GeForce RTX 3090 vs TITAN RTX
Summary
Reasons to consider GeForce RTX 3090 |
This is a much newer product, it might have better long term support. |
Higher theoretical gaming performance, based on specifications. |
Supports PhysX |
Supports G-Sync |
Supports ShadowPlay (allows game streaming/recording with minimum performance penalty) |
Supports Direct3D 12 Async Compute |
Supports DirectX Raytracing (DXR) |
Supports Deep Learning Super-Sampling (DLSS) |
Reasons to consider TITAN RTX |
70 watts lower power draw. This might be a strong point if your current power supply is not enough to handle the GeForce RTX 3090 . |
Supports PhysX |
Supports G-Sync |
Supports ShadowPlay (allows game streaming/recording with minimum performance penalty) |
Supports Direct3D 12 Async Compute |
Supports DirectX Raytracing (DXR) |
Supports Deep Learning Super-Sampling (DLSS) |
HWBench recommends GeForce RTX 3090
Based on theoretical specifications.
Core Configuration
| GeForce RTX 3090 | | TITAN RTX | |
---|
GPU Name | GA102 (GA102-300-A1) | vs | TU102 (TU102-400-A1) |
Fab Process | 8 nm | vs | 12 nm |
Die Size | 628 mm² | vs | 754 mm² |
Transistors | 28,000 million | vs | 18,600 million |
Shaders | 10496 | vs | 4608 |
Compute Units | 82 | vs | 72 |
Core clock | 1395 MHz | vs | 1350 MHz |
ROPs | 112 | vs | 96 |
TMUs | 328 | vs | 288 |
Memory Configuration
| GeForce RTX 3090 | | TITAN RTX | |
---|
Memory Type | GDDR6X | vs | GDDR6 |
Bus Width | 384 bit | vs | 384 bit |
Memory Speed | 1219 MHz
19.5 Gbps effective | vs | 1750 MHz
14000 MHz effective |
Memory Size | 24576 Mb | vs | 24576 Mb |
Additional details
| GeForce RTX 3090 | | TITAN RTX | |
---|
TDP | 350 watts | vs | 280 watts |
Release Date | 1 Sep 2020 | vs | 18 Dec 2018 |
GeForce RTX 3090
189.80 GP/s
TITAN RTX
169.90 GP/s
GigaPixels - higher is better
GeForce RTX 3090
556.00 GT/s
TITAN RTX
509.80 GT/s
GigaTexels - higher is better
GeForce RTX 3090
936.20 GB/s
TITAN RTX
672.00 GB/s
GB/s - higher is better
GeForce RTX 3090
35580.00 GFLOPs
TITAN RTX
16312.00 GFLOPs
GFLOPs - higher is better