“The RTX 4080 12GB is a fantastic graphics card, but it’s not named right,” the company wrote in a short 71-word blog post (opens in new tab). “Having two GPUs with the 4080 designation is confusing,” the post continued, “so, we’re pressing the ‘unlaunch’ button on the 4080 12GB. “The RTX 4080 16GB is amazing and on track to delight gamers everywhere on November 16th,” the post concluded, before boasting about the around-the-block queues that the 4090 attracted at launch.
Confusing, how?
This is a surprising but welcome concession from Nvidia. While the branding of the two RTX 4080 cards suggested that the only difference was the quantity of RAM and the $300 price difference, the cheaper model actually offered considerably fewer CUDA cores, less memory bandwidth and the AD104 chip instead of the AD103. You can see the differing key specs between the three GPUs in the table below. As you can imagine, this would have led to markedly weaker performance, and Nvidia’s own benchmarks (opens in new tab) showed the 12GB model being around 30% slower than the 16GB version. That’s not a problem in and of itself — Nvidia is allowed to release cut-back versions of its cards, but the company has clearly conceded that putting the same 4080 badge on both boxes was a recipe for underwhelmed gamers. It’s entirely possible we’ll see this card reappear at a later date with “GeForce RTX 4070 Ti” written proudly on the packaging instead. But that won’t be in November, and by the time Nvidia does give whatever the 12GB RTX 4080 ends up as its debut, it may find the competition is quite a bit more fierce. AMD will be revealing its RDNA 3-powered line of GPUs at an event on November 3, after all, and you can’t discount Intel, even if its first generation of Arc GPUs haven’t yet forced Nvidia to break a sweat.
title: “12Gb Nvidia Rtx 4080 S November Launch Is Canceled” ShowToc: true date: “2022-12-24” author: “William Kmiecik”
“The RTX 4080 12GB is a fantastic graphics card, but it’s not named right,” the company wrote in a short 71-word blog post (opens in new tab). “Having two GPUs with the 4080 designation is confusing,” the post continued, “so, we’re pressing the ‘unlaunch’ button on the 4080 12GB. “The RTX 4080 16GB is amazing and on track to delight gamers everywhere on November 16th,” the post concluded, before boasting about the around-the-block queues that the 4090 attracted at launch.
Confusing, how?
This is a surprising but welcome concession from Nvidia. While the branding of the two RTX 4080 cards suggested that the only difference was the quantity of RAM and the $300 price difference, the cheaper model actually offered considerably fewer CUDA cores, less memory bandwidth and the AD104 chip instead of the AD103. You can see the differing key specs between the three GPUs in the table below. As you can imagine, this would have led to markedly weaker performance, and Nvidia’s own benchmarks (opens in new tab) showed the 12GB model being around 30% slower than the 16GB version. That’s not a problem in and of itself — Nvidia is allowed to release cut-back versions of its cards, but the company has clearly conceded that putting the same 4080 badge on both boxes was a recipe for underwhelmed gamers. It’s entirely possible we’ll see this card reappear at a later date with “GeForce RTX 4070 Ti” written proudly on the packaging instead. But that won’t be in November, and by the time Nvidia does give whatever the 12GB RTX 4080 ends up as its debut, it may find the competition is quite a bit more fierce. AMD will be revealing its RDNA 3-powered line of GPUs at an event on November 3, after all, and you can’t discount Intel, even if its first generation of Arc GPUs haven’t yet forced Nvidia to break a sweat.
