KitGuru KitGuru.net – Tech News | Hardware News | Hardware Reviews | IOS | Mobile | Gaming | Graphics Cards
Hot on the heels of our RTX 3080 Ti review, today we are back with a look at Nvidia's RTX 3070 Ti. While the core-count has only increased slightly from the RTX 3070, the RTX 3070 Ti is able to deliver a 36% boost to memory bandwidth thanks to its use of GDDR6X memory, instead of the slower GDDR6 modules found on the original model. We put this new GPU through its paces to see what sort of performance it can offer.
As always for a new Nvidia launch, here we will be assessing the RTX 3070 Ti Founders Edition, which actually uses a new cooler design we have not seen before on either the RTX 3070 or RTX 3080 Founders Edition models. As with our RTX 3080 Ti review however, we will also be comparing the Founders Edition against a couple of Nvidia's AIB partner cards.
The first of these is Inno3D's RTX 3070 Ti X3 OC, a dual-slot model with a triple-fan cooler. We also test MSI's beefy RTX 3070 Ti Suprim X, offering a gargantuan cooler, increased power limit as well as dual-BIOS functionality.
| RTX 3090 | RTX 3080 Ti | RTX 3080 | RTX 3070 Ti | RTX 3070 | |
| SMs | 82 | 80 | 68 | 48 | 46 |
| CUDA Cores | 10496 | 10240 | 8704 | 6144 | 5888 |
| Tensor Cores | 328 | 320 | 272 | 192 | 184 |
| RT Cores | 82 | 80 | 68 | 48 | 46 |
| Texture Units | 328 | 320 | 272 | 192 | 184 |
| ROPs | 112 | 112 | 96 | 96 | 96 |
| GPU Boost Clock | 1695 MHz | 1665 MHz | 1710 MHz | 1770 MHz | 1725 MHz |
| Memory Data Rate | 19.5 Gbps | 19 Gbps | 19 Gbps | 19 Gbps | 14 Gbps |
| Total Video Memory | 24GB GDDR6X | 12GB GDDR6X | 10GB GDDR6X | 8GB GDDR6X | 8GB GDDR6 |
| Memory Interface | 384-bit | 384-bit | 320-bit | 256-bit | 256-bit |
| Memory Bandwidth | 936 GB/Sec | 912 GB/Sec | 760 GB/Sec | 608 GB/Sec | 448 GB/Sec |
| TGP | 350W | 350W | 320W | 290W | 220W |
The RTX 3070 Ti uses Nvidia’s GA104 GPU, but unlike the RTX 3070, it is a fully populated die. That means the RTX 3070 Ti incorporates 48 Streaming Multiprocessors (SMs), and thanks to Ampere’s new SM structure with its two FP32 datapaths, each SM houses 128 CUDA cores, giving a total of 6144. This is only 256 more cores compared to the RTX 3070, an increase of 4%
Ampere also places one RT core and four Tensor cores in each SM, giving a total of 48 RT cores and 192 Tensor cores. This is accompanied by 192 texture units and 96 ROPs which are now housed directly within each graphics processing cluster (GPC), with 16 ROPs per GPC, and 6 GPCs in total for RTX 3070 Ti.
Clock speed is another area where Nvidia has tweaked things compared to the vanilla RTX 3070. The RTX 3070 Ti Founders Edition has a reference clock of 1770MHz, an increase of 45MHz against the RTX 3070, though both of the partner cards we test come factory overclocked. We test real-world clock speed behaviour later in this review.
For the memory, this is where we see the biggest upgrade versus the RTX 3070. While we still find the same 256-bit memory interface, in place of the slower GDDR6 modules of the RTX 3070, Nvidia has opted for GDDR6X memory clocked at 19Gbps. This offers total memory bandwidth at 608 GB/s, a 36% increase over the RTX 3070.
Lastly, for total graphics power, Nvidia rates the RTX 3070 Ti for 290W, which is a sizable increase over the RTX 3070's 220W TGP, with most of that likely going towards the GDDR6X memory. We are using our new GPU power testing methodology in this review, so read on for our most detailed power testing yet.
Be sure to check out our sponsors store EKWB here
