Discrete HTPC GPU Shootout
by Ganesh T S on June 12, 2011 10:30 PM ESTHTPC enthusiasts are often concerned about the quality of pictures output by the system. While this is a very subjective metric, we have decided to take as much of an objective approach as possible. Starting with our HTPC reviews, we have been using the HQV 2.0 benchmark for this purpose. The HQV benchmarking procedure has been heavily promoted by AMD, but it is something NVIDIA says it doesn't optimize for. Considering the fact that there aren't any other standardized options available to evaluate the video post processing capabilities of the GPUs, we feel that HQV benchmarking should be an integral part of the reviews.
However, HQV scores need to be taken with a grain of salt. In particular, one must check the tests where the GPU lost out points. In case those tests don't reflect the reader's usage scenario, the handicap can probably be ignored. An example is cadence detection. Only interlaced streams with non-native frame rates (i.e, 24p content at 60i, 25p content at 50i etc.) need this post processing. Even within this, it is streams requiring 3:2 cadence detection that are most common. Streams with 2:3:3:2 and other fancy patterns are almost non-existent in most usage scenarios. So, it is essential that the scores for each test be compared, rather than just the total value.
The HQV 2.0 test suite consists of 39 different streams divided into 4 different classes. In our HTPC(s), we use Cyberlink PowerDVD 11 with TrueTheater disabled and hardware acceleration enabled for playing back the HQV streams. The playback device is assigned scores for each, depending on how well it plays the stream. Each test was repeated multiple times to ensure that the correct score was assigned. The scoring details are available in the testing guide from HQV.
In the table below, we indicate the maximum score possible for each test, and how much each GPU was able to get. The NVIDIA GPUs were tested with driver version 270.61 and the AMD GPUs were tested with Catalyst 11.5.
| HQV 2.0 Benchmark Shootout | ||||||||
| Test Class | Chapter | Tests | Max. Score | NVIDIA GT 430 | MSI GT 520 | AMD 6450 | Sapphire 6570 | MSI 6450 |
| Video Conversion | Video Resolution | Dial | 5 | 5 | 4 | 5 | 5 | 4 |
| Dial with Static Pattern | 5 | 5 | 5 | 5 | 5 | 5 | ||
| Gray Bars | 5 | 5 | 3 | 5 | 5 | 5 | ||
| Violin | 5 | 5 | 3 | 3 | 5 | 3 | ||
| Film Resolution | Stadium 2:2 | 5 | 5 | 0 | 5 | 5 | 5 | |
| Stadium 3:2 | 5 | 5 | 5 | 5 | 5 | 5 | ||
| Overlay On Film | Horizontal Text Scroll | 5 | 5 | 5 | 5 | 5 | 5 | |
| Vertical Text Scroll | 5 | 5 | 5 | 5 | 5 | 5 | ||
| Cadence Response Time | Transition to 3:2 Lock | 5 | 5 | 5 | 5 | 5 | 5 | |
| Transition to 2:2 Lock | 5 | 5 | 0 | 5 | 5 | 5 | ||
| Multi-Cadence | 2:2:2:4 24 FPS DVCam Video | 5 | 5 | 0 | 5 | 5 | 5 | |
| 2:3:3:2 24 FPS DVCam Video | 5 | 5 | 0 | 5 | 5 | 5 | ||
| 3:2:3:2:2 24 FPS Vari-Speed | 5 | 5 | 0 | 5 | 5 | 5 | ||
| 5:5 12 FPS Animation | 5 | 5 | 0 | 5 | 5 | 5 | ||
| 6:4 12 FPS Animation | 5 | 5 | 0 | 5 | 5 | 5 | ||
| 8:7 8 FPS Animation | 5 | 5 | 0 | 5 | 5 | 5 | ||
| Color Upsampling Errors | Interlace Chroma Problem (ICP) | 5 | 5 | 5 | 5 | 5 | 5 | |
| Chroma Upsampling Error (CUE) | 5 | 5 | 5 | 5 | 5 | 5 | ||
| Noise and Artifact Reduction | Random Noise | SailBoat | 5 | 5 | 5 | 5 | 5 | 0 |
| Flower | 5 | 5 | 5 | 5 | 5 | 0 | ||
| Sunrise | 5 | 5 | 5 | 5 | 5 | 0 | ||
| Harbour Night | 5 | 5 | 5 | 5 | 5 | 0 | ||
| Compression Artifacts | Scrolling Text | 5 | 5 | 3 | 3 | 5 | 0 | |
| Roller Coaster | 5 | 3 | 3 | 3 | 5 | 0 | ||
| Ferris Wheel | 5 | 3 | 3 | 3 | 5 | 0 | ||
| Bridge Traffic | 5 | 3 | 3 | 3 | 3 | 0 | ||
| Upscaled Compression Artifacts | Text Pattern | 5 | 3 | 3 | 3 | 3 | 0 | |
| Roller Coaster | 5 | 3 | 3 | 3 | 3 | 0 | ||
| Ferris Wheel | 5 | 3 | 3 | 3 | 3 | 0 | ||
| Bridge Traffic | 5 | 3 | 3 | 3 | 3 | 0 | ||
| Image Scaling and Enhancements | Scaling and Filtering | Luminance Frequency Bands | 5 | 5 | 5 | 5 | 5 | 5 |
| Chrominance Frequency Bands | 5 | 5 | 5 | 5 | 5 | 5 | ||
| Vanishing Text | 5 | 5 | 5 | 5 | 5 | 5 | ||
| Resolution Enhancement | Brook, Mountain, Flower, Hair, Wood | 15 | 15 | 15 | 15 | 15 | 15 | |
| Video Conversion | Contrast Enhancement | Theme Park | 5 | 5 | 5 | 5 | 5 | 5 |
| Driftwood | 5 | 5 | 5 | 5 | 5 | 5 | ||
| Beach at Dusk | 5 | 5 | 5 | 5 | 5 | 5 | ||
| White and Black Cats | 5 | 5 | 5 | 5 | 5 | 5 | ||
| Skin Tone Correction | Skin Tones | 10 | 7 | 7 | 7 | 7 | 7 | |
| Total Score | 210 | 193 | 146 | 189 | 197 | 144 | ||
A look at the above table reveals that there is not much to differentiate between the AMD 6450, GT 430 and 6570. The GT 430 scores in between the 6450 and 6570. However, the GT 520 and the DDR3 based MSI 6450 stand out because of their low scores.
In our GT 430 review last October, we were willing to give it some leeway because it lost out in the bulk of the cadence detection tests. The GT 520 is in a similar situation here. The all-important 3:2 pulldown is performed correctly. However, none of the other cadence detection tests passed. GT 520 also has other issues in general which cause it to get a lower score than what the GT 430 obtained in its initial review. We will take a look at how the GT 520 fares in the other tests before delivering the final verdict.
The DDR3 based 6450 misses out on the bulk of the scores because it is unable to perform denoising in a proper manner. When AMD was contacted about this, they admitted the issue and indicated that they were working on a fix. However, they pointed out that the problem was only for standalone files and not Blu-ray discs. To our surprise, we found that denoising worked properly in PowerDVD irrespective of ESVP when the HQV Benchmark Blu-ray was used! We decided not to let that alter the scores above. Blu-rays are already mastered carefully, and don't need as much post processing as local files from recorded TV shows or camcorder files. The low score of the DDR3 based 6450 will probably improve a great deal after driver updates, but we will consider only playback of files on the hard drive in the rest of this review.
Facebook
Twitter
RSS
70 Comments
View All Comments
enki - Monday, June 13, 2011 - link
How about a short conclusion section for those who just use a Windows 7 box with a Ceton tuner card to watch hdtv in Windows Media Center? (i.e. will just be playing back WTV files recorded directly on the box)What provides the best quality output?
What can stream better then stereo over HDMI? On my old 3400 ATI card it either streams the Dolby Digital directly (the computer doesn't do any processing of the audio) or can output stereo (doesn't think there can be more then 2 speakers connected)
Thanks
BernardP - Monday, June 13, 2011 - link
The inability to create and scale custom resolutions within AMD graphics drivers is, for me, a deal-breaker that keeps me from even considering AMD graphics. It will also keep me from Llano, Trinity and future AMD Fusion APU's. I'll stay with NVidia as long as they keep allowing for custom resolutions.My older eyes are grateful for the custom 1536 X 960 desktop resolution on my 24 inch 16:10 monitor. I couldn't create this resolution with AMD graphics drivers.
bobbozzo - Tuesday, June 14, 2011 - link
In your case, you should just increase the size of the fonts and widgets instead of lowering the screen res.Assimilator87 - Tuesday, June 14, 2011 - link
I wish there was a section dedicated to the silent stream bug. I have a GTX 470 hooked up to an Onkyo TX-SR805 and this issue is driving me insane. For instance, does this issue only plague certain cards or do all nVidia suffer from it? I was hoping the latest WHQL driver (275.33) would fix this, but sadly, no. Otherwise, the article was amazing and I'll definitely have to check out LAV Splitter.ganeshts - Tuesday, June 14, 2011 - link
The problem with the silent stream bug is that one driver version has it, the next one doesn't and then the next release brings it back. It is hard to pinpoint where the issue is.Amongst our candidates, even with the same driver release, the GT 520 had the bug, but the GT 430 didn't. I am quite confident that the GT 520 issue will get resolved in a future update, but then, I can just hope that it doesn't break the GT 430.
JoeHH - Tuesday, June 14, 2011 - link
This is simply one of the best articles I have ever seen about HTPC. Congrats Ganesh and thank you. Very informative and useful.bobbozzo - Tuesday, June 14, 2011 - link
Hi, Can you please compare hardware de-intelacing, etc., vs software?e.g. many players/codecs can do de-interlacing, de-noise, etc. in software, using the CPU.
How does this compare with a hardware implementation?
thanks
ganeshts - Tuesday, June 14, 2011 - link
This is a good suggestion. Let me try that out in the next HTPC / GPU piece.CiNcH - Wednesday, June 15, 2011 - link
Hey guys,here is how I understand the refresh rate issue. It does not matter weather it is 0.005 Hz off. You can't calculate frame drops/repeats from that. In DirectShow, frames are scheduled with the graph reference clock. So the real problem is how much the clock which the VSync is based on and the reference clock in the DirectShow graph drift from each other. And here comes ReClock into play. It derives the DirectShow graph clock from the VSync, i.e. synchronizes the two. So it does not matter weather your VSync is off as long as playback speed is adjusted accordingly. A problem here is synchronizing audio which is not too easy if you bitstream it...
NikosD - Thursday, June 16, 2011 - link
Nice guide but you missed something.It's called PotPlayer, it's free and has built-in almost everything.
CPU & DXVA (partial, full) codecs and splitters for almost every container and every video file out there.
The same is true for audio, too.
It has even Pass through (S/PDIF, HDMI) for AC3/TrueHD/DTS, DTS-HD. Only EAC3 is not working.
It has also support for madVR and a unique DXVA-renderless mode which combines DXVA & madVR!
I think it's close to perfect!
BTW, in the article says that there is no free audio decoder for DTS, DTS-HD.
That's not correct.
FFDShow is capable of decoding and pass through (S/PDIF, HDMI) both DTS and DTS-HD.
And PotPlayer of course!