Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
The Internet

HDTV Over IP 124

Posted by CmdrTaco
from the babysteps-towards-broadband-media dept.
gravelpup writes " NASA Watch has this article about a NASA demo of streaming an HDTV feed over a 20Mbps network from D.C. to California. Suddenly, watching NASA TV at home over a dinky DSL connection doesn't seem so cool anymore." For some reason this just makes streaming high quality video over the net seem even further away to me.
This discussion has been archived. No new comments can be posted.

HDTV Over IP

Comments Filter:
  • When I read an the blurb on this article ant think to myself:

    1. How did they bypass the encryption?

    2. Boy Nasa's gonna get sued..

    • Considering HDTV is a standard, not a product that requires encryption.. no, they aren't going to get sued, and there wasn't any encryption to break.
    • Note to self: Spellcheck a little more closely so I don't sound like Commander Taco.

      This should have read:

      When I read the blurb on this article and think to myself:

      1. How did they bypass the Encryption?

      2. Boy, Nasa's gonna get sued...

  • Maybe IP isn't the appropriate media for high quality video transmission.
  • What's new? (Score:3, Informative)

    by zook (34771) on Tuesday August 07, 2001 @08:47AM (#2109971)
    What's new besides the mutlicast aspect? Hasn't this already been done. [washington.edu]
    • It was a hybrid link in which the signal was uploaded to a satellite, downloaded to a mobile groundstation, and sent across a wireless link into the demo location.
    • Yeah, it was done in 1997, when I demonstrated HDTV content inside (64-bit-sized) Quicktime files, being streamed over IP-on-HIPPI. We (SGI) showed this at IBC that year, as well as at the next NAB. We also showed RICE-encoded video (no-loss compression) on 100Mb ethernet. We were serving it all from an O2000 with a big RAID array for the content. Not only did it play in streaming mode, but you could do frame-by-frame navigation and scrub. A couple of video editing companies were using it in their products, and there was at least one competing network technology for the same result, as well. It was all for sale. No one really wanted it then, either, and I doubt that many people want it now. (BTW, the hard part about VOD isn't the network bandwidth, it's the disks.)
  • by awb131 (159522) on Tuesday August 07, 2001 @09:08AM (#2111275)

    I've seen something like this before -- a 36 Mbit DV stream sent over the Internet2 (IP network instead of Firewire) from Ohio to Pennsylvania. It was just a test, to see if it could be made to work. Latency was in the 150ms range. (Basically it was two FreeBSD machines with Firewire and tuned 100Base-T cards on both ends.)

    At the time, my reaction was "What a waste of bandwidth!" but extremely high quality video streams at relatively low latency are critical for remote instrumentation/manipulation applications. Like moving a robot arm in space, or allowing scientists from all over the place to use one piece of very expensive equipment instead of moving them all to the same location. We also considered using something like that in an on-campus video editing facility for moving footage around from machine to machine. I can see the use for it in some situations.

    But for broadcasting? I don't see the point of using all that spectrum just for a video.

  • by foobar104 (206452) on Tuesday August 07, 2001 @09:07AM (#2111795) Journal
    Just a clarification.

    When people say "HDTV," they can mean either the full-bandwidth uncompressed signal (on the order of 1.3 Gbps) or the broadcast-standard signal (MPEG-2 compressed to the order of 19.7 Mbit, or about 50-to-1, more or less).

    Obviously this test didn't use uncompressed HDTV. Must have used MPEG-2.

    When it comes to standard definition TV, the stuff you get over digital cable or DBS is typically between 4 and 6 Mbit. I think most people would consider 2 Mbit to be unacceptably noisy... but then again, I can ignore an awful lot of softness and artifacting from my TiVo, so maybe even 2 Mbit would be acceptable under the right circumstances.

    Uncompressed standard-def TV, on the other hand, is carried over a 270 Mbit signal.

    MPEG-2 compression seems to be totally acceptable up to 50-to-1, and marginally so up to about 100-to-1. DVCpro 25 (25 Mbit, or about 10-to-1) is widely considered to be crappy by broadcast standards, but looks a damn sight better than my TiVo on my home TV.

    My rambling point (coffee, please) is that "HDTV" is a soft, fuzzy concept. Squeeze it down to 5 or 6 Mbit and it'll still be HDTV, with a thousand lines of resolution on-screen. But it might be so fuzzy or artifact-y that nobody would watch it.

    • The title of this article 'HDTV over IP' is a bit misleading:

      20Mbps over IP is no mean feat. 270MBps (yes, Mega Bytes) over IP, OTOH, is the bandwidth that would be required for full res HDTV. You would need a pipe that gives you roughly the same bandwidth as an OC24!

      You could also achieve these speeds if you used a different protocol, such as HIPPI [hippi.org], for example.
    • by SteveSgt (3465) on Tuesday August 07, 2001 @10:14AM (#2136155)

      The standard for uncompressed HDTV is a document called SMPTE-292M -- it's a 1.485Gb/S synchronous stream typically carried over coax. There is an IETF draft RFC for carrying uncompressed SMPTE-292M over IP HERE [ietf.org].

      The University of Washington, with the assistance of Sony and Enron, presented a demonstration [washington.edu] of seven channels of HDTV compressed to 200Mb/S over an OC-48 backbone at the National Association of Broadcaster's [nab.org] convention in April of 2000. In this demo, they produced a HiDef newscast on the floor on the Las Vegas Convention Center, while the newcasters, cameras, and the broadcast transmitter were all in Seattle.

      I know there were limited demonstrations of highly compressed HDTV over internet protocol almost a year before that. One group that has been working on that is a University consortium called The Research Channel [researchchannel.com].

      By the time the MPEG toolkit compresses a video signal down to 50:1, a LOT of data has been discarded. You see strange artifacts (if you're watching carefully enough) such as arms disappearing while the football player is throwing the ball, or water behind a moving boat looking more like clouds. Yes, for some still images you still get the 1920X1080 resolution, but mostly you get interpolated fuzz lower than the resolution of standard-definition video.

  • ...and you can have your PVR, and the NASA channel.
  • Look at this link: http://www.washington.edu/hdtv/nab/press.html [washington.edu]. The University of Washington sent 4 HD streams down to Las Vegas. In LV, all the video switching, chroma key, and digital fx were added. The resulting stream was sent back to Seattle for broadcast. Each stream was in the 200Mbps range. Nice.
  • That's a press release, not an article. Please don't try to pass it off otherwise.
  • Let's just stay away from using IP for Broadcast. Call it multicast or whatever you want to call it, but it isn't an ideal way to transmit a signal to a bunch of people and it just eats up bandwidth for people who are using the internet for two way communication.

    It may be cool to get everything over IP but it doesn't make sense. Let's get video conferencing working on some sort of standard first.
    • Ah, but it isn't broadcast. It only goes where it's wanted, and two connections on the same subnet use about as much bandwidth as one connection.

      Yes, it may eat bandwidth, but only the same amount as downloading a big patch or video file. (not the same number of bytes, but we're talking bits-per-second here)

      If it's pushing the limits of the IP infrastructure,

      a. there's something wrong with the infrascructure-- let's light some dark fiber
      b. it's unwatchable, and its effects will be limited.

      You know, I used to think it was dumb to try and merge computers and TV sets. Now I don't own a TV-- I watch everything on my video capture card.

      I'm not saying it's coming soon. But it's coming.
  • Yes yes, this is all very cool, but there has to be consumer demand to fuel this. Considering how long it is taking HDTV to become the norm (if ever), and our propensity to easily eat up whatever bandwidth current technology delivers with inane shit, I would truely be surprised if TV starts coming across the 'net anytime soon (ie, 10 years). Somehow, the media providers would have to slip their foot inside the door to reserve bandwidth on a telco's network before it gets eaten up by consumers? I'm just thinking about the case where the consumers already have the link, and are used to having bandwidth X available .. it'd be a tough sell to start piggybacking HDTV on those connections, and tell your consumer base that their available bandwidth will now be X, to make way for multicast HDTV streams you may not even be interested in.

    Actually, I'm more interested in 'friendly off the air' messages in explorer:

    "I'm sorry, the TV show you are trying to watch is unavailable. Please hit 'reload', or try again later."

    That'd be enough to drive me back to my remote 'n good 'ol cable.
  • by alteridem (46954) on Tuesday August 07, 2001 @08:34AM (#2120278) Homepage
    Everyone said high speed Internet access with DSL and Cable would bring in an age of high bandwidth Internet applications like TV over IP. This article just shows that this is not going to be the case for sometime. If HDTV requires a throughput of 20 Mbps, then our fast connections at home are still a long way off. Besides, when is the last time you even managed full bandwidth when doing an upload? Something upstream is always a bottleneck.

    Just think about it, a T1 is 1.5 Mbps, my cable modem max's out at 2.9 Mbps (not that I ever see that.) These bandwidth hungry applications are still a long way away, at least until the next Internet revolution when we all have fiber to the home...

    • Just think about it, a T1 is 1.5 Mbps, my cable modem max's out at 2.9 Mbps (not that I ever see that.) These bandwidth hungry applications are still a long way away, at least until the next Internet revolution when we all have fiber to the home...

      I can't help but think that a lot of undue emphasis is being placed on installing faster internet links. I can remember when RealVideo was shiny and new, and people complained that streaming video wasn't watchable even over a T1. Today this isn't the case, but this is because of the increased complexity of the encoding and not the speed of the link. Lots of current DSL and cable lines are more than fast enough to replace broadcast quality NTSC signals, but only using very recent compression schemes like Sorenson or OpenDivx. Next generation encoding methods will be commoditized and moved into the mainstream long before fiber in the home becomes a reality and will yeild much better video quality than current standards could even with a ten fold bandwidth advantage.

    • Not only do I see this as a problem with the downstream bandwidth I get from my ISP, I also see this as a problem for servers.

      What kind of server can send out 20MBps streams to thousands if not millions of users?

      And we think the net is clogged now. Imagine a HDTV server is streaming something that gets posted as news to /.
      • They are using IP multicast. NASA already uses IP multicast to distribute high data rate telemetry streams from ground stations to control centers and other end users over NASA networks.
      • Yao said ThunderCastIP HDTV is an industrial-strength multicast video
        server for professional enterprise applications.
        ``The software supports live
        or pre-recorded video streaming from 10 to 50 Mbps over ordinary IP-based
        networks.
        Video can be streamed to virtually an unlimited number of users simultaneously via an IP multicast without significantly impacting the
        network.''

        It's not establishing a separate link to each user, it sends out a multicast stream.
        • Maybe I just don't understand how this works, but at some point the signal will have to be separated to each users line right? I mean, its not like a cable wire where every channel is sent out but you only watch what channel your on. So maybe the server that sends out the multicast isn't all that affected, but at some point, some server or router (i dont know much about netowrking- ?) is going to have to send these huge amounts of data down separate lines. Thats where I thing bottlenecks would happen.
          • That is exactly the problem. Multicast has little impact on your network, but as you start going upstream and it starts splitting out to separate networks, then the bandwidth problems begin. Think of multicast like one road leading away from your house, then constantly forking out into new roads. Regular IP is more like thousands of parallel roads leading out from your house to different destinations. The end result is the same, but the single driveway out of your house is a lot easier to build and maintain.

            The only difference between the road analogy and real life is that for multicast, if you sent a truck out from your house (a data packet,) at every intersection, that truck would duplicate itself at every intersection. With regular IP, you would just send out all the trucks at once.

            Of course, all of this is greatly simplified (as analogies tend to be ;o) but I think it gets the point across...

      • by Chundra (189402) on Tuesday August 07, 2001 @08:51AM (#2127387)
        It's called an antenna. Back in the old days we used to use them to broadcast all sorts of signals.

        To millions of people.

        Without lag.

    • by JoeShmoe (90109) <askjoeshmoe@hotmail.com> on Tuesday August 07, 2001 @09:05AM (#2121662)
      Here in Sacramento, Western Integrated Networks [winfirst.com] is doing just that.

      Fiber (well really it's that hybrid coax/fiber system that cable companies already use...just fatter pipes the whole way down and more focus on two-way syncronous connections) to the home with a single connection that does telephone, Internet and digital TV.

      According to a couple techs I've talked to, the telephone service is basic VoIP. Since a T1 is 1.5Mbit and that's I think 32 lines then I don't expect this will take up much bandwidth. Supposedly the interface to the "modem" is going to be 10BaseT (it remains to be seen if I will be able to hook my own hardware into the line at its true 100Mb+ speed or I have to use their hardware) so that isn't a lot of traffic. Now each TV channel is apparently a full 5-6MBps MPEG-2 video stream. This I imagine is going to chew up the majority of their system bandwidth, especially if they plan to offer the same channels as AT&T digital cable or DirectTV.

      It's kinda amazing to think about how much data that single coax from your cable provider carries. In order to provide the equivalent hundreds channels of video, WIN is having to rollout some pretty high powered stuff.

      - JoeShmoe
    • I agree. The one bit of optimism I have is that multicasting IP will be more common on broadband providers. That would really be necessary for cable providers where their outside bandwidth for the whole area can be as little as a T1 today. For one user to be able to uset streaming video the total bandwidth for the area would have be above 20Mbps (DS3/T3 maybe?), and without mbone every additional streaming user would add 20Mbps to that.
    • Everyone said high speed Internet access with DSL and Cable would bring in an age of high bandwidth Internet applications like TV over IP. This article just shows that this is not going to be the case for sometime. If HDTV requires a throughput of 20 Mbps, then our fast connections at home are still a long way off. Besides, when is the last time you even managed full bandwidth when doing an upload? Something upstream is always a bottleneck. Just think about it, a T1 is 1.5 Mbps, my cable modem max's out at 2.9 Mbps (not that I ever see that.) These bandwidth hungry applications are still a long way away, at least until the next Internet revolution when we all have fiber to the home...

      There is no 'will' about it, TV over IP *has* already arrived in the UK. We (www.kitv.co.uk) carry a MPEG 2 encoded PAL signal using an IP transport over ADSL and our system is available to consumers TODAY.

      Using MPEG 2, It's possible to carry a reasonable TV signal in ~2.5Mbps, and a pretty good one in ~4.5Mbps. The picture quality is superior to analogue broadcasts and comparable with cable/satellite systems. In practice picture quality *really depends* on how much you are prepared spend on your MPEG encoders, No 199 pound/dollar mpeg cards here. We use custom built broadcast spec PIXStream encoders which cost 60,000 pounds (~90,000dollars) apiece. It perhaps worth mentioning that PIXStream make probably the 'best of breed' MPEG encoders in the world and choosen because we know we are pushing the envelope.

      The nature of the platform (Multicast IP) also means that an MPEG 4 upgrade is a straight forward exercise, and since MPEG 4 is nearly twice as efficient, we can improve picture resolution and save bandwidth.

      This technology will make Cable Cable uses a ring based network topology, so the bandwidth is shared between connections. ADSL uses a star topology, so the contention is moved back from the MAN backbone to the head end. Our customers get between 4.5-8Mbps each depending on their line attenuation.

      Cable and Satellite are obsolete. They just don't know it yet.

  • Fibre is not necessary, Interactive Digital TV can be delivered at 4.5Mbps over ADSL, for a fraction of the cost.

    How can I be sure?

    We are already doing this commercially. I work on a system, which delivers Interactive Digital Television using IP over ADSL. It includes 60 channels of broadcast TV, Video on Demand using the largest nCube Video Server installation in the world, Web and Email, TV-Commerce, a Local Link which includes links to local services such as Schools and Libraries & Pizza Deliver. Our News on Demand service has won innovation awards including from the Royal Television Society, (which was perversely sponsored by our arch Rival BT :-). We also have an Interactive Virtual Avatar.

    Checkout this link for a detailed case study.

    http://www.broadcastpapers.com/data/KingstonCase St udy01.htm

    BTW I've submitted links to our story 3 times over the last two years, but /. Has never seen fit to publish it, now a much simpler US system gets the coverage, make me wonder.

  • And when will we be able to watch "Geeks in Space" over an HDTV signals ?
  • TV over IP is probably going to herald Video On Demand and you know what that means : pay as you go TV.

    Maybe then though the bandwidth on the TV satellites can be utilised for IP traffic.

    There are so many barriers to success though, that although it's a cool tech achievement widespread deployment will probably have to wait for a paradigm shift in the internet infrastructure. All those ISP's have got a lot of investment in their current hardware that the budgets probably project them for at least 5 years use.

    My DSL provider (ntl:) is also a cable TV provider. The analogue TV & cable modem comes into my house on the same wire. 50 channels of TV & 1 x 512k data. I don't think that they are going to squeeze 20mbps of data through there too any time soon.

    but let's look forward to fibre to the door and then we'll see things happen but probably not for quite a few years yet.

    Our kids will probably get it but by then our eyes will be too dim to notice the difference between HDTV and analogue!

  • The University of Washington (UW, not WU) already did this about two years ago. And using studio quality HDTV to boot. There were two different types of streaming, one which was around 40Mbps and a higher quality one at 200+ Mbps. Check out http://www.washington.edu/hdtv/ [washington.edu] for more info.

    -B
  • 20Mb/s can't even do uncompressed standard definition video (that's with no audio). If you haven't seen HD uncompressed, you may not notice the blurriness of beat-down HD- but the end result is that it looks a hell of a lot like standard def NTSC. But we get to call it HD!

    All of this would be funny if people still had the disposable income they had two years ago. Now it just seems sad.

    It's six times larger, the programs are six times better!! Hurray!!!1! [ridiculopathy.com]

    • While I'm sure uncompressed 1080i is spectacular. it's not realistic for anything approaching a consumer level device.

      As the sales numbers show, consumers are happy with MPEG compressed video. Look at the number of DVD movies and players that are being sold now.

      HDTV takes that wide screen format common to most DVD's, and cranks up the resolution. While you may see some artifacting (use 720p for less of that) everything I have seen in HD looks MUCH better than the equivalent DVD format.

      We always want higher resolutions and newer technology.. But, the difference between broadcast HD and standard definition is HUGE. Anyone who says different has never really seen an HD broadcast.

  • I want to do this over the 100BaseT LAN in my house.

    [Permit me to go off on a tangent..] In fact, I want to allow all my content input/generation devices to broadcast on my LAN, where any other content presenters can pick them up.

    I want to:
    - play cd or tuner from my living room stereo and also hear it in my bedroom or basement
    - when watching the game on tv, also feed the audio to the bathroom :-)
    - when surfing or gaming at my computer, monitor what someone else is watching on tv (or the other way around would be cool: mirror my game on my big screen tv (but that's a different problem))
    - have an audio & video server which can play to any screen or speaker in the house

    Is this possible? impractical? silly?

    --
    no sig, no plan, no clue
  • University of Washington folks have been working on this for a while, take a look at:
    http://www.washington.edu/hdtv/ [washington.edu]
    and the Research Channel:
    http://www.researchchannel.com/hdtv/ [researchchannel.com]
  • by Kagato (116051) on Tuesday August 07, 2001 @09:31AM (#2128255)
    Everytime there is an HDTV story posted I read people WHINE about how bad HDTV is. Just stop it. The reason HD (or DTV for that matter) have not taken off is the MPAA (or rather who they rerosent.) HD content is sparse. Why? Studio's don't want to produce it. And it's not a matter of technology. It's not a matter of costing huge ammounts of money for the studio to make an HD copy. In fact many prime time programs are filmed, then converted to video during editing. Usually drama's, sci-fi, made for TV movies, etc.

    Studio's hate HDTV. Why? Because it ruins a very important Video market. They now count on the fact that VCR's make low quality, grainy copies of on-air content. This means they can make tons of bank on [insert fav show here] box sets. Once you deal with a digital format they are sunk. People can now make a high quality recording for personal use. Hence no reason to buy an over priced box set from the local retailer.

    While people can contend the studios and networks are free to do as they please, I would counter that the networks are allowed to use OUR airwaves for next to nothing. With out over the air content no one would buy a box set show. Like it or not, Timeshifting is legal and is considered a RIGHT we have gained in exchange for allowing networks to use the airwaves.

    HD prices could have dropped like DVD prices by now if the studios didn't stand in the way. The mear fact that hardware vendors keep having to go to the design stage to add new copy protect and transports to please the studios is just crazy.

    HD is $$$ and in it's intfancy because of the DMCA and studio money. It's doesn't matter if you're a Dem. or Rep., because neither party took a stand and did any thing to protect the consumer.
    • Studio's hate HDTV. Why? Because it ruins a very important Video market. They now count on the fact that VCR's make low quality, grainy copies of on-air content. This means they can make tons of bank on [insert fav show here] box sets. Once you deal with a digital format they are sunk. People can now make a high quality recording for personal use. Hence no reason to buy an over priced box set from the local retailer.

      Hell no. Having a family member that worked in broadcasting for 20 years, I can tell you that the reason that no one's producing HDTV equipment is cost.

      You think that the _consumer_ gear costs an arm and a leg? Just try upgrading the cameras, monitors, editing equipment, and mastering equipment. For each studio.

      On the station side, you're going to need a new control room, bloody TRANSMITTER [horribly expensive pieces of equipment], and sometimes a tower to boot, addition to ugrading the news studios and remote trucks (mirowave and satellite links).

      And all that for crappy programming that three people in the entire country own the equipment to see in the native resolution.

      • Cost to the station is not relavent to the process since they can't opt out of upgrading. Although certainly some stations have dragged there feet, by the end of summer most of the major markets will be up.

        So that leaves us with the cost for producing the actual content. That's usually up to the Major Studio's. (WB for example makes ER for NBC). If you looked at a program like X-files, you wonder, "How much would it cost to get HD-X-files?" The answer is not a whole lot. The show is filmed in 35mm, most likely digitized into a non-linear editing suit, then mastered to a D1 (or similar). There's a good chance the suit is already handling the editing in a Hi-Res format. So at the low end all they have to do is change the mastering process, high end a new editing suit. But one episode of Dave D. pay is prolly 20 times as much as that would cost.

        And last, but not least, three people? You're smoking crack. First off the majority of front projection systems sold in the last 5 years have the ability to display HDTV material. All I had to do was add a $400 Dishnetwork 6000 box to the system. Getting Over The Air signal was a matter of adding a $99 module. The number of sports bars alone that had some sort of Front Projection CRT that would jump on a gimick like HDTV should be enough to get content producers going. Let alone Home Theater.
    • First, I would still buy box sets. Second, many shows never had box sets, or they were made in too small a quantity. Twin Peaks comes to mind...
  • Okay, maybe it's not HDTV. (Anyways, I don't own a set, and I don't know anyone who owns one either).

    But for the rest of us, there's VDSL, the DSL on steroids (up to 52mbps on copper). There have been some trials in US and Canada, I have seen the equipment and the thing is amazing. No new wiring, no disruption, digital TV, high-speed internet, plus internet telephony.

    Here [eurescom.de] are some slides that talk about it.

  • by MartinG (52587)
    The kind of bandwidth apps use today amazes me when I think that only a few years ago when I was at university, us irc users we're constantly moaned at for hogging all the bandwidth.

    I wonder if we will ever reach a point where no (ab)use of available bandwidth will be criticised simply because it will hardly even be noticable?
    • It will only get worse.

      I am programmer from the days when 12k of ram was BIG! Yes, kilo!

      Today, I still write small tight routines.

      But most programmers today, throw more hardware at the problem to get better speed, instead of improving their style.

      This idea of HDTV over IP is fine, but compress the single to the 5M it needs, instead of the wasting 4x that.

      The computers are more than fast enought to do that. MIT proved that years ago. IE: HDTV in the same channel space normal TV.
    • Except if you live in Australia where Telstra and competitors will probably still be charging $0.20 per megabyte in 15 years time.
    • "You can't play Elite on the mainframe! Think how many processor cycles you've wasted! Look, this isn't an Altair we've got here, buddy, it has the storage capacity of 125 college textbooks...."
    • You're forgetting the First Law of TeleCommunications:

      Pr0n expands to fill vacant bandwidth.

      How long before goofballs are 'broadcasting' 1080i realtime pr0n, or selling interactive VR Jenna Jameson from their websites?

      Mmmm, interactive VR pr0n....
  • So we're going to have HDTV video from NASA. But the important question is: what about the audio? The video I see on NASA TV (I get it on my cable system) looks pretty good, about what I'd expect from modern video equipment. But whenever they broadcast audio as well (an interview from the shuttle, for example), it sounds the same as audio from an Apollo mission. Why can't they use TV-quality audio when broadcasting TV-quality video from orbit?
    • The Shuttle uses 32 kbps delta modulation for its air-to-ground voice communication links. See this page [lincom-asg.com] for some references. This allows the Shuttle to multiplex 2 voice channels (2 x 32 kbps) and 128 kbps of telemetry into a 192 kbps telemetry downlink. There are newer audio encoding techniques that provide better fidelity, but this stuff was designed in the 1970s. Delta modulation has the advantage of being resistant to degradation caused by bit errors and bit slips in the RF link. It is also is easy to encode and decode, allowing simple and reliable hardware to be used.
  • HDTV has been around for what, 10 years? Only now is it gaining some acceptance. If HDTV is going to be the factor that drives us towards REAL bandwidth (10Mbps+ into your home), it might take some time. For now, I'll be very happy if I can get TV quality video on demand that doesn't suck.

    What a market.. if you could beat the cable providers to good VOD, you could take a lot of business from them. They don't deserve that business either - they've been working on digital cable for years, and it doesn't look any better than crappy NTSC, or give you video on demand. My box doesn't even have S-video or digital audio.

    • The HDTV standards weren't even finalized until '93. The FCC didn't adopt until these until '95 or '96. The first broadcast of Jay Leno in HDTV 1040i didn't happen until April of 1999.

      • I didn't mean HDTV broadcasts.. as you said, the standard was finalized many years ago. It took way too long from the moment the engineers said, "it's done" until there were actual HDTV broadcasts and sets available. HDTV's are still overpriced.
    • You're all wrong. The concept of HDTV (high-definition television) has actually existed since the early 1980s, with some neat proof-of-concepts models made with very expensive analog components. The goal for moving all television to HD was set then way back then.

      Now for digital television (DTV), this is a new phenomenon, and came about in the last 10 years (with MPEG-2), and coupled with its lower cost and lower bandwidth, DTV and HDTV became one and the same.
  • From all the news lately you would think that it was the Code Red I & II worm that were slowing down the Internet. But no, its those "rocket scientists" at NASA wanting to watch HDTV over IP sucking down all that precious bandwidth. Your tax dollars at work, letting geeks watch high definition pr0n from outer space. Jeeez!
  • do people still watch nasa broadcasts anymore anyway? not that i don't love the american public space program, but i was just curious whether this was the best example of viewing television-style programming online considering nobody seems to watch it on their boob-tube.
  • That's what, 3 DSL lines? $60 a month? As long as you put the TIVO (not literally) at the phone company what's stopping us from video on demand at $60/month? 95% of your DSL bandwidth limits happen after you get to the phone company, not before. Hell, I don't need HDTV quality. DVD quality is more like 6Mbps, or $20/month.
    • That's what, 3 DSL lines? $60 a month? As long as you put the TIVO (not literally) at the phone company what's stopping us from video on demand at $60/month? 95% of your DSL bandwidth limits happen after you get to the phone company, not before. Hell, I don't need HDTV quality. DVD quality is more like 6Mbps, or $20/month. Just because you don't like it, doesn't mean it's unconstitutional

      From whom are you buying your DSL, and where can I get some? Full T1 speeds are 1.5MBPS, and that's where most comsumer DSL tops out -- buisnesses can sometimes pay \$$Largenum for Multi-Mb/S RADSL lines, but you're still not going to get 20Mbps for under $300/mo or so.

      Also, most of that cheap DSL shares your voice line -- any dedicated line is going to cost you for the second/third/tenth line.

      • I am using verizon DSL on the slowest plan, and I regularly get 2.5Mbps. The modem tops out at 7.1Mbps download though, the only reason I top out at 2.5Mbps is because of the next hop bandwidth. Think about it, RADSL goes over the same copper wires. The hardware at the two ends is the only difference, and the bandwidth they need to reserve for the next hop. My point is that with video on demand, the next hop is meaningless, if you store the video at the phone company.

        Yes, it shares my voice line. But I still contend that at least half of the cost is coming from the next hop, whether it's voice or data. Direct point to point copper lines don't cost $20/month, no matter how you slice it. They probably cost the phone company about $5/month in recurring costs. I'd bet it's less than that, even.

  • by iforgotmyfirstlogon (468382) on Tuesday August 07, 2001 @09:01AM (#2149896) Homepage
    ... that NTSC over IP isn't all that far off. HDTV has 5 times the information of NTSC, ergo, NTSC over IP should only take up about 5Mb/sec.

    -Freed
  • Streaming MPEG 2 is nowhere near as efficient as some of the MPEG 4 codecs available. Take a look at www.3ivx.com. I got a 12 to 1 compression ratio over an MPEG 2 file which I ripped from a DVD [which I own]. Basically there is almost no loss in visual quality and a 77 MB file dropped in size to around 6.8 MB. Worth a look if you ever wanted to exercise your right to fair use...
    • I think it's just the lag time involved in implementing new algorithms.. standard broadcast HDTV uses MPEG2 encoding, so all HD equipment supports this.

      Also, 'almost no loss in quality' in a small window on your PC monitor is a lot different than good quality on a 34" to 100" HDTV system. I would be very interested in seeing what MPEG4 could do with a 10 to 15Mbps stream at 1080i or 720p resolutions.

      MPEG4 could still be useful in a closed system, like a cable TV provider with a set top box. But, it's too late for broadcast HDTV. Too many boxes are already out there with MPEG2 based HDTV. This includes DirecTV.. there are about 10 different set top boxes that support DirecTV HD.

  • Technology envy (Score:3, Insightful)

    by Alien54 (180860) on Tuesday August 07, 2001 @09:11AM (#2151109) Journal
    There is the old adage I remember from retail, given the way people behaved sometimes. Some people would ask around for a system that would not cost much, and not be obsolete for 4 or 5 years.

    The eventual answer was that "If you can buy it, it is obsolete."

    Technology envy strikes again

  • British Sky Broadcasting (Rupert Murdoch's European satellite broadcaster) recently bought the licence to broadcast digital TV over ADSL in the UK. The service will begin in the next couple of years.

    Note that this is /not/ HDTV -- it's MPEG2 encoded PAL.
    • Murdoch has shown a keen eye for avoiding wrong technologies. When the offical 'right' way to do DTH broadcast was D-Mac, he launched using PAL, and wiped the D-Mac off the table. Now when the 'right' way to do stuff if HDTV, he's sticking to MPEG2 PAL. It's the KISS principle in action. http://www.terramedia.co.uk/Themes/BSB.htm for an analysis of BSB v Sky
    • Actually Sky's licence is only for upto 1001 homes, it's only a trial. Kingston Communications (www.kcom.com), have been doing this for two years, and our service (www.kitv.co.uk) includes Web & Email, Content (News/ Movies) on Demand, frankly KIT blows the Sky system out of the water, in terms of Capability.
  • by Bonker (243350) on Tuesday August 07, 2001 @08:43AM (#2151557)
    I mean, look at all the industry controls and FUD built into HDTV. The format is less than ideal, and all the hardware required to play it is exorbitant.

    This makes getting open source video formats in place even more important so that, in the very near future, we don't have to make a decision like the one we're making right now between OGG and MP3. One format is techincally superior and open, while the other is the 'industry standard'.
  • I just love these scintilating articles about digital cable, HDTV, etc, etc. It must be a wonderful thing to work for NASA and have such resources available, but the community I live in (only a stone's throw from Silicon Valley) was probably one of the last places in the US to get cable, and then expect it to even be upgraded to digital. Nothing stopping this from satellite, except the bill.

    As far as I'm concerned tho, if *I* had a 20Mbs connection, watching TV would come far second. I'd want to engage in some serious virtual mudwrestling with Natalie Portman! =)

  • The linked "article" is a press release by 2netFX, the company that is selling the technology. Please make a note of this and consider the source. I can't tell you how much it irks me when people pass off press releases as news (plus, it means that the marketing drones have fooled the mighty /. nerdhorde).
  • I work at the UC Berkeley Multimedia Research Center. One of the things we have going on here is the encoding and decoding of video streams for use over the Internet/Internet2 using IP Multicast.

    The thing is, you can get really good NTSC signals going at 20Mbps. This is television you would watch and you wouldn't be able to tell the difference without having another TV next to it and pushing your nose up against the screen. I'm not surprised that NASA is able to get a decent HDTV feed at 20Mbps. (If you read the article, you'll see that the signal was "crystal-clear", and not a thousand lines of noise.)

    The problem is you need hardware to get decent framerates. We use LML33 boards to encode and decode NTSC to MJPEG at 30fps. These boards are $400. To encode MPEG-2 at a reasonable frame rate, you need hardware is at least $2000, I believe. If you try to use software encoding/decoding then your mileage varies but I usually get less than 10fps using 1/4 NTSC or CIF in MJPEG.

    The software is here: http://www.openmash.org/

Crazee Edeee, his prices are INSANE!!!

Working...