Nvidia GeForce RTX 3070 8GB
- Cores: 5888 CUDA CORES
- Base Clock: 1500 MHz
- Boost Clock: 1725 MHz
- Memory: 8GB GDDR6
- VR Ready: YES
- Resolutions Supported: 7680x4320
Nvidia GeForce RTX 3070 8GB Features
The Nvidia GeForce RTX 3070 8GB features the Ampere 8nm architecture and has 5888 CUDA CORES. It also has a base clock of 1.50 GHz (1500 MHz) along with a boost clock of 1.73 GHz (1730 MHz). Inside there are 17.4 Billion transistors, 184 texture units and 96 ROPs.
This card has a memory size of 8 GB GDDR6 running at a memory clock effective of 14000 MHz. It also has a memory interface width of 256-bit and a memory bandwidth of 448.0 GB/s.
Nvidia’s GeForce RTX 3070 has a DirectX version of 12 and is G-sync compatible. Moreover, it supports OpenCL 2.0 , OpenGL 4.6 and shader model 6.5. It is not SLI compatible nor does it support RGB lighting but it is VR ready. More technologies include: NVIDIA DLSS, PCI Express Gen 4, NVIDIA GeForce Experience, NVIDIA Ansel, NVIDIA FreeStyle, NVIDIA ShadowPlay, NVIDIA Highlights, Game Ready Drivers, NVIDIA Studio Drivers, NVIDIA GPU Boost, Vulkan RT API, HDMI 2.1, DisplayPort 1.4a.
The maximum digital resolution that can be supported is 7680×4320. As for the I/O ports, the card has a display port 1.4a and HDMI 2.1. The GeForce RTX 3070 has a graphics card power of 220 W and has a PSU recommendation of 650 W. The Nvidia GeForce RTX 3070 has the interface PCIe 4.0 x16 and will require a PCIe 8-pin.
Its dimensions are 9.5″ (242 mm) x 4.4″ (112 mm) x Dual-slot and was released on October 29th 2020 at the starting price of $499.
View more variants of the Nvidia GeForce RTX 3070 8GB here.
Specifications
Purchase This Card
| Amazon | View Price At Amazon |
General
| Manufacturer | Nvidia |
| Type | Desktop |
| Series | GeForce 30 |
| Variant | RTX 3070 |
| GPU Architecture | Ampere |
| Code Name | GA104-300-A1 |
| Fabrication Process | Samsung 8nm |
| Die Size | 392.5 mm² |
| Base Clock | 1500 MHz |
| Boost Clock | 1725 MHz |
| Transistor Amount | 17.4 Billion |
Cores & Texture Config
| Cores | 5888 CUDA CORES |
| Texture Units | 184 |
| ROPs | 96 |
| Ray Tracing Cores | 46 |
| Tensor Cores | 184 |
| Compute Units | - |
| L2 Cache | 4MB |
| Infinity Cache | - |
Memory
| Memory Size | 8GB |
| Memory Type | GDDR6 |
| Memory Clock | 1750 MHz |
| Memory Clock Effective | 14 Gbps |
| Memory Interface Width | 256-bit |
| Memory Bandwidth | 448.0 GB/s |
Performance
| Half Precision FP16 | 20.3136 TFLOPS |
| Single Precision FP32 | 20.3136 TFLOPS |
| Double Precision FP64 | 0.3174 TFLOPS |
Floating Point Operations Per Clock Cycle
| Half Precision FP16 | 2 |
| Single Precision FP32 | 2 |
| Double Precision FP64 | 1/32 |
Fillrate
| Texture (GT/s) | 317.4 |
| Pixel (GP/s) | 165.6 |
Technologies
| DirectX® Version | 12 |
| G-Sync/FreeSync | |
| OpenCL | 2.0 |
| OpenGL | 4.6 |
| Shader Model | 6.5 |
| Vulkan | 1.3 |
| SLI/Crossfire Ready | |
| VR Ready | |
| RGB Lighting | |
| Extra Technologies | NVIDIA DLSS, PCI Express Gen 4, NVIDIA GeForce Experience, NVIDIA Ansel, NVIDIA FreeStyle, NVIDIA ShadowPlay, NVIDIA Highlights, NVIDIA G-SYNC, Game Ready Drivers, NVIDIA Studio Drivers, NVIDIA GPU Boost, Vulkan RT API, HDMI 2.1, DisplayPort 1.4a |
Connectivity
| Maximum resolution | 7680x4320 |
| Max HDMI Resolution | 7680x4320 |
| Max DP Resolution | 7680x4320 |
| Display Port | 1.4a x3 |
| HDMI | 2.1 x1 |
| USB-C | No |
| Dual Link DVI | No |
| Single Link DVI | No |
| VGA | No |
Requirements
| Interface | PCIe 4.0 x16 |
| Power Connectors | 1x PCIe 8-pin |
| PSU Recommendation | 650W |
| Graphics Card Power | 220W |
Dimensions
| Length | 9.5" (242 mm) |
| Width | 4.4" (112 mm) |
| Height | 2 Slot |
| Slot Width | 2 Slot |
Extra Information
| Price (at release) | $499 |
| Weight | 1.1 kg |
| Released | 29 October, 2020 |
Offers
Price History
Reviews
Disclaimer Note
The information provided is for reference only and may not be 100% accurate
Recently Added
Best Selling Graphics Cards