GeekRadar / The comparison desk

COMPARE.

GPU

Nvidia

P102-100

Graphics processor

Video memory: 5 GB^*
Memory bus: 320 bit^*
Rated memory bandwidth: 440.30 GB/s^*

Benchmark results

G3D Mark · version unreported · g3d_mark3,703^* pointsPlatform unreported Compare results
Geekbench GPU (reported compilation) · version unreported · OpenCL49,602^*Platform unreported Compare results
Geekbench GPU (reported compilation) · version unreported · OpenCL49,602^*Platform unreported Compare results
Geekbench GPU (reported compilation) · version unreported · OpenCL49,602^*Platform unreported Compare results

Processing & platform

Architecture: Pascal^*
Process node: 16 nm^*
GPU base clock: 1,582 MHz^*
GPU boost clock: 1,683 MHz^*

Memory & storage

Video memory: 5 GB^*
Video memory type: Reported value 1: GDDR5X^*
Reported value 2: GDDR5X^*
Reported value 3: GDDR5^*
Memory bus: 320 bit^*
Rated memory bandwidth: 440.30 GB/s^*

System configuration

Detailed specifications

Specifications 1 · 8 fields

Product Name: P102-100^*
GPU Chip: GP102^*
Released: Feb 12th, 2018^*
Bus: PCIe 1.0 x4^*
Memory: 5 GB, GDDR5X, 320 bit^*
GPU clock: 1,582^* MHz
Memory clock: 1,376^* MHz
Shaders TMUs ROPs: 3200 / 200 / 80^*

Specifications 2 · 3 fields

G3D Mark

3,703^* points

Scope and conditions

benchmark_context

suite: G3D Mark^*

native_context

GPU model: P102-100^*
Snapshot date: 2026-10-05^*

Model

P102-100^*

Hardware class

Desktop^*

Specifications 3 · 8 fields

Product Name: P102-100^*
GPU Chip: GP102^*
Released: Feb 12th, 2018^*
Bus: PCIe 1.0 x4^*
Memory: 5 GB, GDDR5X, 320 bit^*
GPU clock: 1,582^* MHz
Memory clock: 1,376^* MHz
Shaders / TMUs / ROPs: 3200 / 200 / 80^*

Specifications 4 · 54 fields

Brand: NVIDIA^*
Top BUS WIDTH: 320^* bit
Graphics Processor GPU Name: GP102^*
Graphics Processor Architecture: Pascal^*
Render Config SM Count: 25^*
Graphics Features CUDA: 6.1^*
Graphics Processor Process Size: 16^* nm
Graphics Processor Transistors: 11,800 million^*
Graphics Processor Die Size: 471^* mm²
Graphics Card Release Date: Feb 12th, 2018^*
Graphics Card Generation: Mining GPUs^*
Graphics Card Production: End-of-life^*
Name: P102-100^*
Graphics Card Bus Interface: PCIe 1.0 x4^*
Clock Speeds Memory Clock: 1376 MHz 11 Gbps effective^*
Memory Memory Size: 5^* GB
Memory Memory Type: GDDR5X^*
Memory Memory Bus: 320^* bit
Memory Bandwidth: 440.3^* GB/s
Render Config TMUs: 200^*
Top GRAPHICS PROCESSOR: GP102^*
Render Config ROPs: 80^*
Theoretical Performance Pixel Rate: 134.6 GPixel/s
Theoretical Performance Texture Rate: 336.6 GTexel/s
Board Design Slot Width: Dual-slot^*
Board Design TDP: 250^* W
Board Design Suggested PSU: 600^* W
Board Design Outputs: No outputs^*
Board Design Board Number: PG611 SKU 100^*
Graphics Features DirectX: 12 (12_1)^*
Graphics Features OpenGL: 4.6^*
Graphics Features OpenCL: 3^*
Graphics Features Vulkan: 1.3^*
source file: nvidia2018.json^*
Graphics Processor Foundry: TSMC^*
Board Design Length: 267 mm 10.5 inches^*
Graphics Processor GPU Variant: GP102-100-A1^*
Board Design Power Connectors: 2x 8-pin^*
Graphics Processor Density: 25.1M / mm²^*
Top TMUS: 200^*
Graphics Processor Chip Package: BGA-2397^*
Top ROPS: 80^*
Clock Speeds Base Clock: 1,582^* MHz
Clock Speeds Boost Clock: 1,683^* MHz
Graphics Features Shader Model: 6.8^*
Top CORES: 3,200^*
Render Config Shading Units: 3,200^*
Render Config L2 Cache: 2.5^* MB
Top MEMORY SIZE: 5^* GB
Theoretical Performance FP32 (float): 10.77^* TFLOPS
Theoretical Performance FP64 (double): 336.6 GFLOPS (1:32)
Render Config L1 Cache: 48 KB (per SM)^*
Theoretical Performance FP16 (half): 168.3 GFLOPS (1:64)
Top MEMORY TYPE: GDDR5X^*

Specifications 5 · 5 fields

Device

P102-100^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

Manufacturer

Nvidia^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

Device

P102-100^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

OpenCL

49,602^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

OpenCL

49,602^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

Specifications 6 · 3 fields

Manufacturer

Nvidia^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

Device

P102-100^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

OpenCL

49,602^*

Scope and conditions

benchmark_context

suite: Geekbench GPU (reported compilation)^*
backend_columns: CUDA^*
Metal^*
OpenCL^*
Vulkan^*
test_configuration_verified: No^*
comparable: No^*

Specifications 7 · 27 fields

manufacturer: NVIDIA^*
manufacturer: NVIDIA^*
rop: 80^*
rop: 32^*
igp: No^*
igp: No^*
bus: PCIe 3.0 x16^*
bus: PCIe 3.0 x4^*
memType: GDDR5X^*
memType: GDDR5^*
gpuChip: Antigua^*
gpuChip: GP102^*
productName: P102-100^*
productName: P102-100^*
releaseYear: 2,018^*
releaseYear: 2,018^*
memSize: 5^*
memSize: 5^*
memBusWidth: 320^*
gpuClock: 1,582^*
gpuClock: 970^*
memClock: 1,425^*
memClock: 1,376^*
unifiedShader: 3,200^*
unifiedShader: 2,048^*
tmu: 200^*
tmu: 128^*

* Provisional data