Sign in with
Sign up | Sign in

Is Performance The Only Variable In Play?

ASRock's E350M1: AMD's Brazos Platform Hits The Desktop First
By

An Intel employee once told me, “video transcoded using CUDA looks like shit.” Honestly, I shrugged him off at the time. Companies say stuff like that on an almost daily basis; you tend to see things with a blue/red/green tint after a while, depending on the organization you’re representing. And as an advocate for the Tom's Hardware audience, I've trained myself to take everything I hear with a dash of skepticism. But I had a number of readers ask for quality comparisons in the comments section of my Sandy Bridge review, so I dusted off a couple of high-def clips and started saving the outputs from CyberLink’s Fusion-optimized version of MediaEspresso to see if there was any credence to those claims.

Reference sourceReference source

The comparison here is really very, very basic. I have four test beds: the E-350-powered E350M1, the Athlon II X2 240e-driven 880GITX-A-E, the IONITX-P-E with Intel’s Celeron SU2300, and the Atom-powered IONITX-L-E. The two boards with Nvidia’s Ion chipset are going to deliver the same output as soon as you start involving hardware-accelerated encode and decode. The other two aren’t powerful enough to even enable hardware-based encode support. So, although we’re technically reviewing ASRock’s E350M1 motherboard here, our quality comparisons becomes CUDA versus software versus two AMD platforms that apply hardware-accelerated decode support.

E-350/UVD3: SoftwareE-350/UVD3: SoftwareIon/CUDA: SoftwareIon/CUDA: SoftwareUVD2/880G: SoftwareUVD2/880G: Software

If you download the full-sized (720p) versions of all three of these software-based images and tab through them, you’ll see the sort of quality variation that’d require you to diff each shot—and that’s in a still frame. For all intents and purpose, they’re the same.

E-350/UVD: Dec. Accel.E-350/UVD: Dec. Accel.Ion/CUDA: Dec. Accel.Ion/CUDA: Dec. Accel.UVD2/880G: Dec. Accel.UVD2/880G: Dec. Accel.

The same goes for all three boards with hardware-accelerated decoding applied. We can clearly see that what comes out of the decoder and is then operated on by the CPU during the encode stage is largely identical. Alternatively, you can tab between the software-only shot and the corresponding hardware decode version shown above to see that the decoded content is the same, whether the process happens on the CPU or GPU.

Ion/CUDA: Enc. Accel.Ion/CUDA: Enc. Accel.

And then we let Ion’s CUDA cores handle the encode stage and things quickly get ugly. The examples I’m using here aren’t even the best shots from our 2:30-long trailer. But if you download the screen shot and compare it to any of the six above, the difference is oh-so-evident.

An even better way to tell would be to download the actual video clips themselves. I threw three examples up on MediaFire: CUDA-based encoding, the Ion platform with hardware-accelerated encode and decode turned off, and AMD’s E-350 with hardware-accelerated decoding enabled. Watch them back to back and see for yourself. You'll notice that the issues are most noticeable in scenes with lots of motion. The best way to describe the issue would be latent blocking or pixelation that distorts the output quality.

Verdict

As far as I’m concerned, sacrificing quality for speed is never alright.

There’s a completely different story for another day here, since we can now turn around and run performance/quality tests on Sandy Bridge, AMD’s discrete graphics, and Nvidia’s add-in cards—all of which offer accelerated encoding. And since there are multiple software apps optimized for all three paths, we can really dig deeper in the days to come.

In the context of media-oriented nettops, though, I’d rather not use Ion’s CUDA-based encode acceleration and get a better picture. That takes away much of the platform’s advantage over AMD’s E-350. Hopefully, the OpenCL-based encoders that emerge later this year, utilizing Llano, are written with our quality-oriented concerns in mind.

Display all 82 comments.
This thread is closed for comments
Top Comments
  • 16 Hide
    cangelini , January 14, 2011 6:35 AM
    johnbilickiBiased; it's awesome all-around. Netbooks aren't going to sport 1680x1050 displays. I'd refer people to the site though they don't understand how to objectively interpret biased reporting.


    Netbooks != nettops. This is a motherboard, which goes into a system on a desk/in an entertainment center. Sorry, your princess is in another castle...
Other Comments
  • 2 Hide
    iam2thecrowe , January 14, 2011 4:18 AM
    now they need some devs to take advantage of that apu to see its full potential as a processor.
  • -2 Hide
    reprotected , January 14, 2011 4:27 AM
    Ironic of how Nvidia quality is worse than AMD.
  • 6 Hide
    Reynod , January 14, 2011 4:27 AM
    This is an awesome processor ...

    Chris ... did you manage to overclock it at all?

    Give it your best shot ... call crashman in with the liquid nitrogen if you need to mate !!

    Really impressive stats for such a small piece of silicon.
  • 0 Hide
    dogman_1234 , January 14, 2011 4:29 AM
    So the Brazo is great for media and hard processing I assume. If someone came to me and asked for a good platrom to watch Blu-Ray...I would say get the Brazo APU for them, right?
  • 3 Hide
    sparky2010 , January 14, 2011 4:38 AM
    Nice, things are starting to look good for AMD, and i hope it stays that way as they start unveiling their mainstream and highend processors, because i'm really fed up with intel dictating crazy prices.....
  • 6 Hide
    cangelini , January 14, 2011 5:03 AM
    reynodThis is an awesome processor ... Chris ... did you manage to overclock it at all?Give it your best shot ... call crashman in with the liquid nitrogen if you need to mate !!Really impressive stats for such a small piece of silicon.


    Didn't get a chance to mess with overclocking. If this is something you guys want to see, I might try to push it a little harder over the weekend.
  • 7 Hide
    joytech22 , January 14, 2011 5:08 AM
    cangeliniDidn't get a chance to mess with overclocking. If this is something you guys want to see, I might try to push it a little harder over the weekend.


    Yeah that would be much appreciated, these little chips are so much faster than Atom, let's see if you can get them to perform similarly to a Dual-Core CPU at 1.8GHz
  • 3 Hide
    cangelini , January 14, 2011 5:10 AM
    Alright, I'll see what I can do. A shiny new video card landed this afternoon, so that's going to monopolize the bench for much of the weekend ;) 
  • 1 Hide
    dEAne , January 14, 2011 5:15 AM
    Yes integration is the key to higher performance, lower power consumption and lower price (affordability this is what people really wanted).
  • 0 Hide
    haplo602 , January 14, 2011 5:23 AM
    can you also run gaming benchmarks with a 5670 or similar plugged into the PCIe slot ? Just to have a look how the limited memory interface will bottleneck ...

    also what happens with the intgrated graphics core when you plug in a discreet GPU ? you gave so much detail about this in the sandy bridge review but totaly skip it for Fusion ...

    the board got me interested. I am trying to buy a small "workstation terminal" ... something to code OpenGL/OpenCL on a budget. Seems this is what I am looking for.
  • 2 Hide
    cangelini , January 14, 2011 5:29 AM
    haplo602can you also run gaming benchmarks with a 5670 or similar plugged into the PCIe slot ? Just to have a look how the limited memory interface will bottleneck ...also what happens with the intgrated graphics core when you plug in a discreet GPU ? you gave so much detail about this in the sandy bridge review but totaly skip it for Fusion ...the board got me interested. I am trying to buy a small "workstation terminal" ... something to code OpenGL/OpenCL on a budget. Seems this is what I am looking for.


    Clearly the focus here was on Brazos' place in the Fusion infrastructure (hence the effort to test video decode/encode). Again, I'll look into adding more performance data with discrete graphics as well, though!
  • 1 Hide
    apache_lives , January 14, 2011 5:59 AM
    about F*** time AMD, good show
  • 1 Hide
    lashton , January 14, 2011 6:01 AM
    yeah overclocking would be good and see how it compares with overclocking of similar playforms
  • 16 Hide
    cangelini , January 14, 2011 6:35 AM
    johnbilickiBiased; it's awesome all-around. Netbooks aren't going to sport 1680x1050 displays. I'd refer people to the site though they don't understand how to objectively interpret biased reporting.


    Netbooks != nettops. This is a motherboard, which goes into a system on a desk/in an entertainment center. Sorry, your princess is in another castle...
  • 2 Hide
    Reynod , January 14, 2011 8:11 AM
    Good point about the mobo Chris.

    Yes we would ove to see some overclocking if possible ...

    This kinda takes me back to the days of Dothan ... remember this landmark article Chris ...

    http://www.tomshardware.com/reviews/dothan-netburst,1041-6.html
  • 1 Hide
    cangelini , January 14, 2011 8:19 AM
    Sure do reynod--those were the days, weren't they?
  • 1 Hide
    i_am_aronman , January 14, 2011 8:35 AM
    sparky2010Nice, things are starting to look good for AMD, and i hope it stays that way as they start unveiling their mainstream and highend processors, because i'm really fed up with intel dictating crazy prices.....

    More like ridiculous prices, i understand them having "top" processors but i don't understand their pricing scheme. I am hoping AMD can pull it together with these processors as well because it will mean Intel will have to get it together as well.
  • -3 Hide
    nevertell , January 14, 2011 8:41 AM
    I was hoping for better performance per watt :( 

    I really would like a tegra 2 notebook with a proper and supported OS.
Display more comments