In the fall of 2020, AMD made a welcome return to the top tier of graphics cards. With the Navi 21 graphics circuit, the Radeon RX 6800, RX 6800 XT and RX 6900 XT were formed, the latter of which in some scenarios could master Nvidia’s Geforce RTX 3080. a counterattack to AMD’s unicorn.
Now Videocardz is putting its hands on a screenshot with a reading of the graphics card in the GPU-Z and where it breaks the cryptocurrency Ethereum. Worth mentioning is that Nvidia is expected to stifle cryptocurrency performance with the Geforce RTX 3080 Ti and future deliveries of the rest of the series. Then the graphics card is also not recognized by the GPU-Z and has a high hash rate at 118.9 Mh / s, Videocardz speculates that it may be an early copy without crypto limitation.
Nvidia Geforce RTX 3000 “Ampere”:
RTX 3090 | RTX 3080 Ti | RTX 3080 | RTX 3070 | RTX 3060 Ti | RTX 3060 | |
---|---|---|---|---|---|---|
Technical | 8 nm Samsung | 8 nm Samsung | 8 nm Samsung | 8 nm Samsung | 8 nm Samsung | 8 nm Samsung |
Circuit | GA102 | GA102 | GA102 | GA104 | GA104 | GA106 |
Circuit surface | 628 mm² | 628 mm² | 628 mm² | 392 mm² | 392 mm² | 276 mm² |
Transistors | 28.3 billion | 28.3 billion | 28.3 billion | 17.4 billion | 17.4 billion | 12 billion |
Architecture | Ampere | Ampere | Ampere | Ampere | Ampere | Ampere |
CUDA cores | 10 496 st. | 10 240 st. | 8 704 st. | 5 888 st. | 4 864 st. | 3 584 st. |
RT cores | 82 st. | 80 st. | 68 st. | 46 st. | 38 st. | 28 st. |
Tensor cores | 328 st. | 320 st. | 272 st. | 184 st. | 152 st. | 112 st. |
Texture units | 328 st. | 320 st. | 272 st. | 184 st. | 152 st. | 112 st. |
Raster units | 112 st. | 112 st. | 96 st. | 96 st. | 96 st. | 64 st. |
Clock frequency | 1 395 MHz | 1 365 MHz | 1 440 MHz | 1 500 MHz | 1 410 MHz | 1 320 MHz |
GPU Boost | 1 695 MHz | 1 665 MHz | 1 710 MHz | 1 725 MHz | 1 665 MHz | 1 777 MHz |
Computational power | 35 581 GFLOPS | 34 099 GFLOPS | 29 768 GFLOPS | 20 313 GFLOPS | 16 197 GFLOPS | 12 738 GFLOPS |
Memory amount | 24 GB GDDR6X | 12 GB GDDR6X | 10 GB GDDR6X | 8 GB GDDR6 | 8 GB GDDR6 | 12 GB GDDR6 |
Memory frequency | 19 500 MHz | 19 000 MHz(?) | 19 000 MHz | 14 000 MHz | 14 000 MHz | 15 000 MHz |
Memory bus | 384-bit | 384-bit | 320-bit | 256-bit | 256-bit | 192-bit |
Memory bandwidth | 936 GB/s | 912 GB/s(?) | 760 GB/s | 448 GB/s | 448 GB/s | 360 GB/s |
SLI connection | NVLink 3.0 x4 | ? | – | – | – | |
TBP | 350 W | 350 W(?) | 320 W | 220 W | 200 W | 170 W |
Launch price | 1 499 USD | 999 USD(?) | 699 USD | 499 USD | 399 USD | 329 USD |
The GPU-Z reading, on the other hand, reinforces previous information about the specifications for the Geforce RTX 3080 Ti. The number of CUDA cores is 10,240 and is thus negligibly less than with the flagship Geforce RTX 3090. Brand new is one of the last pieces of the puzzle – the clock frequencies, where the base level is 1,365 MHz and the boost level is 1,665 MHz. This, together with the number of cores, means a theoretical loss of performance of only four percent compared to RTX 3090.
On the memory front is a 384-bit memory bus with 12 GB of GDDR6X memory with an effective clock frequency of 19,000 MHz, for a theoretical bandwidth of 912 GB / s. Here, too, the difference is negligible compared to the RTX 3090, which with a marginally higher memory frequency is 936 GB / s. The only major difference between the models thus seems to be the amount of memory, where the flagship boasts a full 24 GB GDDR6X.
In terms of the graphics card’s power consumption, the Geforce RTX 3080 Ti is in the 278 W screenshot, which is due to the fact that the graphics card is undervolted. Exactly what level its specified power consumption (Total Board Power, TBP) should be is still a question mark, but it is most likely that it, like the Geforce RTX 3090, will be 350 W.
Finally, the price of the graphics card is expected to land at 999 USD, which is exactly the same level as for AMD’s Radeon RX 6900 XT. The launch of Nvidia’s next graphics card, which will be impossible to obtain, will take place in mid-May.