The 4080 12GB shouldn't be called 4070?
The 16GB version is AD103-300 with 72 SM, the 12 is AD104-400 with 60 SM.
The 16GB version is AD103-300 with 72 SM, the 12 is AD104-400 with 60 SM.
The 4080 12GB shouldn't be called 4070?
The 16GB version is AD103-300 with 72 SM, the 12 is AD104-400 with 60 SM.
A name is but a name. Historically, it is true that the xx80 SKU has been the 104-class die, but 104 is just a name. More importantly, it was the 2nd largest die in the stack. There's a 103-class die now, so it should be the 4080, not the 3rd largest die.Gone over this a thousand times already... it would be the 4070 Ti if it was the same as Ampere. But the three previous gens, the xx80 was the full xx104.
Obviously charging $899 to start, they felt like they couldn't use the 4070 Ti.
NVIDIA taking care of their customers, after all they still have FSR and XeSSNo DLSS 3.0 for 30 series
EU price
Correct. I have the exact same skepticism as well, because at the end of the day it is doing interpolation, meaning it can ONLY work with the data available to it, i.e. the interpolated frames will not look any better than the key frames. The only advantage I see is that Nvidia uses more input data to perform the interpolation, meaning particle effects and more erratic motion is probably handled better than what your TV does with motion interpolation. Outside of that, in theory the fps boost is entirely fictional because if the frame is stationary, the fps can be artificially boosted by just inserting as many interpolated frames as desired (limited by GPU power I guess). Why not just advertise "Up to 10x performance increase**" then?I'm extremely skeptical of DLSS 3. It's a pure post process effect, meaning no input but output frames. Essentially this is what many TVs do to smooth low frame rate jitter, by creating intermediate frames:
NVIDIA Introduces DLSS 3 With Breakthrough AI-Powered Frame Generation for up to 4x Performance
NVIDIA today announced NVIDIA® DLSS 3, an AI-powered performance multiplier that kicks off a new era of NVIDIA RTX™ neural rendering for games and applications.nvidianews.nvidia.com
This kind of thing is prone to artifacts, and it should be obvious, that it can't it will only have (at best) the input lag of the original lower frame rate.
Powered by new fourth-generation Tensor Cores and a new Optical Flow Accelerator on GeForce RTX 40 Series GPUs, DLSS 3 is the latest iteration of the company’s critically acclaimed Deep Learning Super Sampling technology and introduces a new capability called Optical Multi Frame Generation.
Optical Multi Frame Generation generates entirely new frames, rather than just pixels, delivering astounding performance boosts. The new Optical Flow Accelerator incorporated into the NVIDIA Ada Lovelace architecture analyzes two sequential in-game images and calculates motion vector data for objects and elements that appear in the frame, but are not modeled by traditional game engine motion vectors. This dramatically reduces visual anomalies when AI renders elements such as particles, reflections, shadows and lighting.
Pairs of super-resolution frames from the game, along with both engine and optical flow motion vectors, are then fed into a convolutional neural network that analyzes the data and automatically generates an additional frame for each game-rendered frame — a first for real-time game rendering. Combining the DLSS-generated frames with the DLSS super-resolution frames enables DLSS 3 to reconstruct seven-eighths of the displayed pixels with AI, boosting frame rates by up to 4x compared to without DLSS.
Because DLSS Frame Generation executes as a post-process on the GPU, it can boost frame rates even when the game is bottlenecked by the CPU. For CPU-limited games, such as those that are physics heavy or involve large worlds, DLSS 3 allows the GeForce RTX 40 Series GPUs to render the game at up to twice the frame rate that the CPU is able to compute the game.
It will do wonders for games like MS Flight Sim, maybe even some strategy games. For anything fast paced though, I foresee soap opera effect. (puns intended)I'm extremely skeptical of DLSS 3. It's a pure post process effect, meaning no input but output frames. Essentially this is what many TVs do to smooth low frame rate jitter, by creating intermediate frames:
The full spec list:
View attachment 67829
MSI is going all out with its full range of APEX series GeForce RTX 4090 & RTX 4080 series graphics cards which include the next-gen SUPRIM X, SUPRIM Liquid, Gaming Trio and Ventus series cards.
OR, when Nvidia advertised Lovelace can do 90 TFLOPs, that was for the full AD102 die.This amount to 82.575 Tflops@450W, 90 Tflops require 2.75GHz.