The graphics card market is fiercely competitive, and AMD aims to capture the attention of PC enthusiasts and gamers with its latest offering, the AMD Radeon RX 9070 GRE. This new graphics card brings a highly anticipated architecture to a wider audience, positioning itself as an powerful yet more affordable choice for competitive 1080p and 1440p gaming and even 4K gaming for those not chasing higher refresh rates. Built around the RDNA 4 architecture, the RX 9070 GRE is designed to bridge the gap between mid-tier efficiency and high-end performance, without the added cost of the more extreme XT variant. Anyone looking to push their frame rates higher in modern AAA titles without stretching to flagship budgets will find a lot to love here, and from our past experience with RX 9070s, you can get a lot of fun for a very reasonable investment.
Features and Performance
The model we have is the Sapphire PULSE, and Sapphire has implemented several distinct if somewhat familiar engineering choices to enhance the reliability and cooling of this cards design. The card utilises their in-house Dual-X cooling technology alongside AeroCurve fan blades to maintain low operational temperatures while keeping noise to a minimium, however, this is nothing new to Sapphire products, as they’ve used a broadly similar cooler design on a wide range of GPUs for many years as part of their PULSE series; if it ain’t broken, don’t fix it. But overall, this should be a good representation of a typical AMD Radeon RX 9070 GRE graphics cards.
Key Features
- 12GB VRAM
- Factory Overclocked
- Dual Fan Cooling
- Designed for 1080p and 1440p Gaming
- RDNA 4
- TriXX Software Suite
- TriXX Boost
While it’s true that this GPU has been on the market for a while now, it’s only been available in Chinese markets. Now that it’s coming to the west, there’s not a lot we don’t know about the GPU already, but it’s the first time we in the West have gotten hands on with the hardware. It’s still built around the 4nm process node, featuring 3,072 stream processors across 48 compute units. This iteration features 12GB of GDDR6 memory deployed across a 192-bit bus, offering a total memory bandwidth of 432 GB/s. Ray tracing and machine learning workloads are addressed through 48 dedicated ray accelerators and 96 matrix cores. Those of you knowledgable of the 9070 series will of course realise this is a cut-down version of the card, trimming some VRAM and cores, which is what makes this card a fair bit more affordable, but it should fill a gap in the market for those wanting a more competitively priced GPU option.
Technical Specifications
- Graphics Architecture: AMD RDNA 4
- Process Technology: TSMC 4nm N4P FinFET
- Stream Processors: 3,072
- Compute Units: 48 CUs with 3rd Generation Ray Accelerators and 2nd Generation AI Accelerators
- Ray Accelerators: 48
- Matrix Cores: 96
- Infinity Cache: 48MB
- Memory Configuration: 12GB GDDR6
- Memory Bus Width: 192-bit
- Memory Bandwidth: 432 GB/s
- Effective Memory Speed: 18 Gbps
- Interface Type: PCI Express 5.0 x16
- Maximum Power Draw: 220W
- Recommended System Power Supply: 550W
- Display Connectors: 1x HDMI 2.1b, 3x DisplayPort 2.1a
- Maximum Digital Resolution: 7680×4320
