Nvidia shelves cheaper RTX 4080, admitting its name was ‘confusing’
polygon.comNvidia will be “unlaunching” the 12 GB model of its RTX 4080 graphics card, the company announced Friday. The news comes just weeks after the company revealed the product as part of a new line of GPUs. In a brief blog post, Nvidia did not provide much context for the decision beyond saying that the card was “not named right.”
Nvidia announced two — well, technically three — new GPUs in late September, the RTX 4090 and RTX 4080. The latter would be available in two versions, Nvidia said: a model with 16 GB of video memory and one with 12 GB.
Based on that, one might think that the RAM would be the only difference between the two RTX 4080 variants. However, that was not the case — the GPUs were to feature considerably different specifications. The 16 GB model has 9,728 CUDA cores, an increase of more than 26% over the 12 GB card’s 7,680 cores. And while both models had GDDR6X memory, the RAM in the 12 GB variant was slower, with lower memory bandwidth. As we put it in our write-up of Nvidia’s 40-series GPU announcement, it might be better to consider the 12 GB model “as an entry-level RTX 4080 that’s closer to a RTX 4070.5” rather than a true high-powered RTX 4080.
Related
“Having two GPUs with the 4080 designation is confusing,” Nvidia acknowledged Friday.
The two RTX 4080 models were priced according to their specs — the 16 GB version would start at $1,199, whereas the 12 GB model would start at $899. Nvidia’s decision to pull the cheaper card leaves only two extremely expensive GPUs in the company’s 40-series line: The RTX 4090, which launched earlier this week, starts at an eye-watering $1,599. It’s a lot to swallow when the steep prices of these new graphics cards have already led PC gamers to express concern over
Read more on polygon.com
