H100 CNX vs GeForce RTX 4070
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 20 | not rated |
| Place by popularity | 32 | not in top-100 |
| Cost-effectiveness evaluation | 60.20 | no data |
| Power efficiency | 24.30 | no data |
| Architecture | Ada Lovelace (2022−2024) | Hopper (2022−2023) |
| GPU code name | AD104 | GH100 |
| Market segment | Desktop | Workstation |
| Release date | 12 April 2023 (1 year ago) | 21 March 2023 (1 year ago) |
| Launch price (MSRP) | $599 | no data |
Cost-effectiveness evaluation
Performance to price ratio. The higher, the better.
Detailed specifications
General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.
| Pipelines / CUDA cores | 5888 | 14592 |
| Core clock speed | 1920 MHz | 690 MHz |
| Boost clock speed | 2475 MHz | 1845 MHz |
| Number of transistors | 35,800 million | 80,000 million |
| Manufacturing process technology | 5 nm | 4 nm |
| Power consumption (TDP) | 200 Watt | 350 Watt |
| Texture fill rate | 455.4 | 841.3 |
| Floating-point processing power | 29.15 TFLOPS | 53.84 TFLOPS |
| ROPs | 64 | 24 |
| TMUs | 184 | 456 |
| Tensor Cores | 184 | 456 |
| Ray Tracing Cores | 46 | no data |
Form factor & compatibility
Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).
| Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |
| Length | 240 mm | 267 mm |
| Width | 2-slot | 2-slot |
| Supplementary power connectors | 1x 16-pin | 8-pin EPS |
VRAM capacity and type
Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.
| Memory type | GDDR6X | HBM2e |
| Maximum RAM amount | 12 GB | 80 GB |
| Memory bus width | 192 Bit | 5120 Bit |
| Memory clock speed | 1313 MHz | 1593 MHz |
| Memory bandwidth | 504.2 GB/s | 2,039 GB/s |
| Shared memory | - | - |
Connectivity and outputs
Types and number of video connectors present on the reviewed GPUs. As a rule, data in this section is precise only for desktop reference ones (so-called Founders Edition for NVIDIA chips). OEM manufacturers may change the number and type of output ports, while for notebook cards availability of certain video outputs ports depends on the laptop model rather than on the card itself.
| Display Connectors | 1x HDMI 2.1, 3x DisplayPort 1.4a | No outputs |
| HDMI | + | - |
API compatibility
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 Ultimate (12_2) | N/A |
| Shader Model | 6.7 | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.3 | N/A |
| CUDA | 8.9 | 9.0 |
Pros & cons summary
| Maximum RAM amount | 12 GB | 80 GB |
| Chip lithography | 5 nm | 4 nm |
| Power consumption (TDP) | 200 Watt | 350 Watt |
RTX 4070 has 75% lower power consumption.
H100 CNX, on the other hand, has a 566.7% higher maximum VRAM amount, and a 25% more advanced lithography process.
We couldn't decide between GeForce RTX 4070 and H100 CNX. We've got no test results to judge.
Be aware that GeForce RTX 4070 is a desktop card while H100 CNX is a workstation one.
Should you still have questions concerning choice between the reviewed GPUs, ask them in Comments section, and we shall answer.
Comparisons with similar GPUs
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
