HTPC Decoding and Rendering Benchmarks: madVR

In the preceding section, we looked at EVR and EVR-CP. Videophiles often prefer madVR as their renderer because of the choice of scaling algorithms available as well as myriad other features. In our original Ivy Bridge HTPC review, I had been very satisfied with HD 4000 and madVR except for a few corner cases involving high frame rate material which also required both luma and chroma scaling (such as 720p60 material). One of the issues in our initial testbed was that we were using DDR3-1333 DRAM. Our current system under consideration uses DDR3-1600. This is more than enough to get madVR working with default scaling algorithm settings for all video material 1080p60 or lesser. Readers interested in seeing madVR in action on the HD 4000 should definitely check out Andrew's excellent piece in Missing Remote comparing HD 2500 and HD 4000 for madVR.

It is not possible to use native DXVA2 decoding with madVR because the decoded frames are not made available to an external renderer directly. To work around this issue, LAV Video Decoder offers three options. The first option involves using software decoding.

LAV Video Decoder (Software Fallback) + madVR
Stream GPU Usage % Power Consumption
     
480i60 MPEG-2 70.84 48.19
576i50 H.264 72.8 50.41
720p60 H.264 75.88 58.23
1080i60 H.264 61.51 59.05
1080i60 MPEG-2 61.22 55.09
1080i60 VC-1 62.22 59.85
1080p60 H.264 73.65 60.91

The second option is to use either QuickSync or DXVA2 Copy-Back. In either case, the decoded frames are brought back to the system memory for madVR to take over. The power consumption profile improves quite a bit, particularly for the 720p60 and 1080p60 streams.

LAV Video Decoder (QuickSync) + madVR
Stream GPU Usage % Power Consumption
     
480i60 MPEG-2 71.37 47.72
576i50 H.264 71.28 49.83
720p60 H.264 75.76 54.92
1080i60 H.264 62.5 56.15
1080i60 MPEG-2 62.02 55.81
1080i60 VC-1 61.86 55.94
1080p60 H.264 66.31 56.58

One of the interesting features to be integrated into the recent madVR releases is the option to perform DXVA scaling. This is particularly interesting for HTPCs running Intel GPUs because the Intel HD Graphics engine uses dedicated hardware to implement support for the DXVA scaling API calls. AMD and NVIDIA apparently implement those calls using pixel shaders. In order to obtain a frame of reference, we repeated our benchmark process using DXVA2 scaling for both luma and chroma instead of the default settings.

LAV Video Decoder (QuickSync) + madVR (DXVA Scaling)
Stream GPU Usage % Power Consumption
     
480i60 MPEG-2 50.33 43.54
576i50 H.264 52.39 44.33
720p60 H.264 57.34 48.82
1080i60 H.264 62.63 55.52
1080i60 MPEG-2 62.34 55.21
1080i60 VC-1 62.06 55.51
1080p60 H.264 65.56 55.33

DXVA scaling results in much lower GPU usage for SD material in particular with a corresponding decrease in average power consumption too. Users with Intel GPUs can continue to enjoy other madVR features while giving up on the choice of a wide variety of scaling algorithms.

HTPC Decoding and Rendering Benchmarks: EVR / EVR-CP Software Interface: XBMC and JRiver Media Center 18
Comments Locked

138 Comments

View All Comments

  • iwayman1001 - Monday, January 21, 2013 - link

    I think most if not all commenters here are completely missing the point on Jriver. Jriver can be your DLNA server, your media cloud server located in your own house. I've tried most MC software, WMC, TotalMedia, NextPVR, BeyondTV, XBMC, etc... None of them can be set up easily as media server cloud so that you can watch your live/recorded TV, all your ripped TV shows, movies, songs, etc over the Internet, your Android phones, your iPhone. You can watch your US live TV, your own movies, recorded TV shows, songs while you'r in Europe provided you have Internet access in the hotel or your smart phone. It take 5 minutes in Jriver to set that up after you build your home Jriver's media library. You do not have to know about public IP/private IP address, etc. Tryi it and you will find out all other MC servers are just for in home, not roaming on the road like Jriver provides.
  • Monkeysweat - Monday, January 21, 2013 - link

    were you using a RC candidate or last stable release (v.11)?
  • ganeshts - Monday, January 21, 2013 - link

    Frodo RC2

    However, I have seen the VC1 issue in previous stable builds too.
  • Iketh - Monday, January 21, 2013 - link

    I haven't been able to respond to a post on this site for a couple months now. Can only make a new thread. I'm using IE9. When trying to reply to someone, it hangs with the working GIF twirling forever. Probably is related to the fact that I can't stay logged in from 1 page to the next either...
  • Laststop311 - Tuesday, January 22, 2013 - link

    Maybe I'm in the minority but I like an HTPC that can game as well. Yeah the htpc cases that can properly operate a full size high end gpu are bigger and louder. But with careful hardware choices and quality noctua fans you can make them nearly silent. You got to have a good furniture set up as well to make gaming with a wireless keyboard and mouse a reality. Proper high end gpu will give you better video playback in some situations as well.

    I can see a place for both. But I don't think ivy bridge is a good choice to jump in on the fanless htpc's. Haswell is perfectly suited for this application, power consumption lowered cpu performance and especially gpu performance increased greatly as this is a tock release which is always the best one. The Haswell version of this pc should run even cooler temps while providing better performance, especially on the gpu side (and intel better have fixed the 24hz bug)
  • gamoniac - Tuesday, January 22, 2013 - link

    Thanks Ganesh, for comparing Win7 and Win8, and Metro vs Silverlight video rendering. I am surprised that the Netflix Metro apps are so much more efficient. Having just switched to Win8 two days ago, after reading this article, I checked and am able to confirm that on Win8, Netflix Metro app uses only 2% of my 6-core AMD 1090T CPU (on SSD), compared to the 8% of desktop IE10 browser (Silverlight), which is still better than Win7.

    Furthermore, the Metro Netflix app better renders in HD than in Silverlight, which periodically fails to render in HD for certain movies. Thanks again.
  • don_k - Tuesday, January 22, 2013 - link

    I probably ask for this in just about every article in this, truly excellent, website so at the risk of becoming a broken record, could we please oh please have linux tests to go along with Windows on these things?

    HTPC, 'enterprise' product tests, file server type tests, all of these are simply incomplete without linux
    testing without going into the reasons why as it will likely result in yet another flame war :)

    I realise Linux isn't exactly your area of expertise but is it really that much more difficult to boot a linux based XBMC[1] live CD than it is to sit through yet another Windows installation?

    Please consider doing this, I would love nothing more than to see Linux tests in my favourite hardware/review website.

    http://mirrors.xbmc.org/releases/XBMCbuntu/xbmcbun...
  • ganeshts - Tuesday, January 22, 2013 - link

    We definitely do Linux testing in our NAS reviews (using a CentOS guest OS). Also, my primary workplace m/c is RHEL 6 :)
  • don_k - Monday, January 28, 2013 - link

    Glad to hear it! :)

    So, are you going to be testing HTPCs with linux based XBMC or..?
  • coolhund - Tuesday, January 22, 2013 - link

    Nice HTPC setup, but Windows 8? Really?
    How much did MS pay for that?

Log in

Don't have an account? Sign up now