Usually, we put off evaluation of products until their premiere and tests, but sometimes you can already see on paper that we’ll deal with disappointment. A good example of this is the GeForce RTX 4080 12 GB, which sparked a lot of controversy from the very beginning. All because of the relatively low specifications, high price and performance expected at the level of the RTX 3090 Ti. Fortunately, the manufacturer has dropped the idea of bringing the card to market.
NVIDIA admits that having two 4080 graphics cards is baffling and the 12GB VRAM model won’t eventually be released. However, we expect the arrangement in question will come back to us in the near future, but with a more favorable flag.
NVIDIA GeForce RTX 4080 16GB offers 30% higher performance than 12GB RTX 4080 chipset
NVIDIA seems to be listening to the community. The Green Party’s latest statement is surprisingly honest. On the official website of the manufacturer, we can read that Having two graphics cards named 4080 is confusing. Finally, the 12GB VRAM model will not be released. By the way, it was confirmed that the “real” RTX 4080 with 16 GB VRAM will be in the hands of gamers and enthusiasts on November 16 this year. Given the great reviews of the RTX 4090, NVIDIA expects a successful release for the next card in the series.
NVIDIA GeForce RTX 4080 12GB should be the GeForce RTX 4070. Looks bad and bodes bad for future generations
Let us remind you that the 12 GB NVIDIA GeForce RTX 4080 graphics card will debut at a price of PLN 5299. The results of recent tests published by the manufacturer showed 30% lower performance compared to the 16 GB GeForce RTX 4080. Despite the cancellation of the premiere, we expect that the system discussed will return to us in the near future, but with a more appropriate tag (such as GeForce RTX 4070 or RTX 4070 Ti). Who knows, maybe we’ll also see a slightly more competitive price?
NVIDIA GeForce RTX 4090 | NVIDIA GeForce RTX 4080 16GB | ||
general engineering | Ada | Ada | |
drawing layout | M 102 – 300 | M 103 – 300 | |
lithography | TSMC “4N” | TSMC “4N” | |
number of transistors | 76.3 billion | 45.9 billion | |
Surface | 608 mm² | 378.6 mm² | |
SM . blocks | 128 | 76 | |
SP . units | 16384 | 9728 | |
FP32 قوة power | 82.6 TFLOPS | 49 TFLOPS | |
RT cores | 128 RT 3rd generation | 76 RT 3rd generation | |
tensor core | 512 fourth generation of motor | 304 fourth generation motor | |
TMUs | 512 | 304 | |
Royal Oman Police Units | 176 | 112 | |
L2 . cache | 96 MB | 64 MB | |
base clock | 2230 MHz | 2210 MHz | |
GPU Boost around the clock | 2520 MHz | 2505 MHz | |
VRAM bandwidth | 21 Gbps | 22.5 Gbps | |
VRAM | 24 GB GDDR6X | 16 GB GDDR6X | |
Memory bus | 384 bit | 256 bit | |
eligibility | 1008 GB/sec | 720 GB/sec | |
PCIe compliant | PCIe 4.0 x16 . slot | PCIe 4.0 x16 . slot | |
TDP (base) | 450 watts | 320 watts. | |
TDP(Max) | 660 watts | 516 watts. | |
DLSS 3 . support | yes | yes | |
Price in PLN | PLN 9699 | zloty 7049 | |
the first show | October 12 2022 | November 16, 2022 | canceled |
Source: NVIDIA
“Prone to fits of apathy. Introvert. Award-winning internet evangelist. Extreme beer expert.”