Old Nvidia gaming GPUs enjoy huge performance uplifts from new mod — DLSS 3 to FSR 3 mod enables frame generation to deliver up to 75% better performance on previous GeForce RTX GPUs
Old Nvidia gaming GPUs enjoy huge performance uplifts from new mod — DLSS 3 to FSR 3 mod enables frame generation to deliver up to 75% better performance on previous GeForce RTX GPUs
AMD Radeon GPU owners are out of luck for the current time being.
You're viewing a single thread.
But everyone here on Lemmy keeps telling me that AI is literally the devil? It's suddenly cool and ok if we use it for better performance in games?
There aren't any people who manually generate frames who's jobs get cut.
Nvidia is making great use of AI with their GPUs imo. Dlss 3.5 is impressive. Frame Gen is impressive.
Haven’t people been blaming DLSS for encouraging publishers to release poorly-optimized games, though?
Yes, the job it’s taking isn’t frame generation
It’s optimization jobs but more to what people don’t like; they are investing in this instead of better hardware
Whose*
If your job was cut because of AI your job was already replaceable. People still have to know how to effectively utilize the tech. Learn it and adapt like humans have done with every increase in automation.
On-device AI is not bad. If you can do tricks to actually improve things with AI, it is a tool like any other.
Stable diffusion is also on device but everyone here calls that bad. Trust me I'm with you on this but the hypocrisy of a large portion of the people I interact with on here is palpable.
"... Everyone here on [the instances startrek.website federated with and communities I subscribe to on] Lemmy ..."
This sort of generalization was already ignorant on Reddit, but it is doubly ignorant on the fediverse.
It's an accurate generalization. This isn't my only account. The larger technology communities are rampant with this thinking and you can even see it here in this thread.
It’s replacing one ai with a better ai so it’s neutral
It's a net positive, as are things like SD based models.
It's because all the AI craze these days are ChatGPT clones tacked onto other existing products, where all processing is done on the cloud. Here, having "AI" is the entire point of it, with no regard for if it actually improves anything. It's a solution in search of a problem.
In contrast, some products use AI as a means to an end. This distinction is important, because the focus is on the end result and AI is just a tool to achieve it. Here, it's a solution to a problem.
Just because you don't understand the problem solved doesn't mean there isn't one.
It is a very piwerful tool and what matyers is it's usecase
It's a great tool but it's still generative image ai. If you're fine with this but not with things like stable diffusion then you're a hypocrite
I mean I am fine with stable diffusion after relising that intellectual property is a lie