I am only posting this because both versions have nearly identical file sizes. The Spark version weighs 10,494KB and the VP6 one comes in at 10,639KB.
Be patient with this one, there's almost 20MB of data to be pulled in. Not suited for slow connections - you best mute the video, wait 10 minutes and then refresh once both files are cached.
Click here for yet another Flash 8 video sample.
Be patient with this one, there's almost 20MB of data to be pulled in. Not suited for slow connections - you best mute the video, wait 10 minutes and then refresh once both files are cached.
Click here for yet another Flash 8 video sample.


...Also what settings ?
I'm assuming that 2-pass VBR was used for the Spark video.
Both videos are encoded with identical settings, 1 pass because that's the only option that this particular piece of software offered.
Sorenson Squeeze 4.1 Encoder
http://www.sorensonstore.com/products/sq4mfwin.htm...
...Still the wavelet compression used in the VP6 CODEC is quite amazing, especially for fast moving images.
( looks very simillar to the MS WMV9 CODEC )
A better test in my opinion is to show the best that the Spark Pro CODEC can do at the same data rate, against the best that the VP6 CODEC can do, otherwise you people a false impression that the Spark Pro CODEC is far worse than it really is, which I consider to be dirty pool.
...I've seen the same thing in the past when comparing the Sorenson 3 Pro CODEC against MPEG4. In the right hands, with the right tools, the Sorenson 3 Pro CODEC can look just as good as MPEG4 at the same data rate.
The Dingo
Check the first post in this series, the Heineken clips use 2pass for both spark and VP6 - with that some people didn't like the fact that VP6 ended up with a larger file and wanted a 'same target size' exercise. Well here it is, same file size for both.
What you now suggest is a test that draws its results from a comparison by the naked eye and mucking about with all sorts of tools - and I am sure as soon as I'd post such a test we'd see voices demanding a different methodology yet again. I never claimed for this to be scientific nor do I want to stop anyone using Spark. But if someone just wants to hit encode then this is what they get.
with F7 Spark, if you compress using a fixed quality setting (say 80) instead of a fixed bitrate you get a much better video quality (i.e. without the "pixel" look) and often smaller filesize.
Also Stefan don't worry about the comments. people whine excessively anyway. there will always be someone who won't be happy with the effort you put into it.
Thanks. I appreciate it
thanks for the encouragement.
one more question I'm hearing that Flash player 8 won't have realtime VP6 encoding only decoding. the encoder will be part of flash communication server
I read that here
http://flashvideo.progettosinergia.com/