Las Vegas' dystopia-sphere, powered by 150 Nvidia GPUs and drawing up to 28,000,000 watts, is both a testament to the hubris of humanity and an admittedly impressive technical feat | PC Gamer
Las Vegas' dystopia-sphere, powered by 150 Nvidia GPUs and drawing up to 28,000,000 watts, is both a testament to the hubris of humanity and an admittedly impressive technical feat | PC Gamer
The Las Vegas Sphere might have an impressive spec sheet, but its energy demands are almost as staggering.
You're viewing a single thread.
Wouldn't just one GPU be enough to run the Sphere, or a I getting something wrong?
I remember hearing about that it's not exactly high resolution, each "pixel" being a bunch of pretty large lamps.
Wikipedia says it's 16,000x16,000 (which is way less than I thought). The way the math works, that's 16x as big as a 4k monitor, so 16 GPUs would make sense. And there's a screen inside and one outside, so double that. But I also can't figure out why it needs five times that. Redundancy? Poor optimization? I dunno.
But wouldn't that be only necessary if it needed to render real-time graphics at such a scale? If I'm correct, all its doing is playing back videos.
I think it's doing some non-trivial amount of rendering, since it's often syncing graphics with music played live.
Even if it's just playing back videos, it still should compensate for the distortion of the spherical display. That's a "simple" 3d transformation, but with the amount of pixels, coordinating between the GPUs and some redundancy, it doesn't seem like an excessive amount of computing power. The whole thing is still an impressive excess though...
Live audio visualization in game engines is definitely a thing ex. https://youtu.be/IZL7VAt97ws?si=H74SwrLZYfsYNTY8
I'm guessing it's the department of redundancy department, is my guess.
Someone elsewhere in the thread suggested it might be a marketing thing on Nvidia's part, and that makes a lot of sense.
I work for a digital display company, and it is definitely redundancy. There will be at least two redundant display systems that go to the modules separately so they can switch between them to solve issues. If a component fails on one side they just switch to the other.
Ah, nice. Thank you for bringing your expertise to my nonsense.
The way I think it, it's possible a really small number of GPUs would be enough to render the framebuffer, you'd just need an army of low-power graphics units to receive the data and render it on screens.
Having a high-power GPU for every screen is definitely a loss unless the render job is distributed really well and there's also people around to admire the results at the distance where the pixel differences no longer matter. Which is to say, not here.