T w e n t y S i d e d
We got so many questions this week that we had to dedicate the entire episode to them. Well, we spent a few minutes of the latest Satisfactory update. But other than that, this episode is all mailbag. I hope you’re happy, The Internet.
Also, this episode is also available on YouTube, if that’s your jam.
Hosts: Paul, Shamus. Episode edited by Issac.
Diecast290
Link (YouTube) |
Show notes:
00:00 This show might be on YouTube?
01:33 Satisfactory Update 3
Link (YouTube) |
11:49 Mailbag: Tomb Raider
Dear diecast (although its mostly for Shamus, sorry Paul)
Since you’ve been hurting for mailbag questions, I’ll try to give you some. I recently sank my teeth in the tomb raider rereboot (2013, rise and shadow of the tomb raider). I remember Shamus being really positive about tomb raider 2013 but less so about rise of the tomb raider, and disliking shadow (if i interpreted his blog correctly). My question is: does Shamus dislike the changes the of the latest TR games, or did the rebooted TR series not sit well over time? What i mean is, would Shamus still be blown away by TR 2013 if it came out today, or was it just impressive in 2013. Should Shamus still enjoy enjoy replaying TR 2013 now?
With kind regards,
Chiris
17:36 Mailbag: Overpowered in a Good Way
Dear Diecast,
You recently praised Risk of Rain 2 for the fact that it allows the player to stack up bonuses without limit and become totally overpowered in a way most games don’t permit. Usually becoming overpowered is boring for the same reason that turning on god mode is boring, but every now and then I find a game like Noita or Slay The Spire where becoming way more powerful than the enemies somehow makes it fun instead of tedious to spend fifteen minutes effortlessly stomping everything. What do you think is the magic spark that makes this difference?
Ninety-Three
25:51 Mailbag: The Forums
Dear Diecast Shamus, (sorry Paul!)
Any update on whether the forums are coming back? If they aren’t coming back, is there any chance of making a data dump of the archives available? I’ve had times where I’ve wanted to refer back to some previous forum discussions since they’ve been down.
—
Relatedly, are there any changes that could be make to the blog to facilitate longer-term discussion? It seems blog discussions currently have a very short shelf-life: once a post is no longer the top post discussion mostly dies.
And when revisiting a comment section, it’s hard to find the new comments posted since I last visited; and to see if anyone has replied to anything I’ve said.
I know this probably means a terrible trip to the wonderful world of WordPress plugins, but if any of these could be addressed – email notifications, or a quick way of seeing new comments – I think it’d help the blog as a platform for discussion.
– Retsam
26:50 Mailbag: Geforce Now
Dear Diecast,
I’ve been watching release of Geforce Now with interest but it seems that the service is faltering somwhat… given that publishers are removing or withholding their games from the service. However, I’m a bit confused as to why this is an issue or why there are even “compatible” games for this service.
Geforce Now appears to effectively be a virtualised OS instance. I’ve checked several EULAs from various publishers and none of them prohibit the use of games on a virtual machine. They prohibit third party software that alters the game experience (mods/hacks/bots), reselling the software, etc. So, on what basis is this “removal” even taking place under? Why does NVidia even need to obtain permission from the publishers to run the games?
Maybe you have some insight into this or maybe it’ll spark a debate in the comments.
All the best,
Duoae
33:56 Mailbag: Heresy!
Dear Diecast,
Not too long ago, I encountered this video. It makes the argument that Half-Life 2 is a bad sequel (though not necessarily a bad game). Now I’ve never properly played any of the Half-Life games (my attempt to play the series began and ended with 1, where my awful sense of direction served me very poorly), so I can’t make a properly informed assessment of the argument, but something struck me about it.One of the points he makes is that HL2 throws away most of the interesting and unique story elements of 1, in favor of a safer and more cliched story. In particular, the game discards the original’s sequel hooks by introducing a new antagonist faction which comes out of nowhere, takes over the world, and is suddenly the most important thing ever. This sounds very familiar.
So I wanted to see what you’d think of it. Given your fondness for the game, what’s your reaction to seeing someone make criticisms of it so similar to the ones you leveled at Mass Effect 2? Do you think they’re reasonable arguments, or do you think his analysis of the material is faulty?– Kestrellius
P.S.: I don’t know if you’ll like the first video I linked by that guy, but you’ll definitely like this one.
40:03 Mailbag: Fake Rivers in Cities: Skylines
sup Diecastermans:
I was recently disappointed by Cities: Skylines. Well, maybe by one of their level designers. I love the game, and I recently started my yearly-ish city building project, selecting a map that had a lovely canyon in which I hoped to build a hydroelectric dam. This dam wasn’t going to win any awards for highest dam in the game, but it was good enough. I built up my population, saved my… sky-moleons?… and built the dam. It wasn’t the best dam, but it was ours.
That, and it was going to double my electricity production.
But it was all a lie.
Turns out, the lovely canyon was just a sculpture. The riverbed at its bottom was nothing but a carved sluice. As soon as the river broke its bank, filling stopped for good. I even tried to augment the flow of the river by diverting my (mostly-filtered) sewage into it.
No good. In fact, the flow arrows showed (and you can see from the… smear…) that the fluid I was pumping into the basin was actually flowing up river, presumably draining off the edge of the map after the Fog claims our vision.
I’ve always enjoyed the way Cities: Skylines simulates “realistic” water flow and lets you build dams based on “actual” hydrostatic pressure, but the map design on this one foiled the system and left me sad. I wound up blowing up the dam and retiring the city shortly thereafter. What games or circumstances have left you similarly disappointed?
Kind regards,
toadicus
47:14 Mailbag: Bioware’s fate
Dear Diecast,
There was a recent interview with (now) ex-bioware writer Drew Karpyshyn where he lamented that the studio became more corporate as it got larger, which affected the game design.
How does this fit into your thoughts on their games and acquisition by EA?
All the best,
Duoae
57:47 Mailbag: Machine Learning and Game Install Sizes
Dear Diecast,
In the past both of you have spoken about ballooning game install sizes and procedurally generated content. Now it seems like there’s more tech on the horizon for reducing not only install size but also development work.
On the one hand, a team at microsoft are working with using machine learning to up-res textures on-the-fly. On the other, Dreams (from Media Molecule) uses some sort of real-time renderer that utilises the async compute abilities of the APU in the playstations. Both of these techniques mean small install sizes and interesting optimisations, and could be an interesting addition to procedural generation.
Further to this we already have the nascent ray-tracing technology…. i think the next few years will be really interesting.
What do you think?
All the best,
Duoae
Link (YouTube) |
Dead Island
A stream-of-gameplay review of Dead Island. This game is a cavalcade of bugs and bad design choices.
Marvel's Civil War
Team Cap or Team Iron Man? More importantly, what basis would you use for making that decision?
Was it a Hack?
A big chunk of the internet went down in October of 2016. What happened? Was it a hack?
Trusting the System
How do you know the rules of the game are what the game claims? More importantly, how do the DEVELOPERS know?
Charging More for a Worse Product
No, game prices don't "need" to go up. That's not how supply and demand works. Instead, the publishers need to be smarter about where they spend their money.
On EA and BioWare: You touch on this in your discussion, but I don’t think the issue is “EA doesn’t understand game development” so much as “Everyone conflates a bunch of different things as game development.” The analogy I like to use is: Imagine a corporation that manages casino properties decides to go into book publishing and acquires an indie publishing house. I imagine the shareholders would be scratching their heads and asking questions about “core competencies”, but when both the casinos and the books are technically video games, nobody raises an eyebrow. Maybe it’s most accurate to say that the money folks are the ones who don’t understand game development.
There’s definitely a messed up problem with Bioware if the crunch is so bad that they apparently hold therapy sessions for their employees
That’s true insofar as the money folks keep funding Bioware’s boondoggles, but Bioware decided all on their own to build Anthem: someone in Bioware leadership has the same confusion about core competencies.
Amen.
The idea of Abthem wasn’t bad. Actually, it sounds as though several of the ideas behind it were really good. The problem was indecisive leadership. That’s not something that excludes some Bioware people being at fault, yet…
Yet…
Choosing, directing, supporting, and if necessary removing leaders is exactly what EA does. The core of EA exists to do that. The Finance, Accounting, HR, Marketing and other departments only exist to support the core function of creating winning team to sell millions of copies of games. EA decides on a lot of things to do that, of course. Real life is never that straightforward. But that is basically the heart of the company.
Also, realistically there is no “Bioware”. Ita just a label EA can use. I don’t mean that as an insult or out of nostalgia but because that’s literally the facts.
The thing about Anthem is that it wasn’t automatically a bad idea of a game to develop. Because over its LONG gestation period, it was like 10 different games. Some of those 10 could have been an extremely good fit for Bioware!
Disagree: Per Jason Schreier, even before they had any details down, Anthem was always an online multiplayer action game with “roleplaying” and “narrative” not mentioned in the design doc. Bioware’s action gameplay has always been mediocre compared to the studios that make real action games, their only* multiplayer experience is the tacked-on wave shooter mode of ME3, and their only experience with persistent online games is their WoW clone The Old Republic. Someone should have noticed from the start that Anthem discarded the one thing Bioware was known for in order to be three things they were either bad or inexperienced at.
I was going to make this reply to Hector above, but I didn’t want to deal with the bullshit asymmetry principle in responding to claims about “literally the facts” backed by zero facts, so thanks for giving me somewhere to say this.
*Technically there were also tacked-on multiplayer modes in DA3 and Andromeda, but those were so unremarkable I’m betting you only remembered them because of this asterisk.
Here’s the thing… every modern Bioware game has multiplayer. Saying that, “They tacked on a multiplayer mode to… every game from the current generation,” is a silly argument. They ARE a multiplayer focused studio at this point. They had an extremely popular microtransaction based horde mode with ME3, the sequel to that mode (that was bad because the game was bad) in Andromeda, an attempt to capture that idea for a different setting in Inquisition, and an MMO. So no shit they had multiplayer in the plan for Anthem. They have multiplayer in ALL of their games since 2011! It’s kind of like you’re saying, “Nintendo doesn’t DO online multiplayer. They’re just not interested in that. Okay, except in Mario Kart. And Smash Bros. And Arms. And Splatoon. And Mario Maker. And Animal Crossing. And Luigi’s Mansion. And all their mobile titles. And…”
Further, some of the ideas from proto-Anthem sounded interesting. A harsh conditions survival game where you had to explore and settle a harsh alien world that was more about traversal rather than third person shooting could have been VERY good, if they had gotten those mechanics right.
So, muscle memory. I’m a left-hander, but I originally learned to use the mouse in my right hand. After several years, I developed an ache in my arm, so I started moving it to the left. Nowdays I’m pretty ambidexterous, depending on what I’m doing. I’d never play a FPS with the mouse on the left. Doesn’t take any effort. Sure, that was 20 years ago, in my late teens. But if you want to relearn the inverted mouse, I believe you can.
The Tomb Raider talk reminds me of an observation about the Craig James Bond movies. He spends three movies becoming the Bond we all know from previously, and then suddenly he’s over it and washed up. Where was the Bond we were promised? Why can’t we just have the female Indiana Jones? OTOH, there have been three Tomb Raider movies and zero Uncharted movies.
The dam in the screenshots looks about as pretty as they ever get. Now, I don’t know much about dams, but is there a reason for the vertical lines? Maybe something could be done with them, some kind of art deco.
Your discussion of short-term hardship for long-term gain sounds like the way I play Torchlight 2. When I find a good gold weapon, I drop it in my shared stash. Next character I create has a build just for that weapon. So I tend to be a bit glass cannony, unless I’m dong a weapon and shield build. Wand and shield is probably the best, because the wand often generates additional effects to straight out damage.
Last year I taught myself touch-typing on Dvorak layout despite using azerty all my life. It was surprisingly easy. I figure switching from inverted control to regular shouldn’t be too hard. But I support proper options in games, so I feel like Shamus should just be able to play with inverted controls.
“can” and “want to” are different things tho
I just realized I made a mess of things, to clarify: I want developers to put in the 10 minutes work to allow Shamus to play his preferred way.
My guess is the vertical lines on the dam, are what you get if an artist that doesn’t understand[1] dams tries to draw a buttress-dam. Those lines should be redirecting force from the water pressing against the dam, down to the ground that’s holding the dam up. Like, they’re the less-materials way of having a dam that’s got all parts fattened down to the base, instead of a few sections at intervals.
[1] I’m not an engineer, but the dam in the picture seems to be pretty dang skinny compared to real dams, from Google image search, and Wikipedia.
Thanks, I see it now.
Re: Noita @ 17:58:
Finnish guy here, your first two pronunciations were good. Perfect pronunciations for comparison (also, don’t look at the translations on that site, go to wiktionary instead).
Re: Discord, what’s involved in taking care of it is just “How much moderation do you want to do?” They take a couple minutes and zero dollars to spin up, and spambots are not a big problem (plus there are some friendly anti-spam bots you can bring into the server if you’re really worried).
I’d be in favour of a Twenty-sided Discord.
I’ve run a couple Discord servers over the years (one massive one for a game, one small one for D&D) and I’ve found that one of the best things you can do for your server during setup is to have a “Reception” channel.
The first text channel that exists when you create the server cannot be deleted and everyone has read access to it. Use this to post the server rules, even if those rules are as simple as “don’t be an asshole”. Limit “Send Messages” permissions to the server’s admins.
Then I add one more channel with “everyone” and “admins/moderators” but prohibit “members” from being there. This is the Reception channel. Newcomers to the server are isolated here until such time as a moderator can verify that (A) this is a human and (B) they agree to follow rules. That’s it; just a short “Hello” and they’re in. This Reception channel serves a couple purposes. First, any spambots or jackoffs trying to raid your server are isolated and can only annoy the people with the power to ban them; second, it’s easy to identify when someone new has joined and say hi first.
Beyond those initial setup tricks, I’ve found that good server design is like good sewer design: Keep the number of channels low enough that there’s some flow in each of them – you don’t want too many that they get stagnant, nor too few that the flow goes too fast. Every once in a while you might have to wade into the stuff and unstick a real problem but your filters on the front end should prevent those from getting in to begin with.
– CMDR GoStu, retired Server Tyrant.
That’s really clever, but you’d need to have moderators on 24/7. You wouldn’t want to leave newbies effectively shadow-banned for an extended period of time. Also, can they read the other channels before they’re verified? If I don’t see anything, I wouldn’t bother saying hello.
I am active on several Discords that do this, and it’s actually fine to have newbies shadow-banned for an extended period of time. You just make sure the landing page rules tell people to pipe up for access.
Also, the default landing channel automatically posts a message about a new user joining.
Shamus – I am a moderator on a 100+ member fan Discord for a woman streamer. Feel free to hit me up if you have any questions; hopefully you still have my socials (or worst-case you have my email.)
You could also handle the “onboarding” via a bot. Discord bots are pretty common in the servers I frequent. Might be able to find one “off-the-shelf” that handles the onboarding, but otherwise it wouldn’t be terribly hard for someone to write one. (I could probably spin one up if we had a list of features we wanted)
I made an edit and my comment was marked as spam so re-posting here:
Wow, didn’t expect you to address all three questions in one mailbag. Especially since I sent in the final question so late. I just sent them as they came to me/my attention.
The thing with the up-res-ing of textures appears to be that you can have the artists generate The Elder Scrolls: Morrowind level of textures which can then be upscaled to 4K. i.e. the time and effort to create 4K art is higher compared to mid-2000s level of art requirements. It also means you can reduce the art pipeline requirement across an entire game – no more 2K textures on models and most scenery, only for the player to encounter a 720p texture on a rock because the artists didn’t have the time/money/reason to spend on that particular texture.
Like you point out in the podcast, you also save on download and, potentially, HDD space. This might not be so important in some places (like major metropolitan areas) but the download quantity is actually quite important in terms of a) personal data caps, b) energy spent storing and transmitting the data from the datacentre.
In terms of a), I think the benefits to the users (all users) are immediate – you get less congestion, smaller data usage and monumentally faster downloads (depending on the amount of reduction in total file size you can manage). In terms of energy usage and the environment, you have fewer demands on server storage (across all CDNs) which reduces no only components and their demands on construction materials (such as rare earth elements) but also datacentre cooling and other management features and this smaller size also costs less energy to transmit the data to the end-user (which can be quite substantial – for example, 1 hr of DVD quality viewing of a stream is [on average in the USA] 0.2 kg CO2 and this cost rises with increasing transmission speeds and, of course, quality levels*).
In terms of a slight delay when installing a game on a client device? I think there’s very little difference between the current installation process (you tend to pre-load games on Steam days in advance and when you don’t the download will take mostly a decent chunk of time and then you still have to go through the process of “installing” the game and associated libraries) and a potential “unpacking” you both thought of in the podcast. Especially since it could be parallelised.
Even better, if there is minimal cost to CPU/GPU operation in a game (i.e. the number of threads of a CPU is vastly outstripping game programmers’ ability to harness them outside of a few edge cases) then there’s little reason why the textures can’t be loaded into RAM (how many games go past 8GB VRAM and then have nothing in the 16+ GB of the “next gen” baseline of system memory?
Essentially, this idea is the whole concept of DLSS from NVidia but in a more general sense. DLSS down-scales the resolution in terms of graphics processing but then upscales that output (with machine learning, in order to not make the textures look like crap) to “mimick” the original intended resolution in order to provide higher fps. It seems to work quite well there….
Link here.
*You can get a good, if slightly outdated overview here: https://iopscience.iop.org/article/10.1088/1748-9326/9/5/054007
************
In terms of Dreams, the tech is essentially performing ray-tracing (actually path tracing but from what I understand, that’s the way the RTX cards do it as well) on the GCN-based PS4 and PS4 Pro. Now, this is not my area of expertise and the documentation and information is sparse and not well-understood so I may be wrong in what I’m about to say: Not only that but the developers somehow got their voxel “splat” renderer to perform subsurface scattering and other effects. Yes, there is not path-tracing across the whole scene but at the same time, it’s sort of like there’s no models in this game, only carving from the “ether” of the world level. It’s like the blocks of ray-traced minecraft were miniaturised so that there was a high-fidelity look to it and then modifiable filters placed over those surfaces.
In essence, from what I can see, there aren’t any textures in the game at all. Everything we see in the game appears to be produced from shaders and lighting effects. This even applies to the sound engine in the game, though this is publicly less well-explained at this point but it seems to be a sample editor for which you can modulate to “infinite” variations and this same sort of philosophy is applied to the graphical side of things too.
Oh, I also came across a more recent less technical overview of the Dreams engine/gameplay after I sent the original email. Here’s the link.
I think that these technologies, path-tracing, DLSS/machine learning up-scaling and procedural generation could really help with content generation and also with the fidelity of an entire game/world. Sure, you can just add in more HDDs in your computer but that’s the brute force method solution to that problem. You can have a sweat shop of people creating 4k x 4k textures for each element of a game but, again, that’s not a great way of solving the issue. It gets WAY harder to maintain an art style with more people onboard inputting their own take on it and it gets way more expensive in terms of manpower AND it gets way harder (and more expensive) to change direction in development once an art style is set in stone.
Take this another way, the next gen consoles (and current gen) don’t have the opportunity of adding 5 HDDs to the system. Nor to low-to-mid end systems (the PSUs likely aren’t installed for that). If developers are wanting to put out 150-200GB games as standard then that’s a problem and it’s a problem all the way up the supply chain from the consumer to the developer.
Also, I forgot to mention that (as per all machine learning) the algorithm is specific to each game. Hence, there’s no “the game did this to my texture when I wanted *this* result”, it’s part of the development process and you can’t just swap the algorithm between games. DLSS works in a similar but maybe less general manner.
OMG, it’s Captain Vowel! I’m your biggest fan! Can I get your autograph?
That’s Dr. Vowel, to you, sir! And i could never countenance capitulating to such a cunning convicted consonant such as yourself!
;)
You’d probably get even better compression if you used machine-learning to compress your textures into parameters for procedural texture generation routines that run really fast on GPUs[1]. The demoscene has historically (and continues to have) games, short videos, etc, that fit into kB, what would take MB in JPEGs or other normal compression methods for images or videos. Cutting game texture sizes by a quarter is nice. Cutting them by 10 times to 1000 times is much better. :)
[1] That is to say, a lot of computations that are independently computable for each pixel, and don’t rely on surrounding pixels.
Yep! That’s why i mentioned procedural generation as also something that could be used in conjunction with these techniques.
I haven’t played a demoscene game in such a long time… that takes me back to around 2000.
“We’re busy thinking about other things”. This is my thought whenever people want to ridicule customers (a la Clerks) for asking questions when there’s a sign explaining that information. Bonus points if there are a hundred other signs pointing out other details as well. At some point, that information just becomes background noise. *You* think it’s annoying and obvious because you work there, and get asked that question a lot. Most of those customers aren’t going to have the same experience with and exposure to that information as the person who works there.
As far as the HL2 story kind of being a digression from HL1 in the story (“While everyone was dealing with X, Y snuck in and took over the Earth!”), while I definitely can see the point, I think that ship already sailed with HL1’s Opposing Force and the Race X?
Besides, weren’t the original aliens defeated in HL1 when Gordon killed Nihilanth? They need to come up with a new antagonist anyway at that point, or somehow say that the original Xen aliens suddenly had a new leader, which isn’t much better from a story perspective.
Opposing force isn’t canon in HL2, although that OF introduced really became a problem for the HL2 story.
As for the original aliens, it was too vague to know what happened. We know the vorginauts were enslaved by them and Nihilanth kept the portal open, and also that Nihilanth was scared of some other force which Gordon (and the G-man) might be working for. Or at least you defeating the Xen forces allowed this third faction to gain more power. Thats kind of the problem i have with HL2, they just drop a lot of open plot points of HL1 and go on with a new story. They couldve given you a new assignment that gave you more info on who the Gman was, or what his organization was, and the 3rd faction the Nihilanth was warning you of.
It seemed pretty clear that the episodes were leading towards a confrontation with the G-Man, considering he throws Gordon into the conflict at the start of HL2, attempts to abduct him and get Alyx killed at the end of HL2, is interfered with directly by the Vortigaunts in HL2 Ep. 1, and then directly interferes to force the worst possible outcome in HL2 Ep. 2. I didn’t bother to read the leaked details of Episode 3 because the format is too different to provide any satisfaction, but from the summaries I looked at, dealing with the G-Man and whatever faction he was a part of was the point of it all. So I don’t think HL2 really dropped anything. Instead, it dramatically increased the complexity and depth of the scenario.
Not a mechanic, but my big disappointment was in Horizon: Zero Dawn. There’s a quest to help a girl get revenge against a slaver. The slaver was an official acting under orders from the old, evil king, and stopped slaving but kept his position under the new, supposedly benign king. He also then claims it’s mistaken identity and they’re actually looking for someone else, while the nearby official says he’d love to arrest the guy but can’t without something major. The political intricacy of the situation held my interest, and I was trying to figure out how we were going to unravel the whole mess.
Then I remembered the game is a YA, and realized the quest was going to resolve with the bad guy marching out into the middle of nowhere for no reason other than allowing us to kill him without consequence. And he did, and he also gave a big villainous speech about how he was even more evil than people thought.
Most stand-out disappointment of a quest that I can think of.
H:ZD does also have one of the best subversions of YA I’ve ever seen, though. I’m thinking of the part where Aloy gets enrolled in Warrior School and the game sets up all of the most thoroughly played-out YA character and plot tropes in, inevitably, a dorm scene. You’ve got the Best Friend/Competition, the Antagonist Jock, etc. Everybody’s getting ready for the Big Trial which will determine who is the best.
Then they all get wiped out in the very next scene, the school effectively ceases to exit, and Aloy’s on her own againThe game did have a couple solid trope subversions. But I didn’t even notice that one because that Antagonistic Jock was so absurd my brain turned off in self-defense. “Remember when I was a child and I threw a rock at another child? I still brag about that.”
Regarding “Overpowered in a Good Way”:
I think the key is needing to *work* at getting overpowered. In both RoR2 and StS, you start the game as a mere mortal, and need a combination of luck and skill to get to the point you flatten everything. This means that watching everything fall before you feels like the reward for a run well done, rather than an empty cheat code victory. It’s also important that these games aren’t designed to go on for dozens of hours (like Fallout or Skyrim) – you play for an hour or two, get ludicrously overpowered, stomp things for as long as it amuses you (probably less than an hour), and then hang it up.
Or, if the game isn’t over in about 1 or 2 hours, you need extra challanges, where you aren’t overpowered or the power doesn’t matter. Final Fantasy X for example: You laugh at the story Endboss, but ragequitt before Black Anima. Or Civilization: Slaughtering stonemen with tanks doesn’t get you anywhere near building that world wonder.
Another alternative is endless modes. Warframe has made the most of its absurd differences in player damage output by building in modes where enemy health and damage increases exponentially over time (and, for one faction, double-exponentially as both health and damage reduction each scale exponentially). You’re only really expected to play these modes for 20-40 minutes, with silly one shot behavior from bullet sponges dissuading you from pushing past the length of a DotA match. Still, you can always use more power to push further, with no upper limit as eventually you’ll reach a point where you just can’t go on.
However, if you lean into the brokenness and use strategies independent of the raw numbers, people manage ridiculous times for bragging rights (displayed on in-game leaderboards).
Agreed. Furthermore I think in a game like Slay the Spire the gameplay loop itself is largely ABOUT becoming more powerful. It’s harder for it to spoil the gameplay when it is the gameplay. If you manage to reach the point of becoming overpowered and stomping things then in a way it is just an extended and more fun victory screen/state.
EA’s full, original name is Electronic Artists for a reason. EA used to love developers. The Digital Antiquarian has an excellent article about it here. Back at the dawn of time, when the company was new, it ran ads asking “Can a computer make you cry?” and tried to portray developers like Dan (later Dani) Bunten and Bill Budge as both auteurs and rock stars. This is both less and more ambitious than it sounds. One the one hand, development teams were so small in those days that most developers were practically auteurs by default. On the other, neither the guys nor the one gal in the “Can a computer make you cry” ad look like credible rock stars to me. The strategy obviously didn’t work. EA pivoted from PC to console games and gradually and ultimately became everyone’s favorite monstrosity unless it’s Activision’s turn again this week.
And now for the portion of the comments section where I shout into the wind.
Yes, EA is awful. Yes, EA is bad at stuff. But I will never accept that everything is always EA’s sole fault. Developers are people, not blameless, innocent automatons. They make decisions. They have agency. Mass Effect may have been released after EA acquired Bioware, but it is very much represents the continuation of trends that were evident well before the acquisition. The studio went from isometric RPGs with painted backgrounds (the Baldur’s Gate series) to a tile-based 3D RPG (Neverwinter Nights) to a third-person RPG full of unique environments (Knights of the Old Republic) to a third-person action-RPG-ish thing full of unique environments (Jade Empire) to Mass Effect. Over time, Bioware’s presentation became more and more cinematic and the size of their teams and the costs of their games increased. In other words, Bioware was getting bigger and “more corporate” before the EA acquisition. I’d argue that Bioware needed to be acquired not only because the founders wanted to cash out but also because Bioware needed publisher money and backing to realize its ever-expanding ambitions.
As an aside, I also want to push back against the notion that developers are somehow only influenced by publishers when they are wholly owned by publishers. Not so. Bioware didn’t self-publish any of its games. As an independent studio, Bioware could in principle walk away from a publisher that it didn’t like, but it couldn’t walk away from all publishers. For non work-for-hire games–Sonic Chronicles: The Dark Brotherhood, anyone?–Bioware either had to find a publisher that liked what Bioware wanted to do or else choose to do what some publisher was willing to fund. The difference between a nominally independent studio and a publisher-owned studio can be less significant than it appears.
Finally, while I accept Shamus’ argument that a publisher is ultimately responsible for what goes on at any of its wholly owned subsidiary development studios in the same way that Captain Kirk was ultimately responsible for the actions of his idiot red-shirted away team members, I’d just like to point out that Captain Kirk was stuck with some real goobers. Yes, it was ultimately Captain Kirk’s responsibility that Ensign Crewcut Squarejaw was eaten by a space-monster, but Ensign Squarejaw really should have known better than to get distracted by mysterious space-rocks, you know? Ensign Squarejaw was a goober and he’d still have been a goober even if he hadn’t been under Kirk’s command.
My first memory of Electronic Arts was receiving a copy of the Bard’s Tale II for Christmas in 1989, which I was mostly happy with. They seem to have dwindled in popularity since then….
My first memory involving Electronic Arts was Jungle Strike on the Sega Megadrive. Such great memories of that game and I watched the publisher/developer splash screens so many times I can almost run through it in my mind*.
*
I was tempted to put a rickroll in that link :DI mean, that depends on how far up the business hierarchy you are, as in any corporate structure. However, if a large company has a history of mismanaged game development, or mediocre games, the blame lies with the people who have the power to hire skilled, knowledgable people, and give them a sound direction to point themselves in.
I hear you. I agree with you. I even said as much . . . right before I launched into an extended Star Trek simile, so I don’t blame you if you missed that part.
The thing that drives me crazy is that the EA narrative used to be that “EA meddling and interference ruins great game development studios!” There’s this attitude among fans that if only EA hadn’t acquired their favorite old studio, whatever studio that may have been, then the studio would have carried on with business as usual and everything would have been great. (I have my doubts, but nevermind that now.) When EA acquires a popular developer, everyone prays that EA will be hands off, that they’ll leave the beloved studio alone. Thanks to Jason Schreier and Kotaku, there is now considerable evidence that this is in fact exactly what happened with Bioware. The fans got just what they wanted, and their beloved studio is still all screwed up. How many times have you heard words to the effect that “Bioware is dead to me” or “the old Bioware is gone” in the comments on this very web page? So now we have the new EA narrative that “lack of EA meddling and interference ruins great game development studios!”
It’s quite a change from the old days, you must admit.
I hadn’t heard at all that they had been entirely hands-off. They weren’t being overly meddlesome, but they were still influencing the company plenty in how Bioware would manage their company going forward. Notably Bioware continued to become more corporate. Not bigger mind, but deliberately corporate. That’s the attitude that dooms the studios in the end. Where they stop focusing on what made them good and start thinking that now that they’re doing well, they need to become like the big boys and so inevitably screw things up. That in part is definitively from EA and how they handle their companies.
I see it more like a push back against the narrative that blames EA for everything. I’ve seen exchanges like
“EA is stupid because they forced Bioware to use the Frostbite engine!”
“Actually, that decision was made by Bioware.”
“Well, EA must have pressured them into it somehow!”
(similar for “making a looter-shooter”, etc.)
I’m not an EA fanboy, the last game of theirs that I bought was Mass Effect 2, but you have to acknowledge that video game studios can make bad decisions all on their own. Of course, it’s also unrealistic to expect a publisher to always save them from themselves.
Imagine if a small, family owned craftsmanship business gets bought out by a bigger shop, with the premise that they will get access to much better tools and resources to make bigger projects. It’s common sense that the big shop doing the buying then needs to… train the small shop to use these better tools and share institutional knowledge about how to manage bigger projects. If they fail to do that, the big shop is very much at fault if the small one burns its operation to the ground desperately struggling to use unfamiliar tools to create a project in a scope they’ve never attempted before. Yes, the small shop started the fire, but they wouldn’t necessarily have if they were given the guidance, support, and training to avoid those mistakes.
(If the analogy seems murky, Bioware was encouraged to use an in-house game engine to save costs and get access to the most modern of technology. It was promised that the creators of the engine would be on hand to help them use it. Then that… didn’t happen at all. Their support tickets were low priority or outright ignored because they weren’t making Madden or FIFA.)
As I understand it EA also went so far as to move Bioware’s Frostbite experts over to FIFA, cannibalizing the talent Bioware already had.
This isn’t really a direct reply to shoeboxjeddy so much as a reply to the whole thread:
This has been an interesting discussion, but I feel like arguments over the validity of EA-bashing have pulled us away from the point I wanted to make in the show. And maybe that’s my fault for the rambling, roundabout way I said it.
At various points in this thread, people have pointed out the terrible decisions made by the leadership of BioWare. I believe John called these people “goobers”.
That’s true, but it’s also true that those goobers were in charge because the original leadership left. Yes, EA leadership took a hands-off approach. That’s fine if you’ve got a strong in-house leadership. But if your original leadership is gone and the publisher isn’t offering any leadership, then you end up with NO leadership. Based on the rumors, this is what happened to Anthem.
Again, this is something EA should know how to deal with. Do they not read gaming news? Did anyone notice the exodus of the profoundly important people that all left the studio in a short time? Does EA understand people and game development well enough to realize that this should be a cause for concern? Did they know to send someone to check up on the studio? Do they have anyone on staff that knows how to do that? Do they know what to do if things are NOT okay?
Movie studios are usually aware that having a director leave a project is really bad. They also seem to know that if the director leaves, you can’t just let the director of photography run the show and expect everything to work out.
EA paid for BioWare, and then they lost the people that made BioWare what it is. They still have the IP, but it only takes a few bad games to turn a top-tier IP into a joke.
Yes, BioWare has made a LOT of mistakes. But as the leader, Andrew Wilson can’t just shrug and say, “It’s not my fault. All the people who work for me are idiots.” Like, making sure you have good people making smart decisions is your entire JOB. If you can’t do it, then who can?
This seems to be an admission that “credit can rise or fall, but blame always goes to the top.” Is there some hypothetical Jason Schreier exposé you could read that would convince you “Oh huh, I guess Andrew Wilson is in the clear on this one”, or is any possible problem at Bioware ultimately going to be EA’s fault for not noticing and correcting the issue?
If Bioware had a strong idea for a game, worked in a reasonable amount of time to create that game, and EA simply published it for them… and then it completely flopped as the market rejected their vision, that would be a sort of classic “Bioware’s fault” type scenario. Bioware floundering through endless rounds of revisions and pre-pre-planning in lieu of releasing a product to market is 100% something EA should be on top of, as a ruthless, cynical producer of consumer goods.
Yes, I’m a big believer in the idea that blame starts at the top. Maybe not for single incidents, but when you’re looking at a pattern of behavior over years, blame has to go to the people running the show.
In the broad sense, EA does so many questionable things that for us to believe that Andrew Wilson is actually a good manager of a game company, we would have to believe there’s a grand conspiracy of many people working to undermine his decisions and feed him bad information.
But in this SPECIFIC case? Yeah, I could imagine a scenario where the downfall of BioWare was a run of really bad luck for him.
Maybe when the Three Doctors (I forget their hard-to-spell names) departed the company, Wilson realized this was a serious problem. He spoke with (or more sensibly, dispatched a trusted expert) to see how the company was doing, who was in charge, and what their plans were. But then the guy in the driver’s seat at BioWare was an incompetent weasel who was really good at putting on a good show and assuaging everyone’s concerns.
This hypothetical Mr. Weasel was a terrible leader, but managed to send reasonable-looking vertical slice demos to corporate every six months. Weasel chose to move to Frostbite without consulting with his team because he wanted to impress his boss, and then he never mentioned any of the problems he was having as the projects matured.
It’s certainly not impossible. Guys like Mr. Weasel aren’t rare or anything. But given how EA seems so clueless about the business they’re in, the most plausible explanation seems to be that this is yet another example of a non-dev, non-gamer trying to run a bunch of companies that develop games.
EDIT: I chose the departure of the Three Doctors as the example scenario, but then I realized maybe they left while Riccitello was running the company. In any case, you asked for a hypothetical where I wouldn’t blame Wilson. This still works as an example of the kind of thing that would exonerate him.
The Tomb Raider reboot is like when you ask a child about something and they give a smart answer – clearly copying someone else they heard – but still smart.
And then impressed, you ask them a second question. But they just repeat the first answer they gave.
Regarding GeForce Now: my suspicion is that regardless of whether or Nvidia could get away with using games regardless of publisher desires, they won’t. For several reasons, they won’t want to antoganize these companies for a brand-new experimental product line. My totally unqualified opinion is that the companies cannot legally prevent customers from using their software in this manner, but that is almost certainly not an issue Nvidia wants to fight over. From their perspective, that would just antagonize their partners. Besides, from Nvidia’s angle, they’re still in the early stages so room exists to grow and they can lever relationships over time.
More to the point, while it may not CURRENTLY be against the EULA(or ToS or what have you), there’s nothing that stops an annoyed publisher from amending those terms.
Just another reason why the legality of click-wrap agreements and EULAs should be challenged in courts around the world. It’s a ridiculous situation that you can’t choose how to run the software you’ve purchased. If another company profits from me wanting to stream my own content to myself, then they shouldn’t be able to ask for a cut of the pie.
Paul and Shamus mentioned something about granting licenses or something in that segment (I forgot what they said exactly and I’m feeling a little lazy to find it right now) but there’s no copying of content or other sub-licensing going on since *I* am accessing the license I’ve paid for on a system I *own* (read: am paying to access).
It’s really a ridiculous situation….
To put it another way – in-home streaming has none of these issues. Logically speaking, there’s zero difference.
Regarding typical character poly count, your ballpark is a bit out of date. Here are some examples from various assets I have to hand (full bodies and heads, no weapons or other gear). XCom 2, which you spend the majority of the time in a high isometric perspective, averages somewhere around 20K tris for its soldier models. Total War: Warhammer II, which is an RTS that spends most of its time at a birds eye perspective, has a similar ~20K poly budget for lords and other important characters. A sample from Black Desert Online gives 40K for a female character. In the Witcher 3, Geralt in his starter gear is pushing 50K tris and some of the added Blood and Wine gear is up around 70+K tris. Battlefront II (2017 version) seems to average around 30-40K tris for the hero models (although it’s pretty variable). Mass Effect 3 looks to be running in the 18-20K tri region for its squad members, and that’s getting on for a decade old and soon-to-be two prior console generations now. Interestingly, Metal Gear Solid 5, which is slightly younger than ME3 but still of the same console gen (although became cross-gen due to prolonged development) appears to be in the order of around 12-13K tris for the likes of Snake and Quiet (even when fully dressed). I’d guess they kept poly counts lower due to being open world.
Regarding the comment system – maybe it would make sense to implement Disqus?
https://disqus.com/
I haven’t ever actually used Disqus. Every time, I click one of the “sign in with X” buttons, then remember that they just use that account to auto-fill all the detail for your brand new account. All the invasiveness of using your real/main accounts, with the hassle of another password anyways! ^^;
I prefer the system already in place for the comments – or are you thinking about the forums?
A system like Disqus, a Subreddit, or similar, could replace both.
If he does decide to start a Discord or Disqus, or whatever, I really hope he doesn’t get rid of the comments here.
Disqus is meant (as far as I know) to be able to be plugged in, right below whatever main content you have, like these comments. The more-like-Reddit or more-like-a-forum, is as far as I know, an extra set of features, or way to set it up.
That’s good to know – making the comment system similar to Reddit is the last thing I’d like to see. Really, the only thing that’s missing in the current system is email notifications for when someone replies to your post.
Why? The only difference is the votes (and the ability to fold a comment tree, which I do miss on wordpress comments).
Yeah but since i don’t use disqus on any other site, i don’t want another login and profile out there. I’ve enough already.
I only ever played Tomb Raider 2013 and Rise of the Tomb Raider but dang the quality sure went downhill which is surprising since gameplay wise it pretty much stayed the same. I feel that they ignored the thing that made Tomb Raider 2013 great which was hunting relics and the supernatural aspect. I still remember coming across the relics and getting the little story blurbs that talked about the WW2 soldiers being picked off by the oni, and learning all about Himiko.
On the other hand Rise focuses on the character interactions and it sucks. Jonah sucks, Lara crying about her dad sucks, and the “revelation” that he didn’t kill himself but was killed by Trinity the super secret hidden organization isn’t a freaking revelation. It’s super obvious and I called it from the very first cutscene. What I want to know about was the Divine Source. You know…the focus for the story? Where did it come from? How did Jacob find it? Why was Jacob able to use it and resurrect perfectly fine but anyone else who was resurrected became a demon creature? All that supernatural stuff became side fluff and never got answered. That’s literally the reason I was there. Very frustrating and made me drop the series entirely. Also, no one cares about Lara’s dad.
One game that looks like a detailed simulation, that is pretty shallow, is Rimworld. A lot of the systems are medium-ish complexity, but the set of systems that are the weakest, are the shallowest, are the tribal / empirical / etc alliances and enemies. The game is billed as a story-generator, but the high-level[1] story of the game always becomes, “everyone who doesn’t love you hates you, and will send ever more militarily advanced forces, to steal your stockpile of turnips”. I really liked the beginning and middle of that game, but once you start trying to do anything on the world-size, it breaks down. All the blips on the map are the same, and everyone is murdering you, no matter how poor you are[2].
[1] The lower-level stuff, like marriages, divorces, mourning for deceased, are all reasonably fleshed-out, even if most of it is generated from a handful of randomized ad-lib scripts.
[2] Raids scale by wealth and time, so if you’re role-playing a tiny village of monks, you’ll be killed just like a fort of super-soldiers…well, faster, but just as inevitably :)
Shamus, I regret to inform you that The Internet is never happy.
The internet was built to house flame-wars.
FLAC files are compressed – it’s just lossless!
Thanks for the Youtube link. I used to listen to the Diecast at work, but a couple months back they started blocking your site.
Re GeForce Now: I don’t know all the details about how it works but it sounds similar enough to Aereo that I can see why their lawyers wouldn’t want to take things to court.
Re: Machine Learning and Game Install Sizes: after seeing that title my mind went in completely the opposite direction, calling to mind AI Dungeon 2, a text adventure so huge it needs research grants and a Patreon to keep itself running.
As I listen to you rag (correctly) on EA’s management and their lack of understanding that engineers are not fungible, I have to say that this is not unusual in senior executives in American businesses. In fact, the opposite is rare and precious.
In my 35 years in Silicon Valley, I’ve worked for many companies, and through friends at various startups I’ve seen many, many more. There seem to be three classes of executives:
1) Business school types. People are fungible, sales depends only on marketing, and quality is only a buzzword. Most common.
2) Former engineers. Technical expertise is prized over understanding the user or market, and the real reason they started their own company is that they are incompetent at actual management – poor people and planning skills. The source of the vast majority of failed startups.
3) Both business and technical background. These can be some of the absolute best to work for, unless they happen to be a sociopath in which case they’re the absolute worst.
Category 3(a) is rare as hen’s teeth. They tend to found successful startups, then either get out themselves or are kicked out by the board when the company is successful at being acquired. At that point, the company almost always stops being innovative, starts being bureaucratic, and becomes a place that good people leave. See figure 1.
Note the use of “sociopath” as a descriptor of executives for category 3(b). It’s a demonstrated fact that CEO’s and other executives show more than double the rate of sociopathy as the general population, and it’s demonstrably true that ruthless backstabbing and credit-stealing will frequently get you promoted, especially in bigger companies. There’s a reason for this.
A lot of the behavior we see from executives, like this from EA, has its roots in a philosophy of business that is based in factories – people are not individually valuable, because people are just machines or machine controllers, anyone can be replaced. This is also the core personality flaw of people on the sociopathy spectrum – they do not perceive other humans as having intrinsic value, only value coming from being used by the sociopath/company. So corporate structures are inherently sociopathic, thus promoting people in line with corporate values excessively promotes sociopaths, and… EA is a prime example of what we get.
One more reason I will never work at a game company that I don’t found myself. :)
Regarding GeForce Now, all I could find was that Activision-Blizzard games got removed. However, this was apparently a mistake, and the lawyers are trying to get things working again:
https://www.techspot.com/news/84031-activision-blizzard-games-pulled-geforce-now-due-misunderstanding.html
The agreement is presumably needed because Blizzard has their own launcher, so they need things set up so that you can use an arbitrary VM and have everything Just Work. They probably have a similar agreement with Valve to use Steam.
The big mechanical disappointment for me was Europa Universalis 4. It has forever wrecked all Paradox games for me. I wrote about this on the forums.*
My plan was to have the Aztecs invade Europe in EU4. Problem is they are extremely heavily penalized on the resource that allows them to learn how to build ships. Still that is what I decided to do. I had this constant pressure to ensure I won the race across the ocean. When I finally had enough points to learn ships, he game Would. Not. Let. Me. I finally made it to the finish line and I wasn’t allowed to cross. All that work, scrimping and saving and nope. I will never trust another Paradox game ever again. I would feel the same way about toadicus’ dam in City Skylines.
*BTW Shamus, you skipped over the part of question about the data dump. I would also like a data dump of the forums. I would have referenced my post here for example.
Shamus, have you tried a macro program as a kludge to your inverted mouse issue? You could rebind the keys with a toggle with an external program. When you hop into the vehicle you could tap a key to swap the up/down. Or you could use the number keys or something to drive.
Sure it grates that a bug like this is in the game and you have to find your own solutions. Regardless it could allow you to enjoy the game.
The question @57:47 seems to hit on what you spoke about last week on the Diecast. Techniques like Animation Bootcamp: An Indie Approach to Procedural Animation and various other techniques along those lines. Which doesn’t seem to have the kind of trade-offs you spoke about this week.
If you do replace the forums with Discord, I really REALLY hope you archive the forums. Chat rooms are not a replacement for a message board.
I think I noticed a bug, this is a test comment. 1 1=2.
Hey Shamus, bug confirmed: You can write posts that use the plus sign (+) just fine, but if you try to edit a post containing one, the symbol will be stripped from everywhere in your post, even if you’re not changing that part of the post. Dunno if you can fix this, but letting you know.
Ah!
I hit the reply button to tell you how odd this is, and now I see the moderation-side editor is all different and fancy. I applied a WordPress update earlier today. The patch notes claimed it was just bug fixes and security fixes, but apparently it had a few surprise features in the mix.
Great. I wonder what else broke.
If anyone else notices any other problems, please do let me know.
Shamus – Bit late to the party and you might not even see this, but for some idea of the crazy stuff AI learning can do now, visit this:
https://www.thispersondoesnotexist.com/
So yes, it would be totally feasible to have AI learning upscale textures and in fact add in details and things to make them look photorealistic. It’s just a question of feeding it enough data.