Human Ingenuity Beats AI’s Machine Precision

By Shamus Posted Wednesday Feb 13, 2019

Filed under: Column 144 comments

My column this week talks about the final game between pro Starcraft II player MaNa and the AlphaStar AI. As you’ve probably figured out from the subtle hints in the title, the human won this time around. My article talks about how MaNa won and gives some historical contextIn this case, me talking about games I was playing 20 years ago counts as “historical context”. to the matchup.

How AlphaStar was developed:

They seeded the system with some of the basics of the game: Building a base, creating units, attacking the opponent. From there:

Continue reading ⟩⟩ “Human Ingenuity Beats AI’s Machine Precision”

 


 

Andromeda Part 17: Welcome to the Bungle

By Shamus Posted Tuesday Feb 12, 2019

Filed under: Mass Effect 82 comments

The next stop on our tour of the Heleus Cluster is Havarl. Like the other planets, the climate is out of whack. Unlike the other planets, this place suffers from an overabundance of life rather than a shortage.

Havarl

Havarl is overgrown with a thick jungleYet another single-biome planet.This is the one planet where you can’t drive the Nomad. and all of the native flora and fauna is exceedingly dangerous. Everything is toxic, poisonous, or filled with murderous intent.

The game says this is the ancestral home of the Anagara and that they had to abandon it when it became too inhospitable, but that doesn’t really fit with what we’re shown. I’d expect to see bits of Angaran cities poking out of the overgrowth. They’re a spacefaring species, so their cities must have been pretty impressive. But instead the “cities” are just endless sections of Remnant architecture that form horizon-smothering black walls around the play area. I imagine this is another detail that would have been ironed out if the game had been given time for polish.

I do like that this world offers some contrast with the other four and isn’t yet another lifeless hellscape for us to colonize.

As with the other worlds, you need to visit the monoliths, then do the vault, and then the game tells you the planet is stabilizing. Then you run around and do fetch quests and kill the mooks congregating around map markers until the planet hits 100%. Whee.

Continue reading ⟩⟩ “Andromeda Part 17: Welcome to the Bungle”

 


 

Diecast #243: JAI, Warframe, Trailmakers

By Shamus Posted Monday Feb 11, 2019

Filed under: Diecast 90 comments



Hosts: Paul, Shamus. Episode edited by Issac.

Show notes: Continue reading ⟩⟩ “Diecast #243: JAI, Warframe, Trailmakers”

 


 

Andromeda Part 16: Morda Meets The Eye

By Shamus Posted Thursday Feb 7, 2019

Filed under: Mass Effect 139 comments

Elaaden is where the Krogan decided to settle when they left the Nexus. There’s some local politics you need to sort out here to stabilize the region. Morda has appointed herself as “Overlord” of all Krogan here. She’s bellicose, confrontational, and prideful. She’s also proof that bringing the Krogan on this expedition was a foolish move, surpassed only in the foolishness of betraying them. I much prefer stories that have a sort of inevitability to them, where mistakes are grievous but understandable. The mistakes made by the Andromeda Initiative are so idiotic that I sometimes lose interest in helping out. I often find myself thinking, “Screw it. These morons deserve everything that happens to them.”

Assuming you’re not going to turn the game off, you’re going to need to deal with Morda. There’s another Krogan, Strux, who is more cunning and less overtly warmongering. I really thought the game was setting up a choice between a warmonger Krogan leader or a devious Krogan leader, but you can’t actually side with Strux. Instead, Strux attempts a coup that brings about his own downfall, and the only choice you get to make is if you want to screw the Krogan yet again. Like the confrontation with the Cardinal on Voeld, it’s like the writer deliberately ignored an interesting choice to offer you a shallow one.

Continue reading ⟩⟩ “Andromeda Part 16: Morda Meets The Eye”

 


 

AI Hasn’t Really Mastered StarCraft II

By Shamus Posted Wednesday Feb 6, 2019

Filed under: Column 137 comments

In my column this week I talk about the AI that recently beat some pro StarCraft II players 10-1. This was a tough column to write because I think that:

  1. The developers that designed the AlphaStar AI have done something really remarkable and groundbreaking, but
  2. …they also overstated their accomplishment.

It’s hard to criticize someone that’s just invented something amazing without coming off like a smug idiot. I don’t really have a problem with their AI, or even with the constraints placed on the match that tilted the game in favor of the AI. My problem is with the claims they made that made it sound like the AI was playing with human-like restrictions on speed and perception, when this simply wasn’t the case.

If you want a much more technical analysis, this article by Aleksi Pietikäinen offers a pretty good breakdown on what AlphaStar was doing during the game and why its performance doesn’t really match with the developer’s claims.

 

 


 

Anthem Demo Impressions

By Shamus Posted Tuesday Feb 5, 2019

Filed under: Video Games 198 comments

I have pre-ordered Anthem, but I do not recommend you do the same. I got the game to feed the content mill for the ever-hungry internet. If I didn’t write about games for a living, then I’d hold off until after the first reviews came in. The public demo ran from February 1-3, and I spent as much of the weekend with it as I couldI was held back by health problems that kept me away from the computer for a lot of the time. Pardon if this entry feels a bit dashed-off, I’m just getting back to work now and wrote this in a hurry.. I managed to hit the demo level cap of 15, and then did some grinding until I was fully decked out in rare gear.

Based on what the demo showed me, this game is basically a showcase for the recent sins of EA and late-stage BioWare.

  1. We think all games should be LIVE SERVICES that require an online connection, except we can’t build a stable network to save our lives.
  2. We want to copy the money-sucking microtransactions that the competition is doing, but we’re too dense to notice that those other games are free-to-pay and our game starts at $60.
  3. Dialog is important! Story is important! But it’s not so important that we’re willing to polish the dialog and make sure it’s worth listening to.

Going into Anthem, I wanted one of two things. Either it should be a great game that recalls the days when BioWare could make a game with heart, or it should be a disaster that punishes EA for taking this “story first” developer and making them produce a “it’s all about the shooting and looting and who cares about story?” style of game. Sadly, the game seems to have fallen into the boring but predictable space between these two extremes. I did occasionally have fun with the demo, but I also occasionally had a really miserable time.

Continue reading ⟩⟩ “Anthem Demo Impressions”

 


 

Spider-Man Part 5: How to Spider-Man

By Shamus Posted Thursday Jan 31, 2019

Filed under: Retrospectives 79 comments

During my Batman Arkham City series, I praised the game for the way it seamlessly wove the tutorials into the flow of the story. It’s a big game with a lot of different systems. You’ve got brawling encounters, stealth encounters, detective mode investigations, traversing the worldStuff like gliding, swinging, and parkour., and using gadgets to solve “puzzles”.

The challenge the game designer faces in these kinds of games is that you want to teach the player as quickly as possible. We want the player to have access to the full open world so they can engage with whatever content seems fun to them. That’s the whole point of having an open world. But we can’t let players off the leash until they know the basics because if they explore the open world and encounter gameplay system they haven’t learned yet, they’ll get confused and blame the game. At the same time, ramming too many tutorials down their throat at once is just as bad. It’s tedious, it gets in the way of the narrative right when we’re trying to get the story off the ground, and players will have trouble remembering the lessons if they’re packed too close together.

What you want to do is introduce a concept, allow them to try it, and then give the player a few minutes of doing something else. Then you remind them of the concept. Then later you give them some sort of “final test” where you present the challenge again, only this time with more pressure. Maybe they’ll have to deal with a time limit, or they’ll have less room for error, or they’ll have to blend this concept with another one. The point is that it sort of follows the rule of three in the form of “setup, reminder, payoff.”

Continue reading ⟩⟩ “Spider-Man Part 5: How to Spider-Man”