H100 PCIe 96 GB vs P102-100
Contents
Primary details
GPU architecture, market segment, value for money and other general parameters compared.
| Place in the ranking | 557 | not rated |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | 0.97 | no data |
| Power efficiency | 2.41 | no data |
| Architecture | Pascal (2016−2021) | Hopper (2022−2024) |
| GPU code name | GP102 | GH100 |
| Market segment | Workstation | Workstation |
| Release date | 12 February 2018 (7 years ago) | 21 March 2023 (2 years ago) |
| Launch price (MSRP) | $599 | no data |
Cost-effectiveness evaluation
The higher the ratio, the better. We use the manufacturer's recommended prices.
Performance to price scatter graph
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 | 3200 | 16896 |
| Core clock speed | 1582 MHz | 1665 MHz |
| Boost clock speed | 1683 MHz | 1837 MHz |
| Number of transistors | 11,800 million | 80,000 million |
| Manufacturing process technology | 16 nm | 4 nm |
| Power consumption (TDP) | 250 Watt | 700 Watt |
| Texture fill rate | 336.6 | 969.9 |
| Floating-point processing power | 10.77 TFLOPS | 62.08 TFLOPS |
| ROPs | 80 | 24 |
| TMUs | 200 | 528 |
| Tensor Cores | no data | 528 |
| L1 Cache | 1.2 MB | 33 MB |
| L2 Cache | 2.5 MB | 50 MB |
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 3.0 x4 | PCIe 5.0 x16 |
| Length | 267 mm | 268 mm |
| Width | 2-slot | 2-slot |
| Supplementary power connectors | 2x 8-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 | GDDR5X | HBM3 |
| Maximum RAM amount | 5 GB | 96 GB |
| Memory bus width | 320 Bit | 5120 Bit |
| Memory clock speed | 1376 MHz | 1313 MHz |
| Memory bandwidth | 440.3 GB/s | 1,681 GB/s |
| Resizable BAR | - | + |
Connectivity and outputs
This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.
| Display Connectors | No outputs | No outputs |
API and SDK support
List of supported 3D and general-purpose computing APIs, including their specific versions.
| DirectX | 12 (12_1) | N/A |
| Shader Model | 6.4 | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 1.2 | 3.0 |
| Vulkan | 1.2.131 | N/A |
| CUDA | 6.1 | 9.0 |
| DLSS | - | + |
Pros & cons summary
| Recency | 12 February 2018 | 21 March 2023 |
| Maximum RAM amount | 5 GB | 96 GB |
| Chip lithography | 16 nm | 4 nm |
| Power consumption (TDP) | 250 Watt | 700 Watt |
P102-100 has 180% lower power consumption.
H100 PCIe 96 GB, on the other hand, has an age advantage of 5 years, a 1820% higher maximum VRAM amount, and a 300% more advanced lithography process.
We couldn't decide between P102-100 and H100 PCIe 96 GB. We've got no test results to judge.
Other comparisons
We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.
