The specifications of NVIDIA’s GeForce RTX 3070 graphics card have been confirmed in a list on Geekbench. The list was noted by TUM_APISAK and shows the next-generation RTX 30-series mobility GPU running in the OpenCL test suite.
NVIDIA GeForce RTX 3070 Mobility GPU to Contain GA104 GPU with 5120 cores, 8 GB memory and 1300 MHz clock in Max-Q variant, similar to an RTX 2080 SUPER
Just a few days ago, we had to see the first offerings for the NVIDIA GeForce RTX 30 Series Mobility Series, as well as some preliminary specifications. It seems that the specifications will possibly be true, as it corresponds to that of today’s leak. Based on the clock speed alone, we can also see that it is the Max-Q variant of the GeForce RTX 3070 instead of the Max-P variant that offers much higher clock speeds.
As for the specifications, the NVIDIA GeForce RTX 3070 Mobility will contain 40 SMs on the GA104 GPU SKU (presumably GA104-770). This chip has a total of 5120 core points and a maximum clock speed of up to 1300 MHz (~ 1290 MHz). The graphics card also contains 8 GB of VRAM running over a 256-bit wide bus interface, but it remains to be seen whether NVIDIA will use 12 or 14 Gbps DRAM modules for its mobility setup. In the case of 14 Gbps, the GPU can offer a maximum bandwidth of 448 GB / s.
NVIDIA GeForce RTX 30 Mobility GPU Setup:
| GPU name | NVIDIA GeForce RTX 3060 | NVIDIA GeForce RTX 3070 | NVIDIA GeForce RTX 3080 |
|---|---|---|---|
| Process node | Samsung 8nm | Samsung 8nm | Samsung 8nm |
| GPU-SKU | GA106 | GA104-770 | GA104-775 |
| SM’s | 24 | 40 | 48 |
| CUDA core | 3072 | 5120 | 6144 |
| Base Clock | ~ 900 MHz | ~ 1100 MHz | ~ 1100 MHz |
| Boost Clock | ~ 1700 MHz | ~ 1620 MHz | ~ 1700 MHz |
| Memory clock | 14 Gbps | 14 Gbps | 14 Gbps |
| Memory type | GDDR6 | GDDR6 | GDDR6 |
| Memory Size | 6 GB | 8 GB | 8/16 GB |
| Memory Bus | 192-bis | 256-bis | 256-bis |
| Bandwidth | 336 GB / s | 448 GB / s | 448 GB / s |
| TGP | 80-115W 60-70W |
115-150W 80-90W |
115-150W 80-90W |
| Configurations | Max-Q Max-P |
Max-Q Max-P |
Max-Q Max-P |
| Start | Q1 2021 | Q1 2021 | Q1 2021 |
Preliminary specifications derived from NotebookCheck also indicate that the Max-Q variant will have a TGP between 80-90W, while the faster Max-P variant will have a TGP of 115-150W. The NVIDIA GeForce RTX 3070 Mobility was tested on an HP Omen 15 laptop that also installed the Intel Core i7-10870H and 16GB of DDR4 memory.
In terms of performance, the GPU scored 110839 points in the OpenCL test. The score puts the GeForce RTX 3070 mobility on par with the GeForce RTX 2080 SUPER, but at the same time it is 17% slower than the desktop GeForce RTX 3070, which can be expected as the variant not only offers lower core but also comes with much lower watches a Max-Q design. It seems that the Max-P variant should reduce the gap and bring the performance difference below 10 percent, but it remains to be seen.
Laptop manufacturers also offer luxurious and enthusiastic configurations of NVIDIA GeForce RTX 30 Amper GPUs with AMD’s Ryzen 5000H ‘Cezanne-H’ CPUs and Intel’s Tiger Lake-H 11th Gen CPU Series at CES 2021.