Diecast #242: Deepmind, Anthem, Win10, Mailbag

By Shamus Posted Monday Jan 28, 2019

Filed under: Diecast 85 comments

Hosts: Paul, Shamus. Episode edited by Issac.

Show notes:
00:00 DeepMind – The learning AI for Starcraft

Since recording this, I’ve watched the DeepMind show match between the AI and a couple of pro players. I think I’ll cover this topic in my column next week.

05:27 Chainmail Bikini Will Return

Like I said on the show, I don’t want to back-date them into the archives. At the same time, I don’t want to post them three times a week or they’ll drown out the rest of the content. Having thought about this some more, I think I’ll be posting the comics once a week, on Sundays.

07:30 Paul introduced his kids to Portal

As an aside, here’s a reference to the Chime problem I had.

14:51 Anthem First Impressions

Link (YouTube)

In the public reception of Anthem, I was hoping for one of two things:

  1. A return to form for BioWare, taking their classic worldbuilding and blending it with modern looter-shooter gameplay.
  2. A buggy disjointed disaster of a game with an embarrassing story and terrible gameplay that punishes the studio for dumping their longtime fans to chase the shooter market.

Instead, it looks like we’re getting the least interesting result. From the the first impressions, it sounds like Anthem is competent but ordinary. It will probably carve out a modest chunk of the Destiny / Warframe market and coast along as a perfectly serviceable entry in the genre. (Until EA closes it down in disappointment because it didn’t make a billion dollars.)

22:15 Fallout 76 Multiplayer System

Link (YouTube)

I keep promising myself that I’ll stop covering this game because there’s no point in beating a dead horse. But damn it, this dead horse was so smug and entitled that I just can’t resist a few more swings.

30:13 Windows 10 on Old Computers

Paul tried to install Windows 10 on these old machines:

  • 2017 Refurbished: HP Compaq 6200 Pro SFF PC Dual Core Intel G620 2.6GHz 2GB RAM 250GB HDD
  • 2007 Intel Core 2 quad Q6600 @ 2.4 GHz, 4GB ram, a 250GB SSD (new 2014), an optical drive, and a Refurbished: EVGA GeForce GT 630 DirectX 11 02G-P3-2639-RX 2GB 128-Bit DDR3 PCI Express 2.0 x16 graphics card (new 2014)
  • 2005 AMD Athlon 64 X2 Dual 4200+ at 2.2GHz, 1.37 GB ram (on three sticks), 2x 300GB hard drives, integrated graphics, a special Sound Blaster X-Fi module, and not one but two optical drives!

43:05 Mailbag: Measuring Story Quality


With all this recent talk about stories, plot-holes and their impact on the quality of said stories, it got me thinking… So, ok, contrivance’s or plot-hole’s impact is negative on the story. But lets imagine there are two stories of equal length, both without plot-holes (for example, they say Alien and The Thing are both without plot holes, and roughly of equal length). What metrics could be used to measure which story is better? I have few ideas (one of them listed below), but I want to hear your (podcast members) take on it as well

I also has another idea (which is more applicable more to movies, than anything else, I gather). So, there are movies with relatively thin plot, but their strength is in great action scenes, that are enabled by said plot (let’s say, Predator, The Raid or half the Jackie Chan movies). So, when I thought about first question, one of the potential metric that popped up in my head, is the amount and the intensity of conflicts presented in the story per time unit. Individual action scene is the short story on itself, isn’t it? I can show you, say, the plane scene in Octopussy, where Bond (and the stuntman) climbs on top the plane and you’ll get the complete idea what’s happening, what’s in stake and how it will affect Bond if he fails. And good action scene is always about how tense the situation for the hero, and bad ones, outside of bad cinematography, have little danger for the protagonist (as an example, in SPECTRE, Bond shoots bad guys in villain’s lair with so much ease, its boring). So, the popular opinion (I think), is that plot in these movies is weak or not great or not worth mentioning. But I’m starting to think that this is the other way around. If the main story enables the writer to give you such amount of tense and inventive visual mini-stories (Bond on the plane, Bond runs on alligators, Bond climbing the mountain and someone cuts his rope), isn’t it the sign of great storytelling? I’m not sure myself yet, but I think that’s an interesting idea to explore.

Best regards, DeadlyDark

56:26 Mailbag: 7/10 Scores

Dear Diecast

I have noticed a trend of inflation when it comes to game review scores. These days even a review score of 7/10 is considered rather poor. What do you think is the cause of this and do you think there is a way to fix this?

Regards Eric


From The Archives:

85 thoughts on “Diecast #242: Deepmind, Anthem, Win10, Mailbag

  1. Lino says:

    I haven’t listened to the whole show yet, but when writing your column, you should definitely read about OpenAI’s AI that plays DoTA 2. In the last International, they had an AI team vs a human team, and the year before that they had an AI go 1 on 1 against some of the best DoTA players in the world.
    It was kind of haunting how good the AI was (even though the matches vs humans had some concessions to account some of the AI’s limitations).

    1. Thomas says:

      Was the AI limited? What is cool about Starcraft is that it doesn’t take more actions or react faster than humans.

      If the DOTA was just reacting faster to win team fights it wouldn’t be quite so cool, but if it was making good positional choices that would be incredible

      1. Lino says:

        The concessions were in terms of making the game easier for both teams. In DoTA, each team needs to buy a courier – it’s an NPC that takes items from a player’s stash and brings it to them. Usually, a team only has one, and the five players need to share it. You also need to be careful when you call for the courier, because the enemy team could easily kill it.
        In the 5 on 5 game, each player started with their own courier, which was invincible. This doesn’t make the game much easier, mind you, and since both teams had it, it balances out. But it is something that people will correct you over if you fail to mention it.
        I don’t think the AI was that much faster than the players, but don’t quote me on that.

        1. Ninety-Three says:

          The original 1 on 1 AI was playing a highly stripped down version of the game and it was not just faster than humans, thanks to API access it was literally faster than physically possible (humans have hardware input delays).

          The final 5v5 bot was playing on heavily restricted, humanlike reaction times. However, it is worth noting that the restrictions they put on the game absolutely work in the AI’s favour: the humans have years of experience playing and learning the real DOTA metagame where there’s one courrier and a hundred heroes and Roshan and and and. The AI has spent all its time practicing the weird stripped-down version of DOTA. It’s like if you changed the rules of chess and then immediately challenged Gary Kasparov: he’d still be good at New Chess, but he’s at an obvious disadvantage.

          1. Lino says:

            Regarding the 1v1, I remember that after TI, a whole lot of pro players managed to beat it. Did it still have API access?

            1. Ninety-Three says:

              At TI, they mostly tried to play fair and got wrecked by superhuman bot micro. People started finding exploits, items the bot’s combat math didn’t account for or ways to confuse it and drag it off the main path.

      2. Zak McKracken says:

        The APM of the AI was only nominally lower than the players’:
        ..so actually, it was probably up to 5 times faster (when it counted) because:

        1: human players “spam-klick”, i.e. make lots of clicks that don’t actually trigger meaningful actions. The AI does that to because it “learned” from humans, but it’s learned to not always do it, especially when it counts.

        2: Because the frequency is measured in 5-second intervals (the limit is for x “actions” per 5 seconds, it can (and does) issue commands at waaay higher frequencies than a human could, at critical points during the game for a few seconds, as long as it reduces the action frequency before and after.

        Another issue is that for the first 10 games (they played 11 in total, and it lost the last one), it was able to “see” the entire map, as opposed to just one part of the screen and having to move the screen over to where the next command needs to be issued.

    2. Henson says:

      The thing that struck me about the DOTA OpenAI was how well it did in the early game and how poorly it did in the late game. And I wonder, what would that be due to? Is it because the late game is more strategic and less micro-intensive? (micromanagement being something that AI would be much, much better at). Or is it because, since the AI learns by doing, it’s had so few opportunities to improve its late game? After all, pretty much only pro teams get past the early game against the AI.

      It’d be interesting to see how it performs this summer.

      1. Lino says:

        I think it the games vs the team at TI showed what makes a professional team so good – a lot of pro DoTA matches come down to the mind games teams play on each other before and during the game. The AI was lightning-fast when it had to adapt to an ambush or a push, but it was very slow to adapt to the other team’s overall strategy. I think that’s why it lost – it gave into their diversions, and just didn’t have an overall superior game plan.
        But yeah – I’ll be also be very interested to see what happens in this year’s TI…

  2. Tizzy says:

    I played portal years after it came out, having no idea what it would be about. I knew it’d be puzzles, but I didn’t expect a funny game. But I thought the game made it pretty obvious early on, and the humour landed for me, definitely.

    1. John says:

      I played Portal and Portal 2 for the first time in 2017. I knew what I was getting in to and they more or less met my expectations. They’re both good games, especially Portal, and full of interesting puzzles, but I don’t think that there’s any question that it’s the humor that makes them memorable. Like Paul, I played them with my child. Unlike Paul, I handled the controls while my child made suggestions and commented on the game. Also unlike Paul, my wife thinks Portal is funny. She’d never played the game–she never plays any games–but she’s a Jonathon Coulton fan and loves “Still Alive” despite having no context whatsoever for it.

  3. Tizzy says:

    So is the Anthem thing a demo or a full beta? Because if it’s a walled-off demo, maybe that’s why we’re not hearing much feedback on the story.

    Edit: My bad, this is actually addressed within the first minute of the video linked by Shamus. The demo is a slice of level 10-15 content, and should comprise complete missions with story. It also means that the demo skips the whole player’s introduction to the world, robbing the missions of some context.

  4. Geebs says:

    Is it my imagination, or has Anthem gone through a major case of “looks a lot less impressive than the trailer” to equal e.g. The Division?

    I don’t think I’d notice, except for the fact that it’s so visually uninspired otherwise that the bling-mapping was the only thing that elevated the original reveal trailer above “tepid”.

    On the other other hand, I think the general lack of inspiration of Anthem as a whole is the only reason why there hasn’t been another “visual downgrade” internet freak-out.

    1. shoeboxjeddy says:

      Are you looking at the demo build with all the High settings on? The real game will also most likely look a little better, as the demo build is almost a month old and they’ve confirmed some changes since then.

      1. Geebs says:

        Just seeing other people playing on youtube. It is at least easy to tell that a streamer is playing on PC because they immediately start bitching about the motion blur.

        I didn’t see anything that wouldn’t have looked out of place in Andromeda, and that’s not the most amazing looking game around by a pretty wide margin. The footage I saw mostly seemed to involve people tooling around in a cave with really flat-looking lighting.

        Was interesting to see that Bioware’s “A” team still appear never to have seen an actual human mouth move, though. I can actually imagine Bioware staffers who didn’t work on Andromeda hearing all the jokes about the animation and going “eh, looks fine to me”.

    2. trevalyan says:

      The demo reviewers I’ve heard of insist that Anthem is eye-poppingly gorgeous, and that even flying around is beautiful. But the point the podcast makes is that extremely important details- the server capacity for the limited audience- is flawed. The cynical notion that “oh no, ths game is so popular the servers are getting crushed” is -stupid- schtick, lads. We know the game is open to a limited population and that you should have sufficient servers. Doing this just confirms you’re stupid -and- that you’re assholes. The game will not be a Destiny killer: at BEST it will be as good. But again, you wouldn’t have to resort to gimmicks if it was good.

  5. Lars says:

    The inflation of game review scores is an outcome of the fact, that only a few titles get the attention to be scored by “professionals”. I never saw a test for say Antigraviator. A lot of the bad stuff falls under the table because testing the good stuff gets more attention by potential readers.
    Fixing that and bring the shown average to 5 would need a massive amount of manpower and page space, which would cause a good rise of costs. These costs would be paid by readers and/or more commercials, at the expense of boring most of the readers by most of the titles.
    Or reviews/previews of good titles have to be dropped to review a lot of visual novel/rpg maker titles at the expense of boring most of the readers by most of the titles.

    1. Hal says:

      I’m not that old, but “back in the day,” PC Gamer magazine would give percentile scores, and they would publish reviews for games that earned single-digit scores. Those were always worth reading.

      Yes, there’s probably something unseemly about dragging a publisher, and it’s a part of that phenomenon where bad reviews are much more enjoyable to read than positive reviews. All the same, I remember a particular review of a paintball shooter that was hilarious.

      1. Shamus says:

        Ultimate Paintbrawl!

        Long after Quake III Arena / Unreal Tournament, someone made a painball game based on the Duke Nukem 3D engine. (Build.) Paintballs would get stuck mid-air, the music was sonic torture, the levels were ugly random mazes, and the reviewer couldn’t actually tell if the AI was playing the game.

        I read that one many times. :)

        1. Hal says:

          Yes, that was it! I knew “paintbrawl” was part of the title.

          I think I still have the physical magazine sitting around somewhere.

          1. shoeboxjeddy says:

            I always used to like reading the reviews for knock offs of games like Tony Hawk. The reviewers were always clearly like “why did I have to play THIS. WHY GOD?!”

        2. Hector says:

          There was also the legendary Big Rigs: Over the Road Racing, a game so incomplete it *didn’t have racing*. The game, I have heard, is terrible. The reviews, however, are hilarious, and it’s in the single-digits on metacritic.

          1. Hector says:

            Sorry, I meant to respond to Hal.

      2. Henson says:

        This may be a change due to method of publishing. When you’re selling a magazine, you’re selling either a whole issue, or a subscription; all the reviews and features come as a package. But when you’re selling a website, your income is based on clicks, and people pick and choose what they want to click on. Given this self-selection method, it’s less likely your audience will actively choose to read a review about some 2-star game from some no-name studio.

        On the other hand, if inflation of scores occurred before the proliferation of internet media, then my above explanation doesn’t explain it at all!

      3. Nimrandir says:

        I got a one-year subscription to Computer Games Magazine when I was in grad school, and I kept those issues for several years just to laugh at the bad reviews. I think Tom Chick wrote most of the ‘winners.’

      4. Sleeping Dragon says:

        Yeah, some publications around here picked that up, sadly it was around the time physical gaming mags were already going into decline but it was still pretty funny.

        One of Loading Ready Run’s more popular streaming series is “Watch and Play” where “Graham inflicts bad games upon Alex”, because they generally play them blind the quality can range from so-bad-it’s-dull to so-bad-it’s-hilarious with an occasional rare gem that can give the impression of being bad but actually turns out to be pretty cool. If you want to check it out it should be archived as a playlist on the Loading Ready Live (not Run) YouTube channel and is an ongoing series they stream on every other Wednesday.

      5. Lars says:

        Reading about hilarious bad games is fun. It is a reason why Jim Sterling is successful. Reading about average middleware is boring. Visual novels aren’t bad per se and they have their fans, but most of the readers/gamers give them a skip.
        Take the shooter genre: About which game do you want to read: Battlefield 1, Lawbreakers, or Ultimate Paintbrawl?
        I bet Lawbreakers as a mediocre game would be last. Following Gauss mediocre is the majority of all published games.

    2. Matthew Downie says:

      Even in the 8-bit era, I don’t recall a time when 7 out of 10 looked like a good score. It’s like, you could buy this, but why would you, when there are plenty of 8+ out of 10 games on the market?

  6. Ninety-Three says:

    I watched all the Deepmind matches, and they were pretty disappointing from a strategic perspective. Obviously it’s a huge technical accomplishment representing a zillion engineer-hours of work and a ton of compute, but ultimately, Deepmind ended up with superhuman micro despite its APM limit (unsurprisingly, it turns out that perfect attention span and perfect map vision are really strong). Its insane stalker micro let it play a pretty aggressive game and coast by on normally weak unit comps, poor engages (fighting up ramps and through chokes) and objectively flawed build orders (it had a weird tendency to oversaturate minerals while leaving vespene open).

    1. Lino says:

      Yeah, when they removed its perfect attention, it lost in a live match vs Mana.

      1. Syal says:

        That one was also scaled back to TLO-game levels.

    2. Thomas says:

      The AI that went up ramps with TLO faced was a lot stupider. Against Mana when it went up a ramp it won – Mana theorised that pro-gamers retreat on the threat of a forcefield when there might be an opportunity. And it took excellent engagements against Mana, often with units coming from multiple angles.

      I really wouldn’t be surprised if the AI was just correct about oversaturating minerals. As soon as it started a new base, it would split the workers between bases correctly. Because it had more workers (which is a relatively small investment) it was completely invulnerable to the early game harass and neither TLO nor Mana could take an economic lead, despite getting adepts into the mineral line.

      Saying that, you are right about it using superior Blink stalker micro with precision, which does seem like a crutch – but this was the most viable strategy when developed against _other AIs_. It doesn’t incorporate player data except for the initial seed. So the AI wasn’t deliberately exploiting superior micro over humans – it’s more likely that when micro’d perfectly Stalkers are overpowered units compared to the rest of the game and the AI was correct going for those Stalker / Phoenix compositions. Which makes sense, because Stalker’s have the highest micro potential when you’re not exploiting super-human prism manoeuvres.

      I think they do need to handicap the AI more, to properly account for the precision advantage. Saying that, the AI had solved all the greatest challenges, it scouted, reacted to builds, deciding when to engage and retreat and could move around the map fluidly – all on only 7 or 14 days training.

      Looking at how slow TLO and Mana were to come to terms with their chances of beating the AI, I suspect we’re going to be handicapping it _below_ human levels before we’re actually satisfied it’s making better decisions than us.

      1. Ninety-Three says:

        Overprobing is, in general, a maybe-correct strategy, what I was talking about was specifically its habit of oversaturating while only being 2/3 on a geyser. That’s pretty indefensible.

  7. Nixorbo says:

    These days even a review score of 7/10 is considered rather poor.

    “These days”? ::Sidelong glance:: This discussion has been going on for well over a decade, right? I’m not going crazy?

    1. Nixorbo says:

      I found search results for “video game score inflation” dating back to 2006. This makes me feel simultaneously better and worse.

      1. PPX14 says:

        Did this happen at any point for films or books? Or are review ‘scores’ themselves a fairly recent thing?

        1. shoeboxjeddy says:

          Film reviews are pretty much all out of 5 stars, so inflation isn’t as keenly felt. Reviewers feel free to use all 5 of the numbers. For book reviews, they’re out there for readers so… they aren’t about the score. They’re about the text. If you don’t want to read a review of a book and want to skip to a score… there might be something wrong with you.

        2. John says:

          I’ve seen several different methods for scoring movies. There’s thumbs up or thumbs down, ratings out of five stars, and sometimes letter grades (as in A, B, C, D, F). In my experience, ratings out of five stars are the most common method, but I don’t assume that my experience is universal. My impression is that most people don’t take movie review scores too seriously, though I have seen people write angry letters to Roger Ebert along lines of “How dare you give Serious Artsy Movie three out of five stars when you gave Wacky Comedy Movie four out five stars! How can you say that Wacky Comedy Movie is a better film than Serious Artsy Movie?” Ebert typically responded with something like “The star ratings mean that Wacky Comedy Movie is better at being a wacky comedy movie than Serious Artsy Movie is at being a serious artsy movie, not that one is a better film than the other.” (Ebert also often said that he’d stop giving out star ratings if his paper would only let him. I get the feeling he was tired after decades of letters like this.) The point is that people have been reading movie reviews and arguing about review scores for much longer than video games have existed and most people have some sort of perspective about it.

          There are several differences between movie reviews and game reviews, but to my mind the most important is that movie reviews are usually published in general interest publications, like newspapers, along with a lot of other content, while gaming reviews are usually published in specialist publications, like gaming magazines or web sites. Your average publication with movie reviews in it is much less dependent on advertising from movie studios than your average gaming publication is on advertising from games publishers. For that reason, I think that movie reviewers are under less direct and indirect pressure to produce favorable reviews. That said, movie reviewing is not without its own controversies. Movie studios sometimes court reviewers with “junkets” where they fly critics, often from smaller organizations like local television stations, to lavish events with the obvious intent of influencing reviews and obtaining favorable quotes that can be used in advertisements.

        3. Retsam says:

          I’m pretty sure this pretty common wherever numeric reviews are used. Simpsons had a joke about this back in 2000: “Sorry Marge, I’m afraid this gets my lowest rating ever. Seven thumbs up”.

      2. Asdasd says:

        I read a lot of games mags in the nineties. Even then a sub-80 score meant ‘this game will be chiefly defined by its flaws’ and a sub-70 score meant ‘do not buy this game’. Everything below 60 was just different gradations of ‘awful’.

  8. Lee says:

    As far as human limitations on AI, the creators claim that they’re limiting the AI to approximately the speed of human input, but this article claims that they’re very wrong about what those limits are in this context: https://medium.com/@aleksipietikinen/an-analysis-on-how-deepminds-starcraft-2-ai-s-superhuman-speed-could-be-a-band-aid-fix-for-the-1702fb8344d6

  9. I almost feel sorry for the developers that works at Bethesda. Sure they are partly in fault for some of the mess, but as always management makes the decisions so the blame lies at them.

    An Todd Howard, in my eyes there is only one way he can handle this issue, imagine a future gameshow. Todd is announced and as he comes out there is a lukewarm reception a few people even booing.

    He’ll have to say something along the lines of “Yeah. I know. WE really screwed up didn’t we. All I can say is that I’m really sorry, and, I hope this makes up for it.”
    Then the lights dim and a really awesome Elder Scrolls 6 trailer plays (revealing a lot more than the teaser did).

    I really can’t see any other way they can get a good gameshow event. After the ES6 stuff they can present some mobile stuff, maybe talk about Fallout 76 changes coming, then present the Starfield stuff (I’ll assume full trailer, and talking and screenshots and some other videos).
    And then at the very end as Todd waves, says thank you and leave the stage, and as he’s off stage the lights dim and a Fallout 5 teaser is shown (ala the ES6 one).

    I hope someone close to Todd and co. sees this (or somebody else tells them something similar).

    Why? Because Bethesda Game Studios (BGS) are one of the few “good” Single Player RPG studios around.
    I mean…
    “it sounds like Anthem is competent but ordinary. It will probably carve out a modest chunk of the Destiny / Warframe market”
    Yeah. They’ve become a slightly better version of Bungie. The BioWare that made Jade Empire, KoTOR and Dragon Age Origins… is gone now.

    Just think about it. The dev team(s?) that worked on Anthem where the same that worked on Mass Effect and Dragon Age. Does Anthem seem anywhere similar to those? Anthem is different, it looks formulaic and “safe”. This is not BioWare exploring and innovating, this is just a “paint a cash cow by the numbers” project.

    Now we may be lucky and Anthem is popular enough to become a cash cow, and if BioWare can them use that money to help fund future Mass Effect and Dragon Age while getting back to the story telling form of yesteryears then that is awesome.
    But somehow I doubt it. BioWare is owned by EA, and there are execs high up and stockholders to answer too. In todays world if you have to balance greed vs art then greed always wins at the expense of art.

    With BGS the Fallout 76 mess is a side project they are trying. With BioWare though, Anthem is their main project and may possibly change future Mass Effect and Dragon Age games. Mass Effect could end up a sci-fi call duty clone or batlefront clone and Dragon Age a fantasy variant of that. With people playing online, a minimal single player “campaign” rather than a proper single player story, and microtransactions.

    I’m actually crossing my fingers for RockStar North now, that with the incredible success that GTA V maybe the higher ups are too nervous to screw up things as a mistake could actually cost them a billion dollars, that they might just let RockStar North do their thing. Which means GTA VI will be fun, big, have interesting characters and be enjoyable to play, have a good single player and a good multiplayer. I do think though that GTA VI’s multiplayer is a expansion of GTA Online (the airport will be used to travel between cities).

    1. Lino says:

      After Andromeda, I don’t think we’ll ever see another Mass Effect game – it was just got so much hate, and couldn’t revitalize the franchise. As for Elder Scrolls, I’ve never really liked the series, Skyrim included. I hope VI comes out well, but given Bethesda’s track record, I’d be very surprised if it isn’t a buggy mess that the modders will have to fix :(

      1. John says:

        Oh, we’ll get a another Mass Effect game. Eventually, maybe a decade from now, whoever has the rights to the IP is going to realize that there’s a lot of Mass Effect nostalgia out there, that something with the Mass Effect name on it is bound to sell a certain number of copies, and that not making another Mass Effect game would be leaving money on the table. There are absolutely no guarantees, however, that the game will be anything like Mass Effect–or Mass Effect 2, Mass Effect 3, or even Andromeda. Or even in the same genre. Or good. Such is the history of older IPs.

        1. BlueHorus says:

          Mass Effect Podracing! ;D

          Or Mass Effect: the Dating Sim? Strip out all that ‘RPG’ nonsense and make a game about ‘romancing’ weird aliens. Naturally, you unlock an awkward underwear-dry-humping scene at the end. Collect them all!

        2. Ninety-Three says:

          At the very least we’ll get a The Bureau: Mass Effect Declassified where a studio is working on a game about shooting aliens and one of the legal team wanders in and says “Hey, we own this IP about shooting aliens, wanna paint it over your game?”

          This is the actual story of how that supremely mediocre X-COM FPS got made.

          1. John says:

            Exactly. See also the Syndicate FPS. Or the Dungeon Keeper mobile game.

            1. Chad Miller says:

              Or the Shadowrun FPS.

    2. Chad Miller says:

      > Bethesda Game Studos (BGS) are one of the few “good” Single-Player RPG studios around.

      I kind of agree, and I really wish I didn’t, because this really sums up why I’m still upset about Fallout 4.

      I don’t care for any of The Elder Scrolls games, never have really, and Fallout 4 wasn’t just a shark jump for an individual series but honestly made me fear for the health of the entire genre. New Vegas is far from perfect but still undefeated at what it does simply because no one else seems to want to do it.

      The good news is that Fallout 76 doesn’t bother me despite being part of my favorite video game franchise, but the bad news is that this is because Fallout 4 dashed my hopes so badly that I was already past caring. I have a hard time seeing the series ever come back from that, although maybe Fallout’s reanimated corpse can live on as “Skyrim with guns” if they ever get their act together.

      1. baud says:

        There’s the Outer World game from Obsidian, who did the New Vegas game. Perhaps it will be good. Or at least it will be more of a single-player Skyrim with guns.

    3. Galad says:

      As usual, indies are the answer to your woes, man.

      Not sure *which* indie games fit the bill for Mass Effect 1 lookalikes, but I’m sure there must be some!

    4. Nimrandir says:

      An Todd Howard, in my eyes there is only one way he can handle this issue, imagine a future gameshow. Todd is announced and as he comes out there is a lukewarm reception a few people even booing.

      It took me a minute to realize you weren’t talking about something like Double Dare, and people were going to get a chance to hit him in the face with a pie or spray him down with seltzer.

    5. Sleeping Dragon says:

      Or, you know, they can do the usual thing and either pretend FO76 doesn’t exist whenever they’re not talking specifically about it, or pretend that it’s actually an awesome game that is doing awesome, you know put on a Stepford smile and just go about how amazing the next update/feature/expansion is going to be as if nothing bad ever happened.

      1. I’m gonna guess that if Fallout 76 was a huge success that they’d make Fallout 89, and Fallout 12 and so on. Each set in a different location. But I think they nuked their own chances of doing that.

  10. Geebs says:

    Windows 10 seems able to install on pretty much anything. I just recently put it on a 2008 Mac Pro without any problems more challenging than burning a DVD, despite this being a computer that’s been EOL’d by its own manufacturer, won’t run the most recent version of its own operating system, has a couple of Xeons mounted on some sort of custom Intel motherboard, and doesn’t have any official Windows 10 hardware drivers.

    It actually ran some 2018 games fine, including VR stuff.

  11. Ninety-Three says:

    On review scores, I will stand by the soft corruption of “Outlets want to be known for giving high scores to AAA games so that they look like good people to give preview copies to”, but it doesn’t have to operate on the level of individual reviewers. Rather than Greg Tito booting up DA2 and thinking “Okay, better give this a high score or it’ll hurt The Escapist’s long-term profitability”, I think The Escapist looked at Greg Tito’s interview and said to themselves “Hm, this guy doesn’t seem too picky, easily satisfied, probably the sort to hand out a lot of perfect scores. Exactly what we need!” The reviewers are honest, just not representative. Heck, even Escapist management don’t need to be consciously choosing Tito-like people: natural selection means that randomly, some review sites will end up with a lot of Titos, some won’t, and the ones that don’t lose out on previews, end up more likely to go bankrupt and leave us with only the high-scorers.

    1. Ninety-Three says:

      Oh, and since I’m talking about review incentives, I should mention that the audience is to blame too.

      Remember those stories when the next installment in some popular franchise comes out and before the game is even released, the internet is furious at a review site for giving GTA or Uncharted or whatever only a 9/10? No one wants to piss off their audience, and while Greg Tito proves you can be too liberal with the perfect scores, in general aiming high is the safe strategy.

    2. Lino says:

      “Outlets want to be known for giving high scores to AAA games so that they look like good people to give preview copies to”

      A couple of years ago there was a leak about EA blacklisting reviewers who hated on them, and these reviewers don’t receive review copies of their games. I think this is a video where Jim Sterling talks about it. Or it could be this one – I don’t know, we don’t have YouTube in the office.

      1. Ninety-Three says:

        Oh yeah, I should have been clear: it is established fact that the industry at least sometimes cares about “how likely do they seem to give us a 10/10” and assigns preview copies accordingly, the only speculation is whether review sites respond to incentives (hint: everyone responds to incentives).

    3. Asdasd says:

      Soft corruption is a good term for it. Don’t forget that, for the majority of its existence, the games media has been in large part funded by ad revenue. Who controls the ad spending? Er, the very publishers whose games the media reviews. Unfortunately such is the culture war that if you point out the obvious potential for a conflict of interest there you’ll almost certainly be dismissed as a political radical and an antisocial malcontent.

  12. Ninety-Three says:

    DA2 wasn’t Tito’s first RPG, he had in fact played DA1. I went and dug up some of Greg Tito’s commentary on his review:

    The job of a reviewer is to weigh everything, and I judged the aspects people complain about in DA2 were outweighed. Heavily.

    DA2 was the first CRPG to me which felt like a tabletop roleplaying experience. You played as Hawke, not a dork on a couch.

    Apparently he just really liked the storytelling in DA2. At this point I feel like Shamus’ comment on David Cage fans: I’m not judging him (I’m judging him a bit).

    1. Sleeping Dragon says:

      I can sort of see where he is coming from though because I think that comment resonates with the things that I liked about DA2. I actually liked that the game was happening in one city over a period of years because it made the world feel more persistent rather than “here’s another set piece” or “we did the forest, the ruins and the town, time to go to the winter area now, we just need a volcano and we’ve got all the boxes checked” attitude. In a similar way I liked how party members found their place within the city and it felt like they were actually living in it. In a somewhat similar fashion the dialogues were heavily focused on establishing Hawke’s personality, out of the three Dragon Age protagonists I remember Hawke as the one I would most describe as an actual character I played rather than just a humanoid turret+camera I used to explore the world. Also, I was a mage in cutscenes, something that Mass Effect failed to acknowledge with my biotic Shepard, not a major thing but it helps with immersion.

      It is unfortunate that all that good stuff was buried under the serious issues the game had mostly related to the near-aborted dev cycle. The “same city” setting resulted in too heavy reuse of locations even when it didn’t make sense (either all the caves look the same or everyone is using the same cave for their nefarious needs), the cool “choose your personality” dialogues were covering the fact that you very often did not get to make decisions other than picking your attitude, especially by the end the game it became increasingly apparent that the writers were desperately trying to justify making the player go through what little content there was (you were pro-mage throughout the game? don’t matter, they went crazy with blood magic and you have to fight them anyway).

      1. Ninety-Three says:

        the cool “choose your personality” dialogues were covering the fact that you very often did not get to make decisions other than picking your attitude, especially by the end the game it became increasingly apparent that the writers were desperately trying to justify making the player go through what little content there was

        To be fair, railroading DMs are totally part of the tabletop roleplaying experience.

        1. Sleeping Dragon says:

          Oh sure, and games are even further limited as far as being reactive to player actions, which is why I appreciated the option to at least determine my Hawke’s personality. That said it was very clear when playing DA2 that in this specific case they did not make enough content to warrant too many choices.

  13. default_ex says:

    Those old computers where you can’t boot from a USB have a silly trick to them. I was in the camp of using USB to install Windows, Linux or whatever the client wanted for longer than USB booting has been viable. What I used to do for such computers is load GRUB onto a CD so that it would boot from CD but GRUB would then hand off to the USB stick using the UUID of the USB stick instead of the partition address. I believe E2B actually provides an option now to burn such a disk, haven’t had to use that trick in awhile but such a tool like E2B and an accompanying disk would have been miraculous back then.

  14. Liessa says:

    I watched some Anthem impressions videos from my favourite YouTubers. I have to say that the gameplay looks monumentally boring to me, but then I find all looter-shooters extremely boring, so it may well be fine if you’re into that sort of thing. I also think they did themselves a disservice by starting the demo in the middle of the game, since it’s impossible to understand what’s going on with the story without context. At least the graphics are nice, I guess.

    The most annoying thing for me was that the ‘hub’ area looks like by far the most interesting part of the game, but half the content was blocked off for the demo (e.g. NPC dialogue). If I were playing the game, I’d want to spend my time exploring that place rather than zipping around with jumpjets and shooting stuff. It just highlights how far this is from being a traditional Bioware game, and that saddens me.

    1. shoeboxjeddy says:

      The first person view city with random ambient dialogue and horribly slow movement is more interesting to you than jetpacking Iron Man team combat? Yikes, this is not the game for you.

      1. tmtvl says:

        I love Starsiege Tribes, and looking at Anthem makes me wonder how games manage to evolve backwards.

  15. Syal says:

    Plot holes are just one aspect of coherence, and a movie’s coherence isn’t as important as its impact.

    I’m trying to think of the elements of impact now. This is probably not complete.

    Writing quality is a big one. The way the characters speak determines what the audience thinks of them; if the dialogue is straightforward and boring, then generally so are the characters. And witty dialogue and action one-liners are highly memorable.

    I’d recommend Every Frame A Painting for how impactful camerawork can be.

    Obviously acting performance matters. Nothing is going to have any impact if it’s constantly delivered with Ryder’s goofy smile.

    Predictability is a factor in both directions. People want to be surprised, but they want things familiar enough to be able to understand the surprises.

    Tone matters. Action scenes want to be loud and hectic, relationship scenes want to be warm and intimate, standoffs want to be cold and static, and none of the scenes should contradict the baseline tone. Color and placement of lights, actor positioning and camera framing, background noise all play into that.

    …I want to say ‘compactness’. Expressing an idea as efficiently as you can. If someone’s feeling one strong emotion, everything in the scene should lean into that. If they’re feeling several conflicting emotions, various parts of the shot should be expressing each of them. (That one’s kind of a toss-up; a movie like Slingblade is very muted and has minutes-long scenes from one camera angle where the actors barely move. Is that compactness? Don’t know.)

    Background objects kind of go with compactness; you can tell a story about a room with the placement of the objects in it, and it gives the viewer something to look at if they’re bored of the actor.

    Pacing is important. Every scene is.a story in itself, not just the action scenes You want some breathing room scenes, but they should still have their own minor arc. The fewer scenes feel like they’re just filler before the next major scene, the stronger the movie will be.

    Art is subjective, there’s no good objective measurement for whether one movie is better than another. But even if I can’t say whether Alien is better than The Thing, I’d watch either of them before I watched a two hour stillshot of someone sleeping on a bed, even though the sleeper has no plotholes whatsoever.

  16. shoeboxjeddy says:

    I had a real bad time with the Anthem demo. Anything that could go wrong, seemed to go wrong. The demo claimed I wasn’t in the VIP group, despite me launching it through the code to be a VIP. The mid-mission loads are obnoxiously long and you get yanked into them without warning. And sometimes the game soft crashes and doesn’t finish the super long loads ever! When I alt tabbed out and killed the process, I had real difficulty starting it up again. I had to kill Origin and restart it to get any consistent start ups going.

    The PC flight controls are inexplicably bad. Like “how did they think this was acceptable??” bad. They claim to have fixed this in the full version, but… great demo that plays differently (worse differently) than the actual game. The controller was better, but the default isn’t flight controls inverted? Who would EVER play that way? The missions as given are a bit tough considering we didn’t go through a tutorial or learn our powers properly. I kept getting stun locked and the starting weapon felt terrible to fire. I would have tried a bunch of weapons, but the loads made me not want to bother. Enemies seemed to spawn without any reason or logic to it, and I was killed a bunch cowering in cover after a single hit destroyed my shields and overheated my jetpack engine. If you die, there’s no respawn timer, you just have to sit and wait until someone revives you or the whole team dies. If you die solo, there’s no recourse but a checkpoint and a LONG LOAD. The end cutscene did not play for me, it just froze up.

    So uhh… I will not be getting this at launch, if ever. Yikes.

  17. tmtvl says:

    I think DD needs his own blog, that’s a pretty massive wall of text for an e-mail to a podcast.

    1. Paul Spooner says:

      Visit the forums. Stand agape in wonderment at multi-thousand-word posts by “Lachlan the Sane” and “The Rocketeer”

      1. Sleeping Dragon says:

        Hey, I’m sometimes in there too, though I try to hold back.

  18. Steve C says:

    Regarding game scores– I convert a score like 7/10 to a letter grade. You’d flunk out of post-secondary education with a 6/10 (C). 7/10 (B) squeaks by if it is your minor and fails if it is your major. 8/10 (A) or higher is necessary is necessary to feel the work you are doing is good and not in danger of being kicked out.

    1. Matthew Downie says:

      My letter / number conversion rate for reviews is:
      10/10 = A
      9/10 = A-
      8/10 = B+
      7/10 = B
      6/10 = B-
      5/10 = C+
      4/10 = C
      3/10 = C-
      2/10 = D+
      1/10 = D
      0/10 = F
      Anyone know what Metacritic uses?

      1. Where’s “A+” ?

        Also, the + and -, are they full grades or half grades? And if they are half grades wouldn’t that mean a C+ and B- overlap for example?

        Also note that I’ve rarely seen anyone give a 0 (be it on a 1-10 scale, or 1-5 scale or 1-5 stars or 1-6 dice eyes). Most rating systems do not let you choose 0 stars for example.

        I once suggested to Jim Sterling that he should add a 0 to his scale for games that do not run at all, but that was a long time ago. I can’t recall if he gave a score of 1 to games that fail to run.

        Another issue is that a score reflect the copy of the game at the time of the review. If I was a developer I’d rather have reviewers review the finished (as the customers would get it) game.
        I might want to give reviewers a alpha/beta that they can play and give feedback on, but won’t be allowed to base their review on.
        This does mean that the game would be sitting finished for a while, waiting for reviews to roll out, but I’m sure reviewers stumble on issues that you’d have to solve and fixes would be made and included before release day (avoiding or making the day 1 patch smaller).

      2. Ninety-Three says:

        That’s an awfully forgiving scale near the bottom. My idea of what most reviewers are saying with numbers looks like this:

        6/10: D
        5/10: D-
        4/10: F
        3/10: FF
        2/10: Super F
        1/10: Super F Second Grade
        0/10: G

      3. Steve C says:

        I think that misses my point that anything lower than 70% is unacceptable in the world of academia. It is technically not a “fail” but a GPA that bad will get you kicked out regardless. It converts to a range of 77% or higher. With 70-77% being a bit of wiggle room if it isn’t a core competency.

  19. baud says:

    Yeah for the return of Chainmail bikini! Thank you Shamus, I had never read them, since I didn’t knew about the blog then. Also shame on me for not thinking about the internet archive.

  20. RFS-81 says:

    Paul, were your kids sad when they found out that the cake is a lie?

    1. Paul Spooner says:

      We didn’t get past chamber 10, so there have been no promises of cake yet, nor gainsaying graffiti.

  21. Marr says:

    Oh Gods, online games that run everything client side with no sanity checking are a scourge of the hobby. I was really into Worlds Adrift last year until it turned out they’d done the exact same thing as Bethesda (Apparently this revolutionary new SpatialOS engine doesn’t solve the age old problem) and their ‘solution’ was to mandate installation of an anti-cheat rootkit Windows service with higher admin rights on my own computer than me. I only have the one serious computer at home, and it hosts my work and personal finances as well as games, so that’s a big old Hell No.

    Somehow you can’t convince any game’s fanbase that this sort of thing is a problem. The reaction is always ‘nothing to hide, nothing to fear’.

  22. Smejki says:

    I think one of the “problems” with review scores is the fact that games operate on more axes of quality than, say, film. The scores 0-5 for games are largely reserved for a wide variety technical failures. There’s really no such thing in movies. If the craft behind the movies renders the whole thing unwatchable it gets 0-2, while the marks 3-10 are mostly reserved for artistic merits only with minor influence if the craft mastery. That said there’s more than enough competently made games and we only pay attention to those.

  23. In case anyone would like to watch, I’ve got Anthem and I’m going to be streaming it on my Twitch channel: https://www.twitch.tv/psyychoblonde

    1. Hmm, this at least makes it SOUND like a couple of bugs with how things were loading were causing MOST of the problems with the servers in the demo http://blog.bioware.com/2019/01/26/anthem-vip-demo-update/?utm_campaign=anth_hd_ww_ic_soco_twt_anthem-day1-bioware&utm_source=twitter&utm_medium=social&cid=55870&ts=1548543884671

Thanks for joining the discussion. Be nice, don't post angry, and enjoy yourself. This is supposed to be fun. Your email address will not be published. Required fields are marked*

You can enclose spoilers in <strike> tags like so:
<strike>Darth Vader is Luke's father!</strike>

You can make things italics like this:
Can you imagine having Darth Vader as your <i>father</i>?

You can make things bold like this:
I'm <b>very</b> glad Darth Vader isn't my father.

You can make links like this:
I'm reading about <a href="http://en.wikipedia.org/wiki/Darth_Vader">Darth Vader</a> on Wikipedia!

You can quote someone like this:
Darth Vader said <blockquote>Luke, I am your father.</blockquote>

Leave a Reply

Your email address will not be published. Required fields are marked *