Conclusion
After reading this article, you likely have a decent understanding of the second-generation HQV benchmark. Let’s have a look at the test class scores and the final total for each card:
HQV Benchmark version 2.0 Results (out of 210) | |||||
---|---|---|---|---|---|
Row 0 - Cell 0 | Radeon HD 6850 | Radeon HD 5750 | Radeon HD 5670 | Radeon HD 5550 | Radeon HD 5450 |
Test Class 1:Video Conversion | 90 | 90 | 89 | 89 | 89 |
Test Class 1: Noise and Artifact Reduction | 54 | 54 | 44 | 44 | 44 |
Test Class 3: Image Scaling and Enhancements | 30 | 30 | 30 | 30 | 15 |
Test Class 4: Adaptive Processing | 27 | 27 | 27 | 7 | 7 |
Totals: | 201 | 201 | 190 | 170 | 155 |
Header Cell - Column 0 | GeForce GTX 470 | GeForce GTX 460 | GeForce 9800 GT | GeForce GT 240 | GeForce GT 430 | GeForce 210 |
---|---|---|---|---|---|---|
Test Class 1:Video Conversion | 90 | 47 | 87 | 87 | 46 | 47 |
Test Class 1: Noise and Artifact Reduction | 20 | 20 | 20 | 20 | 20 | 20 |
Test Class 3: Image Scaling and Enhancements | 30 | 30 | 30 | 30 | 30 | 30 |
Test Class 4: Adaptive Processing | 20 | 20 | 20 | 20 | 20 | 0 |
Totals: | 160 | 117 | 157 | 157 | 116 | 97 |
We’re seeing a fairly natural increase in quality across the Radeon line as you get to more expensive models, with the Radeon HD 5750 acting as the vanguard of maximum Radeon video playback quality. The Radeon HD 6850 does no better, and both are about as close to perfect as you’ll see on a PC. Indeed, all of the Radeons, other than the low-end Radeon HD 5450, have the same driver options. But the GPUs can handle more of the enhancements without stuttering as you get into pricier models.
The GeForce line is a little inconsistent, comparatively. There’s a hiccup, and the GeForce GT 240 and 9800 GT achieve a higher score than the more expensive GeForce GTX 460 according to our observations. This is mostly attributable to poor pulldown detection in some models, and while we’ve asked Nvidia about this disparity, the company hasn't responded.
Overall, when comparing Radeons to GeForces, AMD's cards get the nod for higher overall scores and better results per dollar spent. Frankly, the main reason for the superior results are AMD’s consistent cadence detection, better noise reduction (especially when it comes to compressed video), and a working flesh tone correction feature.
It’s important to note that the GeForce cards don’t suffer from shoddy video quality, and in our opinion, too many points are awarded for obscure multi-cadence detection. Thirty points are applied to this area, and that’s not including scores from important cadences like 2:2 and 3:2 pulldown. If you remove those 30 points, the playing field between GeForce and Radeon becomes much tighter. The GeForce cards offer excellent video playback when it comes to high-definition source material, and all of them handle the important 3:2 cadence without issue. Realistically, if you put a GeForce in an HTPC for DVD and Blu-ray playback duty, you’d probably never guess that it didn’t achieve the top score.
Having said that, the Radeons earn a well-deserved win here. While obscure multi-cadence support might be responsible for the bulk of point advantage, their real strength is superlative noise-reduction options. This comes in real handy with compressed video, so if you plan to play back any files that aren’t optimally encoded at HD resolution, the Radeons have a real advantage. It’s also noteworthy that the sub-$100 Radeon HD 5670 can offer slightly better playback quality than a GeForce GTX 470, even when multi-cadence tests are left out of the mix, and that a ~$120 Radeon 5750 card can boast the same ultimate PC playback quality right alongside more expensive Radeons like the 6850.
On a final note, the second-gen HQV benchmark Blu-ray can be purchased from www.hqv.com for $24.99, if you want to replicate the tests we ran here. There is also a standard-definition version of the benchmark for DVD that can be acquired for $19.99.