Viscous Soda posted:

Watch the video in the linked bug report, the bug seems to make you immune to the airlock "kill zone". They just cannot seem to stop naked men from running around in space.

Dementropy posted:

Trillhouse posted:

Le subtweet

Trillhouse posted:

Le subtweet

Paladinus posted:

Shopping for a new sound card. Should I buy Sound Blaster AWE64 for full in-game experience or would AWE32 do?

Beexoffel posted:

Star Citizen: They just cannot seem to stop naked men from running around in space.

Edit: Astrokat taxxe.

Dementropy posted:

Mirificus posted:

Mirificus posted:

Tippis posted:

Mirificus posted:

Hav posted:

Can I point out that simply upvoting this for visibility is brigading, and you should have an entirely *legitimate* reason for wanting to see Romero's response to that.

Mirificus posted:

Mirificus posted:

Foo Diddley posted:

Not the answer that Chris would have given

Foo Diddley posted:

Not the answer that Chris would have given

Armadillo Tank posted:

https://kotaku.com/in-the-future-hideo-kojima-says-his-studio-will-make-f-1839603527

So kojima is doing the chris roberts thing and wanting to make movies. Also early reviews of his singular vision video game are either "This is like liquid gold" or "This is loving piss".

Dementropy posted:

Scruffpuff posted:

I'd lose my poo poo if Romero responded along the lines of "as a game developer, Star Citizen is not my area of expertise." The entire question isn't because the poster wants John's opinion, he just wants to link the phrase "game developer" to the phrase "Star Citizen" in a bid to legitimize it.

Scruffpuff posted:

So I played MGS5 phantom pain due to all the reviews assuring me it was just incredibly amazing you guys. So I install and play, crawl around, go through tutorials, send sheep up on balloons, shooting people, the whole 9. After what seemed like 40 hours in I started to think "this is the longest goddamned video game tutorial I've ever goddamned experienced, I hope the game starts up in earnest soon." Eventually I realized I had been playing the game proper the entire time, it just felt like a tutorial because everything was all presentation with no substance, and a meaningless story that is "deep" if you're the type who can drown in a cup of water, but insipid if you're more your standard regular person. No technical quibbles that I can recall, just sleep-inducing as poo poo, and an army of people ready to tell you that you don't "get it" if you think the story is anything less than stellar. Hot tip for those people: if you "get" something then by logical deduction it's necessarily simple.

Parallel Component Loading and Initialization posted:

With all the groundwork done we have:
•Levels split into building block (the Object-Container)
•Entities (which make up a large part of the Object-Container) implemented without LUA and as components
•All entity runtime state organized in Serialized-Variables for easier state communication via network
•Multi-threaded creation of engine-side resources (textures, meshes, characters etc.)

The next steps are building on that foundation. Our first goal was to actually load entities in parallel to the game simulation. This first step didn’t include any high-level logic of what to load or unload or when yet, so it was a very dumb streaming system. Nevertheless, it already reduced the runtime stuttering (e.g. spawning a ship at runtime no longer required us to run the initialization code of the ship’s entities as part of the game simulation).

At that time, we had roughly 300 to 400 different component types. If we had tried to execute all those in parallel from the start, we would have drowned in bugs. Therefore, we had to develop a system allowing us to incrementally execute more and more component types in parallel to the game simulation. The more component types we made safe for parallel loading, the less the game simulation would stutter.

The solution we have chosen is utilizing Fibers. A Fiber is an execution state where we can control exactly when and where we want to execute it (parallel loading or game simulation). While Fibers can be very tricky to use, they provided exactly the control we needed. With Fibers, we could move the logic for resource loading between concurrent loading threads and game simulation threads, depending if the component type supported parallel loading or not. And with that, it was possible to step-by-step adjust more and more code to run in parallel while ensuring that everyone uses (and thus tests) the already parallelized code. Those changes were partially rolled out with Alpha 3.2, where they reduced the stuttering caused by loading entity resources at runtime, like spawning ships.

Abstract posted:

Fibers (sometimes called stackful coroutines or user mode cooperatively scheduled threads) and stackless coroutines (compiler synthesized state machines) represent two distinct programming facilities with vast performance and functionality differences.

This paper highlights efficiency, scalability and usability problems of fibers and reaches the conclusion that they are not an appropriate solution for writing scalable concurrent software

Case Studies posted:

We have accumulated more than a quarter century of experience with fibers across variety of programming languages and platforms. In the 90s fibers and N : M scheduling looked promising, now, with improvements in hardware, operating system kernel and painful experience of trying to make the fibers work has resulted in a recommendation: DO NOT USE FIBERS! Use threads instead and/or write your code using asynchronous APIs with hand-crafted state machines.

Fiber Use on Windows posted:

Current recommendation is to avoid using fibers and UMS. This advice from 2005 remains unchanged:

“... [I]nstead of spending your time rewriting your app to use fibers (and it IS a rewrite), instead it's better to rearchitect your app to use a "minimal context" model -instead of maintaining the state of your server on the stack, maintain it in a small data structure, and have that structure drive a small one-thread-per-cpu state machine.”

Solaris posted:

After exploring N:M programming model for 7 years, Sun Microsystem abandoned it in favor of simpler 1:1 threading.

Linux posted:

Similarly, after exploring N:M model for a bit, Linux kernel adopted 1:1 threading since kernel 2.6

POSIX posted:

After providing facilities for user-mode stack switching (ucontext_t) for 8 years, POSIX deprecates and removes it.

Facebook Experience posted:

Facebook has deployed fibers internally for 5 years and now “... looking to migrate away from it as soon as we are confident of the direction [stackless] Coroutines will go in in the standard. We already try to dissuade non-experts from using them as much as possible, but in the absence of anything better they occasionally do.”.

Programming Language survey posted:

The only programming language among top 10 most popular languages according to TIOBE index [TiobeIndex] that provides fibers is Go. Go programming language uses fibers successfully, but incurs a very high cost (~160ns) whenever it has to interoperate with C libraries, due to the need to swap in normal stack [GoOverhead]. That is not acceptable for C++ where the ability to seamlessly and efficiently interoperate with existing code is very important.

Conclusion posted:

While fibers may have looked like an attractive approach to write scalable concurrent code in the 90s, the experience of using fibers, the advances in operating systems, hardware and compiler technology (stackless coroutines), made them no longer a recommended facility.

Mirificus posted:

CIG: Jump Point - The Road to Object-Container-Streaming


Nishanov: Fibers Under the Magnifying Glass

Blue On Blue posted:

With all the ground work done

I stopped reading after that

These loving dev's have been finishing the ground work on everything for 7 years, and they never go past that point

Blue On Blue posted:

With all the ground work done

I stopped reading after that

These loving dev's have been finishing the ground work on everything for 7 years, and they never go past that point

Mr.PayDay posted:

Maybe the Kickstarter should have added an asterisk where they honestly wrote at the end that they will need to learn game and engine development for the first decade after the stretch goals are locked.

Mirificus posted:

CIG: Jump Point - The Road to Object-Container-Streaming


Nishanov: Fibers Under the Magnifying Glass

quote:

The solution we have chosen is utilizing Fibers. A Fiber is an execution state where we can control exactly when and where we want to execute it (parallel loading or game simulation). While Fibers can be very tricky to use, they provided exactly the control we needed. With Fibers, we could move the logic for resource loading between concurrent loading threads and game simulation threads, depending if the component type supported parallel loading or not. And with that, it was possible to step-by-step adjust more and more code to run in parallel while ensuring that everyone uses (and thus tests) the already parallelized code. Those changes were partially rolled out with Alpha 3.2, where they reduced the stuttering caused by loading entity resources at runtime, like spawning ships.

quote:

TECHNOLOGICAL LIMITS OBJECT-CONTAINER-STREAMING MUST OVERCOME
But what implications does that goal have for the engine technology? Foremost, a videogame must refresh its screen at least 30 times a second to give the impression of a fluent experience.

quote:

Each frame, the game simulation asks the character manager to update all loaded characters. And here it becomes tricky. If our character resource loading finishes in parallel to the character update, it wants to put the newly loaded character into the character manager’s list, so that the character will receive updates in the future. Putting the character into the list modifies it. If the game simulation is using this list at the same time to update the characters, this can result in a data corruption and crash the game

Zaphod42 posted:

WHAT




WHAAAAAAAAAAAAAAAAAAAAAAAT

NO

NOOOOO


THAT'S NOT, THAT'S WRONG

WHY


Ah yes, targeting 30fps, very nice SC

Mirificus posted:

CIG: Jump Point - The Road to Object-Container-Streaming


Nishanov: Fibers Under the Magnifying Glass

Foo Diddley posted:

Not the answer that Chris would have given

trucutru posted:

Translation: on-the-fly loading and unloading of complex multiplayer entities to servers is hard, instead of developing a proper architecture before we started making a billion entities, we have instead decided to use extremely complicated and inefficient *computer magic* to figuratively pound a square peg into a round (rear end)hole to solve the issue. Early days

Scruffpuff posted:

So I played MGS5 phantom pain due to all the reviews assuring me it was just incredibly amazing you guys. So I install and play, crawl around, go through tutorials, send sheep up on balloons, shooting people, the whole 9. After what seemed like 40 hours in I started to think "this is the longest goddamned video game tutorial I've ever goddamned experienced, I hope the game starts up in earnest soon." Eventually I realized I had been playing the game proper the entire time, it just felt like a tutorial because everything was all presentation with no substance, and a meaningless story that is "deep" if you're the type who can drown in a cup of water, but insipid if you're more your standard regular person. No technical quibbles that I can recall, just sleep-inducing as poo poo, and an army of people ready to tell you that you don't "get it" if you think the story is anything less than stellar. Hot tip for those people: if you "get" something then by logical deduction it's necessarily simple.

Saladin Rising posted:

CI!G: "We are doing a new technology, FIBERS! Surely this will be the magic silver bullet we have been searching for! What do you mean 'Linux, Facebook and Sun Microsystems tried to make them work for years and then moved to something else'? Bah, they just didn't try hard enough."

This post is great, and it's CI!G in a nutshell: every single thing about fibers is a warning that they're too much of a pain in the rear end and not worth it, and that you shouldn't use them. So what is CI!G's new plan? Well obviously they're going to bash fibers into their overloaded zombie of a code base and hope it works somehow. Then in a few months (or years) when they slowly realize fibers won't do what they're blindly hoping they will do, CI!G will eventually abandon them and move to the next snake oil fix. Meanwhile their code base has become that much more impossible to fix because they blindly jammed fibers into it hoping it would magically fix everything.

Chris Roberts really is stuck in the loving 90's, and we're now in year 8 of him being confused that things didn't just scale linearly from the last time he actually worked on a Wing Commmander game or Freelancer, or that DVD-quality cut-scenes do not single-handedly make a game impressive anymore (if they ever did).