Amnesia is now open source | Hacker News Reader

Amnesia is now open source

frictionalgames.com

902 pointsjsheard205 comments

The monster's AI is amusing to read, and its behavior has a surprising amount of depth. Really explains its scary, clever and unpredictable nature.

It has over 15 different behavior states:

Idle, GoHome, Wait, Patrol, Investigate, Alert, Search, Eat, Hurt, Hunt, HuntPause, HuntWander, AttackMeleeShort, AttackMeleeLong, AttackRange, BreakDoor, Dead, Flee, Stalk, Track

States:

https://github.com/FrictionalGames/AmnesiaTheDarkDescent/blo...

Implementation:

https://github.com/FrictionalGames/AmnesiaTheDarkDescent/blo...

There's also naming gems like "mindfuckevents"

https://github.com/FrictionalGames/AmnesiaTheDarkDescent/blo...

And the macros used for the implementation are available here: https://github.com/FrictionalGames/AmnesiaTheDarkDescent/blo...

Scary stuff.

The most scary part is the usage of if(0).

That's common for generated code, though. It makes it so that the other code can always use "else if".

I'm also amused that such gameplay scripting logic is implemented in C++, but not on some scripting language like Lua or Python.

isnt the man pig the enemy from the DLC ‘a machine for pigs’? or is that coincidence

Pretty impressive. These games have sold millions of copies.

Here's the GitHub the repo for Amnesia: The Dark Descent, https://github.com/FrictionalGames/AmnesiaTheDarkDescent — looks like a custom C++ engine called HPL using SDL for input, Newton Dynamics for 3d physics, and OpenGL for graphics? The engine also supports AngelScript for scriping in-game events & logic: https://www.angelcode.com/angelscript/

> The engine also supports AngelScript

Heh, if I didn't know already that FrictionalGames was a Swedish studio, this would have given it away

Why is that? Do the Swedes use Angelscript pretty exclusively?

I’m Swedish and I’ve never heard of it, but I also don’t work in the games industry. Apparently a Swedish guy made it according to Wikipedia, so I’m assuming it’s popular among Swedish game developers or something?

The few devs I know who use it are all Swedish and working in game dev. IIRC it's used in at least one of the game dev university courses over there, that might be part of it

Heh, HPL... Fitting name

Lovecraftian game engines, who wouldve guessed

Cyclopean code blocks and sky-flung monoliths, all dripping with green ooze and sinister with latent horror...

Screamingly sentient, dumbly delirious, only the devs that were can tell. A sickened, sensitive reference writhing in code that is not code, and whirled blindly past ghastly midnights of rotting defines, corpses of dead memory with sores that were variables, charnel winds that brush the pallid pointers and makes them flicker null. Beyond the memory vague ghosts of monstrous constructs; half-seen columns of unsanctified statements that rest on nameless line numbers beneath space and reach up to dizzy vacua above the spheres of light and darkness. And through this revolting graveyard of the repository the muffled, maddening beating of keys, and thin, monotonous whine of blasphemous beeps from inconceivable, unlighted chambers beyond the kernel; the detestable pounding and piping whereunto dance slowly, awkwardly, and absurdly the gigantic, tenebrous ultimate gods—the blind, voiceless, mindless functions whose soul is Nyarlathotep.

(from Nyarlathotep by HP Lovecraft)

Anyone who's looked at game engine code...

Gargoyles of the cathedral of computing.

non-euclidian

Minor correction: I think you mean SDL, not STL.

This absolutely thrilled me. Right up until I realized that there's a closed-source, binary-only dependency (FBX SDK). I guess my dreams of playing this game (which I do, in fact, already own a copy of - twice over, actually) on Linux/aarch64 are still a ways off. AutoDesk indicated as of ~2017 that Linux/ARM support for the FBX SDK wasn't on their road map, and nothing seems to have changed since then.

Still- major kudos to Frictional for doing this. It's 100% an appreciated move, and very much in the right direction. The choice of FBX SDK is just, in hindsight, unfortunate, but that doesn't diminish the helpful nature of what they've done here.

(fingers crossed - maybe FBX SDK is just needed for the editor? Hope springs eternal...)

With it being open source now, someone may be able to write a new asset loader using either GLB or UDZ and convert any FBX assets to one of those formats.

Just got the HPL2 engine from TDD to build on Linux/aarch64 right up until it needs to link against the FBX SDK. Only real PITA was getting Newton on ARM going. After that it was a few cmake hacks in HPL2, and a few symlinks in the lib/linux directory of the dependencies archive.

Looks like the culprits are "MeshLoaderFBX.cpp" and "LowLevelResourcesSDL.cpp" - any workarounds for FBX SDK would need to rework and/or replace those.

Does it actually use the FBX SDK during runtime? Most engines I've worked in just use the FBX SDK to convert assets from a DCC into engine specific assets. If that's the case you might just be able to disable it?

This all sounds very fixable, and one of my first programming projects was actually interacting with the FBX SDK. I’ll take a look at it tonight, could be fun.

The Assimp library can load FBX. Considering how well Frictional's code is architectured this should be pretty easy to fix.

That... was actually very helpful. Thanks - I'm having a look at it now.

I wonder if box86 [1] could be made to emulate only calls to FBX.

1: https://github.com/ptitSeb/box86

Wow. That's a neat project. It should also be a good way to run Windows binaries on Wine on ARM.

Indeed people do exactly that to run Steam on Raspberry Pi. Check this video by LowSpecGamer: https://www.youtube.com/watch?v=wxxpNfOeOCE

Curious to know why aarch64? Desktop gaming on linux is still very much x86_64 from my experience.

I thought GPL wasn't compatible with closed source binaries?

The copyright holder can do whatever they like with their code. They can multi-license it, add exceptions to the GPL for their own code, whatever.

GPL restrictions apply to you, since you're a third party, a licensee. You can only use the code in compliance with the GPL.

So, if YOU take Amnesia's source code, and redistribute a binary, you MUST redistribute the source code as well.

If you write a software "foo" that links to a GPL-licensed library, then "foo" must be GPL as well.

This old trick was (maybe still is? I don't know) used by QT. There was a free, open source GPL version of QT; and there was a commercial version.

You could develop and distribute QT-based applications as long as they were GPL; if you wanted to write a commercial, closed source applications, you needed to pay for a QT license.

and who is going to sue them? They are the copyright holders.

It works in the other way. A closed source program can't rely on GPL code (without special permission) but a GPL program can rely on closed source code since you need permission from the closed source license anyway.

GPL just sets a base level of permission that you get, the rights owner is still free to grant any exemptions they want.

No, you can't distribute a binary under the GPL if it's been linked to GPL-incompatible (e.g. closed source) code. Unlike the LGPL, it doesn't make any difference whether/which part of the program is included from a library.

However, the Amensia devs aren't distributing the compiled game under the GPL, they're only distributing the source code, which doesn't include FBX. You're free to distribute code under the GPL which can't actually be compiled/linked.

That said, what you wrote doesn't make sense to me anyway, maybe I misunderstood it.

I think you are mostly right, only that they can distribute the binary since they have the rights to do whatever they want with their own code.

If I fork it I think I could add a new proprietary dependency and release the code for it but once it is built the binary would be unsharable as you say.

What I was trying to say originally is that if I was writing proprietary code I would not be able to add a gpl library and distribute the binary but I could do it the other way around.

> No, you can't distribute a binary under the GPL if it's been linked to GPL-incompatible (e.g. closed source) code.

No, GP is right. Your GPL license covers software "foo". GPL says what can be done with "foo" and things that depend on "foo"; it says nothing about what "foo" can or should depend on. "foo"'s dependencies can (and should) have their own licenses.

TL;DR: If you're writing a software "foo", you can choose what people do with YOUR OWN software (either directly or by linking into it), not with upstream dependencies.

EDIT: I think that what you mean is that, if you're a GPL licensee (not the copyright holder) for "foo", you cannot add a proprietary dependency to it and redistribute the binary. But the copyright holder CAN do this, and it CAN add exceptions to the GPL (it's the copyright holder who can choose what to do with its code)

> GPL says what can be done with "foo" and things that depend on "foo"; it says nothing about what "foo" can or should depend on.

False, the GPL (when applied to binaries rather than source code) applies to entire copyrighted works -- entire programs, with the exception of "major essential components" of the OS (ie system libraries). No other limits are placed (unless the author adds them): it's designed to stretch as far as the concept of "copyrighted work" will allow, which is presumed to include dynamically linked dependencies, and commonly presumed to not include other binaries called by your program (e.g. calling the Steam API using RPC is GPL-compatible, doing it using the SteamWorks client library is not.)

I wrote "you can't distribute a binary under the GPL..." which is true regardless of whether you wrote the program. Of course if you own the software you can instead distribute it under a different license, or add exceptions to the GPL.

To reword: basically, you're right in that if you own the code, you can license it under GPL with an exception to allow linking to anything you want. If you don't add necessary exceptions and distribute binaries, you've provided an invalid license to code you don't own, and those binaries aren't lawfully redistributable.

See https://www.gnu.org/licenses/gpl-faq.html#GPLIncompatibleLib...

(Would have edited my post to this, but the noprocast timeout got me)

I think we agree and we understand the license.

I was trying to clarify the point that the GPL is binding for the licensee, not the copyright holder, since that seemed tonbe CursedUrn's doubt.

You are right that plain vanilla GPL (rightfully) places restrictions about what you can link to from your program.

So you could fork this repo and add dependencies on other closed source libraries?

https://www.gnu.org/licenses/gpl-faq.html#GPLPluginsInNF

It's complicated, but essentially yes, you can. But you may not be able to distribute the compiled work, just the source code, without bundling the closed source libraries.

As long as those other libraries allow you to

https://www.protondb.com/app/57300

Does Proton have ability to convert x86 code to aarch64?

Their own website incidcates Linux support:

FBX Python SDK Windows

FBX Python SDK Windows Mac

FBX Python SDK Mac Linux

FBX Python SDK Linux

He meant Linux/ARM64.

I only played the first Amensia game. I felt it was kind of a weird fit. It was a horror game in that you were in a horror environment with entities that were scary. But the lack of consequence for death made it feel kind of gamey. Horror is hard to get right. The amnesia monsters are horrifying, yes, but in the context of a horror game, not especially memorable in my opinion. One thing I think the resident evil series has done well especially in number 7 and the number 2 remake, is realize that players won't be scared for the whole game. In the first half of these games, you tend to be weak and there are genuinely frightening moments. Towards the end you're pretty desensitized to the nature of that game's fucked up shit, and you just want to mow them down or get by the efficiently.

My absolute favorite horror moment is in the original Silent Hill. You enter this giant hospital and you know its going to be a huge daunting task if its full of monsters, but its not. So you go through the elevator to floor one. It's empty. Floor 2 is empty. The final floor, floor three... is empty. So returning the the elevator you're feeling like you missed something and then, ta-dah, the elevator eerily has acquired a bloody fourth floor button letting you know you're about get shat on.

IMO "consequence" is the epitome of "gamey". Which is why Amnesia didn't seem like it was targeted at the resident evil/silent hill players who were used to "mowing" through enemies. It was for people who wanted to be scared, and have a scary experience, so they didn't give you any way to defend yourself besides hiding and covering your eyes. More like a movie or...a haunted hay ride. Very Lovecraftian.

That silent hill example sounds cool. They used anticipation and your own imagination to scare you. Amnesia is like 95% this.

By consequence, I mean, you have some sort of reason to want to avoid death. Amnesia as I recall had very frequent checkpoints, and dying set you back a minute or so, only to encounter the same scare puzzle shortly after. Doing the same scene twice in the same pattern breaks any immersion (which is, again, a very difficult problem to solve in game design). One of the design principles in the movie Alien was that you do not want to show the monster. I think that's a good rule, and is doubly true in games where death means you do it again. An enemy you can only hide from is cool, but frustrating when said enemy follows a scripted pattern and its secretly a puzzle rather than something where you actually feel like you're hiding from a being (Alien Isolation does this well). Haunted hayride is a good descriptor. Resident evil and silent hill have tried to tackle this problem by inducing resource scarcity. Most scary encounters will not kill you, but they'll force you to expend resources which make you feel more hopeless about the future. It's far from perfect but it sort of works. It is vulnerable to players taking the opposite mentality, whereby they purposefully die to try and clear sections with lower resource loss.

Silent hill games (the good ones...) aren't action games btw. Most people consider the second one to be best psychological horror game ever made.

Horror's pretty dope these days. Next Resident Evil lookin slick https://www.youtube.com/watch?v=JSapXD9vxYA

I vaguely remember a dev commentary or interview about Amnesia.

If I remember right, one thing they said was close to what you said about "you do not want to show the monster" -- but changed to "you do not want to show what happens when the monster gets you".

If the player dies, you either have a gamey consequence (you are set back a long way, you lose resources), or nearly no consequence (you restart quite close to where you died with the same resources). Either way though, the player's fear of this monster is now based on that game mechanic, and could be better described as annoyance than fear. You've lost the visceral suspension of disbelief where on some lizard-brain level the player still thinks "this thing is going to GET me".

So it leads to kind of an ironic balance where the devs are trying to come up with things that are very effective at scaring the player while not really being that difficult to evade for a reasonably competent player. Their ideal is for the player to get through the whole game always feeling like they're in danger and just barely surviving, but never actually dying.

Again, wish I could actually find the source, because it's been years, but the idea stuck with me. I want to say they gave an example of some trick they pulled on the memorable "water monster" level, to amp up the terror while not actually increasing the player's chances of failing. Maybe something like the water monster moves slower while you're looking away from it, so when you hear it behind you while you're turning a crank you think it's about to get you but you actually have enough time to succeed. I don't remember.

That is interesting. I do remember the water monster. That was a great segment. Probably the best part of the game. Sounds like a good framework, but that's pretty darn tough to execute.

Letting the world end in Majora's Mask comes to mind for me. As a kid I was terrified of the ever present timer and never let it hit 0, even once. There's probably something about agency there. I wouldn't ever choose to let the monster get me in Amnesia, but when I did it was because I misjudged something and got myself into an unwinnable situation. If you're going for the strategy they're describing, I think you need to reduce the tree of possible mistakes that leads the player without them realizing they're taking a risk...

One mechanism that really got me ... maybe not "horrified" but uh, anxious as hell, was the Dark Souls games.

The "you better survive until you find your corpse or you lose everything" mechanism worked really well when I was new at the games.

Also in the "not horrified but anxious as hell" gameplay was the original Bioshock and seeing the Big Daddy character... still remember hearing him before he came around the corner... so cool. :-)

I didn't find that aspect of Dark Souls too tense or stressful, maybe because I'd played salt and sanctuary and Hollow Knight first and was used to it.

Dark Souls did have some good creepy moments though. The first time entering the catacombs with the giant skeletons, the first time you face the ghosts of anor londo especially if you stumble in there early, seeing the guardian at the top of sen's fortress and knowing you're going to have to face him after you get by the archers and giants after getting through that hellish trap filled nightmare. Going back to the undead asylum and falling through the floor to face that demon. The giant rat before blight town. There's more, Dark Souls had a lot of those that worked well on the first playthrough. The did lose some of their impact on subsequent ones though.

The Dark Souls one is interesting, because losing a big stack of souls sucks. But at some point you're probably gonna find an easily grindable stack of enemies and just farm them for a few minutes for 15 minutes and get way more. So it's not that big of a deal.

But it still works pretty well in spite of that. The emotional harm of loss stings.

> Amnesia as I recall had very frequent checkpoints, and dying set you back a minute or so, only to encounter the same scare puzzle shortly after.

It worked great for Limbo, Inside and I loved those. Just different tastes.

TBH I only died twice on my first and only Inside playthrough, zipping through it in a couple of hours: it was enjoyable, and I found my experience of it very different from Limbo.

At first I was disappointed, but then I also vividly remember feeling fully immersed during those few hours since I basically did not run into checkpoints.

Have any games tried frequent savepoints + only allowing the player to reload 24 hours after dying? Seems like that would make reloads feel a bit more fresh/consequential, and also prevent addiction.

24h sounds like a turnoff, but in general, I think that's a great concept. You just need something that the player can do while unable to try to scary part again.

To relieve the frustration, it would need to be a game where a) death truly feels like your fault, not some jump scare QTE, and b) with enough information to leave you with a way to think about how to go through the challenge while not being in front of the game during the downtime.

this would be boring as hell because it relates to mobile games: they will tell you can't do something for X hours, unless you pay them some dollars. Even if you don't charge, the memories will flow I can guarantee you

I never played any of the silent hills. I played resident evil 4 and 5 years ago, and recently played resident evil 7 - which i felt had a lot more horror elements then emphasis on shooting/exploding enemies like 4 and 5 did.

I REALLY enjoyed RE7. I felt like it kept me on my toes and I played it way faster than any other single player game in recent times.

What is the best entry point and order for the silent hill series?

Resident evil is a weird series the branches into different genres over time. The first few were survival horror and played on the dread of not being well equipped enough to get through the rest of the game. Then Numbers 4-6 become action games with unreasonably sexy badass characters. They were good games, and some great co-op, but not scary at all. 7 was a horror masterpiece (or at least the first half). 8 looks like it'll continue along that trend. The recently reimagined Resident Evil 2 is phenomenal and finds an unexpected sweet spot between all 3 of them where it's still action-y, but has a bunch of things that are suspenseful and scary.

Silent Hill struggled to get out of the original playstation and hasn't had anything good in years. Remakes and remasters of the originals are generally perceived to be lower quality. I would try to grab an emulator and just play the first 4 in order then pretend the rest doesn't exist. Number 1 and 2 are the treasures of the series.

> Number 1 and 2 are the treasures of the series.

3 and the constant pressure of the big bad guy is great too, the pace contrasting massively with 1 and 2. Code Veronica is really nice too. I don't like 4.

But the "tank controls" really are some relic of the past.

I just played SH1 for the very first time on an emulator, last month, and had a great time with it. Getting used to the controls and game mechanics took a little bit, but I got up to speed quickly. Finished the game in about 7 hours.

It's creepy, weird, kind of funny, and ultimately an interesting game. Definitely recommend a play through.

I think maybe you met the game mechanics too quickly and that might have ruined it for you. For me it was very non-gamey because I had no idea what the consequences of meeting one of the monsters was so I spent almost the entire game petrified. Once I had died a couple of times and learned how the monsters worked, the spook factor wore off quickly.

But I still far preferred that to the classic weapons based horror games where the spookiness is derived from how tough an openent is in combat.

You should really, really try SOMA though. They dialed in the non-combat spookiness and even once you figure out how the monsters work, the few there are, it's the story and atmosphere that drives the spookiness anyway so it doesn't detract from the experience. Much less gamey, and quite a good mystery mixed with some existential dread.

+1 for SOMA. That game was amazing emotionally. It was left me thinking for days what i went through in it.

SOMA was a great game, especially the last hour or so of game play kept me on my toes. The monsters were not the focus of the game at all I recall you could even disable them entirely.

I enjoyed how scared I felt in Amnesia until the first time I died and reloaded. After that the edge was gone.

The original Silent Hill was a masterpiece of horror games. But mainly because it was all in your head. The creatures weren't that bad, but the sensation that they were coming at you is what it counts. Also, the radio that started every time you had one close to you was excellent, because it made you feel horrible even if you weren't doing anything.

I played a few games after that one in the next years (I was only 11 when I played Silent Hill) and nothing made me feel the same.

The first couple Penumbra games were the same way. You knew there were monsters about, but mostly you crept around and if you heard them the only sane strategy was to run like hell.

I think the people who enjoyed the games were the ones that could get sucked into that mental model early and didn't want to die in game.

The dogs in Penumbra: Overture can be killed with just the pickaxe - it's a bit akward due to the controls but not too hard. Not sure about the spiders, but they are only a small part. The giant worm is unkillable afaik but it's only there for one chase sequence.

Definitely takes the edge off the horror aspect when you figure that out. But then again, I like both the Penumbra series, Amnesia: A Machine for Pigs and SOMA more as exploration/puzzle games with a scary atmosphere than for the actual horror aspect.

*"the radio that started every time you had one close to you was excellent, because it made you feel horrible even if you weren't doing anything."

Haha, same with Aliens vs Predator, where you had either radar or night vision, but not both at the same time.

The beeping made you crazy.

Have you tried Alien Isolation? It's a masterpiece, similar mechanic to that in place too.

I think the last Alien game I tried was "Alien: Colonial Mariens".

It was so bad, that I stopped playing games from thaz franchise

Alien Isolation is seriously my favorite in the franchise next to AvP. It's a bit derivative of Amnesia series but I still highly suggest picking it up. The atmosphere alone of the game is scraped from Alien 1 where it's got this retro futuristic setting. You won't regret it but you can find it for cheap in a sale just in case.

The most horrifying moment for me was the hospital level in the last of us 2. This and the night where you play abby and save the girl and the boy. I had to switch from hard to easy or I was going to have PTSD. Didn’t remember that the first game was really scary.

The hospital basement is such a creepy level, and the rat king was terrifying.

I bought the game in some bundle and then watched it on youtube. It worked well as horror that way.

I had no motivation to play it after and ended in some entance hall right after start.

I'd love to do this with my VR game, but it contains at least a dozen thirdparty paid assets -- some of which contain heavy modifications. I'd have to withhold these assets from a release, making it unbuildable, unless someone could prove with invoices that they also own them.

Kind of bums me out, because there aren't really any released projects from commercially successful Unity games. And then there is the legal ramifications of having your source out there. What's stopping a bad actor from trying to find something trivial they could sue you over because of vague patents?

If you're doing it for community contributions then yeah you're stuck. But if you just want to let other people see & learn from your code you're perfectly free to release whatever you can.

For example DOOM, the thing ported to freaking everywhere, was released without working sound on DOS because of licenses ( https://github.com/id-Software/DOOM ):

> The bad news: this code only compiles and runs on linux. We couldn't release the dos code because of a copyrighted sound library we used (wow, was that a mistake -- I write my own sound code now), and I honestly don't even know what happened to the port that microsoft did to windows.

And also without any game art assets - no textures, no levels, no sounds, no etc... Just the code, and just the code that wasn't covered by a license agreement.

I have a policy with the games I'm working on that I won't include any resource, engine, code, etc... unless it's Free (Libre) or I personally own the IP. I often wonder if I'm too strict about that, but seeing comments like this makes me feel like it's the right move.

I don't want to be in a situation where I want to do something with my game and can't because I would need to track down the rightsholder. As a consumer I want to have some semblance of ownership over the things I buy, but even more than that as a creator, I really want to own the things I make.

This is just my opinion, so grain of salt.

But I think this is an untenable restriction for a tiny indie team if you want to ship something across multiple platforms in a relatively short amount of time.

We were ten months from start to finish on PC SteamVR, a couple months to Oculus, and then four to PS4. With two people. My game is built on top of literally millions of man hours of development. This game would basically never have happened without a ton of proprietary software.

That all said, it's awesome to see when people have the opportunity to go full custom engine like they did with Amnesia or Noita.

> My game is built on top of literally millions of man hours of development. This game would basically never have happened without a ton of proprietary software.

Sure, starting from scratch will only be realistic for few projects but there is also millions of man hours worth of non-proprietary software that you can build on. If that would have been possible for you is impossible to say without knowing the specifics, but you should not jump to the conclusion that it is impossible for everyone. VR specifically is a field where this may be more difficult than elsewhere because it's new and niche so good open source solutions have not yet settled.

Even if you go use something like godot, have fun getting it working on PS4. Due to legal requirements of working with their platform, there cannot be any open implementations that work with it. For Godot you are stuck needing to work with a contracting firm, using their proprietary engine modifications. Or figure out how to do all that work yourself once you have a relationship with Sony and have access to their SDK. And then, even when you do you will not be allowed to share that code without violating the agreement.

Sure, closed platforms suck - making you give up your freedoms to use them is kind of a given. But you can limit that to those closed platforms. With your Godot example you might not be able to distribute all your source for the PS4 version but you should still be in full control of the PC version.

This is fair. I've been working on my current project for nearly 2 years (although I haven't been working on it full time, and I've been working as a single dev).

I think a policy like this changes the types of projects you undertake.

With my current game, I would have built a custom engine for it anyway, since the design of the game kind of requires me to have a lot of control over low-level details. So it wasn't a huge stretch for me to go from that decision to, "well, if I'm building my own engine, I sure as heck want to actually own it."

There are tradeoffs in both directions. I'm definitely conscious of the fact that there are things I want to take advantage of that I can't, and I already know console ports are probably going to be horrible to work on.

Kudos! Also, I just checked out your personal site and WOW does "Reset Hard" (2018) sound amazing!

https://danshumway.com/blog/reset-hard-announcement/

On my phone, no time now to look closer but plan to soon.

Thanks! I've been really bad over the past year at updating the site about progress, but the game is still being worked on and I'm still pretty excited about it. Just still in the middle of development.

> I'd love to do this with my VR game, but it contains at least a dozen thirdparty paid assets

Note that the release doesn't contain any assets. It's buildable in the sense that you can build the binary (and even then it looks like that doesn't actually work quite yet for any random setup).

IME this is common for OSS release of commercial games, generally you have to provide the game assets to get a playable artifact. Though I guess it's more complicated if building the game assets is part of your build pipeline.

This is one reason I'd love to be able to fully open it up. It's great to read code and see how things are done (or shouldn't be), but without it being playable it kind of comes and goes.

At this point with my game I'd be happy to have the whole damn thing up such that people could build and play it, and even make new levels if they wanted to. The long tail on my PC sales have dipped enough that any potential loss of income would be irrelevant to me because of the cool stuff people might come up with.

Seems like it would be easier to open it up earlier than later with something like a complete listing and documentation of assets? That way if people are interested they can assist in re-building “unencumbered” assets.

Hobbyist gamedev, kind of a concern to me too. There is no way to make such a project FOSS to any degree without treading on license obligations from the Engine or third party assets (most of which only allow redistribution within a clearly defined team).

A compromise I've seen (and what I do with my hobby projects) is to publish source code to github without a FOSS license, so at least the code I write has some value to others to look at, binary assets (including third party) are included in the project over git LFS so its possible to deny access to those by not providing LFS credentials.

An example along the lines is a project I follow a lot, the developer publishes their core gameplay mechanics code as MIT[1] and it's been very useful to me in a UE4 context (for things like understanding good practices re: building larger systems in UE4).

I guess I mean realistically impossible for you to release projects as FOSS but IMO there is still value if you can push any content you do have ownership to in an open way (at least for educational purposes for others).

1 - https://github.com/alanedwardes/Estranged.Core

Well, in theory, if you had used a free libre engine, you would not need to fear such a thing. That would have been an ahead of time consideration, when choosing your tools.

Whether or not you could have done what you have with a differently licensed software is a different question, which is about licensing of big companies and their ethics.

At hobbyist scale, it's usually more of a problem with assets rather than the engine itself, or with tools.

For example, in a game jam you might want to be able to push something to a public Git repo that anyone can pull from and build, and license it so people on the team have no question about whether they can share the code with other people. Easy way to do that is to have an agreement like, "Let's license our code under MIT and make the art CC BY 4.0. Agreed?" Then drop copies of the license text in the repo with everyone's name on it, and make the repo public.

As soon as you use third-party assets then you probably can't do that any more.

Just speaking from a practical perspective here... I'm working with people and want them to freely use the code that I've written and the assets I've made.

Yep, as far as I'm aware none of the larger engines UE4 or Unity mind if you release your source code under a non viral FOSS license.

But almost all third party assets have restrictions, it's not like the assets can have DRM to prevent the paid for asset being redistributed and used in other works.

Are they the kind of assets you can stuff into a folder and if absent fallback to some dummy assets?

No unfortunately not. The most important ones are code assets.

For instance I use heavily modified versions of:

- PuppetMaster from http://root-motion.com/ - Chronos from https://ludiq.io/chronos

Trying to fake the API surface area of these assets would be... hard. And without them the game is entirely non-functional.

I've seen other devs open source their games and simply release it in a non-functional state with statements clarifying what's not included, why, and what it would take to get things up and running. Just about anyone looking to build a modern game from this sort of source release is going to know what they are getting into with or without these assets and may still get value out of the release even if it doesn't provide all the necessary bits to create an executable.

You could still release what you can. Might be helpful as a reference or for modders of your game.

Could you theoretically supply patches to those libraries alongside your code? I'm curious what the legal implications of that are.

It's possible, but the patches would be tied to a very specific version of the asset. Unity has made this easier recently by making asset store assets part of the package manager, but my stuff predates all of that.

I’m not sure if this release contains assets either, it looks like it is the engine only.

Frictional Games has always had good Linux support as well. I wish other companies would follow suit and release some of their older, abandoned (not that I would call Amnesia either) titles to the open source community.

Warzone 2100 had something similar happen and it gained a community revival.

Edit: I can't believe Amnesia is 10 years old!

> Here is everything you need to build Amnesia: The Dark Descent.

But I think the assets are missing, so it is not everything. Maybe they released those somewhere else?

> But I think the assets are missing, so it is not everything.

It's everything you need to build the game, which is different than everything you need to play the game.

> Maybe they released those somewhere else?

No, studios rarely to never release assets as OSS. In part because most artists don't sign off on this, and in part because especially smaller studios will license existing assets from stores and the like which they literally can not, legally, relicense.

Rarely do they release the assets with the open sourcing of the engine. If you want those they tend to want you to pay for the game and then pull the assets out to put in your build.

That ok. Is great that they open sourced their engine, however they don't even mention the assets and that you need to buy their game to have them, which is a bit strange. And they even say here is everything you need for the whole game.

> Is great that they open sourced their engine, however they don't even mention the assets and that you need to buy their game to have them, which is a bit strange.

When it comes to open sourcing game code, it is pretty standard.

AFAIK most open sourcing of 'games' state that they open sourced only the engine and not the complete game. And maybe provide a guide that shows you how to copy the original assets into a game compiled by yourself.

I just think this is misleading and probably just a marketing act for the next release of their game.

Personally I find it great that they released their engine again, the more open source code the better, however they should have just been more careful with their press release. Some reporters that don't look to close to the code might think that Amnesia is now a open source game like 0AD or recently unvanquished, that actually tries to make their assets fully open source compliant.

> assets fully open source compliant

It's hard to interpret that phrase but it certainly seems to apply to their art assets. You could ship a GPL reimplementation or total conversion mod of their game tomorrow and use the game assets, as long as the user already owns the game. Just like all the very popular Quake and Doom stuff, for starters.

You just won't get the free game that you want.

Everything to build the game. Which is true.

No. A 'game' can be played, and without any assets it cannot be played. It is just a 'engine'.

Build - you get an executable, which is true. Run or Played is a different thing.

Building != playing.

Building -> a functional exe is produced.

Playing -> Many EXEs expect a set of resource files to be in the directory to further their execution.

The announcement on their website noted that you still have to own the game itself. Maybe that should be added to the readme.

They also own their code. Releasing it as open source did not change that.

They don't actually state that you have to own the original game on their website.

It's implicit in their having "only" open-sourced the code: you can't actually play the game without the assets, and accessing the assets legally requires buying the game. Unless you want to remake all assets by hand I guess.

Maybe "to build" means more specifically "to build the binary of". Then yes, you don't need data assets for that. It particularly doesn't say this is everything needed to "build and _play_ the game".

I think it's targeted at modders. So they just get the code and take it from there. You get the misleading article.

> I think it's targeted at modders.

Modders have been modding for years (Amnesia has over a thousand mods). They'll be happy for sure, and certainly are one of the targets:

> We are all really excited to see what comes out of it! The modding community has been incredibly creative over the years and it will be fun to see what it can do with the full source code at its disposal.

but they've never needed the code either.

It could also be useful to, well,

> anyone wanting to create their own engine or just wanting to learn more about game programming. While the code is not the greatest in places and the tech used is not the latest, it is a fully contained game engine in a fairly easy-to-understand package. It is also a testament that it is possible to do this sort of thing, even with a very limited team.

> but they've never needed the code either.

That's true, but having the code of a game usually makes modding it easier and allows mods to go even farther

Maybe they weren't allowed to.

So can you fully build the game and run your executable build, such that we could start seeing these games showing up on Flathub and the like? Or is it the Doom-style "You can build the engine, but you still need to supply your own assets from a purchased copy of the game"?

You guessed right, the GitHub repositories only include the code, and the README mentions (emphasis mine):

> All code is under the GPL Version 3 license.

And the linked page:

> Very important note: This doesn’t mean that the game is suddenly free. It just means that people are free to use the source however they want as long as they adhere to the GPL3 licence. The game and all of its content is still owned by Frictional Games. Just like before.

Doesn't look like any assets are included in the github repo.

The README.md states:

> Here is everything you need to build Amnesia: The Dark Descent.

But as you said, the assets seem to be missing. Maybe they release them somewhere else?

I think "build" in this context means "compile," as opposed to "run."

Well a 'game' can be played, otherwise its not a 'game' its just a 'engine'.

Just because it's not complete doesn't make it a game 'engine' (though there's likely an engine in the code somewhere). It's more than a game engine, it's a game—it's just has no art assets. So, no, you can't make a full release from this repo, but you can build the executable.

This is also how the Doom 1, 2, and 3 sources were released. You get the code but you need to buy the game to legally get the other assets.

For anyone that hasn't yet seen it, I'd highly recommend watching Ars Technica's video interview on Amnesia and how it created its unique atmosphere: https://www.youtube.com/watch?v=sMl2la8-3-o

Along with the rest of their War Stories videos, for that matter. The interview with Andy Gavin on Crash Bandicoot is great, even though I never played the game: https://www.youtube.com/watch?v=izxXGuVL21o

SOMA sources would be nice. Just last night I ran into this bug under Linux when pulling the lever at the relay station in delta sector the game reproducibly crashes with an trap divide error in libfmodevent....

One of the selling points of at least the later games was the availability tools to build your own "stories" with the same base mechanics, so it's cool to see these be made more widely available and some of these fan creations evolve into more.

I never played any of these games but the first one looked pretty scary from the videos i've seen of it. I might have to download it, it is fall after all and i'm in a mood to be scared.

SOMA was my favorite. Amnesia launched the genre and is more famous, but SOMA not only scratched my hard(ish) sci-fi itch, it elevated its own horror by doing so.

Sci-fi horror usually focuses on relatively silly premises like "what if the monster gets out" or "what if we open a portal to hell." SOMA really rises above those tropes. It's legitimately unsettling to have a game confront you with troublesome implications of future technology that you had been subconsciously avoiding.

SOMA was wonderful. Maybe it helped that I didn't have really high expectations going in -- I pretty much thought "more horror like Amnesia, but in a new setting". So I went into it blind.

But it ended up being one of those experiences that you keep thinking about for a few days afterwards. And it's a rare video game that I think would have lost its impact as a movie or a book, despite not having much gamey stuff to do (it's practically in the walking simulator category).

The philosophical questions it brings up may not be anything new, but the game does an excellent job of putting the humanity in them and making you really consider what it would feel like to be in the circumstances of its characters -- not just the playable ones or the ones they interact with directly, but the ones you find audio logs and transcripts from.

Edit: I should also note that I played The Talos Principle shortly after playing SOMA and was struck by a couple similarities they have. Talos is a much less story-driven game, you can almost ignore the story and just focus on the puzzles if you please. But playing it right after SOMA I couldn't help but enjoy how it almost covered the same type of scenario from another -- perhaps more optimistic -- angle.

We had a really similar experience, including playing Talos directly afterwards. They both had a really incredible solitude to them, which I think resonated because of the same ideological reasons but in two different contexts like you said. They managed to evoke pretty much the same emotions from me, except for the horror from SOMA.

There was an uncanniness to the Talos robots and the terminal entries which was really creepy to me though, I found them really unsettling at first.

Seconded. I'm still impressed by SOMA, a couple of years after playing it. Some of its dialogue has remained stuck in my mind: "I woke up in my bed today -- a hundred years ago".

What works for me is that it's a genuine scifi story, even with its horror motifs removed -- some people even prefer it that way! It's a nice scifi story with interesting existential questions and compelling characters. I like how key aspects of the ending are foreshadowed by previous events, and how it's completely consistent that the main character remains blind to the implications.

I was really, really impressed by SOMA.

I'm always surprised at how people can vary so much in opinions. I didn't play an hour of Soma before I felt bored and infuriated at the gameplay mechanics, quit and never played it again.

Do you remember roughly at which point you left the game?

SOMA feels very by the numbers horror-survival at first (think: Bioshock, System Shock, etc), but this begins to unravel after a while. The first few situations seem standard, you think you have the plot figured out, and there is one escape-from-the-monster situation which is infuriatingly difficult.

However, I'd say if you give it a chance you'll discover it's not really in the survival horror genre -- some people play it with monsters disabled! -- and is in fact an exploration of consciousness and the sense of the "self". And quite interesting, too. There are some pretty poignant moments I wouldn't expect from a videogame.

I know every game likes to say this about itself. I, for example, found the plot twist and self-proclaimed "deep" plot points about Bioshock very unimpressive. But SOMA feels closer to something like A Mind Forever Voyaging in my opinion...

I strongly second this opinion. After playing SOMA (and being seriously frightened by it - you need to play it in the dark at night, with headphones, full immersion style) I was an instant fan of the studio and tried Amnesia. But the entire historic-fantasy-style theme turns me off pretty hard. I'm having a hard time continuing to play it, though I want to give it another chance sometime.

But SOMA? Really, really great game. I'm considering repurchasing it on PS4 and replaying it in my home cinema setup for even more immersion. Though knowing the story beforehand probably takes quite a bit out of the experience.

Yes, try to experience SOMA without spoilers. "Figuring it out" is part of the experience.

If you're 5 minutes in and worry that you've figured it out and aren't impressed, don't worry, you haven't seen anything yet.

I found the themes in A Machine for Pigs to be more believable and thus scary than A Dark Descent, but neither hold a candle to SOMA thematically and most people regard A Machine for Pigs to be less good than it's predecessor.

> most people regard A Machine for Pigs to be less good than it's predecessor

I think that's mostly a problem of expectations. TDD focuses on being a horror game while A Machine for Pigs is much closer to a horror-themed walking simulator. If you expect Amnesia II out of A Machine for Pigs then I can see how you would be disappointed but that doesn't make the game worse it just means using the Amnesia name might have been a bad choice since the games are in different subgenres and thus target somewhat different audiences.

It's got some brilliant writing and voice acting as well. It's also not too long a game, I played about an hour a night and completed it in a week. It's not a giant time sink like some games try to be.

I'm definitely interested to hear about similar games - I was never able to find anything that quite compared. Amnesia looked quite dated and unpolished by comparison after playing SOMA and I couldn't quite get into it.

I've never found anything that compared either.

I recently had another pleasant hard sci-fi surprise, though: Horizon Zero Dawn. I didn't expect it to take the backstory (or even the main story) seriously, but it did, and it did a brilliant job of it.

I had exactly the same prejudice. "Shoot robot dinos with a bow and arrow" doesn't sound like a premise that could possibly have a decent underlying story, but they actually pulled it off.

There is nothing like the story and atmosphere of SOMA, but some notable games in the ballpark of "unique, low combat, engaging story" are:

Obduction, The Talos Principle, Alien Isolation (some combat), INFRA (creepier than expected)

Frictional's first game, Penumbra: Overture also fits in my opinion (although not if you already consider Amnesia to be too dated).

I found Kairo to be a little gem in this style.

I'm actually a bit disappointed their next game is set in the Lovercraftian Amnesia universe. Before the SOMA released I considered the Penumbra series to be above Amnesia, despite the later being the one which gave them recognition, and after SOMA I'm convinced they are much better at sci-fi horror than at cosmical/mythological one.

Although maybe the reason is that I find the Penumbra/SOMA setting to be much more believable and therefore unsettling.

Ouch. Color me disappointed as well.

After SOMA, I'd say they are good at sci-fi, period. Even with its horror elements removed, SOMA has some interesting scifi situations.

How did you feel about A Machine for Pigs? I felt the real story there was the way the father went crazy, and the role the church had, two really believable plot points. Not really the man/beast moster and so on which was less believable.

I do agree though, for what it's worth they have two projects in development right now so maybe one of them is more scifi.

I also think that Amnesia: TDD was Frictional's weakest game (not bad but just not as good as the others), in particular due to the more mystical "ancient alien" plot.

Personally I prefer A Machine for Pigs over TDD for that reason too but I was always under the impression that it was mostly The Chinese Room's creation and that Frictional's involvement was limited to the more technical and publishing aspects.

Nitpick but the genre existed before - Amnesia is a first person horror adventure game and the Penumbra series was their first attempt with almost identical mechanics. However even before there was Call of Cthulhu - and Thomas Grip's (Frictional's programmer) own Unbirth game[0] which was cancelled but they demo uploaded (the downloads do not work but you can find it at the archive[1]). Unbirth feels like a very unfinished Penumbra without the physics stuff (but the horror adventure game elements are still there).

And of course there were similar games going back in time, e.g. Realms of the Haunting [2] and i'm sure you can find many others.

[0] https://unbirth.frictionalgames.com/

[1] https://archive.org/details/unbirth_alpha

[2] https://www.youtube.com/watch?v=wzT27LXpoyc

Penumbra: Black Plague remains my favorite. It's good enough on its own you don't even need to play the first Penumbra episode if you're time constrained, though the first Penumbra was pretty good too. I played Amnesia later and it was good, but not as much. Haven't gotten around to Machine For Pigs.

SOMA was disappointing. It wasn't bad, but I don't think of it as horror, and its story and characters were the weakest in the lot. But maybe that's just because I'm too familiar with sci-fi covering the same things much better and with better characters, and there's even a non-fiction book analysis of a future which assumes the central tech comes about that's deep enough in exploring implications to put the sci-fi to shame on that metric.

I can definitely agree to disagree on which game mechanics are better but calling the SOMA story the weakest among the Frictional games just confuses me. If anything I found the magic ancient alien plot of Amnesia to be super meh. Guess you're not much of a Sci-Fi fan?

On the contrary, I like sci-fi. I'll repeat myself that I think seeing the themes explored in other places more thoroughly and with better characters (I'll give one of many anime examples: Ghost in the Shell; two of many fiction books: Permutation City and to an extent The Golden Age trilogy; a video game: Nier Automata), including a deep non-fiction analysis (http://ageofem.com/) makes me more dismissive of SOMA than perhaps I ought to be. In particular near the ending of SOMA where the thing happens to Simon and he freaks out, despite the same thing happening shortly beforehand and freaking him out, was the most annoying bit that really sealed my distaste. The resolution or not of the sideplot driving the 'horror' was also lazy...

I'll agree Penumbra and Amnesia don't have world-class stories either, but what they do have is told well with better characters, the horror aspect is better, and I more enjoyed the rest of the non-story elements that make up a video game. SOMA easily wins on graphics and slightly less clunky control, if it changed nothing else but its main character it may have even won me over, but as it is I can only call it not bad...

(spoiler warning)

Self-delusion is one of the major themes of SOMA. Simon and Catherine partake. I can understand how it came across as annoying and lazy writing, respectively, but it's not. Well... if the writing was annoying or lazy, it wasn't for those reasons.

As for the book comparison, I also think that's slightly uncharitable. Books go far deeper because the written medium facilitates doing so, but interactive media have their own unique benefits: they can force you to make decisions, prohibiting the kind of lazy fence-sitting that's so easy with passive media. I wouldn't object if someone found the book analysis more valuable, or vice-versa, but without a specific disambiguating goal in mind I don't think it's fair to judge one by its strengths and the other by its weakness.

> In particular near the ending of SOMA where the thing happens to Simon and he freaks out, despite the same thing happening shortly beforehand and freaking him out, was the most annoying bit that really sealed my distaste.

I agree that that was a weak point but not because it's unrealistic for Simon to be surprised - that continuity of Simon did after all jump along with every copy so he did not really experience the same thing before. Being so focused on getting on the Ark I find it quite believable that Simon would not think too much about that he would also be staying on the sea floor.

Forcing that reaction onto the player was a mistake though. I think it would have been better use of the medium if Frictional had not added the Ark ending and instead left the player stranded under the ocean with nothing to do and leaving you to come up with your own reaction.

For me that ending didn't take away from the rest of the game though.

> the horror aspect is better

As weird as it is to say about Frictional's games, the actual horror aspect is the least interesting part to me. I guess that also explains why I'm a bit biased against TDD. I do agree that the Penumbra games have great characters though - Red is one of my all time favorite video game characters.

I played through the first and part of the second. First one is quite scary and has an engaging story!

Somewhere almost at the end there's a 'level' where the player need to walk across the ocean floor under a rough 'weather', be sure to check that part out, it's a highly immersive experience.

There's some effort to port it to VR, I can't even begin to fathom the kind of feelings that could emerge from such an immersion (pun intended, the game sets stage on the ocean floor).

Horror games in VR are way too much for me. I can't handle games like HL:A and the walking dead one. Its just too stressful. A shitty horror game on PC becomes terrifying in VR.

The sound design in SOMA was incredible. They sold that scene so well I felt exhausted after it.

Maybe trying to generate interest for their upcoming game, Amnesia: Rebirth? Honestly, their stuff is quite good if you like horror games-- Amnesia: Dark Descent and SOMA are excellent!

I’m not sure open-sourcing an older game would gain them many eyes, since this news will likely only reach programmery types and the most hardcore of gamers.

Maybe (given the timing), but Frictional is known for open-sourcing their older games. The Penumbra sources are on their Github, too!

Frictional Games also developed SOMA. It's also a horror game, but the psychological horror in the game is just a gameplay element where the philosophical topics are embedded.

I consider myself pretty thick-skinned when it comes to being scared but I * checks notes * still have only played 27 minutes of Amnesia according to Steam.

My grandma has this super old wooden house in the middle of nowhere. One summer I was staying there with her, I went downstairs, turned off all lights and played Amnesia. I honestly thought it'd be fun, but I stopped after ~1 hour. It was an experience.

Amnesia needs to come out with a mobile version so I can go camp out in the woods alone.

Not sure if it's my connection, but the website seems to not responds.

https://archive.is/Ij2zA

It's not working for me either, I guess we killed it :) I couldn't find a cache link neither...

I had problems as well. 9/23/2020 @ 10:54 AM (EST)

Awesome! I can also recommend their earlier Penumbra series. Bit dated by today’s standards, but still very enjoyable horror games.

IIRC the game was impressive in the way you could slightly open doors and drawers, never played it but I remember that.

That's the right way to preserve past games.

Absolutely. This practically guarantees there will be a port of this to every OS available for years to come.

Frictional Games and Loiste Interactive are my two gaming sweethearts. Them and Cyan Worlds are the only studios I keep an eye on to make sure I don't miss a game.

Cyan Worlds, makers of Myst, are releasing games again for those like me who are late to parties like this.

Loiste Interactive released a game called INFRA which I have replayed like 4 times already. There is no other game atmosphere like it, similar to how SOMA, Amnesia and Myst are without rival. There are games as good as them, but few if any games that are like them.

Cyan's Obduction was pretty good but had its flaws too.

I was just a kid when Myst was a hit, and never finished it. But I remembered being fascinated by the strange worlds to explore in that game and the uncanny mix of fantastic and realistic elements within them. Obduction definitely hit that same tone. I was absolutely thrilled with the exploration side of it and most of the puzzles were pretty enjoyable too. I enjoyed the unique look of each world so much that I was really disappointed when one turned out a bit boring -- though of course it makes sense in context.

However, like Myst, its open-world puzzle style gets frustrating when you run out of obvious paths and aren't sure what to do next, and end up running around everywhere trying to find something you haven't played with yet. That's a double-edged sword because the freedom and discovery is part of the appeal of those games, and I would never want them to remove it. But it does get annoying sometimes and I wish they would have considered that and chosen NOT to have areas to find with stuff that NEVER becomes useful. Especially when one of them has switches to flip (that don't affect any of the games actual puzzles). I wasted probably half an hour trying to find some purpose to the submarine cave and tiki bar while stuck, thinking they would provide a path through what was blocking me when in fact I needed to look elsewhere.

The other thing is that the load times in Obduction are a little long. That's annoying given how often navigating between worlds is required to complete a puzzle. This is probably getting better with faster and faster SSDs these days.

It was great overall though and I wouldn't hesitate to recommend it to fans of the genre.

I'll have to check out INFRA.

I've found things published by Annapurna Interactive to be disproportionately interesting.

I just started playing Outer Wilds today which was released by them, and you are absolutely right

You're in for a treat! It leans on tropes a bit in some areas, but I'm not sure that's the wrong call - it can be hard on your audience when you're too original everywhere. There's definitely much innovative stuff, and it's a heck of a ride.

I also very much enjoyed Gorogoa, Donut County, and What Remains Of Edith Finch. All very different (which is easier for a publisher than a studio), but all compelling in their own way.

From reading the title, I thought this game was being open sourced: https://en.wikipedia.org/wiki/Amnesia_(video_game)

I never got very far in that game, but it made quite an impression on me for the gritty reality in the game that wasn't present in other games of that era, with you having to beg or wash car windows for money in order to buy food.

Oh I totally misremembered, I thought it was infocom!

But I remember playing. It was hard.

I would pay a lot of money to be able to play these games on my smartphone.

There could be a market for a studio that licenses desktop games IP and makes mobile games out of them.

I have to doubt that Amnesia would "work" on a smartphone. The tiny screen would remove a lot of the immersion. Amnesia is not (just) about finishing the game - it's about immersing yourself in the horror.

sure - i have a chromecast as well. i can mirror my phone screen on my TV in full HD as well.

A phone is far far far more powerful than you think. And the new Snapdragon 865 are beasts.

Isn't there significant latency when screen casting a phone?

People play Fortnite/pubg on their TV by casting through the mobile - very popular. Remember, these are twitchy FPS games. Stuff like Amnesia is much much slower (in fact most games)

https://www.youtube.com/watch?v=lhfxeXxvN6s

Note that this video is not casting through a Chromecast wirelessly, but over HDMI.

In my experience, screen mirroring over WiFi to a Chromecast is barely good enough to watch a video (even then it stutters occasionally), so playing a game would be a bad experience. My phone and Chromecast are both severel years old though - perhaps it works better on newer models.

I'm pretty sure that's Aspyr Media's (https://www.aspyr.com/about) business model. At least, I know they're the ones who ported Fireaxis' Civilization games to OSX and Linux.

You can do that with the Steam Link app if you have a copy on your PC, and Microsoft now has (will have?) a similar thing to play in their cloud if you have an xbox copy/if it's on game pass. Playing on your local network isn't bad for latency, though these sorts of games aren't especially latency sensitive so I imagine the cloud experience wouldn't be terrible either.

If it's matter of size ,is already ported to the Nintendo Switch . https://www.nintendo.com/games/detail/amnesia-collection-swi...

I think there is a large community on android for this kind of stuff DIY style.

Probably why monkey island was released on ios only, so it has some drawbacks too.

I think you can play monkey island on Android with ScummVM. And all the games that work with ScummVM.

Not talking about an emulator. It's fundamentally taking the assets and art and making a mobile first gaming experience.

For example, take Braid - that whole time skipping mechanic. I'm pretty sure there are ways to do that mechanic on the mobile...but you can't emulate.

It could make for a good business model. You are taking out the cost of design, story, art direction, everything.. because you're licensing it.

You are basically creating a game engine company that licenses assets, story and gameplay.

There already are. Feral Interactive is one I can think of off the top of my head.

Browsing the source brought me back to the bad old days of using C macros to fake things like exception handling, messaging etc.

    #define kLuxOnMessage(x)   return true;} else if(aEvent == eLuxEnemyStateEvent_Message && apMessage && apMessage->mType == x){

Debugging misplaced braces, parentheses (or anything) in stuff like this was not fun.

Am I the only one who's kind of terrified at the inconsistent formatting throughout the codebase?

Maybe keeping the devs terrified helps making terrifying games? :D

No Arch Linux package yet!

> We encourage everyone to contribute code to this project

Why though?

Reply on news.ycombinator.com