Sign in with
Sign up | Sign in

CyberLink Powers Up

CUDA-Enabled Apps: Measuring Mainstream GPU Performance
By

The workstation world may be full of CUDA-friendly visualization apps, but so far in the consumer world, transcoding is king for CUDA. Aside from Folding@home and SETI@home, every single application on Nvidia’s consumer CUDA list involves video editing and/or transcoding. Dig into Nvidia’s CUDA Zone site and you’ll discover other projects that promise good things for CUDA’s future in speech recognition, digital surveillance, audio editing/mixing, and gaming (see the UT3 PhysX mod pack with three CUDA-accelerated levels—Lighthouse, Tornado, and HeatRay). Today, though, we have video. Hopefully, we can extrapolate from the early results seen with these apps a similar sort of benefit waiting for a wider diversity of apps in the future.

I say that with some reservation. CUDA isn’t a magic wand that can accelerate everything. Even within the specific field of transcoding, only certain types of operations, such as motion compensation and discrete cosine transform (DCT), lend themselves to rampant parallelization. Many functions don’t. Developers don’t simply say, “hey, let’s coda for CUDA,” have a good chuckle over their wit, and get a 20x performance boost two or three weeks later. The application must contain functions that can leverage parallelism in a way that jibes with CUDA’s architecture.

CyberLink’s PowerDirector 7 was the first application to use the CUDA Video Encoder Library. Today, that library only supports H.264 encoding—an important point, as we’ll see in a minute. The excellence of H.264 as a codec has been amply documented, easily trouncing the likes of MPEG-2 for efficiency and image quality. The trade-off is that encoding and decoding with H.264 takes a crushing amount of processing power. This is why you haven’t seen integrated graphics chipsets try to tackle H.264 until very recently. The decoding load shoved onto the CPU in order to play a Blu-ray disc more or less redlined the system. Is it any wonder that encoding a Blu-ray rip into MPEG-4/H.264 can consume a modern PC for an entire day?

In the end, of course, H.264 is only useful if you can play back files in MPEG-4/H.264 format, such as on an iPod or a PC. Vendors like to assume that you’ll be using H.264 encoding for your unencrypted home videos, and a few of you just might. But whatever your content is, the fact remains that a lot of us now own high-def TVs and the days of disc media are numbered. The ways in which we’ll be able to enjoy HD media files will grow ever wider. If you’re not an everyday transcoder yet, your time may be coming soon, and in the foreseeable future, H.264 will play a big role in that.

Ask a Category Expert

Create a new thread in the Reviews comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display all 56 comments.
This thread is closed for comments
  • 1 Hide
    SpadeM , May 18, 2009 7:04 AM
    The 8800GS or with the new name 9600GSO goes for 60$ and delivers 96 stream processors. Would it be correct to assume that it would perform betwen the 9600 GT and 9800 GTX you reviewed?

    Other then that great article, been waiting for it since we got a sneak preview from Chris last week.
  • 6 Hide
    curnel_D , May 18, 2009 7:08 AM
    And I'll never take Nvidia marketing seriously until they either stop singing about CUDA being the holy grail of computing, or this changes: "Aside from Folding@home and SETI@home, every single application on Nvidia’s consumer CUDA list involves video editing and/or transcoding."
  • 0 Hide
    Anonymous , May 18, 2009 7:15 AM
    As more software will use CUDA, we will not only see a great boost in performance for e.g. video performance, but for parallel programing in general. This sky rocket this business into a new age!
  • 4 Hide
    curnel_D , May 18, 2009 7:18 AM
    l0bd0nAs more software will use CUDA, we will not only see a great boost in performance for e.g. video performance, but for parallel programing in general. This sky rocket this business into a new age!

    Honestly, I dont think a proprietary language will do this. If anything, it's likely to be GPGPU's in general, run by Open Computing Language.(OpenCL)
  • 4 Hide
    one-shot , May 18, 2009 7:23 AM
    Are we both thinking about the same "Pirates 2"? Or am I missing something...
  • 2 Hide
    IzzyCraft , May 18, 2009 7:35 AM
    Who knows it's just a clip he used he could be naming it anything for the hell of it.

    CUDA transcoding is very nice to someone that does H.264 transcoding at a high profile and lacks a 300+ dollar cpu who would spend hours transcoding a dvd on high profile settings.

    Else from that CUDA acceleration has just been more of a feature nothing like a main event. Although can easly be the main attraction to someone that does a good flow of H.264 trasncoding/encoding.

    Encoding/transcoding in h.264 high profile can easily make someone who is very content with their cpu and it's power become sad very quickly when they see the est time for their 30 min clip or something.
  • 0 Hide
    Anonymous , May 18, 2009 7:38 AM
    I'm using CoreAVC since support was added for CUDA h264 decoding. I kinda feel stupid for buying a high end CPU (at the time) since playing all videos, no matter the resolution or bit-rate, leaves the CPU at near-idle usage.
    Vid card: 8600GTS
    CPU: E6700
  • 0 Hide
    IzzyCraft , May 18, 2009 7:49 AM
    Well you lucked in considering not all of the geforce 8 series supports H.264 decoding etc.
  • 2 Hide
    ohim , May 18, 2009 8:01 AM
    they should remove Adobe CS4 suite from there since Cuda transcoding is only posible with nvidia CX videocards not with normal gaming cards wich supports cuda.
  • -2 Hide
    adbat , May 18, 2009 8:05 AM
    CUDA means Miracle in my language :-) I it will do those
    The sad thing is that ATI does not truly compete in CUDA department and there is not standard for it.
  • 0 Hide
    JeanLuc , May 18, 2009 8:26 AM
    I was only really interested in the Badaboom benchmarks and I was fairly impressed but I seem to remember the last time you guys done an article based on GPU accelerated apps (Cuda vs Stream) Badaboom suffered from output quality issues something that hasn't been mentioned in this article. It's all very well a 9800GTX being able to encode HD video content in half the time if the final product is no good.
  • 1 Hide
    cangelini , May 18, 2009 8:56 AM
    Jean,

    Actually, I don't believe we've done a comparison between the two. However, I have read that comparison at other sites, and it's actually ATI's Stream app that has the quality issues. Version two of the software is on the way, and it purportedly fixes the quality issues (though it still isn't demonstrating much GPU scaling, from what I've seen thus far).
  • -1 Hide
    ohim , May 18, 2009 9:17 AM
    cangeliniJean,Actually, I don't believe we've done a comparison between the two. However, I have read that comparison at other sites, and it's actually ATI's Stream app that has the quality issues. Version two of the software is on the way, and it purportedly fixes the quality issues (though it still isn't demonstrating much GPU scaling, from what I've seen thus far).
    yeah but chose your words carefouly since readers could be misslead on this one :)  the quality of the transcoding is related to the aplication used not to the computing technology like cuda or stream.
  • 0 Hide
    Anonymous , May 18, 2009 9:27 AM
    Cangelini, Badaboom definitely has lower quality output compared to the newest x264 builds. I'd definitely like to take advantage of my 9600 GT, but not unless I can use it with Handbrake or some other app on my own terms (NOT BASELINE OR MAIN PROFILE.)
  • -8 Hide
    stlunatic , May 18, 2009 11:46 AM
    I can haz chezberger?

    ATI

    CUDA

    CONA
  • 1 Hide
    randomizer , May 18, 2009 12:34 PM
    SpadeMThe 8800GS or with the new name 9600GSO goes for 60$ and delivers 96 stream processors.

    The 9600GSO has 2 versions (ignoring VRAM variations), one with only 48 SPs (essentially a castrated G94, not G92).
  • 0 Hide
    Anonymous , May 18, 2009 1:12 PM
    There is a plugin for people who do audio engineering/recording/mixing/mastering from this guy:

    http://www.nilsschneider.de

    It runs on CUDA, but TBH, it has not manifested itself as anything special just yet, it's more a "proof of concept". However, as someone who's been doing that kind of thing for years, any quad-core ever made is good enough for real-time audio work, so there's not much point in CUDA acceleration.
  • -5 Hide
    jgoette , May 18, 2009 2:31 PM
    Measuing? Do you not even have spellcheck now?
  • 0 Hide
    Anonymous , May 18, 2009 2:38 PM
    I enjoyed the article, and just like in the dual-core versus quad core debate, there remains few applications that can fully exploit CUDA.

    By the way, I have quick correction. The author writes, "...that can leverage parallelism in a way that jives with CUDA’s architecture." The correct word is "jibe" not "jive."
  • -2 Hide
    1raflo , May 18, 2009 2:51 PM
    CUDA is mostly about hype. Nothing really else.
Display more comments