GeForce RTX 5070 vs GeForce RTX 5090

[ VERDICT ] Winner: GeForce RTX 5090 (96/100)

GeForce RTX 5090 leads overall with a score of 96/100 compared to GeForce RTX 5070 (61/100), offering superior measured performance and technical specifications.

Benchmark Performance Comparison

Normalized Lab Scores
Benchmark measurements and scores across compared devices
Benchmark GeForce RTX 5070 GeForce RTX 5090
Efficiency 48.0 Poor 48.0 Poor
Gaming 64.0 Average 100.0 Excellent WINNER

Detailed Technical Specifications

3D Mark

3D Mark specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
Fire Strike 58324 87569
Night Raid 172064 206648
Port Royal 14498 38202
Solar Bay 107004 236689
Time Spy 22708 46873
Wild Life Extreme 43879 108751

API

API specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
CUDA 12.0 12.0
DLSS DLSS 4 DLSS 4
DirectX 12.2 12.2
DisplayPort 2.1b 2.1b
OpenCL 3.0 3.0
OpenGL 4.6 4.6
Ray Tracing Yes Yes
Vulkan 1.4 1.4

Games

Games specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
1080p High 192 299
1080p Ultra 162 254
1440p Ultra 128 207
4K Ultra 74 125
CoD: Modern Warfare III 17817113689 334322260184
Counter-Strike 2 29321016081 36326520299
Cyberpunk 2077 1641449943 26123316077
Far Cry 6 16414512882 237202190121
Forza Horizon 5 19714013097 336247233187
Ghost of Tsushima 1231078958 221185175113
Hogwarts Legacy 1401179251 24919916697
Margin of Error Low Low
Shadow of the Tomb Raider 25323818998 329302238128
The Witcher 3 21618313171 364327237123

GeekBench 6 ML

GeekBench 6 ML specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
Backend DirectML DirectML
Depth Estimation (Q) 58300 85572
Face Detection (HP) 62662 114685
Framework ONNX ONNX
Image Classification (SP) 12270 18948
Image Segmentation (HP) 30237 55109
Image Super Resolution (Q) 35585 58478
Machine Translation (HP) 6233 8184
Object Detection (SP) 16709 27802
Pose Estimation (Q) 145190 344729
Style Transfer (SP) 246598 687080
Text Classification (SP) 4198 4464

GeekBench 6 OpenCL

GeekBench 6 OpenCL specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
API OpenCL OpenCL
Background Blur 254.4 img/sec 333 img/sec
Edge Detection 10.2 Gpixels/sec 33.7 Gpixels/sec
Face Detection 180.6 img/sec 298.5 img/sec
Feature Matching 1.84 Gpixels/sec 2.53 Gpixels/sec
Gaussian Blur 10.7 Gpixels/sec 34.5 Gpixels/sec
Horizon Detection 7.91 Gpixels/sec 21.7 Gpixels/sec
Particle Physics 27816.5 FPS 59146.9 FPS
Stereo Matching 975.7 Gpixels/sec 2850 Gpixels/sec

General

General specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
AMD Equivalent -Radeon RX 9070
Architecture Blackwell 2.0 Blackwell 2.0
Build Discrete Discrete
Case Desktop Desktop
GPU Codename GB205-300 GB202-300
Launch price (MSRP) $549 $1999
Purpose Gaming Gaming
Recommended CPU -Intel Core Ultra 5 245Kor above -Intel Core Ultra 9 285Kor above
Released January 7, 2025 January 7, 2025
Segment Mid-range High-end
Vendor Nvidia Nvidia

Graphics Processing Unit

Graphics Processing Unit specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
Base Clock 2160 MHz 2017 MHz
Boost Clock 2512 MHz 2407 MHz
Compute Units (Pipelines) 48 170
Instructions Per Cycle 2 IPC 2 IPC
L1 Cache 128KB per cluster 128KB per cluster
L2 Cache 48MB shared 96MB shared
Ray-tracing Cores 48 170
Render Output Units (ROPs) 80 176
Shading Units 6144 21760
Tensor Cores 192 680
Texture Mapping Units (TMUs) 192 680

Memory

Memory specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
Bus 192-bit 512-bit
ECC No No
Effective Memory Speed 28000 Mbps 28000 Mbps
Memory Clock 1750 MHz 1750 MHz
Memory Size 12 GB 32 GB
Memory Type GDDR7 GDDR7

Passmark Graphics

Passmark Graphics specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
DirectX 11 280 FPS 341 FPS
DirectX 12 106 FPS 179 FPS
G2D Mark 1291 1410
GPU Compute 14310 Ops/s 24360 Ops/s

Physical

Physical specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
Die Size 263 mm² 750 mm²
Fabrication Process 4 nm 4 nm
Interface PCIe 5.0 x16 PCIe 5.0 x16
Manufacturing TSMC TSMC
TGP 250 W 575 W
Transistor Count 31.1 billion 92.2 billion
Transistor Density 118.25 MTr/mm² 122.93 MTr/mm²

Raw Performance

Raw Performance specifications comparison
Specification GeForce RTX 5070 GeForce RTX 5090
Pixel Fill Rate 201 GPixel/s 424 GPixel/s
Texture Fill Rate 482 GTexel/s 1637 GTexel/s

Looking for More Comparisons?

Browse all available head-to-head comparisons or select new devices in the comparison hub.

Browse Graphics Cards Comparisons → Launch Full Comparison Hub →