Eli5 why cant we create video game graphics as good as cinematics? shouldnt the texturing be the same as making video games?
In: Technology
For CGI in movies, an entire team of artists spend hours or days on each scene, tweaking the animations by hand (or using motion-capture on actors) to look fluid and natural. They then send the data to a cluster of supercomputers that can spend hours, or even days, rendering a single frame of the resulting video (*Toy Story 4* took between 60 and 160 hours to render each frame of the movie, and there are 24 frames per second).
Compare that to a video game, where the animation has to be programatic (so that what the character does can respond to user input), so there are a limited number of actions and movements that a character can make, and all of them have to be predetermined. Plus, each fame has to be able to be rendered on a fractions of a second on consumer-grade hardware.
There’s just no possible way to get movie-quality visuals in a real-time medium like a video game, because the game will always be running on hardware with a fraction of the power, and won’t have a team of artists hand-animating character movements (or motion-capturing actors) in response to every single button press.
Texturing is indeed the same in gameplay vs cut-scenes. However in the cutscene the developers can set everything up ahead of time, the camera angle, the background, the lighting shadows on the characters faces. The scene in the cutscene doesn’t need to be a fully realized environment in which characters can walk around and interact with things because it’s scripted.
In contrast, during actual gameplay, your hardware has to calculate everything as you go.
The thing about 3D movies is that they are done in a few steps. Create high resolution models and environment, animate the models, record the scenes, the transfer to 2D frames for the movie. Each different scene is made separately and put together eventually to make the final movie.
This make the movie high quality and smooth in the end, as they can adjust the movie frame by frame too if need be, thereby, they can remove the lag.
In the game, you can’t have that high resolution model and environment without lagging due to poor optimization. And the average player does not have movie studio quality beefy PC.
In fact, I bet if a game with movie quality models and textures are made, even a gaming pc will struggle to run it.
A frame in a video game must render on a $500 computer in 1/30th of a second or less.
A frame in a movie is rendered on a six or seven figure rendering supercomputer and can take however long it needs.
Cinematics are pre-rendered, the game is essentially just playing a movie clip. This isn’t very taxing for the computer. The game’s real graphics have to be rendered in real time, which is very taxing. You have to draw all the textures, calculate all the lighting, particle effects and so on 30/60 times per second.
By cinematics, I assume you mean movies. The main difference is that video games need to render frames in real time, whilst movies are filmed either with a real set, or CGI which takes many hours of work for a few seconds of footage.
Cinematics (in video games) and CGI (in movies) are pre-rendered. That means they can build out the scene and graphics before the movie is released.
The graphics you see in video games as you are playing them are rendered in realtime, so you can’t get as much detail.
As an example, the movie Final Fantasy: The Spirits Within has 141,964 total frames. Each frame took roughly 90 minutes to render, compared to 0.015 seconds that a single frame from your favorite video game at 60 fps has to render.
its not about the texturing.
Rendering for a videogame has to be done in Real-time, and in order to have a playable experience you need ot maintain an acceptable number of frames drawn per second, and you must od this while also processing the game’s logic, and input(calculating physics and other player’s input as well depending on the type of game).
this is extremely processor intensive on matter how you put it, so the compromise is ot lower image fidelity.
this is not an issue in cinematic rendering because everything is being pre-processed and then rendered as a video file as long as you have the time and the detail you can render as complex of a scene as required.
Latest Answers