[opendtv] Re: Netflix's Move Onto the Web Stirs Rivalries
- From: "Manfredi, Albert E" <albert.e.manfredi@xxxxxxxxxx>
- To: "opendtv@xxxxxxxxxxxxx" <opendtv@xxxxxxxxxxxxx>
- Date: Mon, 29 Nov 2010 15:39:48 -0600
Craig Birkmaier wrote:
> There have been a number of advances in this area.
>
> First, most of the current streaming media servers have been developed
> since 2005, including Flash, QuickTime streaming server, Real, Microsoft
> et al. They employ a variety of techniques to improve performance and
> reliability. This site has some good reference material and studies of
> performance conducted between 2006 and 2010:
> http://www.iupui.edu/~nmstream/
Let me start by saying this, to put things in their proper perspective:
I've already noticed that rate adaptive streaming servers or not, my 1.5 Mb/s
broadband connection is getting to be marginal at best. It won't be long before
it simply will not keep up with servers, and will have to interrupt and buffer
all the time (as opposed to just occasionally). So where is all this huge
improvement you claim for H.264, Craig? What I notice is, 1.5 Mb/s USED to be
just fine for streaming media when I first got connected, and now it ain't.
This the the reality.
There is nothing surprising in these pieces you URLed, Craig. Yes, they went to
H.264 and they went to AAC audio to improve (marginally) quality at lower bit
rates, but this doesn't mean that streaming wasn't possible before. It just
means that quality can now be better, in principle, at a given bit rate. The
most significant point is that broadband connections improved to the point that
the higher quality streaming could actually be used. And streaming media are
CONTINUING to demand higher bit rates, as time goes on.
Here is the quote:
"Global broadband adoption has seen big growth in the last several years. A
report by the Organization for Economic Co-operation and Development (OECD)
from June 2007 showed that 60% of member countries' web users were on
broadband. In the United States, over half of the households (53%) had a
broadband connection as of July 2007, which accounted for 72% of all home
Internet subscriptions, up from 60% in 2006, as revealed in a Leichtman
Research report. Although U.S. average broadband connection speed is still slow
(4.9Mbps)compared to those of the top countries, such as Japan (63.6Mbps),
South Korea (49.5Mbps), etc., according to The Information Technology &
Innovation Foundation, it is already fast enough to handle HD streaming videos,
which are mostly encoded with a bit rate in the 2Mbps range. ABC.com's full
episode HD video streaming with a resolution of 1280x720, 60fps is a good
example."
So, take a close look at the last part of that paragraph. Even a 1.5 Mb/s
broadband connection today is not capable of keeping up today's so-called HD
streaming media over the web. Do you still wonder why Netflix couldn't make
their stuff work in 2003? If your premise was true, and the article's premise,
then this much better quality would have been possible without requiring higher
broadband rates than initial broadband speeds. But that's just NOT THE CASE.
> Channel Adaptive streaming techniques have contributed to performance
> improvements in several areas:
Channel adaptive techniques were possible from the very start of this. I'm
talking, the early to mid 1990s. They were built into RTP/RTCP from the start.
I'm not saying that these techniques haven't been tweaked over the years, that
server software hasn't improved. Of course, they have. I'm just saying, you and
the author of the article are simply not seeing the elephant in the room!!
> Sorry, but h.264 streaming rates are typically 30 to 50 percent less than
> the equivalent quality with MPEG-2. It is true things begin to even out at
> higher bit rates, such as those used in broadcast channels. But for
> streaming you will not see much MPEG-2.
Certainly, if the receiver is upgradeable to H.264, that's the preferred
choice. But even your 30 to 50 percent improvement over MPEG-2 (which by the
way is mostly a measure of quality and not a can/cannot do criterion) pales in
comparison to the really significant improvement in this equation. Roughly an
increase of more than 6.5X in the average broadband connection from 2003 to
now, let alone the increase for dial-up users which is close to 100X, and who
were then still the majority and are now a rarity.
> THe quality of streaming sucked until the middle of the past decade, and
> has improved dramatically in the past 2-3 years. ONly a small portion of
> this is due to improvements in delivered bit rates from my AT&T DSL
> service.
Big picture: If the vast majority of Internet users can only get dialup or low
speed broadband, the servers will be transmitting low quality streaming media.
Even if rate adaptive, the servers will only provide quality to a level they
feel is going to be used by most receivers. When the majority of Internet users
have broadband, and the average speed (according to the piece you URLed) is 5
Mb/s, the streaming media servers will gradually ramp up quality levels.
So, the fact that you had broadband early on, and yet you only noticed an
improvement subsequent to getting broadband, proves nothing more than the
servers hadn't yet accommodated higher speed broadband. Not that they COULDN'T.
They didn't, because it wasn't worth their while.
Bert
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at
FreeLists.org
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.
Other related posts: