Searching for an Ideal Streaming Technology report by Dr. Edgar Huang and Clifford Marsiglio of Indiana University-Purdue University.
Based on their Methodology, the Findings page ranks the codecs in the following order (best to least): VX30, Flash, WMV, Real, QuickTime.
I’ll be honest: I never heard of VX30. Checked out their website, but couldn’t find any customers (though their showcase worked well and looked fine). More searching quickly turned up a sordid story on about the codec and company. Curious indeed.
Not sure how relatively unknown VX30 made it to the list, MPEG4/H264/3GPP did not.
I was also surprised by the report’s finding that Flash was noted for the best image quality (Flash 8 On2 VP6).
The study’s scope does not include the ability to work with the files for further enhancements such as synchronization, clipping/quoting, or other Web-based video tools in development to do things with “opaque” media. It also doesn’t cover how flexible or “open” they may be to transcoding, or their basis in open standards. To this end I also wish Ogg Theora was on the list.
Regardless, at first glance this is an interesting comparison worth further investigation.
August 23, 2006 at 4:29 pm |
Hi Obadiah,
Sorry out study didn’t go far enough for you.
It was intended to compare a few commonly used technologies that we were pretty sure would be able to run on a good percentage of our audience — for example, I can pretty much guarentee you that there isn’t even 10% of the computers that could run Theora without having to install something new. I’m a geek and I know this technologies limitations.
But we were studying technologies and packages that would work — the VX30 is actually used in places that you wouldn’t notice, but has a large following. Mainly because of its ‘playerless’ technologies. I’ll leave the sorid stories to someone else because while I’m a FOSS fan, our study didn’t involve politics or otherwise.
As for “MPEG4/H264/3GPP” — those are not plugin technologies, they are encoding types. The QuickTime encoding I used was actually H264. Its a nice standard. I’m pretty sure that the streams could be played on a Linux or BSD box — but I haven’t tried it out. My personal preference is using open standards over other technologies but again, thats not what this study was about.
Thanks for taking the time to read though our publication and feel free to send a note privately if you want to discuss this further!
August 28, 2006 at 4:09 am |
Hi Clif,
The study’s great. I was just thinking about other factors that go into the decision of going with a particular codec. You obviously know a lot more than me about the technologies. As I said, some of the findings were surprising, ie., Flash rated the highest quality and QT the least. I think about things that I want to _do_ with the media down the road and try not to get locked in to something inflexible. To that end, I’ve been leaning toward Flash for ease-of-use and relative ease to build interactive tools around the media (countered by it’s price), and QT for it’s H.264 or MPG-4 encoding for download. At first glance the report might not give a newbie a sense of these other factors critical in this expensive decision (yes, free is costly if it locks you in). But as I said, I understand that this is out of scope. Cheers, and let’s be in touch.