AMD's Trinity : An HTPC Perspective
by Ganesh T S on September 27, 2012 11:00 AM ESTIn the last few HTPC reviews, we have incorporated video decoding and rendering benchmarks. The Ivy Bridge review carried a table of values with the CPU and GPU usage. The Vision 3D 252B review made use of HWInfo's sensor graphs to provide a better perspective. In the latter review, it was easier to visualize the extent of stress that a particular video decode + render combination gave to the system. Unfortunately, HWInfo doesn't play well with the A10-5800K / Radeon HD 7660D yet. In particular, GPU loading and CPU package power aren't available for AMD-based systems yet.
The tables below present the results of running our HTPC rendering benchmark samples through various decoder and renderer combinations. Entries in bold with a single star indicate that there were dropped frames as per the renderer status reports in the quiescent state, while double stars indicate that the number of dropped frames made the video unwatchable. The recorded values include the GPU loading and power consumed by the system at the wall. An important point to note here is that the system was set to optimized defaults in the BIOS (GPU at 800 MHz, DRAM at 1600 MHz and CPU cores at 3800 MHz).
madVR :
madVR was configured with the settings mentioned in the software setup page. All the video post processing options in the Catalyst Control Center were disabled except for deinterlacing and pulldown detection. In our first pass, we used a pure software decoder (avcodec / wmv9 dmo, through LAV Video Decoder) to supply madVR with the decoded frames.
| LAV Video Decoder Software Fallback + madVR | ||
| Stream | GPU Usage % | Power Consumption |
| 480i60 MPEG-2 | 38 | 77.9 W |
| 576i50 H.264 | 24 | 68.2 W |
| 720p60 H.264 | 49 | 106.6 W |
| 1080i60 H.264 | 81 | 128.1 W |
| 1080i60 MPEG-2 | 85 | 115.4 W |
| 1080i60 VC-1 | 84 | 131.7 W |
| 1080p60 H.264 | 51 | 116.6 W |
madVR takes up more than 80% of the resources when processing 60 fps interlaced material. The software decode penalty is reflected in the power consumed at the wall, with the 1080i60 VC-1 stream consuming more than 130W on an average. The good news is that all the streams played without any dropped frames with the optimized default settings.
The holy grail of HTPCs, in our opinion, is to obtain hardware accelerated decode for as many formats as possible. A year or so back, it wasn't possible to use any hardware decoders with the madVR renderer. Thanks to Hendrik Leppkes's LAV Filters, we now have a DXVA2 Copy-Back (DXVA2CB) decoder which enables usage of DXVA2 acceleration with madVR. The table below presents the results using DXVA2CB and madVR.
| LAV Video Decoder DXVA2 Copy-Back + madVR | ||
| Stream | GPU Usage % | Power Consumption |
| 480i60 MPEG-2 | 44 | 76.8 W |
| 576i50 H.264 | 24 | 66.2 W |
| 720p60 H.264 | 54 | 102.4 W |
| 1080i60 H.264 ** | 72 | 111.1 W |
| 1080i60 MPEG-2 * | 82 | 111.8 W |
| 1080i60 VC-1 * | 84 | 111.6 W |
| 1080p60 H.264 ** | 64 | 110.4 W |
There is a slight improvement in power consumption for the first few streams. We still have a bit of power penalty compared to pure hardware decode because the decoded frames have to get back to the system memory and then go back into the GPU for madVR to process. An unfortunate point to note here is that none of the 1080i60 / 1080p60 streams could play properly with our optimized default settings (rendering their GPU usage and power consumption values meaningless). We did boost up the memory speeds to DDR3-2133 and saw some improvements with respect to the number of dropped frames. However, we were unable to make the four streams play perfectly even with non-default settings.
EVR-CP :
For non-madVR renderers, we set Catalyst 12.8 to the default settings. The table below presents the results obtained with LAV Video Decoder set to DXVA2 Native mode. All the streams played perfectly, but the power numbers left us puzzled.
| LAV Video Decoder DXVA2 Native + EVR-CP | ||
| Stream | GPU Usage % | Power Consumption |
| 480i60 MPEG-2 | 26 | 78.1 W |
| 576i50 H.264 | 22 | 78.1 W |
| 720p60 H.264 | 38 | 90.1 W |
| 1080i60 H.264 | 69 | 103.9 W |
| 1080i60 MPEG-2 | 69 | 102.2 W |
| 1080i60 VC-1 | 69 | 104.2 W |
| 1080p60 H.264 | 60 | 98.4 W |
For SD streams, the power consumed is almost as much as madVR with software decode. However, the HD streams pull back the numbers a little. This is something worth investigating, but outside the scope of this article. However, we wanted to dig a bit into this, and decided to repeat the tests with the EVR renderer.
EVR :
With Catalyst 12.8 in default settings and LAV Video Decoder set to DXVA2 Native mode, all the streams played perfectly with low power consumption. All post processing steps were also visible (as enabled in the drivers)
| LAV Video Decoder DXVA2 Native + EVR | ||
| Stream | GPU Usage % | Power Consumption |
| 480i60 MPEG-2 | 27 | 60.6 W |
| 576i50 H.264 | 25 | 60.1 W |
| 720p60 H.264 | 35 | 65.7 W |
| 1080i60 H.264 | 67 | 80.1 W |
| 1080i60 MPEG-2 | 67 | 80.6 W |
| 1080i60 VC-1 | 67 | 82.5 W |
| 1080p60 H.264 | 59 | 79.2 W |
A look at the above table indicates that hardware decode with the right renderer can make for a really power efficient HTPC. In some cases, we have more than 20 W difference depending on the renderer used, and as much as 40 W difference between software and hardware decode with additional renderer steps.
Facebook
Twitter
RSS
49 Comments
View All Comments
Denithor - Tuesday, October 2, 2012 - link
Which explains 1600p on 30" monitors, right?Granted, most of us don't sit 3-4' from our TV but I know even on my 50" 1080p barely cuts it (text is hard to read sometimes, fuzzy if I zoom in enough to read easily).
OCedHrt - Thursday, September 27, 2012 - link
Meaning, setting it to 16-235 means to discard 0-15 and 236-255 and expand the remainder to full RGB.Obviously I don't have a Trinity setup so I'm just speculating, but on my HD6400 there is a different parameter on the display configuration section to tweak screen output range - which I set to RGB full range.
ganeshts - Thursday, September 27, 2012 - link
I think you are referring to the pixel format output which is YCbCr 4:4:4 / YCbCr 4:2:2 / RGB Limited / RGB FullThe dynamic range aspect is orthogonal to the pixel format output over HDMI.
The screenshot posted is that of a video playing in the background. Sorry if that wasn't clear. I am not sure about AMD's terminology here, but any user setting the dynamic range to 16-235 would expect NOT to see values 0 - 15 and 236 - 255.
OCedHrt - Friday, September 28, 2012 - link
Yes I was referring to pixel format output. I use RGB Full. I was under the impression that YCbCr cannot display the ranges 0-15 and 236-255 but I think I might be wrong on this one. It is YV12 / YUY2 colorspaces that lack these ranges.And what you're saying about dynamic range is exactly what I'm saying is happening. If you select 16-235, 0-15 and 236-255 from the video is filtered out and the remaining is expanded back to 0-255. Thus a video decoded to YV12 / YUY2 space played on a full range display would have a greyish black or white without selecing 16-235 range. Meaning, the wording on AMD's UI is correct, just the whole idea behind it is confusing.
superccs - Friday, September 28, 2012 - link
Have all of our expectations of their new hardware dropped considerably? I am an AMD fan as much as the rest of you, but it just seems like we are trying so hard to find their stuff useful.CeriseCogburn - Saturday, October 13, 2012 - link
Good deal, another fanboy zombied out for years by the marketing hatred and hype useful idiots collective has shown a glimmer of light, hope that the slave mind can break free from the dirty chains.The new test is this: Would you put up with this crap from any other company or vendor ?
Hardcore69 - Friday, September 28, 2012 - link
- HTPC box: No point. A G540 + GTX 650 if you really want MadVR and 23.976.- Office box: No point. A G540 is enough for a basic everyday system
- Gaming box: No point. A dedicated card is still the answer for 1080p High/Ultra gaming i.e. real PC gaming.
Well? APU's are rather pointless. All this accelerated media crap, HD 1000 can do that too.
Medallish - Friday, September 28, 2012 - link
- HTPC Box: that's passively cooled: An A10-5700 would work great in there and be a nice upgrade!- Office/Workstation Box: GPU acceleration can make a lot of difference, not to mention people have different needs.
- Gaming Box: For someone who wants to game but don't want to shell out the money needed to get 1080p Ultra graphics, or as I see it, a gaming starter kit.
Well? APU's have plenty of point if you're not an out of touch Intel fanatic. Also did you even read the review? There was encoding and decoding that the APU did really well.
btw. I have a passively cooled HTPC, and a Laptop I use for office work, both based on APU's(Currently Llano, the HTPC is getting a Trinity upgrade though.) and I wouldn't want them any other way.
ssj3gohan - Monday, October 1, 2012 - link
Passively cooling a 130W box? Really?I'd like to see AMD trying a bit harder to keep their power consumption down, because in the end the reason for me to choose an i5-3570K was that like AMD it offered 'enough' GPU power, but at a much lower max power. My computer runs at well under 10W idle and about 75W max (OCCT+Furmark), more like 45W in normal use. I wouldn't be able to get near that kind of power consumption with equally-featured Trinity parts (aside from the lower CPU performance, which isn't really a big deal tbh).
(by the way, my 5.9W core i5-computer: http://ssj3gohan.tweakblogs.net/blog/8217/fluffy2-...
Medallish - Friday, October 5, 2012 - link
Yup, I've been working on my own little HTPC project(Although not as cool as yours :D). The Streacom FC5-OD is surprisingly good at cooling down even a 100W APU, right now I'm using a 3870k, I'm planning on getting the A10-5700 asap, and the final touch I plan on adding is the 6670, connect it to the opposite cooling ribs, however right now I'm running into a PSU limit, that I plan on countering by getting a slightly better PSU(250W CarPC PSU instead of a 150W picoPSU)But yeah despite the slightly higher load, the fact is on idle, and most likely average, AMD have really brought down power consumption with Trinity. But I like your setup, and will probably borrow a few ideas from there.