Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently May Sun 31, 2020 2:27 pm


All times are UTC [ DST ]





Post New Topic Post Reply  [ 29 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Tue 04, 2020 6:23 pm 
Member
User avatar

Joined: Aug Tue 28, 2018 9:22 pm
Posts: 2007
Location: Sanford Fla 32771 (USA)
Hope this is in the right forum and not a duplicate.
While I have a good understanding of analog TV and bandwidths, I have wondered what the digital TV bandwidth was compared to the analog TV. It turns out (if I understand it correctly) it is approx. still 6 Mhz wide.
At any rate I have often wondered how the digital TV data format was done. Here is an interesting article on digital TV. I am still working my way thru it without total brain freeze.

https://www.mathscinotes.com/2012/05/hi ... sion-math/

_________________
Paul of Florida ….I had my patience tested. I’m negative.
https://paulsironhorse.smugmug.com/


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Tue 04, 2020 6:50 pm 
Member

Joined: May Thu 14, 2015 4:15 pm
Posts: 2178
Location: Dallas, TX
Yes, digital data transmission is a whole different animal. A key point is the data for a video frame is not transmitted during that frame, only the change from the last frame. That leaves "room" to transmit other video frames that are on other digital (virtual) channels on the same RF channel.
In other words it's "computer stuff".

_________________
Tim
It's not the Destination, It's the Journey.


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Tue 04, 2020 7:11 pm 
Member
User avatar

Joined: Mar Sun 01, 2009 10:27 pm
Posts: 5256
Location: Ann Arbor, MI
On the high definition sets that used a CRT, the bandwidth of the signal going to the CRT was quite a lot wider than 6 MHz.

_________________
Tom


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Tue 04, 2020 7:23 pm 
Member
User avatar

Joined: Nov Thu 11, 2010 6:03 pm
Posts: 1224
Location: Pewaukee, WI
Actually they don't have to (and probably don't) specifically compress the motion data of one channels frames to slip another's in between frames. It is more like each sub channel independently does its own IIRC MPEG based compression of motion and creates a compressed data stream then each subchannels data stream gets combined into the final stream the transmitter puts to air. They probably limit the bit rate (which is directly proportional to bandwidth Iin the digital realm) of each encoder so no one stream can ever consume enough bandwidth to crowd out other sub channel streams.

Since the transmitter signal is a digital video data stream (or rather several sharing a TX signal) you literally can record the digital bits directly of the RF demodulator save them to a file and play that file with a media player on your PC (such as the player you might be using to watch DVDs on your PC). Many ATSC DVRs work this way and create "lossless" recordings of the ATSC broadcast data stream....but not lossless relative to the studio program source since the transmitter has compression losses baked into the ATSC transmission standard.

Effectively compression allows a high bandwidth video signal to be able to be shoe horned into a lower band width RF carrier and expanded back to normal bandwidth at the reciever with little to no loss of quality....


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Tue 04, 2020 9:00 pm 
Member
User avatar

Joined: Mar Sun 01, 2009 10:27 pm
Posts: 5256
Location: Ann Arbor, MI
Electronic Memory wrote:
Actually they don't have to (and probably don't) specifically compress the motion data of one channels frames to slip another's in between frames. It is more like each sub channel independently does its own IIRC MPEG based compression of motion and creates a compressed data stream then each subchannels data stream gets combined into the final stream the transmitter puts to air. They probably limit the bit rate (which is directly proportional to bandwidth Iin the digital realm) of each encoder so no one stream can ever consume enough bandwidth to crowd out other sub channel streams.

For each sub-channel, the encoders send a complete picture every few seconds and then send changes the rest of the time. The amount of data per unit of time is variable depending on what is needed. The encoding processes for the sub-channels work together so that one sub-channel can send more data when another can send less. The result it that the full bandwidth of the full channel is used in the most optimum way.

_________________
Tom


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Wed 05, 2020 3:16 am 
Member

Joined: May Sun 07, 2017 11:35 am
Posts: 840
Location: Belrose, NSW, Australia
I was fortunate enough to have attended a lecture and workshop on MPEG2, just after it first appeared. It was given by the guys who had developed it.
There is some really clever math used in DTV codecs. It blew me away at the time just how many disciplines were brought together to make MPEG2 and just how much future-proofing was built into it.
We got to model motion vectors with Matlab - very slow!!
At that stage (early '90s), MP3 and MP4/H264 were just being mooted.

It was thought at that stage that you would always need dedicated hardware to encode MPEG in real time.

_________________
Wax, paper, bitumen, cotton, high voltages - what could possibly go wrong?


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Wed 05, 2020 5:55 pm 
Member

Joined: Nov Wed 30, 2016 7:35 pm
Posts: 6909
Location: Sunbury, Ohio 43074
By 'bandwidth'.... are you referring to how much spectrum is consumed during the transmit process, or what is actually being displayed on the screen? They are vastly different things.

There is no way to compare, in any really meaningful way, what you see on a digital LCD screen today with what you saw on an analog CRT in the good ol' days.

I'll go one step further ... what you see in a TV studio, production truck, or other origination facility would blow your sox off compared to the digital "junk" you end up seeing on your cable TV. The HD picture coming directly from a broadcast grade HD camera to an HD LCD monitor running in native mode, is stunning. It's so good you'd swear it was 3D, and it's not.

By the time it gets to your home, it's been compressed, decompressed, processed, fiddled with, repeat..... and then you cable provider does it again. And then your display does it again...... to format it for its own unique screen format. Between bits discarded, compressed frames missing .... it's garbage.

That said, the garbage is "usually" better (to your eye) than old analog was.

So, back to the original question ..... the actual bandwidth of the transmission process for HD is not overly much more than it used to be. If it's 4K uncompressed? figure about 8x the transmission bandwidth of HD.

What you see on screen? It varies, and depends on all the processing along the way, and then the processing in your house. The math would make your head explode, and long for the good ol' days of analog transmission. ;-)

_________________
Preserving the hist. of electronics, one boat anchor at a time! :)
https://www.bbtvtestequipment.com


Last edited by Barry H Bennett on Feb Wed 05, 2020 6:26 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Wed 05, 2020 6:01 pm 
Member
User avatar

Joined: Aug Tue 28, 2018 9:22 pm
Posts: 2007
Location: Sanford Fla 32771 (USA)
I was referring to the bandwidth used in the transmission, which is nearly the same 6 Mhz, plus or minus a few percent.
FRom: https://www.mathscinotes.com/2012/05/hi ... sion-math/


Attachments:
zzzUntitled.png
zzzUntitled.png [ 92.97 KiB | Viewed 1464 times ]

_________________
Paul of Florida ….I had my patience tested. I’m negative.
https://paulsironhorse.smugmug.com/
Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Wed 05, 2020 6:28 pm 
Member

Joined: Nov Wed 30, 2016 7:35 pm
Posts: 6909
Location: Sunbury, Ohio 43074
Yes, as far as that goes, you are correct. They are nearly the same as far as transmission bandwidth is concerned. The difference mostly lies in what is being transmitted, in this case digital data on an analog carrier.

_________________
Preserving the hist. of electronics, one boat anchor at a time! :)
https://www.bbtvtestequipment.com


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Wed 05, 2020 7:18 pm 
Member
User avatar

Joined: Mar Sun 01, 2009 10:27 pm
Posts: 5256
Location: Ann Arbor, MI
According to Figure 7, the digital transmission has about 5.38 MHz of usable bandwidth. Now for analog transmissions the sound carrier is 4.5 MHz away from the visual carrier. I expect that means that the usable video bandwidth is not much more than 4 MHz. And that assumes the use of filters quite a bit better than what was available when NTSC broadcasting first started. Note that NTSC requires the visual carrier to be 1.25 MHz away from the low edge of the channel. That was the best that they could suppress the lower sideband with the filters available back then.

_________________
Tom


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Wed 05, 2020 7:20 pm 
Member
User avatar

Joined: Nov Thu 11, 2010 6:03 pm
Posts: 1224
Location: Pewaukee, WI
The way you modulate a carrier with digital information also affects bit rate (digital bandwidth). Most digital transmission scheme are some spin on an analog transmission scheme such as CW, AM, QAM, FM, PM, etc.... Some of those schemes can pack twice or better the number of bits per second into a fixed bandwidth RF channel of some other schemes just with 0 or 1 binary input.... schemes that allow more than 2 levels of modulation can achieve higher bit rates by sending symbols...ie if you have 8 distinct modulation levels you can send a single symbol that represents 3 bits in a the same carrier cycle that you might have tried to send a single in....in some schemes that causes increased noise vulnerably.

There's also freaky black magic modulation schemes like CDMA where multiple transmitters and recievers are running on the same channel at the same time and each applies a digital code to each packet. Each TX rx pair has a different code and unless you add that code into the RX all the signal appears to be random noise, and applying the code makes all other coded signals random noise that is ignored infavor of the correctly coded signal...
I had a course on signal modulation schemes in college that was somewhat digital focused, and even after having it explained to me CDMA still seems like black magic to me.


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Fri 07, 2020 1:27 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1668
Location: Phoenix, Arizona
ATSC3.0 fits into the same 6 MHz channel width as Current ATSC1.0 and former NTSC.

The ATSC3.0 digital pedestal looks exactly like ATSC1.0 except there is no pilot on 3.0, as viewed on a spectrum analyzer.

Jim


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Sat 08, 2020 6:02 am 
Member
User avatar

Joined: Apr Mon 16, 2012 5:31 am
Posts: 179
As a television broadcast engineer still working after more than four decades, I have been fortunate to manage the technology from analog NTSC and PAL through the digital transition. As pointed out above in the US, the same RMA 6 MHz wide channel is still in use since it was first used for RCAs first public demonstration of television in July 1936. The 6 MHz channel was used for 343 line in 1936, 441 line 1938-1941 and from 1941 to 2009 525 line analog TV.

The ATSC system adopted in 1996 continued to use of the 6 MHz channel for delivery using standards introduced by MPEG (The Motion Pictures Experts Group) between 1992 and 1994. The video in the broadcast station is applied to an MPEG2 digital video compressor for the different program streams or "sub-channels". The sub-channels are multiplexed into a stream of digital packets known as the MPEG 2 transport stream. The MPEG2 transport stream conveys the multiplexed compressed sub-channels to the transmitter, where it is modulated to go over the air, transmitted, is received at home where the TV set demodulates and then demultiplexes the transport and decompresses for display. The MPEG standard MPEG 2 digital video compression is one of a number of standards and this was used exclusively for ATSC 1.0 for standard definition video (525 line or with blanking removed 486 lines of active video and 720p as well as 1080i). ATSC 3.0 uses a different MPEG compression and transport technology which must be described separately.

Consider that in the analog days, the actual video transport was the video itself. This means that the maximum video resolution was determined by the channel bandwidth e.g. the 6MHz channel would contain 4MHz video and the rest of the space was for sound and to provide guard room for adjacent channels. For digital broadcasting, the physical transport is separated from the actual video itself. This means the channel was optimized for a reasonably robust carrier of data and the resolution of the compressed digital video woul be limited by the efficiency of the compression algorithm. Because digital compression is so efficient, multiple compressed video streams can be fitted or multiplexed into a single 6 MHz channel.

Although the ATSC adopted the MPEG2 video compression scheme for standard definition and high definition video, the MPEG2 transport stream has been almost universally used for delivery over the air, via satellite and digital cable. As MPEG2 compression is over 25 years old, newer more efficient compression techniques are being used for satellite and OTT (Over The Top). Note over the air, satellite and cable each use different RF modulation schemes.

The modulation scheme for OTA is 8VSB which can convey a payload of 19.39 Mb/s maximum. Cable 256QAM conveys 38Mb/s. Both use the 6MHz channel. Cable can convey more because the wired cable system is a more benign path less prone to electrical interference.

Lastly the studio resolution of standard definition broadcast is 6 MHz, up from 4MHz in analog days. And because standard definition is component rather than composite video, the resolution is subjectively greater still. HD studio resolution for both 720p and 1080i is 30MHz. Of course both SD and HD must be heavily compressed to be broadcast but subjectively this can be done these days with minimal loss.

I read the article you linked us to and it is good but missing bits and pieces of the puzzle. If you need any further explanation as to what any of it means, I will be glad to help decipher it for you.


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Fri 21, 2020 11:22 am 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 4334
Location: Lincoln City, OR 97367
Greetings to the Forum:

Electronic Memory wrote:

Since the transmitter signal is a digital video data stream (or rather several sharing a TX signal) you literally can record the digital bits directly of the RF demodulator save them to a file and play that file with a media player on your PC (such as the player you might be using to watch DVDs on your PC).


I don't believe this is true. If one were to apply an actual digital signal directly to the HDTV modulator, the spectrum would end up being "bunched... with lots of energy at certain frequencies and not much at others... effectively limiting the bandwidth further. Therefore, the 19 Mb transport stream (already MPEG 2 encoded) is further encoded in the transmitter exciter to form a spread spectrum signal. Spreading the spectrum (which can be employed as a form of encryption) causes the signal to appear as more or less uniform noise across the occupied bandwidth. However, without the key to undo the spread spectrum, the resultant bit stream would appear random. The key is, of course, standardized and built into the decoder circuitry in the receiver.... but just recording the raw bits from an 8VSB signal would get you psuedo-random noise.

BTW, the 8VSB signal is actually 8 levels of amplitude, which means that three bits are sent with each "symbol". You need a demodulator capable of discerning between the 8 discrete signal levels to determine the symbol value (3 bits).... something that is helped along by the pilot signal. Just dumping a raw demodulated signal into a digital port would get you nowhere.... at best case, half of the signal levels would be below the logic threshold level and would be interpreted as zero, while the other half would be above the threshold level and would be interpreted as one.

HDTV demodulators are magic.... without the modern revolution in microprocessors, HDTV would not be possible.

Regards,

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Fri 21, 2020 1:48 pm 
Member
User avatar

Joined: Aug Tue 28, 2018 9:22 pm
Posts: 2007
Location: Sanford Fla 32771 (USA)
It is starting to make some sense, I think.
Quote:
or OTA is 8VSB which can convey a payload of 19.39 Mb/s maximum.

Quote:
6MHz channel would contain 4MHz video and the rest of the space was for sound and to provide guard room for adjacent channels.

So based on a 4 Mhz bandwidth I would guess that 5 subchannels would be the maximum, or is it 4 based on full channel bandwidth of 6 Mhz ? ?
EDIT: Just checked local channel lineups and see up to 6 subchannels.

_________________
Paul of Florida ….I had my patience tested. I’m negative.
https://paulsironhorse.smugmug.com/


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Fri 21, 2020 4:22 pm 
Member

Joined: Jun Mon 24, 2013 3:00 pm
Posts: 1453
Location: Champaign IL 61822
Jthorusen wrote:
Greetings to the Forum:

Electronic Memory wrote:

Since the transmitter signal is a digital video data stream (or rather several sharing a TX signal) you literally can record the digital bits directly of the RF demodulator save them to a file and play that file with a media player on your PC (such as the player you might be using to watch DVDs on your PC).


I don't believe this is true. If one were to apply an actual digital signal directly to the HDTV modulator, the spectrum would end up being "bunched"

Regards,


The ATSC 1.0 digital signal "massaged" into a stream that has a uniform random noise like spectrum
before transmission. One step is Viterbi encoding, which is local (in time), and alone is
"almost" enough, but another nonlocal method is applied too. These methods are uniformly
applied, so are trivially undoable at the end. All the necessary info is in the standard.

There is no RF demodulator in ATSC. One heterodynes the signal down to some modest IF and
digitizes that, phase locked to the carrier. I should add that a phase lock is possible even if
there happens to be a narrow dip in the frequency response all the way to zero at the carrier.
(This is done, effectively, by full wave rectifing and phase locking to the second harmonic
of the carrier.)

That digitized signal could then be stored, though that would be inefficient. In practice the equalizer
would be applied and then it would be Viterbi decoded down to baseband digital before recording.

OF course one could analog filter at IF and then heterodyne the carrier down to about 500 kHZ and
analog record that and it would work fine.

Disclaimer: I owe parts of three patents involved in the process of signal equalization and Viterbi
decoding ATSC 1.0, and wrote a working digital equalizer/decoder which successfully decodes
analog recorded ATSC signals, and is about as good as the better ones normally used, though the
"best" ones are now a bit better.

I would add that "new" full frames are transmitted several times a second ... this is really "necessary"
as otherwise it would take that much extra time for a stream to get started when newly
tuned in. I don't remember what the standard says about the max time between new frames.
The "usual" equalizers are slow enough to start up to be annoying in any case.


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Fri 21, 2020 5:08 pm 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 4334
Location: Lincoln City, OR 97367
Greetings to dtvmcdonald and the Forum:

I never worked on the receiving end of things; but if one is to use an SDR approach to demodulating the signal, then one must lock to the symbol rate and then successfully discover one of 8 amplitudes at that time and thereby derive three bits of data. My point in replying to the original post (I may have misunderstood what he was saying) is that you cannot turn an 8VSB signal into a watchable digital data stream by simply recording the transmitted RF. 'Taint that simple.

BTW, the Harris exciters we used employed what the manufacturer described as "trellis" encoding to spread the spectrum. Also, in decoding 8VSB, timing is everything. The signal is unmitigated garbage except at very discrete time snapshots. One must sample the level at the correct time in order to decode the signal. I'm sure that this can be and is done with a software approach, but I would think it would be a non-trivial exercise.

Also, I'm not sure what you mean by "carrier". There is no carrier as such; only a pilot.

I also don't understand this:

Quote:
OF course one could analog filter at IF and then heterodyne the carrier down to about 500 kHZ and
analog record that and it would work fine.


500 KHz is too low a frequency to carry the digital data load. Of necessity, even after using tricks like 8-level coding or COFDM, you would need at least the original 6 MHz bandwidth to convey the information.

Regards,

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Fri 21, 2020 5:49 pm 
Member
User avatar

Joined: Nov Thu 11, 2010 6:03 pm
Posts: 1224
Location: Pewaukee, WI
Jthorusen wrote:
Greetings to the Forum:

Electronic Memory wrote:

Since the transmitter signal is a digital video data stream (or rather several sharing a TX signal) you literally can record the digital bits directly of the RF demodulator save them to a file and play that file with a media player on your PC (such as the player you might be using to watch DVDs on your PC).


I don't believe this is true. If one were to apply an actual digital signal directly to the HDTV modulator, the spectrum would end up being "bunched... with lots of energy at certain frequencies and not much at others... effectively limiting the bandwidth further. Therefore, the 19 Mb transport stream (already MPEG 2 encoded) is further encoded in the transmitter exciter to form a spread spectrum signal. Spreading the spectrum (which can be employed as a form of encryption) causes the signal to appear as more or less uniform noise across the occupied bandwidth. However, without the key to undo the spread spectrum, the resultant bit stream would appear random. The key is, of course, standardized and built into the decoder circuitry in the receiver.... but just recording the raw bits from an 8VSB signal would get you psuedo-random noise.

BTW, the 8VSB signal is actually 8 levels of amplitude, which means that three bits are sent with each "symbol". You need a demodulator capable of discerning between the 8 discrete signal levels to determine the symbol value (3 bits).... something that is helped along by the pilot signal. Just dumping a raw demodulated signal into a digital port would get you nowhere.... at best case, half of the signal levels would be below the logic threshold level and would be interpreted as zero, while the other half would be above the threshold level and would be interpreted as one.

HDTV demodulators are magic.... without the modern revolution in microprocessors, HDTV would not be possible.

Regards,


Jthorusen wrote:
I never worked on the receiving end of things; but if one is to use an SDR approach to demodulating the signal, then one must lock to the symbol rate and then successfully discover one of 8 amplitudes at that time and thereby derive three bits of data. My point in replying to the original post (I may have misunderstood what he was saying) is that you cannot turn an 8VSB signal into a watchable digital data stream by simply recording the transmitted RF. 'Taint that simple.


You totally misinterpreted what I was trying to say...

What I was saying is that after the signal has been correctly demodulated and it's digital data pay load made made back into plain digital data that data stream in ATSC 1.0 is just an MPEG2 data stream which many media players on computers are designed to handle. I was not trying to imply there that the raw RF level passes through a DAC can just be dumped into VLC or windows media player and give video (as you seem to think I was saying).

However something similar can be done SDR (Software Defined Radio) hobbiests and engineers can record a raw data file of a pair of DACs recording the RF signal in quadrature and configure a computer math program to demodulate, decode and recover the data pay load of any transmission format if enough is known or can be deduced about it's specifications. You would be amazed what consumer grade SDRs can do, professional ones are even more amazing.


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Fri 21, 2020 6:06 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1668
Location: Phoenix, Arizona
As I am sitting here this morning directly in front of a full power digital TV transmitter that is on the air . . . . . .it is amazing to think of all the math flowing through the thing vs the end result coming out the far end.

Car wrecks, bad politicians, roberies, drug deals and no rock-n-Roll music.

For the love of technology.

Jim


Top
 Profile  
 
 Post subject: Re: Old anolg TV bandwidth verses new digital TV.
PostPosted: Feb Sat 22, 2020 2:33 am 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 4334
Location: Lincoln City, OR 97367
Greetings to Electronic Memory and the Forum:

Yes, I did misunderstand what you were saying. Thank you for the clarification.

Regards,

_________________
Jim T.
KB6GM


Top
 Profile  
 
Post New Topic Post Reply  [ 29 posts ]  Moderators: 7jp4-guy, Mr. Detrola Go to page 1, 2  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: ACORNVALVE, madlabs and 4 guests



Search for:
Jump to:  


































-->


Privacy Policy :: Powered by phpBB