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 Some of the information may be out of date.
**********************************************Tech Notes
Published by: Larry Bloomfield & Jim Mendrala
(408) 778-3412 or (661) 294-1049
E-mail = larrybend@aol.com or J_Mendrala@compuserve.com
July 18, 1999
Tech Note - 034
***************************************************
Talent does what it can -- Genius does what it must!
Our Mission: Sharing experiences, knowledge, observations, concerns, opinions or anything else relating to Electronic Cinema, DTV, etc., with fellow engineers and readers. We publish when there is something to share. The Tech Notes are only sent (BCC) directly to those who have asked to be on the mailing list, however feel free to forward them, intact, to anyone who you think might be interested. There is no charge for this Newsletter, no one gets paid (sigh), there is no advertising and we do not indorse any product or service(s). The ideas and opinions are those of the individual authors. We still administer everything manually. We don't use any "majordomo" automatic servers. We do hope that everyone will participate with comments, experiences, questions and/or answers. We now have 425 subscribers and growing. This is YOUR forum!
Past issues are available at: WWW.SCRI.COM
**********************************************
Subj: Digital Projection Questions
By: John Fode >>> johnfode@earthlink.net <<<
(Ed Note: The responses are by Jim Mendrala)
I enjoyed your excellent article on your (Jim Mendrala's) experience of the Film & Digital Projections of Phantom Menace. I was present at the Burbank AMC along with others in our industry at the Friday 1:30 PM opening show (Digital Projection). I thought it was terrific.
Q: Was the data presented to the TI in 1280x1024 ?
JM: The data was presented to the Texas Instrument projector from the Pluto Technologies server via a SMPTE 292 High Definition Standard Digital Interface or HD-SDI. The displayed data was formatted within the center 1280 x 1024 pixels of the available 1920 x 1080. An anamorphic 1.9:1 projection lens was used to project the squeezed anamorphic image onto the screen and un-squeeze the image.
* * *
I certainly saw no lack of picture quality or details. I sat directly in the middle about 3-4 screen heights back. I also saw the Film release about 2 weeks prior, but only watched about 45 minutes before leaving (Too many kids in the audience, and I did not particularly enjoy the movie itself).
Q. The Pluto server delivered at 24 fps?
JM: The Pluto server delivered the data at 30i fps. The 3:2 pulldown was removed at the projector through the use of time code. The data was recorded so that all "A" frames fell on time code 0's and 5's. The time code was feed to the projector so that only the original 24sF frames would be used. Synthetic frames derived from 3:2 pulldown were discarded. The TI projector then reconstructed the segmented fields back into the original 24p frames.
* * *
Q. I saw no flicker.... Was it doubled to 48 fps, or tripled to 72 fps?
JM: Actually each individual pixel was turned on and off using a form of Pulse Width Modulation (PWM) and the mirrors switched on and off at around 50,000 times per second.
Q. The end credits looked noticeable soft. How much better can we expect Digital Projection to get (Other than less expensive)?
JM: Yes, the end credits were noticeably soft. This was also true on the film print. The electronic cinema credits were at the absolute end and they appeared sharper.
Q. In what areas can we expect it to get better?
JM: I think the resolution and colorimetry will get better. Today the resolution is limited by the projector and the colorimetry is limited by the masking parameters used in the telecine whose output is tailored to SMPTE 240M specifications. Pixel bit depth is already at 10 bits log or 1,024:1 contrast range, so I don't expect much improvement in this area.
Q. I am a bit confused with the issue of aspect ratio.... You wrote: "According to a paper handed out by CineComm, the screen was said to be 47 ft. x 27 ft. The projector was a Hughes/JVC model 12K projector. The throw from the projector to the screen was 77 ft. The image aspect ratio on the screen was 2.35:1." If I divide 47/27, I get a factor of approximately 1.74. Pardon my ignorance, but how do you derive 2.35:1?
JM: The screen aspect ratio is one thing. The image displayed is another. The 2.39:1 aspect ratio image was projected in the 2.35:1 format onto the 1.74:1 screen. Black velvet masks were brought in from the top and bottom so that it looked like the screen was tailor made for the presentation.
John, I hope this answers your questions.
Jim Mendrala
***************************************************
Subj: Lead, Follow or get out of the way!
By Larry Bloomfield
The cable industry has done a fine job of convincing the American public that they are the only source of television signals worth considering. Over the years, they have been responsible for building codes, CC&Rs in fancy neighborhoods and the list of their intrusions into our lives goes on. Go back twenty or thirty years and look around at the rooftops of most any American town and it looked like a porcupine on aluminum steroids. Slowly, but surly, the TCI's, Cablevisions, Falcons etc. began offering additional services which the average viewer wanted, but was unwilling to put up a 10-12 foot satellite dish and it's associated equipment, to receive. I'm not telling you anything you probably don't already know; there just aren't many TV antennas poking their directors, reflectors and driven elements into the sky to capture the signals of local TV stations anymore and those that are still up are usually in disrepair.
This scenario has caused the tuner industry to all but atrophy. Until a small tuner company in Plano, TX came onto the scene a while back, not much has been done to improve tuners in any way over the years. (Check out www.Microtune.com, Microtune's web page.) I've seen older TV sets with more receive abilities then many of todays's super big screen behemoths. The incentives have just not been there. Why put bucks into making a better tuner when cable has been dumping tons of signal into the back of subscribers TV sets? The set manufacturer's are in business to make money. There aren't too many successful Don Quixote's in business. Cutting costs is part and parcel to their corporate bottom lines. There just hasn't been any challenge.
The sad part to all of this is that cable's overall performance has deteriorated to the point that you can usually get a better signal off the air than from the local copper signal peddler. When you haven't got anything to compare with, viewers have grown accustomed to the increasingly poorer and deteriorating performance of cable. Thank God for Direct-to-Home (DTH) Satellite services which should go a long way to give them some competition. This is not said as a cantankerous comment, but rather in the spirit of fostering good old Yankee competition. Isn't that what has built our country in the past?
In addition to DTH, there are a number of things that should be giving cable a wake up call. (DTH is digital.) If we can ever get the various issues of modulation settled, digital television promises to bring crisp clean pictures into our homes once again either over the air or via cable, but the latter has some unresolved issues.
It doesn't take a seafaring salt to realize that the cable industry is still adrift with no direction when it comes to resolving what is becoming "The Cable TV Digital Dilemma." Many in the broadcast industry have hoped that something positive would come out of the cable industry's big Chicago whoop-tee-do, the National Cable Television Association (NCTA) convention back in June, but the expected commitment to deliver OpenCable specifications for retail-ready advanced digital set top boxes (STBs) was not forthcoming.
This is an important issue to broadcasters. The latest surveys indicate that nearly 80 percent of all television sets get their signals from one kind of cable service or another and time is running out. With an FCC deadline of July 2000, someone's got to burn the midnight oil and stop the lollygaging so the Application Programming Interfaces (APIs) needed to for the interoperability of the STBs can be developed. These APIs are crucial to the industry. If not in place on time the ripple effect will impact the arrival of interactive services for everyone.
In a letter dated July 1, 1999, signed by Consumers Electronics Manufacturers Association (CEMA) President, Gary Shapiro to FCC Chairman William Kennard, CEMA expressed its concerned. Shapiro promised the FCC Chairman that will every effort is being made to "reach agreement with the National Cable Television Association (NCTA) by October 31, 1999 on the necessary technical and operating specifications for digital receiver-ready cable systems and cable-ready digital receivers." Shapiro also said: "We also will work to complete "build-to" standards based on the agreement by the end of 1999." More than just a few hope this is not just more political smoke and mirrors.
Two of the biggest players in the behind-the-scenes tug-of-war have been, none other than Microsoft Corp. and one of their bigger of several nemesis, Sun Microsystems, Inc. It is generally accepted here in the Silicon Valley that each of these two software giants have sought to promote their own software varieties for television sets attached to anything, let alone cable. Microsoft's "It's ours -- all or nothing" track record does not help to promote a congenial platform where everyone can contribute to the cable STB brew. Sun, one of the few in the software industry who can stand up to Microsoft, doesn't have a much better track record.
There is no question that a decision of this magnitude has to be made carefully and it takes a lot of consultation and decision-making. A cooperative effort on everyone's part would be ideal, but a pie-in-the-sky concept in these neck-of-the-woods. Cooperation is a nearly forgotten or unheard of concept in the boardrooms of San Jose, Cupertino, Sunnyvale, and the other contributing communities to the Silicon Valley megalopolis. The road to define software specifications, that would meet industry OS-agnostic and CPU-agnostic principles, should be an easy one. There is no shortage of talent available in this part of the country, but corporate egos are usually the greatest impediment to progress.
There is little doubt that the high-priced help in cable industry are not letting all this happen without some serious consideration from their standpoints. Cable operators have a lot to lose if they cannot come up with something by the rapidly approaching deadline. First their sanctioned territorial monopolies may be confronted with the god-awful concept of competition. There are those who wouldn't shed much of a tear should that happen. Just think: offering the consumer a choice. How novel!
CableLabs chief technical officer, Richard Prodan told regulators that the software aspects of the OpenCable spec are a "high priority." Prodan expects there should be some sort of standard established by the end of this year, and rightly so, but voices from the PC and consumer-electronics industries are not quit as optimistic, saying it may take a year or more to sort out the various issues.
We look to our digital brothers for answers, but Silicon Valley is young and undisciplined. It's time for the computer and software geeks there and elsewhere to wake up and get to work before noon. Perhaps then they can then get some work done or be available when the rest of the electronics industry is conducting business. It's also nice for them to be in their offices before most of the rest of the United States is headed home for the day. Then there is the issue of learning about the industries who need their cooperation to move on to new technology.
Living and working in the Silicon Valley certainly has been an experience for this broadcast engineer. It is not difficult to emphasize with others in our broadcast industry who try to get things done only to run into very well camouflaged roadblocks. The lack of understanding for the applications that non-computer industries need is appalling. Typical of this is the software company I recently worked with who were attempting to develop a product for the broadcast industry. By the numbers, nearly all of the some 85 people who were writing the software had never stepped foot into a broadcast facility, let alone understand how it worked.
STBs in Trouble Too
In light of all this, it is not difficult to understand why it has become necessary for the cable industry's business expectations for OpenCable-compatible digital set-tops, to dramatically shift during this past year. Discussions with many of the STB designers and projections for impressive specifications and applications for the first wave of the new generation of STBs that could fully get things going for the two-way cable infrastructure have all but been scrapped for now. The current genre of STB specification dictates nothing more than a plain-vanilla box with an electronic programming guide (EPG) for digital pay-per-view and not much else.
It was expected that with the deregulation of STB ownership by the FCC, droves of new features would attract a whole new market for this kind of device, but it is not difficult to see why this has just not happened. The public will not pay the several hundred dollars for an STB that does little more than give them what they already have. As Alan McCullough, president and chief executive officer of Circuit City Stores Inc. (Richmond, Va.) has said on several occasions, retailers want something with competitive functionality to sell that encompasses more than just basic functionality.
Don't look for all the fancy bells and whistle features expected to be part of the bi-directional digital cable business from STB makers as they are few and far between. Referring back to the days, a couple of years ago, when Open Cable first saw the day of light, James Bonan, vice president of new-business development for the Consumer Audio/Video Products Group of Sony Electronics Inc., says the industry is "right back where it started - defining what interactivity means and what it needs to do."
When you leave the door open, don't be surprised at who might come in. With the acquisition of TCI cable by AT&T, there comes to the table a whole new bunch, who play ball a different way. With all this inactivity and to keep its head above water, it doesn't come as any surprise that the industry has to reassess its definition of interactive TV. One concept that snuck in the door with AT&T is a concept called voice-via-cable touted as the wave of the future by cable's newest "big" entrant.
AT&T chairman C. Michael Armstrong, has been very clear about what AT&T plans to do with all its subscribers: sell them local, long-distance and international telephone service (de ja vue) along with Internet access on the side. If you've also observed which cable companies AT&T has gone after, they've been those that have concentrated subscriber bases that lend themselves to AT&T's plan. Nearly every one of the AT&T acquisitions are volume focused and will provide lots of subscribers in the more densely populated areas. Admittedly this gives cable a solid and dependable monthly income by providing a single source for "all" home external electronic services. One cannot help but ponder, where's Judge Green when you need him most?
Nothing has been said so far about the consumer who wants more than just an STB with OpenCable compatibility. For the consumer electronic companies that build terrestrial digital television, cable and Internet devices, the concerns go much farther than how to build the OpenCable compatibility into the set. Concerns must be addressed relative to maintaining the integrity of devices based on the dissimilarity of the various platforms.
In addition to the two software opponents mentioned earlier, the API debate has split the remaining part of the industry into two factions. In truth, if the two sides lineage is traced back to its genesis, you'd find the same two competitors at the bottom of all this. One side wants Java Virtual Machine-based interoperability for advance digital STBs and terrestrial digital television sets. The other faction wants to achieve a common content spec across cable, satellite, terrestrial and Internet transmissions.
The Java group is synonymous with Sun Microsystems and a few cable operators, with a host of consumer-electronics companies under the Advanced Television Systems Committee (ATSC)'s DTV Application Software Environment (DASE) group. The group is forging key elements now for a common Java TV API for cable and terrestrial TV.
The competing group is led by the supporters of the Advanced Television Enhancement Forum (ATVEF), which was founded by none other than Microsoft, Intel Corp., and a number of media companies. Go figure!
The crux of the issue is the scalability of interactive data services. Broadcasters whet to NAB '99 looking for ways to help pay for the transition to digital television by finding solutions in the area of Datacasting. Not many had their expectations satisfied. Depending on the complexity of the programming and execution environments designed for TV data broadcasting, the emerging datacast spec will profoundly affect next-generation TV receiver and set-top architectures. It would be best for broadcasters to continue their supplications to "the Force" in hopes their prayers are ultimately answered.
If all this isn't enough, probably the best of the hidden agenda beyond the API battle for some is Microsoft's emergence as a power ball within the cable industry. If you have been following the financial pages of your local newspaper, let's hope the editor was astute enough to pick up on the story about the deal between Microsoft and AT&T to deploy up to 10 million Windows CE-based cable STBs over the next five years.
Laurie Priddy, senior vice president for advanced technology at AT&T Broadband and Internet Services, the part of AT&T headed up by former TCI big-shot Leo Hindery, insisted that her company's deal with Microsoft will "absolutely not" affect the industry's debate on OpenCable issues. "The focus on middleware allows each cable operator or consumer-electronics company to choose its own OS," she said. At this point of development, no one appears willing to sacrifice that flexibility. Remember that compromise is doing something that neither party wanted to do.
Nothing would make Microsoft happier than to have the cable industry standardize on a product that would reprise their track record with DOS & Windows, but now in the cable TV industry. On the other side of that coin, major cable operators are very paranoid about letting Microsoft control anything.
Common sense says that we'll eventually see a series of API recommendations, but the new API will almost have to take into account currently in use legacy APIs.
With the deadline rapidly counting down, perhaps cable could take a lesson from its big brother television and strike up arrangements like RCA-NBC once had or more recently CBS-Mitsubishi.
Perhaps the saddest part of this whole story is the FCC's roll or should I say non-roll. Instead of helping resolve these issues, the FCC will continue to press the cable and consumer-electronics companies, telling them they should try to work out their digital television/cable compatibility issues. This should be no surprise as we all know that picking up a pencil to help with something of this sort might give one of those governmental bureaucrats a hernia.
Instead of being the FCC of a bygone era; the FCC that helped develop the American broadcast industry into the technically finest on the planet; that helped the industry when issues of this import and magnitude, came up, they will most likely take their typical "wait and see posture while the ship bobs aimlessly around.
The FCC is supposed to be the stewards of the public airwaves. It is sad! This is the same FCC that has evolved from a fine, respected governmental organization into the money grubbing pawns and puppets of governmental bean counters who are deathly afraid of their own shadows when it comes to setting technical policy and standards. Like every other nautical vessel, perhaps all antenna structure should be required to display two black balls dangling from their mast. It means we're underway with no away.
***************************************************
Subj: Should Hollywood Be Satisfied With HDTV?
By Jim Mendrala
With all the hoop-la surrounding the digital presentation of George Lucas's Film "Star Wars - Episode I, The Phantom Menace" and the Miramax film "The Ideal Husband", Disney is readying "Tarzan" for digital projection at several venues.
Starting on July 30th, the "Tarzan" film will be shown at AMC's Pleasure Island multiplex at Disney World outside of Orlando, FL. The film will also be projected digitally at AMC's Media Center North 6 in Burbank, CA and at the Edwards Irvine Spectrum complex in Irvine, CA. All will be using Texas Instruments' DLP Cinema projector.
Details are scarce at this time, but it is rumored that the theater playback will be from a Pluto Technologies Server using the Panasonic HD D-5 compression format that compresses the picture 4:1. Uncompressed 5.1 channels of audio will be played in sync from an MMR8 just like the "Star Wars" presentation.
The major difference with the "Tarzan" movie is that it was created and animated on computers. It is also rumored that instead of doing an HDTV film to tape transfer from an Inter Positive (IP) like "Star Wars" and "The Ideal Husband", the computer generated images (CGI) will be down converted directly from the digital CGI files to the 1280 x 1024 pixels within the 1920 x 1080 HDTV raster.
Now, since the broadcasters are having second thoughts about HDTV DTV transmission in the ATSC 8-VSB (8 Vestigial Sidebands), and Sinclair Broadcasting is advocating testing COFDM (Coded Orthogonal Frequency Division Multiplexing) as an alternative RF transmission standard, Hollywood seems to be satisfied to run digital HDTV film to tape transfers for the theaters.
This is a shame because HDTV was designed for the home on screens that could be as big as 10 ft. wide by 5.6 ft. high. (Floor to ceiling is 8 ft. in most homes today and room depth is about 15 to 20 ft.)
The few of us that have seen DTV HDTV projected onto a big theater screen know that the images are okay for those that sit three or more screen heights back. But for those theaters that have big screens and stadium seating, and the back row of seats in the theater are only 3 screen heights back, HDTV digital projection leaves a lot to be desired.
Resolution is less of an importance as one gets further from the screen. An HDTV 1920 x 1080 image is okay for viewing at three or more screen heights. However, at less than three screen heights, the image is soft. At around two screen heights, the pixels start to become obvious. At one screen height, all you see are pixels.
Pixels are fixed. They don't move. Film grain however is random; it moves from frame to frame, sort of like dithering. The eye tends to average out the grain, and the detail in the image is more readily seen. Film typically resolves about 55 line pairs per millimeter (lp/mm). HDTV digitally projected by a projector with 1280 x 1024 pixels is just barely able to resolve 20 lp/mm. That's less than half of what a typical 35mm film print can resolve.
At this time, electronic cinema is in its infancy. The infant has not even started to walk yet. It seems that Hollywood would want to start with a clean slate and address the problems within the theater environment and come up with a system that exceeds the 100 year old film system that we use today.
It is true that digital projection is cleaner, i.e., no dust and dirt. It has more saturated colors. There is much less weave. Focus is more consistent. There are no scratches. Sound is clearer. However, the one thing that is lacking is a sharp sparkling picture that stands up in the theater environment like the 35mm image.
***************************************************
Subj: A military strength ATSC antenna system from a not so retired observer
By: Nicholas Bodley >>> nbodley@tiac.net <<<
(ED Note: The results of several 8VSB tests have shown the need for tall, large, rotating antennas for any measure of successful reception. This new subscriber addresses this issue with an interestingly different and humorous prospective. Just think what you'd have on your roof if the military where to design your antenna structure.)
Just discovered your DTV tech newsletter (a happy find!) from a Web search. Although I'm retired, I still find the ATSC/COFDM situation to be fascinating as well as sad. Through unusual circumstances, I was in D.C. last Sept. or so for the NAB gathering, and became aware of the situation firsthand. It seems to be The Dirty Big Secret of US DTV.
Plainly, directional antennas, and often-big ones, are needed for 8VSB. My background in naval gun fire control, where multi-ton gun mounts require fast, accurate position servos to keep them on target with a rolling and pitching ship, as well as recoil of twin and triple guns that tries to rotate the mounts, made me propose a spoof: Have reinforced antennas rotated by high-torque position servos; assume that they are pointed to the desired azimuth in about 1/4 second with about 1 degree of arc accuracy (guesses).
The reinforced towers for these could cost $thousands, and might have problems with zoning; the servos might draw peak powers enough to dim the lights. Eventually, the servo cost might be brought down to a few hundred dollars. Accumulators for electro-hydraulic servos are unlikely. Of course, the surfer's channel selector hardware would contain antenna-angle data for each non-co-located transmitting antenna.
Seriously, I don't expect to see a news report about a chimney being twisted apart by excessive channel surfing. I can, however, imagine an animated cartoon showing just that happening. Done well, it could be quite funny, except to ATSC advocates.
Regards, Nicholas Bodley
***************************************************
Subj: Broadcast / Pro Video Facilities Show Increasing Usage of Video
Networking
From: SCRI
(Ed Note: The following SCRI note is our way of saying thanks for them letting us use their web site.)
According to SCRI's recent Broadcast / Pro Video Industry Trends Report ('99-2000), the majority of broadcast and production facilities are already using local area networks - currently about one in seven (68.4%), going up to eight in ten by the year 2000. With 12.4% of facilities unsure, the actual penetration of LANs is likely to be even higher. "As production and broadcast facilities gear up for full digital production and transmission environment, the use of video networking has shown a concomitant increase" commented SCRI's Research Director, Des Chaskelson.
"This report provides manufacturers with a roadmap of the shifting Broadcast and Professional Video Marketplace as we move into the new millennium. The report tracks all the key technology issues, like DTV, video networking and transport, video formats, equipment budgets and production activity trends, incidence of traditional and new video applications like webvideo and CD-ROM and DVD Production."
Contact SCRI at: info@scri.com
***************************************************
ED Note: We have several stories that wouldn't fit into our limited 30,000 characters and spaces. If you have anything to contribute, we'll be glad to include it in our #35, which will be coming out sometime this week.
************************30*************************
The Tech Notes are published for broadcast professionals, and others, who are interested in Electronic Cinema, DTV, etc., by Larry Bloomfield and Jim Mendrala. We can be reached by either e-mail or land lines (408) 778-3412, (661) 294-1049 or fax at (661) 294-0705. News items, comments, observations, opinions, etc. are encouraged and always welcome from our readers; material may be edited for brevity, but usually not. Tech Note articles may be reproduced in any form provided they are unaltered and credit is given to both Tech Notes and the originating authors, when named. If they are to be used by a publication that normally compensates their writers, please contact us first.
END
