HD Video Decode Quality and Performance Summer '07
by Derek Wilson on July 23, 2007 5:30 AM EST- Posted in
- GPUs
HD HQV Image Quality Analysis
We have already explored Silicon Optix HD HQV in detail. The tests and what we are looking for in them have not changed since our first round. Fortunately, the ability of NVIDIA and AMD hardware to actually perform the tasks required of HD HQV has changed quite a bit.
Both AMD and NVIDIA told us to expect scores of 100 out of 100 using their latest drivers and hardware. We spent quite a bit of time and effort in fully evaluating this test. We feel that we have judged the performance of these solutions fairly and accurately despite the fact that some subjectivity is involved. Here's what we've come up with.
The bottom line is that NVIDIA comes out on top in terms of quality. We've seen arguments for scoring these cards differently, but we feel that this is the most accurate representation of the capabilities offered by each camp.
On the low end, both AMD and NVIDIA hardware begin to stumble in terms of quality. The HD 2400 XT posts quite a lack luster performance, failing in noise reduction and HD deinterlacing (jaggies). But at least it poorly deinterlaces video at full resolution. We excluded tests of NVIDIA's 8500 series, as their video drivers have not yet been optimized for their low end hardware. Even so, we have been given indications not to expect the level of performance we see from the 8600 series. We would guess that the 8500 series will perform on par with the AMD HD 2400 series, though we will really have to wait and see when NVIDIA releases a driver for this.
With video decode hardware built in as a separate block of logic and post processing being handled by the shader hardware, it's clear that the horrendous 3D performance of low end parts has bled through to their video processing capability as well. This is quite disturbing, as it removes quite a bit of potential value from low cost cards that include video decode hardware.
Both AMD and NVIDIA perform flawlessly and identically in every test but the noise reduction test. AMD uses an adaptive noise reduction algorithm that the user is unable to disable or even adjust in any way. NVIDIA, on the other hand, provides an adjustable noise reduction filter. In general, we prefer having the ability to adjust and tweak our settings, but simply having this ability is irrelevant in HQV scores.
The major issue that resulted in our scoring AMD down in noise reduction was that noise was not reduced significantly enough to match what we expected. In addition to the tests, Silicon Optix provides a visual explanation of the features tested, including noise reduction. They show a side by side video of a yellow flower (a different flower than the one presented in the actual noise reduction test). The comparison shows a noisy video on the left and a video with proper noise reduction applied on the right. The bottom line is that there is almost no noise at all in the video on the right.
During the test, although noise is reduced using AMD hardware, it is not reduced to the level of expectation set by the visual explanation of the test. Based on this assessment, we feel that AMD noise reduction deserves a score of 15 out of 25. Silicon Optix explains a score of 15 as: "The level of noise is reduced somewhat and detail is preserved." In order to achieve a higher score, we expect the noise to be reduced to the point where we do not notice any "sparkling" effect in the background of the image at all.
By contrast, with NVIDIA, setting the noise reduction slider anywhere between 51% and 75% gave us a higher degree of noise reduction than AMD with zero quality loss. At 75% and higher we noticed zero noise in the image with no detail loss until noise reduction was set very high. Tests done with the noise reduction slider at 100% show some detail loss, but there is no reason to crank it up that high unless your HD source is incredibly noisy (which will not likely be the case). In addition, at such high levels of noise reduction, we noticed banding and artifacts in some cases. This was especially apparent in the giant space battle near the end of Serenity. It seems to us that computer generated special effects seemed to suffer from this issue more than other aspects of the video.
While, ideally, we would like to see artifacts avoided at all cost, NVIDIA has provided a solution that offers much more flexibility than their competition. With a little experimentation, a higher quality experience can be delivered on NVIDIA hardware than on AMD hardware. In fact, because NVIDIA sets noise reduction to default off, we feel that the overall experience provided to consumers will be higher.
We have already explored Silicon Optix HD HQV in detail. The tests and what we are looking for in them have not changed since our first round. Fortunately, the ability of NVIDIA and AMD hardware to actually perform the tasks required of HD HQV has changed quite a bit.
Both AMD and NVIDIA told us to expect scores of 100 out of 100 using their latest drivers and hardware. We spent quite a bit of time and effort in fully evaluating this test. We feel that we have judged the performance of these solutions fairly and accurately despite the fact that some subjectivity is involved. Here's what we've come up with.
| Silicon Optix HD HQV Scores | ||||||
| Noise Reduction | Video Res Loss | Jaggies | Film Res Loss | Stadium | Total | |
| AMD Radeon HD 2900 XT | 15 | 20 | 20 | 25 | 10 | 90 |
| AMD Radeon HD 2600 XT | 15 | 20 | 20 | 25 | 10 | 90 |
| AMD Radeon HD 2600 Pro | 15 | 20 | 20 | 25 | 10 | 90 |
| AMD Radeon HD 2400 XT | 0 | 20 | 0 | 25 | 10 | 55 |
| NVIDIA GeForce 8800 GTX | 25 | 20 | 20 | 25 | 10 | 100 |
| NVIDIA GeForce 8600 GTS | 25 | 20 | 20 | 25 | 10 | 100 |
| NVIDIA GeForce 8600 GT | 25 | 20 | 20 | 25 | 10 | 100 |
The bottom line is that NVIDIA comes out on top in terms of quality. We've seen arguments for scoring these cards differently, but we feel that this is the most accurate representation of the capabilities offered by each camp.
On the low end, both AMD and NVIDIA hardware begin to stumble in terms of quality. The HD 2400 XT posts quite a lack luster performance, failing in noise reduction and HD deinterlacing (jaggies). But at least it poorly deinterlaces video at full resolution. We excluded tests of NVIDIA's 8500 series, as their video drivers have not yet been optimized for their low end hardware. Even so, we have been given indications not to expect the level of performance we see from the 8600 series. We would guess that the 8500 series will perform on par with the AMD HD 2400 series, though we will really have to wait and see when NVIDIA releases a driver for this.
With video decode hardware built in as a separate block of logic and post processing being handled by the shader hardware, it's clear that the horrendous 3D performance of low end parts has bled through to their video processing capability as well. This is quite disturbing, as it removes quite a bit of potential value from low cost cards that include video decode hardware.
Both AMD and NVIDIA perform flawlessly and identically in every test but the noise reduction test. AMD uses an adaptive noise reduction algorithm that the user is unable to disable or even adjust in any way. NVIDIA, on the other hand, provides an adjustable noise reduction filter. In general, we prefer having the ability to adjust and tweak our settings, but simply having this ability is irrelevant in HQV scores.
The major issue that resulted in our scoring AMD down in noise reduction was that noise was not reduced significantly enough to match what we expected. In addition to the tests, Silicon Optix provides a visual explanation of the features tested, including noise reduction. They show a side by side video of a yellow flower (a different flower than the one presented in the actual noise reduction test). The comparison shows a noisy video on the left and a video with proper noise reduction applied on the right. The bottom line is that there is almost no noise at all in the video on the right.
During the test, although noise is reduced using AMD hardware, it is not reduced to the level of expectation set by the visual explanation of the test. Based on this assessment, we feel that AMD noise reduction deserves a score of 15 out of 25. Silicon Optix explains a score of 15 as: "The level of noise is reduced somewhat and detail is preserved." In order to achieve a higher score, we expect the noise to be reduced to the point where we do not notice any "sparkling" effect in the background of the image at all.
By contrast, with NVIDIA, setting the noise reduction slider anywhere between 51% and 75% gave us a higher degree of noise reduction than AMD with zero quality loss. At 75% and higher we noticed zero noise in the image with no detail loss until noise reduction was set very high. Tests done with the noise reduction slider at 100% show some detail loss, but there is no reason to crank it up that high unless your HD source is incredibly noisy (which will not likely be the case). In addition, at such high levels of noise reduction, we noticed banding and artifacts in some cases. This was especially apparent in the giant space battle near the end of Serenity. It seems to us that computer generated special effects seemed to suffer from this issue more than other aspects of the video.
While, ideally, we would like to see artifacts avoided at all cost, NVIDIA has provided a solution that offers much more flexibility than their competition. With a little experimentation, a higher quality experience can be delivered on NVIDIA hardware than on AMD hardware. In fact, because NVIDIA sets noise reduction to default off, we feel that the overall experience provided to consumers will be higher.
Facebook
Twitter
RSS
63 Comments
View All Comments
Wozza - Monday, March 17, 2008 - link
"As TV shows transition to HD, we will likely see 1080i as the choice format due to the fact that this is the format in which most HDTV channels are broadcast (over-the-air and otherwise), 720p being the other option."I would like to point out that 1080i has become a popular broadcast standard because of it's lower broadcast bandwidth requirements. TV shows are generally mastered on 1080p, then 1080i dubs are pulled from those masters and delivered to broadcasters (although some networks still don't work with HD at all, MTV for instance who take all deliveries on Digital Beta Cam). Pretty much the only people shooting and mastering in 1080i are live sports, some talk shows, reality TV and the evening news.
Probably 90% of TV and film related blu-rays will be 1080p.
redpriest_ - Monday, July 23, 2007 - link
Hint: They didn't. What anandtech isn't telling you is that NO nvidia card supports HDCP over dual-DVI, so yeah, you know that hot and fancy 30" LCD with gorgeous 2560x1600 res? You need to drop it down to 1280x800 to get it to work with an nvidia solution.This is a very significant problem, and I for one applaud ATI for including HDCP over dual-DVI.
DigitalFreak - Wednesday, July 25, 2007 - link
Pwnd!defter - Tuesday, July 24, 2007 - link
You are wrong.Check Anand's 8600 review, they clearly state that 8600/8500 cards support HDCP over dual-DVI.
DigitalFreak - Monday, July 23, 2007 - link
http://guru3d.com/article/Videocards/443/5/">http://guru3d.com/article/Videocards/443/5/http://guru3d.com/article/Videocards/443/5/Chadder007 - Monday, July 23, 2007 - link
I see the ATI cards lower CPU usage, but how is the power readings when the GPU is being used compared to the CPU??chris92314 - Monday, July 23, 2007 - link
Does the HD video acceleration work with other programs, and with non blueray/hddvd sources? For example if I wanted to watch a h.264 encoded .mkv file would I still see the performance and image enhancements.GPett - Monday, July 23, 2007 - link
Well, what annoys me is that there used to be all-in-wonder video cards for this kinda stuff. I do not mind a product line that has TV tuners and HD playback codecs, but not at the expense of 3d performance.It is a mistake for ATI and Nvidia to try to include this stuff on all video cards. The current 2XXX and 8XXX generation of video cards might not been as pathetic had the two GPU giants focused on actually making a GPU good instead of adding features that not everyone wants.
I am sure lots of people watch movies on their computer. I do not. I don't want a GPU with those features. I want a GPU that is good at playing games.
autoboy - Wednesday, July 25, 2007 - link
All in wonder cards are a totally different beast. The all in wonder card was simply a combination of a TV tuner card (and a rather poor one) and a normal graphics chip. The TV tuner simply records TV and has nothing to do with playback. ATI no longer sells All in wonder cards because the TV tuner card did not go obsolete quickly, while the graphics core in the AIW card went obsolete quickly, requiring the buyer to buy another expensive AIW card when only the graphics part was obsolete. A separate tuner card made so much more sense.Playback of video is a totally different thing and the AIW cards performed exactly the same as regular video cards based on the same chip. At the time, playing video on the PC was more rare and the video playback of all cards was essentially the same because no cards offered hardware deinterlacing on their video cards. Now, video on the PC is abundant and is the new Killer App (besides graphics) which drives PC performance, storage, and internet speed. Nvidia was first to the party offering Purevideo support, which did hardware deinterlacing for DVDs and SD TV on the video card instead of in software. It was far superior to any software solution at the time (save a few diehard fans of Dscaler with IVTC) and came out at exactly the right time, with the introduction of media center and cheap TV tuner cards and HD video. Now, Purevideo 2 and AVIVO HD introduce the same high quality deinterlacing to HD video for mpeg2 (7600GT and up could do HD mpeg2 deinterlacing) as well as VC-1 and H.264 content. If you don't think this is important, remember that all new satelite HD broadcasts coming online are in 1080i h264, requiring deinterlacing to look its best, and new products are coming and exist already if you are willing to work for it, that allow you to record this content on your computer. Also, new TV series are likely to be released in 1080i on HD discs because that is their most common broadcast format. If you don't need this fine, but they sell a lot of cards to people who do.
autoboy - Wednesday, July 25, 2007 - link
Oh, I forgot to mention that only the video decode acceleration requires extra transistors, the deinterlacing calculations are done on the programable shaders of the cards requiring no additional hardware, just extra code in the drivers to work. The faster the video card, the better your deinterlacing, which explains why the 2400 and the 8500 cannot get perfect scores on the HQV tests. You can verify this on HD 2X00 cards by watching the GPU% in Riva Tuner while forcing different adaptive deinterlacing in CCC. This only works in XP btw.