Hot on the heels of last month's Radeon HD 4890 release which set a new performance record for ATI as their fastest single GPU yet, the company has another new graphics card for us today, this time targeting the sub-$100 market.
The new Radeon HD 4770 is the world’s first 40nm graphics processing unit, and although it's not designed to set any speed records like the flagship 4890, the 4770 aims to deliver the best possible value in a neat cost-effective package.
It has become common practice for both ATI and Nvidia to make their initial migration to a new design process using a mid-range or even low-end GPU, allowing time for refinement before they push it to high-performance parts. That said, the Radeon HD 4770 is far from a simple affair as the GPU has a transistor count of 826 million, or just 14% less than the Radeon HD 4890, ATI’s most complex GPU yet.
As we often enjoy testing the latest and greatest products such as the Radeon HD 4890, it is easy to forget that the demand for graphics cards priced over $300 is comparably very small. While owning the performance crown does award bragging rights, providing the best value mainstream solutions is likely more prosperous.
As mentioned before, ATI will be targeting the $100 price bracket with the Radeon HD 4770, placing it in direct competition of the Nvidia GeForce 9800 GT. AMD claims that based on current pricing the GeForce 9800 GT ($115) provides 5.1 GFLOPS per dollar whereas the Radeon HD 4770 will deliver an incredible 8.8 GFLOPS. Furthermore, the Radeon HD 4770 will achieve 12.0 GFLOPS per watt, while the GeForce 9800 GT is good for just 4.8 GFLOPS.
But of course, they would never say otherwise, so naturally we will be checking out how these two graphics cards compare in real-life scenarios. Before we look at the numbers we will also take a quick look at the Radeon HD 4770 card design in more detail...
- Ad-free TechSpot experience while supporting our work
- NEW Get a free copy of Superhot: Mind Control Delete, courtesy of GOG
- Our promise: All reader contributions will go toward funding more content
- That means: More tech features, more benchmarks and analysis