There are 64 squares in chess, and each piece can only move so many ways.  In go there are 361 spaces to play in, and any space can be played at any time.  That's just too much to predict.

As a result, go playing programs were never very good.  They can beat a beginner easily, they can challenge a mid level amateur, and even a lower ranking master player could beat them with no real difficulty.  The grandmasters never bothered playing a go program before (at least not in tournaments, I'd be surprised if they hadn't played a few in less public arenas just to see).

AlphaGo wins not by predicting but by actually playing, thinking, more like a human does.  It's a combination of two neural nets (one for strategic evaluation, one for local fights), some old school heuristic AI programming, and yes as much prediction as they could manage (humans try to predict too).

And it just beat one of the very best human go players in the world 4-1 in a five game match.

The team behind Deep Mind says that they're going to try turning their attention to StarCraft: Brood War.  Not StarCraft II, because of all the DRM and such they can't hook their program directly into SC2, while building a custom AI for SC:BW has been possible since it was released.

I have little doubt that they'll be able to make a very good StarCrat player out of Deep Mind.  There are more options in SC than in go, but at the same time its actually simpler from a certain standpoint.  And, of course, any computer will have a micro game that is just plain unbeatable.  I suppose for the sake of fairness they'll have to cap Deep Mind's APM, but even with an APM cap it'll doubtless have a better micro game than any human.  The only question then is strategy, and if Deep Mind can master the strategy of go, I think it'll win in StarCraft no problem.

Don't misunderstand, I love StarCraft, but compared to go it just isn't very strategically deep.

Personally I'd love to see how it does at Civilization.

Leaving aside the potential for developing actual general purpose AI, that is a human level self aware mind, what this means in the shorter term is the possibility of a genuinely challenging AI in single player games.

A much less complex version of the same sort of neural net program that they use with AlphaGo, in a decade or so when computer power makes that possible on your home rig, would allow for a single player opponent that is challenging without cheating.

Right now the only real way for game AI to be a challenge is by cheating.  StarCraft II, for example, allows its AI to have more resources, do more damage, take more damage, etc.  In Civilization the AI gets free resources, city bonuses, etc.  Name a game, if there's an AI that tries to be challenging, it almost certainly does so by allowing the AI to cheat.

Some games make this a central gameplay mechanic (AI Wars for example).  Others try not to talk much about it.  But they all do it.

For now anyway.

Right now Deep Mind is a hugely expensive, highly experimental, result of centuries of programmer time, and it requires a truly staggering amount of CPU just to work.  But give it a decade or so and all that will be possible to package into a game you buy for $60.  Or heck, a bit less than that for cloud based AI.

The Flash (2014)

The Flash
Grade: C
Genre: Comic book, action adventure
IMDB

Having a kid makes you watch a lot of shows you normally wouldn't.  The Flash is one that turned out not to be awful.

Fitting into the continuity that DC and the CW already built with Arrow (about the Green Arrow, naturally) the Flash takes the Arrowverse more in the direction of DC comics by introducing metahumans, time travel, multiple universes, and so forth.

When I read comics, for whatever reason I drifted to Marvel first and never really paid much attention to DC, and as a result most of the villains they show on the Flash are newish to me.  I'm vaguely familiar with the DC universe from watching the old 1980's Justice League cartoon series and more recently Batman The Animated Series and Batman the Brave and the Bold with my son.

As a result, I came into the Flash with minimal expectations, and was pleasantly surprised when the first season turned out to be quite well done and the second season (so far) turns out not to suck too badly.  The writing team either changed or got worse in season two, but its still watchable.

The cast mostly does a credible job, but the real stars are Tom Cavanagh and Jesse L. Martin, who respectively play Dr. Harrison Wells and Detective Joe West.  In the first season, Tom Cavanagh brought a delightful air of subtle menace to his role, managing to seem amiable enough but threatening and keeping a dark secret as well.  Jesse L. Martin brings Joe West to life, taking a role that in less competent hands would be merely mawkish or cliche ridden and turning it into a rounded and interesting character.

One of the nicer aspects of the Flash is that it fits today's world so well, though it never really takes a closer look at what it portrays.  Joe West and his daughter Iris, Barry Allen's adoptive family, are black.  In the real world that sort of adoptive situation is rare because its actively discouraged.  Race is simply not discussed in the Flash, ever, which considering Joe West's position as a police detective is difficult to justify.  I don't demand that every show be The Wire, nor do I demand a deep examination of race in America from a show aimed mostly at teens and young adults.  But the total silence seems at first slightly odd and then almost oppressive.

So far two characters are acknowledged to be gay.  Police captain David Singh (who doesn't look Sikh to me) once grumpily justified a bacon cheese burger on the grounds that his boyfriend was trying to make him eat healthier, and later had wedding jitters as he prepared to marry his boyfriend.  It was handled well, Singh was established as a character before his sexuality was brought up.  He was always a hardass police captain who was gay, not a gay police captain.  And that is a sign of progress, even just five years ago the fact that he was gay would have needed a lot more explanation and would have been a much larger part of his character.

The other openly gay character is a brilliant supervillain, who like Singh is more than just a gay villain.

The Flash is a monster of the week type show, with a season long story arc that threads through all the individual episodes.  The characters are mostly well developed and interesting enough, and while most of the villains were one shot a few are recurring and become (as villains inevitably must) more interesting than the heroes.

Barry Allen isn't as angsty as a typical X-Man, but the writers try to wring every drop of pathos out of him that they can.  Ultimately he's sort of shallow, because superheros almost inevitably are.  He runs in circles around bad guys, discovers that he isn't fast enough, finds the inner willpower to go faster, and runs in faster circles around bad guys which defeats them.

But Wentworth Miller's Leonard Snart manages to rise above the constraints of being the guy in a parka with a cold ray, and become the sort of villain you can cheer for.  And considering how cheesy Captain Cold is in most other portrayals that's saying something.

Overall, the Flash is a solid TV show delivering an average amount of entertainment.

On Trump supporters

One of the constants in liberal politics is an unwillingness to take conservative voters at face value.  The book What's the Matter With Kansas is a near platonic example of this.  The basic thesis being that the voters in Kansas *should* have their own best financial interest at heart.  But they voted against their financial best interests so therefore they must have fallen victim to the lies and manipulations of the economic elites who are duping the average conservative voter into hurting themselves.

The conclusion is that conservative voters are basically misguided and if only they can be brought to see the truth (that they'd do better economically under a liberal government) they'd vote for liberals because that makes the most economic sense even if they don't like the liberal social agenda.  For a long time this view made sense to me and that was how I tended to think of conservative voters, as basically well meaning but misguided or duped individuals.

I've lately come to think of this viewpoint as being not merely bullshit, but paternalistic condescending bullshit.

If you actually listen to what conservative people say they're quite open about what they want and how they think and what they believe.  The supporters of Donald Trump may be the best example of this, but the Tea Party hasn't exactly been shy about their beliefs either.

And while they'd like to do better financially, it isn't the biggest item on the list, and moreover they see doing better financially via the sort of things liberals propose as cheating.  In fact, for many of them doing better financially, at least by liberal means, is an option they're willing to sacrifice in order to get what they actually want.

What Trump has done is simply rip off the veil of bullshit that conservative politicians have been trying to cover up the actual drives of conservative voters.  Lee Atwater, describing Nixon's Southern Strategy described the obscuring bullshit politicians found necessary to deploy rather than simply openly addressing the desires of their voters:

You start out in 1954 by saying, “Nigger, nigger, nigger.” By 1968 you can't say “nigger” — that hurts you. Backfires. So you say stuff like forced busing, states' rights and all that stuff. You're getting so abstract now [that] you're talking about cutting taxes, and all these things you're talking about are totally economic things and a byproduct of them is [that] blacks get hurt worse than whites. And subconsciously maybe that is part of it. I'm not saying that. But I'm saying that if it is getting that abstract, and that coded, that we are doing away with the racial problem one way or the other. You follow me — because obviously sitting around saying, “We want to cut this,” is much more abstract than even the busing thing, and a hell of a lot more abstract than “Nigger, nigger.”

Race has always been one of the core issues of American politics, but since the 1960's it's an issue that conservative elites have been uncomfortable with.  But the conservative base has never really felt right about the need to obfuscate the racism.

A few years ago when a conservative complained about being silenced by "political correctness" they were almost always talking about liberals who they imagined would be annoyed at blatant racism or sexism.  These days when a conservative complains about "political correctness" they almost always mean conservative elites shushing the open racism and urging the use of dog whistle terms.

Trump is turning Atwater's advice on its head.  There is a hard core of conservatism in the USA that is not merely xenophobic, racist, and sexist, but is proudly xenophobic, racist, and sexist, and they're tired of being shushed, tired of being told to keep it away from the cameras, and tired in general of the politicians they vote for being embarrassed about who voted for them.

86.9% of Trump voters in one poll agreed with the statement "people like me don't have any say about what the government does".  Donald Trump is popular, in other words, largely because he sounds like your racist grandpa does when he spouts off about topics he knows nothing about.  Illegal immigration?  Build a wall and make Mexico pay for it!  Problems in the Middle East?  Bomb the shit out of 'em!  Complex matters with Muslim refugees, immigrants, and Muslims native to your nation?  Ban all Muslims!

The Trump supporters may also believe that implementing Trump's policies will benefit them economically, but that concern is a far distant second to the more important, more pressing, concerns they have.

Ultimately the desire is for a return to a status that never was, a desire for America to become like they imagine the 1950's were like.  A mythic time when American meant white, and everyone who wasn't white kept their heads down and didn't make waves if they knew what was good for them.  A mythic time when men worked and women stayed home to clean, cook, and care for the children.   A time, in short, when straight, white, Christian, men were unquestioned as the top of the heap and everyone else acknowledged it.

There's a reason why Trump's supporters are so overwhelmingly white, and a reason why people like David Duke and other white supremacists keep endorsing him.

The simple fact is that Trump is saying openly what Republicans have been urging their followers to say privately for decades.  He is the ultimate expression of Nixon's Southern Strategy, and the end product of years of Republican support for hate radio.  That this embarrasses the establishment Republicans who want to keep egging on white resentment for easy votes but don't want to be painted as racists, and especially don't want to actually implement programs that the resentful white voters want, is irrelevant.

To expect a Trump voter to reject him based on economics, or really facts of any sort, is to miss the point entirely.  Trump supporters are the product of resentment, they are not voting for a political agenda but for a tribal one.  It isn't about money, though the financial pinch they feel may well help amplify their resentments.  When they say they want their country back, when they say they represent Real America and by implication that anyone who isn't like them doesn't count as being really American, they mean it.  And that's how they'll vote.

Factorio, or why I may never see my family again

Factorio
Grade: B
Platform: PC, Mac, GNU/Linux
Genre: Building, supply chain management, resource management, sim
Price: 20€
Site
Steam

Odds are good that when you saw the genre entry for this review, you had one of two reactions: puzzlement, or joy.  For the puzzled among you while its true that many (if not all) games have an aspect of resource management some games are more focused on that part of the game than others.  Factorio is really nothing but resource management.

You have to mine iron to make iron plates.  But you have to mine coal to fuel your smelters which turn raw iron ore into iron plates.  You have to use factories to turn iron plates into iron gearwheels.  You have to use iron gearwheels and iron plates to make conveyor belt segments.  You have to use conveyor belt segments to......

If your eyes glazed over there, then Factorio is probably not for you.  Try the demo anyway, its free and you might find you harbor a previously unknown love of supply chain management.  If, on the other hand, you're the kind of person who really started loving Dwarf Fortress when you discovered how detailed the game was when it came to making goods, then Factorio is right up your street.

Factorio is about building a factory complex, shuffling the raw materials around to make finished goods, and shuffling the finished goods around to make even more complex finished goods.  For a game of this sort it has a startlingly small number of resources: copper ore, iron ore, wood, water, coal, and oil.  That's it.  But they're combined, refined, reprocessed, and produce a plethora of goods and intermediate products, and every single thing needs to be moved from where it was produced to where it is needed.

You'll discover that to avoid making a factory as tangled as a plate of spaghetti, and one that's nearly impossible to expand, you'll have to make your factory sprawl.

There's alien bugs who hate pollution and will attack, but they're mostly included just to add more challenge and to force you to expend resources defending your factory.  Combat is meh at best, in the traditional sense, but it fits the theme rather well since the whole point of the game is automation so it makes sense that the focus of combat is on automatic turrets (so far I've found the easiest way to beat the bugs is to "walk" turrets to their hive).

The problem with games of this nature is that their appeal largely rests in the learning curve.  Figuring out how to make a factory work is the challenge and once you have it cracked you either go completely obsessive and refine everything to the utmost, start trying to see if you can build a Turing Machine in game (you can), and otherwise really get into it, or you lose interest and wait until the next game comes along and can consume you until you've learned all its secrets as well.

Factorio hopes to avoid that by encouraging a vibrant modding community, as a result the game is very mod friendly and has a nice interface and lots of modding support.  So far the mods are looking good.

All of which is doubly amazing since technically Factorio is still in pre-release and the developers don't think its anywhere near finished.  Despite having been available through factorio.com for quite a while, it is only very recently available on Steam and that only through the early access program there.

However, unlike most early access games, which often seem like varying degrees of scam, Factorio is polished enough it could be called a finished product right this second.  There's lots more the developers intend to add, but if they were all hit by the hypothetical bus tomorrow Factorio would still be well worth the price.

Man of Steel (2013)

Man of Steel

Grade: B
Genre: Action, Alien Invasion, Comic Book

When I first saw Man of Steel, back at its release, I was underwhelmed.  I hadn't intended to bother watching it again, but my son has been on a Superman kick and since its really bloody easy to rent a movie on Google Play, I loaded it up for him.

And I found that on rewatching it was a much better movie than I'd thought the first time I saw it.  Zack Snyder did good.

There were bad parts, the Kryptonian costumes looked uncomfortable and like costumes rather than like clothes people actually wore.  Only Lara Lor-Van (Supes' mom) looked like she was wearing clothes that a person would wear in the real world.

Plus the absurdity that Jor-El, a man bred for science and presumably not trained as a soldier, was able to be billy badass and defeat not just Zod's mooks, but Zod himself.

But leaving that aside, the movie worked well.  Flashing back and forth in Clark's timeline worked as a good way to establish his character and history while not dragging things out too much as all origin stories will if the director isn't extremely careful.

I did wind up positively hating the character of Johnathan Kent.  The problem wasn't Kevin Costner's acting, he did an excellent job.  The problem was that as written Johnathan Kent was not really a nice or good person.  He instilled fear and doubt in his son at every turn, worked actively to undermine Clark's natural desire to do good and help others, and generally was the closest to a villain I've ever seen Johnathan Kent portrayed.  By comparison, Diane Lane's Martha Kent was wonderfully done, and quite refreshing after the American Gothic cypher that she was turned into for the 1978 Superman.

But the best and most interesting thing about Man of Steel was the fact that Snyder, for a first in Superman movies, really looked at Kal-El through the lens of him being an alien.  That shift in viewpoint changed the whole tone of the movie in a positive and interesting way.

I'm also not sure whether it was a deliberate choice, or simply a remnant of Snyder's love of action sequences done by fast forwarding then ultra slow motion for the arrival of the punch or sword or what have you, but this time around I noticed that the first major power Zod and his soldiers really exploited and used was super speed.  Kal-El, by comparison, first developed his senses, his strength, and his flight.  But any good soldier knows that mobility is the key to victory, and so it really made sense that the first power Faora-Ul and the others would develop would be their newfound speed.

At heart, Man of Steel is a story of aliens visiting and invading the Earth, and Snyder managed to capture the awe at the arrival of aliens, and the fear at them being hostile perfectly.  Its more in the family with Independence Day, and The Day the Earth Stood Still than it is with the first series of Superman movies, or even with Superman Returns.  By choosing that approach to the movie, Snyder manages to instill a sense of awe and wonder at parts of the Superman story that I'd have thought were mostly worn out.  I'd been of the increasing opinion that retelling the origin of the more famous superheroes needs to end.  We know how Spidey gets his powers.  We know that Clark is the last survivor of a doomed world.  We know the Fantastic Four got their powers through irresponsible science.

But by changing not the content, but the viewpoint, of Clark's origin, Snyder developed something new and well worth watching.  Yes, we know Clark is the last survivor of a doomed world and most people who do Superman stories gloss over that part.  But for all that he grew up in Kansas that makes him an alien.  And that aspect of his origin turned out to be well worth exploring.

On rewatching Man of Steel, I find that I intend to watch Batman vs. Superman which I'd previously not intended to do.  Overall, it was a fine movie.

War For the Overworld

War for the Overworld
Grade: C-
Platform: PC, Mac, GNU/Linux
Genre: 4X
Site
Steam

War for the Overworld is really Dungeon Keeper 3, it can't be called that for copyright and trademark reasons, but it's DK3.

And that's both good and bad.  Hardcore DK fans like me have been clammoring for DK3 ever since DK2 was released, so it seems a bit churlish to give it a less than stellar rating when it finally does get released.  But I've got my reasons.

WftO has a lot going for it, to begin with they got Richard Ridings the original voice actor for the mentor from DK1 and DK2 to voice this game as well, and his silky yet sinister tone was always one of the little joys of playing Dungeon Keeper.  The WftO designers also must have been working hand in glove with the legal team to skirt as closely as possible to violating the trademarks and copyrights on Dungeon Keeper without actually stepping over the line. It looks like Dungeon Keeper, it feels like Dungeon Keeper, it sounds like Dungeon Keeper. 

So why a C- instead of a C or a C+?

A variety of reasons, beginning with poor support for post-campaign play.  There is a not very well designed sandbox mode, the inevitable PvP mode, and the promise, or perhaps threat, of DLC in the future.  Compare to DK2's rich post-campaign play and it doesn't hold up.

There's also the problem that while they did innovate a bit, much of what they came up with was mediocre at best.  I did like some changes.  The mana pool is better managed than it was in DK2, they made mana regenerate at a flat rate and took out the abuse inviting mechanic of allowing prayers in the temple to generate mana.  I honestly think DK was better without mana, the DK1 mechanic of paying for spells with gold seemed like a better option, but if we must have mana there is no doubt that WftO handles it better than DK2 did.

Unfortunately that's the only new thing in WftO that was a real improvement.  I like the idea of the alchemy lab, but found the implementation and the range of potions wanting.  I liked the idea of rituals from the dark temple, but like the alchemy lab I found the implementation wanting (how did sacrifices work exactly, it was never really easy to tell) and the rituals themselves a bit bland.

Then there were the unnecessary add ons, the tech tree was kind of pointless really.

Where the game really suffered was in a failure to take risks.  It isn't so much DK3 as it is a loving recreation of DK2 with a few pointless add ons, and a short campaign.  Worse, they kept the parts of DK2 that weren't all that great, and the parts of DK2 where it became less innovative than DK1.

The best part about DK1 was the inversion of aggression.  In most games the player is the aggressor, the one who dictates the tempo of the game, the one invading the enemy and crushing them.  But in DK1 all that was reversed.  The player was a passive defender, the game dictated its own tempo for the most part, and with a tiny few exceptions the player was the one being invaded (though if they planned well of course they were party doing the crushing).  DK2 and WftO changed that, it is always the player doing the invading and taking the active role.  The joy of building a dungeon filled with traps and lurking monsters waiting to destroy would be heroes who would dare try to steal your gold is not a part of DK2 or WftO and I miss that and I miss the bold innovation and role reversal it entailed.

They also kept the Dungeon Keeper combat system, which is both understandable (imagine the wailing if they changed it and it wasn't very fun) but also disappointing.  Combat is easily the worst part of the entire series, it was never anything more than grabbing your monsters and dropping them into battle then hoping you outnumbered the enemy (or in DK2 hoping you'd managed to get all the Dark Knights while leaving your opponent stuck with the goblins).  Combat in WftO is exactly the same, and that's a point really driven home in the level where you have to fight off several other Keepers (ahem, sorry, "Underlords") and the whole thing is basically a bunch of dropping and occasional smiting of enemies with lightning.  Of all that parts of DK that cried out for change and innovation it was combat that called the loudest, and those calls were ignored.

WftO does away with the DK2 mechanic of a shared pool of potential recruits that all Keepers must compete in, but it makes getting hordes of monsters somewhat difficult. Each building will only summon a tiny number of minions, and buildings cost gold and land.  In a way this is good because it can force you to throw minions into combat you otherwise wouldn't, but its also bad because it makes expansion more a matter of deciding which minions you need the most and building more of their rooms than any other consideration.

I'm iffy about the plot of WftO as well. It was short, had an abrupt end with no real demounement (not something I object to, but something of a risk and apparently one that many other players did object to), and while neither DK1 or DK2 had much in the way of player choice in the plot, the way it was presented felt more natural and less forced than it does in WftO.  You had a home realm that seemed like it'd be fun to decorate and fix up with your newly acquired rooms, but trying to do so resulted in the dark god harassing you every few minutes to get back and finish the campaign which kind of detracted from the potential fun there.  Through the entire game you're the pawn of the dark god, and not an unwitting pawn but rather a pawn who is constantly told that they are a pawn, which got old after a while.

What's worse, the plot sometimes had you do something it claimed was important (beating up the other Underlords to make them serve you) but then apparently forgot about that and their servitude was never mentioned again, you never used them for any purpose, it was just sort of left hanging.

Overall, if you want to relive DK2 with better graphics I'd say WftO is a good buy. It's also pretty fun on its own merits, if nothing spectacular.  Personally, I'm loading up DK1 and DK2 from GOG and playing them again.

Superman (1978), a look back at an old movie

Superman - The Movie

Grade: D

Genre: Action(?), Comic Book

My son has been on a bit of a DC kick of late, what with the Flash on Netflix, so I decided to show him the old 1978 Christopher Reeve Superman.

To say it didn't age well is to vastly understate the problems with this movie.

I recognize that times and tastes change, and that modern moviemaking has advanced.  But yeesh.  

It was so **SLOOOOOWWWWW**. Begin with the flipping credits, which were done in that godforsaken era when they had to show all the bloody credits before the movie started, everything from the stars to the assistant to the key grip. Five minutes and 23 seconds of nothing but John William's excellent music and a bunch of names flying onto the screen and off again before they let the movie actually start.  I realize that's a relic of the era, but it set the stage for the pacing of the rest of the movie.

The Krypton sets and costumes did age well, the sets are minimal and don't look awful, the costumes were good.  In fact, I'd argue that the Krypton costumes looked better in the 1978 movie than in Snyder's 2013 Man of Steel.

But it started with the setup for the second movie.  Before Jor-El can actually get to the part where he says Krypton will be exploding any moment now and that the leadership is ignoring a crisis at their peril, he has to be the final judge, and apparently executioner, of Zod and his minions.  This develops plot, but for the second movie, not this one.  Couldn't it have been dealt with in a flashback in the second movie?  Even if they wanted to film it for the first, just to avoid having to get Marlon Brando back or whatever?  And dang even that went slowly.  The bit where Zod & Co were actually being put into the Phantom Zone just dragged, it was many long tens of seconds of reaction shots and a long dragging shot of the 2D plane that would be their prison slowly zooming across the screen and it felt like it just went on and on for no reason.

Finally, 11 minutes and 15 seconds from the start of the movie, the plot starts moving with Jor-El putting out his plea to the council to evacuate Krypton.

But it kept on being so slow.  There were long psychadelic sequences with baby Supes in the escape pod, growing up and being educated by Jor-El's computer ghost.    Then long dragging scenes in Smallville.  Then another bloody 2001 A Space Odyssey style psychadelic scene with fragments of Jor-El lecturing the now teenage Supes.

It was 48 minutes from the time the movie started until Clark arrives in Metropolis. Seriously, no joke. Out of a 2.5 hour movie, they spent almost an hour just setting things up to get the main character to the place where the action takes place.

And the slow pacing didn't let up then either.  Long sequences with nothing actually happening on streets or in offices, long pans across street scenes with random people waking around, interminable times when character wasn't developed and plot didn't advance and basically you just sat there waiting for something to happen.   

They harkened back to the old TV series with a few scenes of Supes fighting crime, but even those dragged and had no real action movie style pacing.

Perhaps its just my modern audience viewpoint intruding here, having Supes fight most crime seems kind of pointless.  He captured a cat burglar, and my thought was "doesn't he have something better to be doing?"   And the bit where he captured a gang of robbers seemed a bit too little and too late, there were dozens of stray bullets flying around and likely hitting bystanders during that whole sequence.

The actual plot, the part where Lex is going to use a nuke on the San Andreas fault to make California fall off and turn his useless desert property into valuable beachfront property is basically glossed over, after wasting the first 1.5 hours of the movie on a boring nothing, they pared down the actual plot to a few brief, but still somehow very drawn out and slowly paced, shots of incompetent and sexy sidekick changing some missile codes, then an extended (and SLOW PACED) action sequence of Supes reversing the damage and the weird flying around the Earth to make it spin backward bit.

Don't misunderstand, slow pacing can at times be good.  I loved Kikujiro, a movie that is nothing but long slow scenes with little action happening.  But it was better done, with a whole lot of subtext going on during all the long slow scenes, and there was artistry there.

The sequence with Lois and Supes flying around wasn't bad, the idea was to try to catch the wonder of flight and the truly amazing thing about Supes being Supes.  It worked well and if it had been tightened up I think it would have broken the mood they were trying to set.  But most of the long slow sequences had none of that purpose, and weren't so well done, and were just long for the sake of being long.

There were elements where it was absolutely brilliant.  Christopher Reeve looked great and acted perfectly as Superman, always the moral core of DC's universe and he played it well with humor and a caring undertone that kept his Perfect Boy Scout routine from seeming cloying and smug.  Reeve's bumbling mild mannered reporter Clark Kent was so spot on that it set the standard and still looks amazing even today.

Overall though, I'm amazed it did well. I think, if someone aggressively edited it, you could actually get a decent hour and a half movie out of that steaming mess. Obviously tastes have changed since 1978.