Ed's View - Threats, Volume 2

[Ed Milbourn]

In Volume 1 of this dissertation regarding various factors that threaten the viability of HDTV, I discussed those that, in my view, are the top three: Compromised Production Values, Bandwidth Conservation, and Spectrum Super Packing. The next three are somewhat less critical, but nonetheless threaten to compromise DTV in general and HDTV in particular.

Interconnectivity Convolution
Much of the enjoyment of HDTV depends on...

[url=http://www.hdtvmagazine.com/articles/2006/11/eds_view_threat_1.php]Read the Full Article[/url]

[Richard]

Hi Ed,

To solve this problem (and it can be solved), the industry must develop a one connector "plug-n-play" A/V interconnection standard that is media transparent.

Once HDMI becomes the standard connector on all of our products doesn't that qualify? Is there another standard trying to take it's place?

Thanks

[Ed Milbourn]

Thanks for your question. HDMI is designed as a baseband secure digital A/V point-to-point interface connector technology to couple a display system to a baseband source. The HDMI protocol does not support packetized compressed data, have a user control mechanism, have a device discovery portocol, support an open middleware layer nor any network protocol. HDMI does a great job doing what it is supposed to do, but it is not a universal media interface. Ed

[Richard]

What product(s) would I buy or subscribe to making these other parameters an issue?

Thanks

[Ed Milbourn]

Essentially, any source of unidirectional or bi-directional A/V systems would be compatible, ranging from games, disc, broadband (wired and/or wireless), Cable, OTA, DBS, camera, camcorder . . .the list seems to be endless.

Basically, the idea is this: Once the user acquired the device, it would automatically become part of a personal network, with either one connection (if wired) or no connections if wireless. When the device is connected to any other device on the network, a "discovery" process takes place, which identifies the device and acquires its signal source. Each device provides its own unique uses interface to the display system as "applets" to the display controller. Of course, each device would work independent of the network if the user so chooses.

This type of network operation, similar in concept to USB but in a network configuration, is not new. The technology has been developed and proven, and it works as advertised. But the CE industry, with all of the complex IP, licensing, security and standardization issues has been unable to agree to a compresensive standard. Ed

[Richard]

I get the wireless part but I guess I am missing something...

Isn't HDMI unidirectional already for the application you have described? While not plug and play yet that only requires software from the source to take the EDID from the display and configure itself.

[Ed Milbourn]

Yes, certainly you could use the HDMI physical layer for a hard wired network, but the rest of the layers would require additions and extensive modifications. In addition, HDMI would not be extensible to a wide area network if the traffic were to carry more than one HDTV baseband data stream.

However, there are HDMI ring network protocols written and proposed as well as modifications of the SDI systems used by many television broadcasters. These have not been seriously considered for a non-commercial universal network for cost and robustness reasons.

In addition, there have been proposals to use a modified version of the HDMI protocol, but change the physical layers to fibre optics for hard wired applications and Ultra Wide Band (super spread spectrum) for wireless. Sometimes the best is the enemy of the good. Ed

[Richard]

Hey Ed,

Thank you for your patience. I think I finally get it...

You are referring to multimedia networks such as Voom as a simplisitc example where 5 people could feasibly be watching something different but using one server for all. Another crude example being Mitsubishi's push for firewire networks over 4 years ago that would have allowed daisy chaining of products rather than individual direct connections. If I still don't get it please let me know.