Video Transcoding Performance

x264 HD 3.03 Benchmark

Graysky's x264 HD test uses x264 to encode a 4Mbps 720p MPEG-2 source. The focus here is on quality rather than speed, thus the benchmark uses a 2-pass encode and reports the average frame rate in each pass.

x264 HD Benchmark - 1st pass - v3.03

x264 HD Benchmark - 2nd pass - v3.03

CPU based video transcode performance is as good as it can get from AMD here given the 2/4 module/core setup of these Trinity APUs. Intel's Core i3 3220 is a bit slower than the A10-5800K. We're switching to a much newer version of the x264 HD benchmark for our new test suite (5.0.1). Some early results are below if you want to see how things change under the new test:

x264 HD 5.0.1 Benchmark
  1st Pass 2nd Pass
AMD A10-5800K (3.8GHz) 33.5 fps 7.41 fps
AMD A8-5600K (3.6GHz) 32.2 fps 7.12 fps
Intel Core i3 3220 (3.3GHz) 35.2 fps 6.61 fps
Content Creation Performance Discrete GPU Gaming Performance
POST A COMMENT

178 Comments

View All Comments

  • creed3020 - Tuesday, October 2, 2012 - link

    I would really appreciate it if a similar test of Trinity as what was done to Llano regarding GPU Performance vs. Memory Speed was completed (http://www.anandtech.com/show/4476/amd-a83850-revi... I am curious if the trend has remained the same, improved, or decreased.

    I am in the process of building a new Office PC for family whose needs are basic and Trinity fits the bill quite well, especially the A8-5500 or A6-5400K. I want to purchase memory that compliments the GPU well.

    On another GPU note I find it strange that there was no test of AMD Radeon Dual Graphics (http://www.amd.com/us/products/technologies/dual-g... as that a native scaling of GPU platform for this APU, not a high end discrete GPU. The latter usage scenario just doesn't seem that common considering the target market for the APU.
    Reply
  • Hubb1e - Tuesday, October 2, 2012 - link

    Go with 1866 or 1600. 1866 is about 5% faster GPU performance if that matter much in your use case. Reply
  • creed3020 - Wednesday, October 3, 2012 - link

    My current HTPC uses an A8-3850 with 1866 memory so I am aware of the benefits, my question is more about getting an understanding of this phenomenon with Trinity. I am curious if it was has become less important or perhaps even more so.

    I'm not gaming on my APU so there is no concern to squeeze every drop of FPS out of the GPU. I am more curious from a research and review standpoint.
    Reply
  • Moizy - Tuesday, October 2, 2012 - link

    Anand, you mentioned several times that Trinity holds the integrated graphics and overclock
    advantage, while Intel holds the single-threaded and power consumption advantage. To me, though, the A10-5700 attempts to address the power consumption advantage by offering a lower tdp without cutting down the clocks too much (while sacrificing overclockability though).

    Throwing in the A10-5700 at some point in the future, assuming you can get your hands on one, would provide an interesting comparison for those interested in Trinity's gpu and competitive power consumption.
    Reply
  • ewilliams28 - Tuesday, October 2, 2012 - link

    apologies if it's been covered but i would like to know exactly which cards work in this mode. I have heard that if you go too high they don't work together. it's my understanding that 7670 are OEM only and i can't believe that the 6670 that i can buy is still the best i can do. i plan to use 1080p since 1920x1200 has basically gone the way of the dodo bird. but i do like to crank up the settings. luckily the most complicated game i play is World of Warcraft. i will probably fold with this box though.

    Reply
  • creed3020 - Tuesday, October 2, 2012 - link

    http://www.amd.com/us/products/technologies/dual-g... Reply
  • halbhh2 - Tuesday, October 2, 2012 - link

    Overall, a Trinity laptop would do fine during idle, which *is about 85%* of what 90% of laptops do when they are on.

    That matters.

    So, an interesting test for real-world use for *most* consumers (wife, kids, most of the people most of us know) would be a run-time battery life test for leaving the computer on, and surfing to 25 web pages, and playing a couple of modest games for 45 minutes, and then watching a streamed movie.

    That would be real world use for 90% of laptops.

    In view of that, for people that aren't using their laptop in a demanding way, a good question is how much does it cost, and how long does it run until you need to plug in. That's all.
    Reply
  • jfelano - Tuesday, October 2, 2012 - link

    Why does Anandtech still use 1280x1024 and 1680x1050 as their bencmarks? Is this still 2008? Reply
  • Beenthere - Tuesday, October 2, 2012 - link

    Answer: Because most people still use these screen resolutions. This review is for a desktop APU, thus the appropriate screen resolutions. Reply
  • silverblue - Tuesday, October 2, 2012 - link

    Hehe... you try playing anything remotely recent at a higher resolution on an IGP... Reply

Log in

Don't have an account? Sign up now