Archived Tech-Notes
Published by: Larry Bloomfield & Jim Mendrala      The following are our current e-mail addresses:
E-mail = hdtvguy@garlic.com or J.Mendrala@ieee.org
 We have copied the original Tech-Notes below as it was sent out.  Some of the information may be out of date.
**********************************************

DTV Tech Notes

% Larry Bloomfield & Jim Mendrala

(541) 385-9115 or (805) 294-1049

E-mail = larrybend@aol.com

J_Mendrala@compuserve.com

============

April 16, 1998

============

DTV Tech Note - 017

============

     Sharing experiences, knowledge or anything else relating to DTV, HDTV etc. with your fellow engineers: That's what we are all about. We will send this to anyone asking, just E-mail us. Welcome to all the new subscribers. We hope you will participate with question, answers and/or comments.

*************

(Ed note: The following press release was received during.)

 NBC ANNOUNCES PLANS TO PROVIDE -- 1080 HIGH DEFINITION TELEVISION VIA ALL-DIGITAL PLANT   NBC will begin broadcasting a 1080 interlaced high-definition digital signal to its owned stations and affiliates in the fall 1998 season.  The announcement was made at NAB in Las Vegas, Nev.  Initially, NBC plans to provide a high-definition signal during prime time with a 480 progressive digital signal during other dayparts.  

NBC is building an HDTV facility for "The Tonight Show with Jay Leno." This arrangement is an extension of a previously announced agreement, where NBC and Sony have teamed up to develop broadcast products and applications for the next five Olympic Games.  NBC expects to begin providing "The Tonight Show" in 1080 format in 1999.  Additionally, NBC announced that it will offer "Men in Black" and "Titanic" in high-definition and expects to announce plans for longform and series programming after the fall schedule is in place.  

NBC announced the rollout of the first new all-digital Broadcast Operations Center.  This tapeless facility is the backbone of NBC's digital.   NBC's enhanced digital broadcasting initiatives include soon-to-be-unveiled Microsoft Web TV for Windows entertainment programming, along with existing efforts with Wink Communications and Intel's Intercast.  These data broadcasting projects are compatible with both digital and analog broadcast and cable signals.  

NBC is committed to a rapid rollout and is scheduled to begin providing digital signals in four top-ten markets by: November 1, 1998;  KNBC, Los Angeles; WRC, Washington D.C.; WCAU, Philadelphia; and KXAS, Dallas.

May 1, 1999:  Owned stations in New York (WNBC) and Chicago (WMAQ) are scheduled for May 1, 1999, followed by San Diego (KNSD), Miami (WTVJ) and Raleigh (WNCN).    

May 1, 2002, the remaining NBC-owned stations in Columbus (WCMH), Providence WJAR), Hartford (WVIT) and  Birmingham (WVTM) will broadcast a digital signal.

*********************************************************

Subj:  Fox

By:  Larry Bloomfield

 

It was hard to get to anyone at FOX to make a statement and in the brief few minuets I spoke to Andy Setos, top engineer at their network, I got the impression that they were going to go 480p for the time being and possible 720p on some specials.  The attitude I got was a wait and see, but that progressive was far superior to any interlace.  Can't argue with that.  Broadcasting & Cable did an exclusive just before NAB on them and what they plan to do.  They did a follow up on all the 4 major networks, after NAB.  You may find it interesting reading.

***********************************************

Subj:  WB UPN and HDTV

By:  Larry Bloomfield

 

Got a call from one of the engineers at the Warner Brothers (WB) network as the result of inquires as to what they'll be doing.  I was informed that since they provide programming to all the networks, they will do what the respective network the show is for will do.  I got the impression that they are looking at 480p for their own network, but the "we're going to wait and see" attitude was most pronounced.  What did interest me was the smaller than 100 DMA thing that WB will be doing.  Since I'm possibly going to do a story for Broadcast Engineering on that, I'll not say much more there.  They are calling  it "station in a box"  and WEB.  More later.    Probably the most intelligent thing I've heard said to date on this whole matter is that if a 1080 progressive scan algorithm could be developed so it could fit into a 6 MHz. bandwidth, that would be the Holy Grail of TV and I agree.  

The United Paramount Network (UPN) never seems to return calls.  If anyone has a clue as to what they'll be doing, please let us know.  I do know they are busy trying to strike a deal with Disney to fill in they're day parts, but since that has nothing to do with DTV, I didn't pursue the information. Common sense says they'll probably do fairly much the same as WB is doing as far as shows on other networks go, but for their own UPN, Who knows?

*********************************************************

The following was provided by ABC-TV

  FREQUENTLY ASKED QUESTIONS

 

1.      WHY HAS ABC SELECTED PROGRESSIVE SCANNING FORMATS?

* Experts agree that progressive scanning is the best format for capture, transmission and display.

* The inventors of the FCC's DTV Standard, the Grand Alliance, agreed that interlace in the transmission path should be eliminated and that they would work towards that goal.

2.      HOW DOES ABC's SELECTION OF PROGRESSIVE RELATE TO THE FUTURE OF HDTV SETS

FOR CONSUMERS?

* We believe that new flat panel displays, which are just now being introduced, represent the future of true mass consumer adoption of HDTV. Those flat panel displays are inherently progressive. ABC has elected to aim for the second generation of consumer sets in selecting its HDTV format. In the meantime, the first generation glass tube and projection digital television sets are designed to decode and display all ATSC formats, including 720P.

  3.      WILL PROGRESSIVE SCANNING BE COMPATIBLE WITH DECISIONS MADE BY OTHER BROADCASTERS?

* Absolutely. This decision is completely in accordance with the FCC's mandated DTV Standard adopted in 1996. All broadcasters must comply with this standard. As a result, ALL digital televisions being designed for introduction this fall will receive and display either the progressive or interlaced formats. There is no need for concern from consumers that they might have to pick amongst their favorite sources of entertainment.  

4.      WHAT IS THE DIFFERENCE BETWEEN INTERLACED AND PROGRESSIVE?

* Progressive formats represent the most advanced digital broadcasting technology available. Progressive scanning (used in all computer monitors and the advanced flat panel displays) "paints" every line of the picture sequentially every time the screen is scanned. It takes only 1/60th of a second to scan one complete picture. By contrast, interlace scanning, invented over 60 years ago, first paints the odd numbered lines every time the picture is scanned, followed by the even lines. 

* With interlaced formats, any horizontal motion in the original source between frames results in blurring. Similarly vertical motion results in flicker (the "venetian blind" effect). These defects are eliminated in progressive scanned systems. For example in 1080I, a moving football quarterback's image is dissected into 1080 lines ONLY HALF OF WHICH are painted each time the screen is scanned. The portions of the quarterback's body represented by the other half of the lines remain back where he was before he began to move. By contrast, progressive scanning re-paints the quarterback's entire image EACH time the screen is scanned. As a result, progressive formats do a better job of rendering moving pictures.  

5.      IS 720P "REAL" HDTV?

* The Advanced Television Standards Committee (ATSC), the group that established the standards for digital television, defined HDTV to include both 1080I and 720P. The Consumer Electronics Manufacturers Association (CEMA) has likewise defined HDTV to include both 1080I and 720P.  

6.      ARE 720P AND 1080I EQUIVALENT HDTV? 

* Actually, we think 720P is better. To focus solely on the "number of lines" greatly over simplifies the complexities of the human visual system and threatens to mislead and confuse consumers. Contrast and brightness have a greater impact on the human visual system than does resolution. The 720P picture is brighter and has greater contrast than the 1080I picture. 

* The number of lines of resolution in progressive and interlace pictures are not an "apples-to-apples" comparison. In the time it takes 720P to paint 720 lines, 1080I paints only 540 lines. And, by the time 1080I does paint 1080 lines, 720p has painted 1440 lines. 

* In side-by-side subjective testing performed by the Advanced

Television Test Center under the auspices of the FCC's Advisory Committee on Advanced Television Services (ACATS), it was shown that 720P had "no artifacts" under a variety of conditions, while 1080I, under the same conditions, showed "increasing quantization noise and blockiness..." Nevertheless, these distinctions are slight, and the ACATS took pains to note that there was no substantive difference in picture quality between the two formats.

  7.      WHICH SCANNING FORMAT IS BEST SUITED TO THE FUTURE?

* Whereas interlace scanning is analog friendly, progressive scanning is digital friendly. Progressive scanning greatly simplifies conversion to all other scanning formats. It is the preferred scanning format for digital signal processing. Even today, standards conversion, digital video effects such as squeeze, zoom, flip, etc., are all done in progressive. 

* Progressive is also better suited to multimedia applications,  creating a rich, new environment for creative storytelling in the converged world of computing and television.

* The 1080 x 1920 (1080I) interlace format specified in the ATSC standard CANNOT be compressed to fit in a 6MHz channel without creating objectionable artifacts and it has been recommended that the 1920 pixels be sub-sampled to 1440 to reduce compression artifacts. Therefore, encoder manufacturers have elected to discard approximately 25% of the picture for over-the-air transmission. This compromise is not required for 720P. More of the original picture information remains through the transmission chain.

  8.      WILL THERE BE EQUIPMENT AVAILABLE TO BROADCASTERS TO ADOPT 720P?

* Until very recently, based on engineering investments made many years ago before the advent of digital signal processing, most professional broadcast equipment was based on interlace scanning. However, in the past few years, there has been an explosion of interest in progressive format equipment. ABC has identified manufacturers for each of the critical components required to introduce HDTV broadcast to our network and therefore to consumers. Panasonic and other equipment manufacturers have accelerated their 720P production schedules based on ABC's decision.  

9.      WHAT ABOUT CABLE COMPATIBILITY?

* The progressive formats selected by ABC are compatible with the digital set-top device being developed by Cable Labs and the cable industry. ABC's goal is to seek voluntary cooperative arrangements with cable operators for the benefit of consumers.

*********************************************************

THIRD PARTY QUOTES IN SUPPORT OF PROGRESSIVE SCAN

 

William F. Schreiber, former Director of the Advanced Television Research Program at MIT (Home Theater, 12/97): "Let's be perfectly clear about this. Interlaced scanning is a really dumb thing to consider in any new video format. Interlaced scanning was not even a good idea when the U.S. B&W standards were defined in 1941! As your readers are probably aware, interlaced scanning can generate very bad artifacts in a video image, things like 30-Hertz interline flicker, motion errors, and reduced vertical resolution. This is why computer monitors are not interlaced -- the interline flicker would be unbearable. The reason why there are still proponents of interlacing escapes me, especially when considered in the context of digital broadcasting. Progressive scanning is simply more "digital friendly" than interlaced. When digitally coded, a progressive signal requires no higher data rate than an interlaced signal that has half the bandwidth. I have a hunch that the continued advocacy of interlaced equipment originates from foreign-owned consumer electronics companies that are trying to get back the substantial investments they foolishly made in obsolete technology."    Department of Defense (DOD) Reasons for Interest in Progressive Scanned Video and Square Pixel Formats: "We believe that the use of Progressive (Non- Interlaced) scanned video and square pixel formats will greatly improve our information management capabilities. The presentation anywhere in the world of Geospatially related information from multiple sources should support improved decision making. Use of these technologies will provide for improved communications between United Nations coalition teams. Future video systems using these attributes will improve risk management in life threatening situations in the fields of Defense, Medicine, and mitigation of Global Disasters. Progressive scanned video and square pixel formats will also improve visualization in our energy, space programs and others, and greatly enhance the DOD's massive training and modeling & simulations programs."  

Proceedings of the IEEE, Vol. 83, No. 2, Feb. 1995, pgs. 158-174

The Grand Alliance System for U.S. HDTV: "However, the GA will work toward the elimination of interlace in the transmission path in the future."

(Note:  3rd party comments will be carried over to issue #18)

***********************************************************

Note:  Those who attend NAB'98, could see that there are a significant number of manufacturers who can and will provide 720p equipment.  Other points of view are welcome.

***********************************************************

Subj: NAB '98

By Larry Bloomfield

 

For the benefit of those who couldn't attend, there is no question that NAB '98 was the year of "Digital TV."  The buzzword was "solution!" and the way everyone, drifted, at times, far from the way Webster had defined its use. Encoders of about every flavor and variety were there.  MPEG-2 servers were offered by several MFGs.    Had the chance to see both the HP and Tek spectrum analyzers for the 8VSB signal; very interesting and educational.  Get your local rep to give you a demo.  They can do it using recorded material, which makes it nice for looking at your signal after the fact.  The 8 levels are shown and any error is vividly displayed.  Each have their own features and it  would be personal preference as to choice.  Larcan had the HP hooked up to their (Zenith) exciter.   

Saw a re-recordable CD carousel like player much like the old 45-RPM players. If anyone else saw it, please let Larry know who the MFG was as the material has gotten lost between LV and Bend.  The idea of doing your long form programming on it and using a small server for commercial insert in small market stations would go a long way to save bucks.  I was surprised to find the same capacity server we bought in San Francisco at Channel 38 from HP available today for about 1/3rd the cost.  That was only 2 years ago.    If you saw anything at NAB that particularly impress you, either drop us a line (e-mail) or call one of us so we can share that info with others in our field.

**********************************************************************

Subj:  Progressive vs. Interlace - Never the twain shall meet!

By:    From several sources

 

Most engineers are aware that the standards promulgated by the FCC for digital television include 18 different video formats; some interlace and some progressive.  We've probably all heard arguments for both.  HDTV can be broadcast in either 1080i or 720p.  If you are one of the few who don't know what that means, it means that the active lines of video are either two fields of 540 lines interlace to equal 1080 total or 720 lines progressive scan.  

It doesn't take a college grad to figure out that there will be Hundreds of billions, and maybe even trillions, of dollars at stake, during the next five to ten years.  It stands to reason, that if you are an executive working for a company with vested interests in the future of television, you'd do what it took to get your system adopted as "the standard." 

  Many people believe that the standards decided on by the FCC have cast in stone the degree of interoperability between multimedia computers and the television industry.  These DTV standards will determine the viability of a National Information Infrastructure, (commonly referred to as "the information super highway"), and will seriously impact the course of television; cablevision and satellite direct broadcasting in the future.  

The story of how it all happened has been told in many places.  It is, however, necessary to retrace some of the history which brought us to where we are today in order to illustrate why and how this debate came about and besides I like telling it.  

The players in this debate seem to be polarized into two groups.  On one side, the players consists of the consumer electronics industry, the four major U.S. broadcasting networks and trade associations who believe the net results will be thousands of jobs manufacturing DTV sets.  On the other side is a group of American computer companies that have banded together under the name Computer Industry Coalition on Advanced Television (CICATS) and many of the world's to  filmmakers in conjunction with the organizations that represent them.  

As Joel Brinkley so aptly pointed out in his book "Defining Vision," that "the battle for the future of television" has been raging for years.  My best guess is that it all began as far back as the early 1970s, when NHK, the National  Japanese Television Network and the Japanese consumer electronics industry launched an ambitious initiative to design a single High-Definition television standard for the world.  Although it is in place and working today, it has not  gained wide acceptance.

This early on HDTV system featured a wide-screen format of a 15:9 aspect ratio, which was a perfect match for the European 1.66:1 cinema aspect ratio. It also features stereo sound and 1125 lines of resolution.  Japan, like the United States, uses 525/60 NTSC.  The Japanese HDTV system called for an interlaced scanning architecture and analog transmission. The system was unveiled during the early 1980s but failed to secure a place as a  worldwide standard.

  The Japanese efforts didn't go unnoticed, however. The Society of Motion Picture and Television Engineers (SMPTE) looked into the Japanese system for six years. In 1988, SMPTE came up with their own HDTV format which was called 240M and bore a striking resemblance to the original Japanese system with one difference; it called for an aspect ratio of 16:9 (or 1.78:1).  One of the members of the SMPTE study group told me that 16:9 was chosen as it was a compromise between the 1.66:1 and the 1.85:1 aspect ratios used in motion pictures in the United States.  The other commonly used U.S. aspect ratio 2.35:1 was disregarded by their committee.    

The FCC, in 1987, appointed an Advisory Committee on Advanced  Television Services (ACATS) to function as an advisory group.  The membership of ACATS has come under fire as not representing all aspects of the entertainment industry as it relates to television As can be deduced from Brinkley's book, mentioned earlier, only those companies with vested interested in the outcome, could afford the dollars to participate.  Testing of the proposed systems and standards was done by the Advanced Television Systems Committee (ATSC) at the Advanced Television Testing Center (ATTC). .  ACATS and ATTC held  court in Washington, DC.

  In 1989, at an industry conference, Joe Flaherty, CBS's senior vice president for technology who said:  "We'll have digital television the same day we have an antigravity machine."  Just four years later, Woo Paik, a Korean born, MIT graduate engineer, working for General Instruments (GI) in San Diego, California, proposed and demonstrated a capability for transmitting HDTV signals digitally. Soon after Woo Paik's breakthrough, GI was quickly joined by others who also demonstrated more complex, but similar capabilities.

 

It didn't take much of a wake up call, after that, at the FCC for them to pronounced:  "HDTV transmissions will be based on digital technology." Shortly after this, the top contenders in the race formed the "Grand Alliance."  

Nearly all television, that isn't either sports, news or live, originates on film.   The American Society of Cinematographers (ASC) felt they had a vested interest in determining the standards for HDTV, so they formed a group, consisting of members and technology leaders in the film industry, to study the ramifications of the digitizing of television.  By the end of 1993, ASC submitted its proposals to ACATS and the FCC.  The plot thickens.  

"HDTV was designed to make dramatic improvements in image and audio quality," said ASC's John Hora. "Once it became digital, it added layers of possibilities which could evolve from linking the computer to the TV set and making them interoperable. It became possible to transmit and display movies in their original aspect ratio."    From the early days when television and computer technology were joined together for the purpose of "colorizing" black and white movies, filmmakers have taken an interest in what the television industry was up to. Since the new aspect ratio offered in HDTV permits films to be displayed in their original aspect ratio, filmmakers claim that "panning and scanning," inherent to showing U.S. movies, destroys the artistic integrity of their work by altering both composition and camera movement.  Keep in mind that we engineers provide the canvas, the paint, the brushes etc.  It is the artists who use these tools to generate their works-of-art.  

The ASC specified that all films be shown in their original aspect ratio on a screen with a 2 by 1 aspect ratio, and that interlaced scanning be  replaced with a single progressive (non-interlaced) architecture. ASC's reasoning is that this would result in a dramatic improvement in image quality and build an essential bridge to the computer industry.

  While all this was going on, the computer industry was asleep.  Apple, Microsoft and others in the computer industry were making their case for progressive scanning with square pixels.  Their argument is that this approach would measurably improve the readability of text, and convergence of computers and ATV displays.    

What the FCC adopted is basically the ACATS proposal: A 16 by 9 aspect ratio for display, a choice of (SDTV) standard definition (the same as the current NTSC 525 horizontal lines), or high-definition (720p or 1080i lines) resolution; stereo digital 5.1 channels of sound; a choice of 18 standards (four are interlaced and 14 are progressive); a data output and all this transmitted digitally.

  As it stands now, individual broadcasters may select the standard they want to use for transmission from the 18 options. DTV sets will be equipped with decoders that enable them to receive transmissions in all 18 standards. Older NTSC sets will require an external device to do the decoding down to a signal the set can use.  

Which brings us to why the ACATS proponents are being challenged.   As recently as a year ago, a new way to digitally compress images was developed. The technique is called "layered compression," which consists of baseline and enhanced layers of images.    

Gary Demos, president-CEO of DemoGraFX began his involvement back in1989, when he started attending SMPTE HDTV subcommittee meetings. After getting up to speed on what was going on in the wonderful world of DTV, it seemed to him that the subcommittee was merely planning to scale up TV display principles. There was more talk about company positions than technical discussions.    

"I had been working with multi-thousand-line high resolution since 1976," he said. "I knew a lot about handling and processing images and how to make the images come alive. I felt they were wasting a ton of effort on obsolete ideas. I kept thinking, this doesn't really make any sense. It won't ultimately prevail."  

Concurrently, Demos became aware of the other players and that the real decisions were being made by ACATS and the ATSC, back in Washington, DC. He was invited to attend an ATSC subcommittee, where there would be a definitive debate on key issues.  

Demos and associates from MIT joined forces with members of the computer industry community to form the Committee on Open High Resolution Systems (COHRS). Included in COHRS was about 70 people from the film and computer industries and influential academics. COHRS convinced the government that television was going digital.

  By early 1991, there was a committee headed by Will Stackhouse, of Jet Propulsion Labs, discussing interoperability issues, including digital architecture and the ability to transmit films in their original aspect ratio, the elimination of interlaced scanning and the need for square pixels to make text more easily readable.  

Demos recalls that the report written by Stackhouse was published by the SMPTE Journal in 1992. That's significant because one of the complaints that ACATS proponents recite when they criticize Demos is that he raised all of these issues at the last moment. He responds that all of these issues were discussed and left unresolved five years ago.

  In August 1992, Demos' status changed from interested party to consultant to Apple, when that company committed to fighting for the convergence of ATV with multimedia computers. That made it economically feasible for Demos to really engage in the discussions that were taking place in Washington.  

In late 1993, DemoGraFX contracted with ARPA to establish a high resolution, motion imaging, interoperability laboratory. "We researched the fundamental issues related to advanced television," said DemoGraFX vice president Alan Peach, who played a key role in this project. "That's when we began shooting film and testing our theories about compression technology. The first thing we shot was a series of frame rate tests."  

The groundwork was laid in October 1993, when ASC and computer industry representatives met at an interoperability review session sponsored by ACATS. "It was an incredible opportunity and experience for us to start working with these great filmmakers," said Demos. "One of the things that struck me was that they felt so passionate about preserving the integrity of their work. Yet, they weren't getting any respect from the ACATS people. I told them about  our ARPA study, and several top cinematographers agreed to shoot the test film for us. We scanned it at very high resolution. One of the things we wanted to see was the trade-off between frame rates and bit rates at very high resolution. We were scanning at 4K using a film scanner at Pacific Title  Digital. We processed the image data with the Indigo 2s, an Onyx computer at POP in Santa Monica and another Onyx at Hughes."

  By then, Demos, Peach and others at DemoGraFX felt that the layered  compression software they had developed fit the film industry's ideas like a glove. With financial support from Apple and Microsoft, they produced a new demonstration film, displayed in its native aspect ratio within a 2:1 image shape, with progressive scanning and square pixels. The camera work was done by ASC's Steve Poster and Bob Primes. FCC Chairman Reed Hundt and commissioners Susan Ness and Rachelle Chong visited DreamWorks SKG, and Chairman Hundt and Commissioner Chong viewed a demonstration of the DemoGraFX layered compression system. After those individual screenings, both Hundt and Ness called for representatives from the three factions to meet and resolve their differences. 

  Demos showed the demo film to Larry Irving, assistant secretary of commerce for Communications and Information. Irving is also administrator of the National Telecommunications and Information Administration, which advises President Bill Clinton and others in his administration on telecommunications and information policy.  

He has written to the FCC, and has asked the commissioners to limit rule setting for DTV to "essential elements" and has urged the interested parties to reach a consensus. Chairman Hundt has publicly said that the commission will not set a standard unless all parties agree.   In the absence of such agreement, Chairman Hundt has suggested that congressional hearings may be appropriate. Meanwhile, William F. Schreiber, professor emeritus at M.I.T., and former head of its DTV effort, has written to the FCC urging the appointment of an independent group of experts who have no financial interest in the outcome. He predicts that the interested parties in the consumer electronics/broadcasting, computer and film industries won't arrive at a compromise that serves the  best interests of the public. Stay tuned.

**********************************************************

The DTV Tech Notes are published for broadcast professionals who are interested in DTV, HDTV etc. by Larry Bloomfield and Jim Mendrala. We can be reached by either e-mail or land lines (541) 385-9115, (805) 294-1049 or fax at (805) 294-0705.  News items, comments, opinions etc. are always welcome from our readers; letters may be edited for brevity. >>>   larrybend@aol.com     ---------   J_Mendrala@compuserve.com <<<    DTV Tech Note articles may be reproduced in any form provided they are unaltered and credit is given to the DTV Tech Notes and the originating authors, when named. 

END