THE HDTV INBOX — AUGUST 2008
Metallica
Q. I am an “antenna head.” I have a Zenith DTT901 DTV converter and Channel Master 0068 amplifier. We get lots of channels out in these flatlands (Lancaster, PA) from Baltimore, Philadelphia, Scranton/Wilkes-Barre and local stuff, of course. I will send you a full report.
I am in process of comparing reception of 5 antennas at my location for UHF and VHF Hi DTV band (channels 7-13) As your 2002 showdown suggests, we are looking for a flat response up to channel 50. On my roof is the discontinued long Channel Master 4257 yagi (a bigger model 3023, a.k.a. the old model 4248).
Also, a CM 4228 8-bay, a CM 4221 4-bay, a JFD corner reflector (nobody makes this, but I might bring it back), a CM 1163A VU quantum combo (discontinued, but there are lots of these still on rooftops in these parts), and a CM 3679A VU crossfire combo (available and there are lots of these still on rooftops in these parts).
A. Yep, it sounds as though you have “aluminum fever!” Some of the models you mentioned are still available from Channel Master (4221 and 4228), plus they’ve introduced new 2-bay and 4-bay compact models for UHF. CM has also announced a new line of antennas optimized for high-band (channels 7-13) and UHF reception, which I hope to test soon. Since only 40 or so stations have opted to stay on channels 2-6, a 7-51 design could appeal to many parts of the country, including New York City and Los Angeles.
Stacking ‘Em Up
Q. I hope it was OK to contact you here, as I found no other contact links on the HDTVexpert pages. The articles you did consisting of (DTV reception) out on the fringe and the outdoor antenna reviews were helpful to me. I have also visited the TVFool site (www.tvfool.com) for my available stations. I live in the country on the fringe.
My specific question is in regards to the Yagi style antenna. Do you think it may be beneficial to "stack" multiple (maybe 4 in a square pattern) antennas? Maybe pointing in a slight spread pattern? I have noticed that bowtie antennas also seem to be configured in multiples. My dual-purpose VHF/ front end UHF is motored on top of a 30' tower. As for now I don't think I will make any major changes before the Feb 09 change. Thank you again for all the valuable information you have provided.
A. You can stack arrays of yagi antennas. To correctly match them and clean up their signal patterns, they should be stacked about 2.5 feet apart vertically and horizontally. (That’s an oversimplification of what is a more complex calculation, but it should work fine for you.) To best connect them, a precision power divider should be used. When stacked the correct distance apart, you will see about 2.5 to 3 dB gain for each array, i.e. a second antenna adds about 2.5 to 3 dB gain over the first, and a second array of 2 antennas stacked adds another 2.5 dB to 3 dB to the first array.
What you’re accomplishing by stacking antennas is creating a narrower aperture for accepting signals, which is where the gain comes in. The patterns in both the antenna’s horizontal (H) and vertical (E) plane become tighter. If you connect things with a conventional 2-way splitter, you’ll retain the directivity but losses in the splitter will offset signal gain.
There’s a great discussion of yagi stacking theory on the American Radio Relay League’s Web site. Look for this article http://www.arrl.org/tis/info/HTML/stacking-yagis/, which is slightly technical but has some useful rules of thumb. It’s also short.
The 600-Pound MPEGorilla
(Editor’s note: I was quoted by correspondent Eric Taub in a recent story in the New York Times about Dish Network offering 1080p content to consumers. This letter is in response to that story.)
Q. I was surprised to see bit rate touched upon in the Dish Network story, but I guess they might be aware of compression issues. I have been working HD encode for ESPN / TNT / etc… for the past few years, and I am becoming very sensitive to bit rate, and it seems to be the 600 lb. gorilla that no one wants to talk about. And the average consumer has NO IDEA how compression impacts their viewing experience.
I’m familiar with MPEG-2, but have no experience with MPEG-4, except to hear that you get about 50% more compression. I do know that 12 Mb/s 8VSB into my home delivers an unacceptable picture, 19 Mb/s is noticeably better, but still not terrific, and 35 Mb/s of MPEG-2 yields a picture where only rarely do I see artifacts, but I’m sending mostly live sporting events.
Flames and reflections in water are two torture tests for compression, which even the average consumer can understand. If you can’t discern blocking in those scenes, you have enough bit rate to have a decent picture, in my book. Obviously, extra bandwidth is nice, but costly somewhere.
Do you see MPEG-4 as being about twice as efficient as MPEG-2? One of my mentors recently commented that, at higher bit rates, MPEG-4 is roughly equal to MPEG-2, but at low bit rates, MPEG-4 shines.
Did you notice that DISH is looking for 1080p material shot in 24? This actually lessens the bit rate demand, and allows a 1080 compressor to work with progressive images, from off-the-shelf HD VTR’s. The 1080p/24 HDCam from Sony is 1080pSF, which during acquisition is displayed as 48 “fields”, 540 lines per field, but it does show the full 1080p, without motion artifacts. On a CRT (common until the past few years) the 48i flashed similarly to viewing a PAL signal, and the strobing went away when the studio VTR output a 1080/60i image formed from the 24p. It seems to me that a 40-minute tape held about 50 minutes of 1080/24p because the actual bit rate slowed down some.
DISH is looking to market 1080/24p, but I suspect that ANY material captured from film is essentially 24p, and 1080/60i conveys 24p very nicely. Could this just be a marketing game that DISH is playing, allowing the consumer to think they’ve made a technological jump, when in fact the real jump would be LIVE programming, 1080/60p? ESPN had plans to be 1080/60p this year, but seems to have shelved the idea, probably because it’s so costly as to be unworkable just yet.
A. You’ve touched on a lot of issues with native carriage of 1080p content. It’s pretty much agreed that 1080i encoded as MPEG-2 needs as many bits as possible to look free from compression artifacts, and that sending 1080i/30 out at 10, 12, and 14 Mb/s isn’t good enough, especially on big screen 1080p HDTVs.
1080p refreshing at 60 times per second (technically, 59.94 Hz) is presenting twice as much picture information in the same time interval as 1080i/30! It stands to reason that 1080p would need twice the bandwidth of 1080i for equivalent picture quality, one possible reason why the maximum bit rate for Blu-ray and HD DVD was set at 36 Mb/s. Coincidental, eh?
MPEG-4, which can do sub-pixel compression, in theory is twice as efficient as MPEG-2. If so, then 8–9 Mb/s should be sufficient for 1080i compression, and 16–18 Mb/s would be the “best practices” minimum for 1080p/60. However, experience has shown that DirecTV and Dish compress MPEG2 HD harder than terrestrial broadcasters and some cable systems. I’ve heard of examples where 1080i was sent out at 10 and 12 Mb/s.
Switching to MPEG-4 should improve 1080i picture quality noticeably at that bit rate. But moving to 1080p demands more bits, and DBS companies are all about favoring the bucks vs. the bit rate. 1080p/24 native transmission wouldn’t need nearly as many bits and might work out fine at 12 Mb/s, but the set-top box would have to convert that to 1080p/60 or even (heaven forbid) 1080i/30 for older, incompatible HDTVs.
