GeekRadar / The comparison desk

COMPARE.

GPU

NVIDIA

GeForce RTX 4070 Ti SUPER AD102

Graphics processor

Video memory: 16 GB^*
Memory bus: 256 bit^*
Rated memory bandwidth: 672.30 GB/s^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Ada Lovelace^*
Process node: 5 nm^*
GPU cores: 8,448^*
GPU base clock: 2,340 MHz^*
Graphics interface: PCIe 4.0 x16^*
Reported chip die area: 609 mm²^*
Transistors: 76,300 million^*
GPU boost clock: 2,610 MHz^*
GPU L2 cache: 48 MB^*
Graphics processor: AD102^*
Texture mapping units: 264^*
Raster output units: 96^*
Streaming multiprocessors: 66^*
Tensor cores: 264^*
Ray tracing cores: 66^*
Chip foundry: TSMC^*
Card or module format: Triple-slot^*
Declared GPU TDP: 285 W^*
Suggested system PSU: 600 W^*
External power connectors: 1x 16-pin^*
DirectX / feature level: 12 Ultimate (12_2)^*
OpenGL: 4.6^*
OpenCL: 3.0^*
Vulkan: 1.4^*
CUDA compute capability: 8.9^*
Shader model: 6.9^*

Memory & storage

Memory options: 16^*
Video memory: 16 GB^*
Video memory type: Reported value 1: GDDR6X^*
Reported value 2: GDDR6X^*
Reported value 3: HBM2e^*
Reported value 4: GDDR6X^*
Reported value 5: GDDR6X^*
Memory bus: 256 bit^*
Rated memory bandwidth: 672.30 GB/s^*

Power specifications

Thermal design power: 285 W^*

System configuration

Detailed specifications

Specifications 1 · 31 fields

proc transistors: 76,300 million^*
proc die size: 609^* mm²
cores: 8,448^*
tensor cores: 264^*
rt cores: 66^*
sm count: 66^*
base clock: 2,340^*
boost clock: 2,610^*
memory clock: 1,313^*
ram: 16^*
name: NVIDIA GeForce RTX 4070 Ti SUPER AD102^*
ram type: GDDR6X^*
bus width: 256^*
bus interface: PCIe 4.0 x16^*
ram bandwidth: 672.3^* GB/s
fp16 performance: 44.1^*
fp32 performance: 44.1^*
fp64 performance: 0.689^*
tdp: 285^* W
suggested psu: 600^* W
slot width: Triple-slot^*
power connectors: 1x 16-pin^*
api cuda: 8.9^*
is retail board: No^*
gpi value: 18.91^*
vendor: nvd^*
manufacturer: NVIDIA^*
arch name: Ada Lovelace^*
card release date: 2024-06-10^*
proc foundry: TSMC^*
proc process size: 5^* nm

Specifications 2 · 40 fields

CUDA compute capability

8.9^*

DirectX / feature level

12 Ultimate (12_2)^*

OpenCL

3.0^*

OpenGL

4.6^*

Shader model

6.9^*

Vulkan

1.4^*

Architecture

Ada Lovelace^*

Graphics interface

PCIe 4.0 x16^*

Memory bus

256^* bit

Reported shader processors

8,448^*

Reported theoretical FP16 peak

44.1^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP32 peak

44.1^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP64 peak

0.689^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Declared L1 cache configuration

128 KB (per SM)^*

Declared GPU L2 cache

48 MB^*

Reported display outputs

1x HDMI 2.1, 3x DisplayPort 1.4a^*

Theoretical pixel fill rate

250.6 GPixel/s^*

External power connectors

1x 16-pin^*

Reported chip die area

609 mm²^*

Chip foundry

TSMC^*

Process node declaration

5 nm^*

Transistors

76,300 million^*

Graphics processor

AD102^*

Reported video memory

16^* GB

Rated memory bandwidth

672.3 GB/s^*

Video memory type

GDDR6X^*

Raster output units

96^*

Ray tracing cores

66^*

Reported shading units

8,448^*

Card or module format

Triple-slot^*

Streaming multiprocessors

66^*

Suggested system PSU

600 W^*

2:4 tensor sparsity

Yes^*

Declared GPU TDP

285 W^*

Tensor cores

264^*

Theoretical texture fill rate

689.0 GTexel/s^*

Texture mapping units

264^*

GPU cores

8,448^*

Video memory

16^* GB

Video memory type

GDDR6X^*

Specifications 3 · 62 fields

Brand: NVIDIA^*
Top BUS WIDTH: 256^* bit
Graphics Processor GPU Name: AD102^*
Graphics Processor Architecture: Ada Lovelace^*
Render Config SM Count: 66^*
Graphics Features CUDA: 8.9^*
Render Config Tensor Cores: 264^*
Graphics Processor Process Size: 5^* nm
Graphics Processor Transistors: 76,300 million^*
Graphics Processor Die Size: 609^* mm²
Graphics Card Release Date: Jun 10th, 2024^*
Graphics Card Generation: GeForce 40^*
Graphics Card Successor: GeForce 50^*
Graphics Card Production: Active^*
Name: GeForce RTX 4070 Ti SUPER AD102^*
Graphics Card Bus Interface: PCIe 4.0 x16^*
Clock Speeds Memory Clock: 1313 MHz 21 Gbps effective^*
Memory Memory Size: 16^* GB
Memory Memory Type: GDDR6X^*
Memory Memory Bus: 256^* bit
Memory Bandwidth: 672.3^* GB/s
Render Config TMUs: 264^*
Top GRAPHICS PROCESSOR: AD102^*
Render Config ROPs: 96^*
Theoretical Performance Pixel Rate: 250.6 GPixel/s
Theoretical Performance Texture Rate: 689.0 GTexel/s
Board Design Slot Width: Triple-slot^*
Board Design TDP: 285^* W
Board Design Suggested PSU: 600^* W
Board Design Outputs: 1x HDMI 2.1 3x DisplayPort 1.4a^*
Board Design Board Number: PG141 SKU 323^*
Graphics Features DirectX: 12 Ultimate (12_2)^*
Graphics Features OpenGL: 4.6^*
Graphics Features OpenCL: 3^*
Graphics Features Vulkan: 1.3^*
source file: nvidia2024.json^*
Graphics Processor Foundry: TSMC^*
Graphics Card Predecessor: GeForce 30^*
Recommended Resolutions: Good:1920x1080 | Good:2560x1440 | Good:3840x2160^*
Graphics Card Launch Price: 799 USD^*
Board Design Length: 310 mm 12.2 inches^*
Graphics Processor GPU Variant: AD102-175-KEF-A1^*
Board Design Power Connectors: 1x 16-pin^*
Graphics Processor Density: 125.3M / mm²^*
Board Design Height: 140 mm 5.5 inches^*
Board Design Width: 61 mm 2.4 inches^*
Graphics Card Reviews: 40 in our database^*
Top TMUS: 264^*
Top ROPS: 96^*
Clock Speeds Base Clock: 2,340^* MHz
Clock Speeds Boost Clock: 2,610^* MHz
Graphics Features Shader Model: 6.8^*
Top CORES: 8,448^*
Render Config Shading Units: 8,448^*
Render Config L2 Cache: 48^* MB
Top MEMORY SIZE: 16^* GB
Theoretical Performance FP32 (float): 44.1^* TFLOPS
Theoretical Performance FP64 (double): 689.0 GFLOPS (1:64)
Render Config L1 Cache: 128 KB (per SM)^*
Theoretical Performance FP16 (half): 44.10 TFLOPS (1:1)
Top MEMORY TYPE: GDDR6X^*
Render Config RT Cores: 66^*

Specifications 4 · 14 fields

manufacturer: NVIDIA^*
rop: 160^*
igp: No^*
bus: PCIe 4.0 x16^*
memType: HBM2e^*
gpuChip: GA100^*
productName: GeForce RTX 4070 Ti SUPER AD102^*
releaseYear: 2,024^*
memSize: 16^*
memBusWidth: 256^*
gpuClock: 765^*
memClock: 1,215^*
unifiedShader: 6,912^*
tmu: 432^*

* Provisional data