Broadcast approval for camera

[Augusto Manuel]

However, the UVW100 is an exception of low quality Betacam SP. While for DV there are so many inexpensive units which do not compare to the quality of the medium grade Betacam cameras.

About the DSR-500 and being better than any of the Betacam SP camcorders until you get to the 600, that is highly debatable. I would settle for footage coming from a Sony PVW-D35 or PVW-D50 over a DSR-500. Same with the BVW-400a.

Originally Posted by Bill Pryor

And, some of that shows a total ignorance of production. Accepting Betacam SP over DV, for example. That would mean you could use something shot with a UVW100 over a DSR500, when the DSR500 recording to DVCAM looks better than any of the Betacam SP camcorders until you get to the BVW600. Like so many who should know better, they seem to confuse the format with the camera. I'm sure they've got to be running stuff shot DVCAM and other DV25 formats that have been delivered on Digibeta masters.

[Bill Pryor]

It's true the UVW100 was an exception, but that simply makes my point about specifying format over camera quality. As for the DVW35, it's the same camera head as the second generation of the DSR500, and the 30 was the same as the first run of the 500. They are dockable and you can use a DVCAM or Betacam back with either. I don't know of the 50 is the same as the 570 but it most likely is. Nice thing about the dockables is you can control all 4 audio channels separately. There are also lots of cameras that are not camcorders that were (and still are) used with Betacam decks. Some of them are comparable to the current crop and some aren't. There is very little difference between DVCAM and Betacam in the final result. I've run the 26 pin cable from the DSR500 to the Betacam SP deck and recorded simultaneously onto Betacam SP and DVCAM, and it's almost impossible to tell the difference, except when you get into very saturated low light areas, and the DVCAM looks just a bit better.

[Augusto Manuel]

Thats all debatable !

Originally Posted by Bill Pryor

It's true the UVW100 was an exception, but that simply makes my point about specifying format over camera quality. As for the DVW35, it's the same camera head as the second generation of the DSR500, and the 30 was the same as the first run of the 500. They are dockable and you can use a DVCAM or Betacam back with either. I don't know of the 50 is the same as the 570 but it most likely is. Nice thing about the dockables is you can control all 4 audio channels separately. There are also lots of cameras that are not camcorders that were (and still are) used with Betacam decks. Some of them are comparable to the current crop and some aren't. There is very little difference between DVCAM and Betacam in the final result. I've run the 26 pin cable from the DSR500 to the Betacam SP deck and recorded simultaneously onto Betacam SP and DVCAM, and it's almost impossible to tell the difference, except when you get into very saturated low light areas, and the DVCAM looks just a bit better.

[Tommy James]

I dont think that HDV will ever achieve broadcast quality acceptance unless HDV can demonstrate its ability to compete with high end high definition head on. If HDV can demonstrate its ability to produce ultra high definition images then there will be no question that HDV not only rivals high end cameras but rather blows them out of the water. A start is the ultra high definition display technologies being developed by JVC namely the JVC line of 1500i CRT projection televisions that upconvert and display all signals to the full 2560 x 1440i resolutions. Now if an HDV camera were hooked up to that type of display would there be any doubt whether or not HDV is true high definition ? Now I am not saying that ultra high definition will be needed on a daily basis as it would be a bandwidth hog but the capability to produce these images would look good on the HDV resume. And remember unless HDV can demonstrate Rolls Royce quality the broadcasters will reject it. Also another step in the right direction is the HDV cameras ability using pixel shifting technology to shoot 3 megapixel JPEG images continously. Could this be a step towards ultra high definition if we change the JPEG to MPEG?

[Tommy James]

What I would like to clarify is that the formats 720p, 1080i and 1080p for television broadcasting are probably set in stone and are not likely to change for a long time.

Nevertheless formats like 1440p and 2160p are viable Digital Cinematography formats that are used to replace 35mm film. The next Star Wars will be shot in 2160p.

If an HDV camera is capable of shooting in these ultra high digital cinematography resolutions with the addition of proper recording and encoding equipment can a television station manager dare say that these cameras are not broadcast quality ?

[Graeme Nattress]

How is HDV, a format that is already too compressed, going to cope with recording a massively greater amount of detail. The whole thing about HDV is that it's cheap - that's it's unique selling point among HD cameras!

You can't put more resolution on a small chip because of noise, and because the lenses are not up to it. You can't make the chips bigger as they become more expensive, and bigger lenses cost more too. You can't record more data as it's using miniDV tape, which is not reliable above the data rates they're using now. The only option would be MPEG4 compression rather than MPEG2, but then it's not HDV anymore. This would improve it's quality somewhat, although I don't think realtime MPEG4 encoders are cheap either.

HDV is a transitionary format to bring affordable HD to the DV masses, and to ease them over to shooting at a higher definition. It seems it's been totally successfull in meeting that goal. Why try to make it something it isn't, and really, cannot be?

Graeme

[Kevin Dooley]

Originally Posted by Tommy James

The next Star Wars will be shot in 2160p.

Ummm.... there won't be another one. Lucas has the licensing locked up for eternity basically and he's said that he's done. Unless one of his kids spends all of his inheritance on space flights and blow (and trust me, that's a lot of coke) after Lucas passes on, there's no chance in hell that any one will ever be allowed to shoot another one...

[Tommy James]

1. First of all the mini dv tapes of the HDV camera are not the only way to record images. The JVC HD100 cameras feature optional hard drives and they feature a 720p60 uncompressed analog output that can be captured and recorded at much higher bit rates using HD-SDI

2. Not only can you triple the resolution of a small one megapixel chip JVC is already doing that with the JVC HD100 camera. By using pixel shifting technology JVC cameras have the ability to record continously shot 3 megapixel equivalent JPEG images. And it is probably technologically feasable to convert JPEG images to quadruple high definition video. Frame rates would have to be increased but this can be done using interpolation technology. Pixel shifting technology is not only an acceptable way to increase resolution but in theory this technology can be used to resolve infinite detail with the resolving power of the glass being the only practical limitation.

[Graeme Nattress]

If you're not recording to DV tape, then it's not HDV, but something else. You might be recording MPEG2 transport streams to your hard drive, but that's not HDV, it's something else, no matter how similar it may seem. As you point out, the 720p60 mode is available on component analogue output. This is only uncompressed in the sense that it's not been through a digital codec. It is not uncompressed in the sense that either a true HD-SDI output (avoiding codec) is uncompressed, or even a dual link 4:4:4 HD SDI output is uncompressed.

"Not only can you triple the resolution of a small one megapixel chip JVC is already doing that with the JVC HD100 camera." That sentence doesn't read properly. Can you explain what you mean more clearly??

"By using pixel shifting technology JVC cameras have the ability to record continously shot 3 megapixel equivalent JPEG images". As we previously discussed, PixelShift is in practical terms, going to give you a 30% resolution advantage over not using it. As you pointed out, you could move the CCD very rapidly and get multiple shifts, but although this may be a fun lab experiment, it's not practical, will incur a massive shutter speed increase as each shift is made, lower the low light capabilities, and basically, just not work in a real practical camera. It might be a fun experiment to integrate the images for a stills camera to increase the resolution, but that's not going to work for moving images. In any case, you'll hit lens limits way before you get any resolution advantage from the pixel shift. The limits of the glass are a very real and practical limitation.

Frame rate interpolation is one of the trickiest image processing tasks you can perform. It takes massive processing power, and with some hand-tweaking can work ok for a cool slomo effect, but is not practical for an entire movie, and it would look "odd" in the extreme, as even the best algorithms currently produce visible artifacts.

Uprezzing technology is, however, practical. It can do a very good job from a high quality SD image, making it look very HD indeed. However, if you have a good uprezzer, you could turn a HD image into a super-HD image, or a super-HD image into a mega-HD image. It does not level the playing field but bumps all formats up to the same degree.

Graeme

[Augusto Manuel]

I am in complete agreement with your explanation. Excellent clarification on the matter. Dont take me wrong but the previous poster needs to take a class on video engineering.

Originally Posted by Graeme Nattress

If you're not recording to DV tape, then it's not HDV, but something else. You might be recording MPEG2 transport streams to your hard drive, but that's not HDV, it's something else, no matter how similar it may seem. As you point out, the 720p60 mode is available on component analogue output. This is only uncompressed in the sense that it's not been through a digital codec. It is not uncompressed in the sense that either a true HD-SDI output (avoiding codec) is uncompressed, or even a dual link 4:4:4 HD SDI output is uncompressed.

"Not only can you triple the resolution of a small one megapixel chip JVC is already doing that with the JVC HD100 camera." That sentence doesn't read properly. Can you explain what you mean more clearly??

"By using pixel shifting technology JVC cameras have the ability to record continously shot 3 megapixel equivalent JPEG images". As we previously discussed, PixelShift is in practical terms, going to give you a 30% resolution advantage over not using it. As you pointed out, you could move the CCD very rapidly and get multiple shifts, but although this may be a fun lab experiment, it's not practical, will incur a massive shutter speed increase as each shift is made, lower the low light capabilities, and basically, just not work in a real practical camera. It might be a fun experiment to integrate the images for a stills camera to increase the resolution, but that's not going to work for moving images. In any case, you'll hit lens limits way before you get any resolution advantage from the pixel shift. The limits of the glass are a very real and practical limitation.

Frame rate interpolation is one of the trickiest image processing tasks you can perform. It takes massive processing power, and with some hand-tweaking can work ok for a cool slomo effect, but is not practical for an entire movie, and it would look "odd" in the extreme, as even the best algorithms currently produce visible artifacts.

Uprezzing technology is, however, practical. It can do a very good job from a high quality SD image, making it look very HD indeed. However, if you have a good uprezzer, you could turn a HD image into a super-HD image, or a super-HD image into a mega-HD image. It does not level the playing field but bumps all formats up to the same degree.

Graeme

[Steve Gibby]

"Broadcast quality" - the elusive butterfly, the great debate, etc. A left-brained pragmatist will quote copious amounts of tech specs, while a right-brained artist will lean on dreamy filmic visions of aesthetic wonder. A mid-brained fence sitter will see the dual necessity of tech and aesthetics. The FCC and NTSC specs will quantify minimum IRE levels for "legal" broadcast standards. ATSC specs somehow reduce HD and DVCProHD originated HDTV broadcasts to highly-compressed 19.2mbps (or lower) MPEG2. Low-res satellite phones suddenly become "broadcast quality" when the economics of TV news dictate that public hunger for images trumps resolution and composition. And on and on and on ad naseum...

Does advanced technology negate the bedrock need for talent and experience in order to create compelling motion media images, whether they be in film or video? Would you rather watch content produced in a lower resolution format that was created by superb artists, or high resolution schlock created by wealthy techies that can afford the highest level of technology, but are creatively bankrupt? Which is more important to creating a masterpiece painting - the artist or the paintbrush?

No matter how creatively and aesthetically inclined we are, if we ignore the economics of the motion media marketplace we are dooming ourselves to decline and fiscal failure. Even media giants must maximize technology and minimize overhead to compete effectively. It's even more critical for small boutique facilities to do the same.

How have the dynamics of this economic model affected the television industry? Enormously. Before the mid-1990's there was very little latitude available to producers who self-syndicated their television programs. The industry was dominated by a few "gatekeeper" programming executives who cracked the whip on all the proletariat masses. In the early '80's they dictated 3/4" and then 3/4" SP. Then in the mid-80's Betacam revolutionized the industry, followed by Beta SP. In the mid-90's, along came 3-chip DV, DVCPro, and DVCAM - right at the time when cable and satellite was geometrically expanding the need for cost-effective programming to fill the many new channels. Station and network executives who had traditionally been the "gatekeepers", now had a dilemma: they wanted to maximize their legacy investments in Beta SP, but they also wanted to economically compete with the new hordes of producers that could now afford inexpensive DV25 equipment and emerging non-linear editing systems. So what was the "gatekeepers" declared mantra? "Broadcast quality". It mattered little that the DV25 formats met the technical specs for FCC mandated NTSC broadcasting - and that what the DV25 formats lacked could be easily corrected in the new NLE systems (IRE levels, etc.). Even though network "Producers Guidelines" routinely specified that analog and digital formats above DV25 were mandatory for acquisition, independent producers were increasingly having the audacity to use 3-chip DV25 cameras in the field, editing their programs, and simply mastering them seamlessly to Beta SP and DigiBeta for delivery to the networks. How dare they attempt to maximize technology and minimize overhead!! Television was democratizing at the speed of light and the "good ole boys" that traditionally kept the washroom keys were bewildered and frightened. The self-proclaimed elitists and purists were shuddering in disgust...

Fast forward to 2005. Satellite and digital cable television now exhibit an insatiable demand for content. Acquisition and post technology is extremely affrodable. The new 3-chip HDV and DVCProHD cameras obviously meet or exceed the bitstream requirement of HDTV broadcast standards - and yet a new generation of supposed "gatekeepers" raise the banner of "broadcast quality"! Would you rather watch a high-definition program created by master artists working with a Z1, HD100, or HVC200 - or a program created by inexperienced but wealthy also-rans that simply had a big enough budget to rent F900 for their production? Does the old phrase "Garbage in - garbage out!" only apply to lines of resolution? Can gifted storytellers, DPs, and directors create a masterpiece of motion media in a lower bitrate hi-def format? Better yet, do the general public viewers of HDTV programs quible over formats, bitrates, MTF curves, editing software nuances, etc.? No - they simply want interesting programs that look good on their HDTV sets. All our endless online techno-babble isn't going to change that. The DV25 history of the mid-90's is repeating itself with the low end hi-def cameras and formats of 2005. Pressure on independent producers - and networks - to hold overhead down is even greater now.

Does HDV meet the tech specs to be "broadcast quality" under the ATSC guidelines? Technically speaking, the answer is yes. Do the same elitists and former gatekeepers that fought tooth and nail against the broadcast of DV25 formats in the 90's now look down on the HDV users from their lofty HDCAM and Varicam perches and cry that HDV is not "broadcast quality"? Yeah. Will HDV content be broadcast widely on HDTV stations and networks? Yeah. Will the cycle of Bourgoisee Elitists and Proletariat Scrimpers repeat itself? Yeah. Time after time after time after time...

So what drives this elitist cycle? Ego? Economics? Techno-pride? Investment maximization? Resistance to creative evolution? Media democratization backlash? Greed? Jeolousy? All of the above?

Rather than use the term "broadcast quality", perhaps we should use separate terms of "broadcast tech standards" and "broadcast creative values". If "broadcast quality" is only determined by gatekeepers and techno-egoists, the entire economic vaibility of the television industry is in jeopardy. To fill the geometrically burgeoning programming needs of HDTV channels, high-caliber, but cost-effective, lower bitrate productions are an absolute necessity! Does that invalidate or infringe on the business model of HDCAM and Varicam users? Not necessarily. Will there be a giant, ongoing viewer demand for limited-budget, low bitrate hi-def productions? Yeah. Will networks like Discovery have to re-evaluate their position on acceptance of HDV-originated content? Yeah. What will force that? Economics of the HDTV programming marketplace. Is Discovery Network's position on HDV a throwback to their analysis of the 1-chip JVC HD10U's performance? Probably. Has Discovery really taken the time to analyze the capability of the Z1 footage edited with one of the outstanding NLE systems? Probably not. Should Discovery take the time to analyze the possibilities with the new JVC HD100 for HDTV broadcast? Yes. Will Discovery take the time to test the new HVC200 or say "Well, the bitrate is higher, but it's small, inexpensive, and only has a fixed lens". Who knows.

If a format meets the minimum tech requirements to be broadcast in a designated country or region, isn't it "broadcast quality"? Can we all respect the creative aesthetics of each others work without getting hung up on endless circumlocution about formats, bitrates, and chroma sampling? Aren't we all in the same worldwide creative family? Respect is contagious. Send some out and you'll be surprised how much comes full circle back to you. After minimum broadcast tech standards have been met, "Broadcast quality" should imply the broadcasting of quality productions - meaning good production values, storyline, etc. If "broadcast quality" is only about lines of resolution, our collective myopia is incurable...

[Bill Pryor]

Wow. That's a good one with some good points.
For me, however, it's basically irrelevant. Nobody at any TV station has ever asked what cameras we used or what format the original tapes were on. We just give them the type of dub they ask for.

[Steve Gibby]

Right on Bill. That tells me that you do good work in the field, edit good, and master correctly for who you're delivering to. That will keep any TV station or almost any TV network happy. If your master is conformed to their specs and your production values and content are spot on, everybody is happy. That said, there are some networks I produce for that have 5" thick Producer Guidelines that give you little "wiggle room". Some of them even resort to unannounced spot checks of your source tape to check for compliance with their acquisition format guidelines! These are almost invariably the networks that use packagers to do contract production, where they keep the copyright to the programs. The self-syndication world is much more lenient on acquisition guidelines. I work both ways, but I prefer the creative freedom of independent production and self syndication (time buys, barter, etc.). As channels and networks that use self-syndicating producers profilerate, the economic pressure increases on networks that use packagers. They're all competing for the same sponsor and advertiser dollars. If the self-syndicator user networks get their equal quality programs by giving affordable rates on airing time buys, and independent producers are holding overhead down by maximizing overhead while still delivering quality programs, the pressure will continue to mount on packager-using networks to be more lenient on their acquisition format guidelines. Economic necessities drive the whole engine...

[Steve Gibby]

Originally Posted by John M Burkhart

I've done some programming with Discovery and NatGeo (in Asia) and here are Discovery's Deliverables Requirements regarding formats (from 2004):

Delivery:
SD: Digtal Betacam
HD: HDCAM (1080i60)

Acquisition:
SD:
Betacam SP
Digital Betacam
DVCpro 50
IMX 50

HD:
HDCAM
HDCAM SR

Formats NOT permitted for general programming aquisition are:

SD: DVCPro25, DVCAM, MiniDV, BetacamSX, Umatic (SP or otherwise), SVHS, VHS, Hi8, Betamax, and 8mm.

HD: HDV, HDVCpro (No Varicams, I'm not sure if this applies to 1080i footage from HDVCpro, but definately to 720p)

From their document: "There are some acceptable uses for small amounts of non-broadcast quality footage. These are "newsworthy events" that were soley captured on amateur video, or in situations which physically demand the use of smaller cameras, e.g. in a cave expedition or high up a mountain."

"Please keep in mind that the [Discovery]Networks are distributed on compressed digital satelite feeds, and non-broadcast footage usually has poor chroma response which produce sub-standard results when transmitted on these services."

As an aside:

If you want to pitch an idea to Discovery, the best way is through e-submission (they really do read them). Go to http://producers.discovery.com

It takes them between 6-8 weeks for them to get back to you.

Hi John. You're in a great location to get good images! Southeast Asia is loaded with good production possibilities.

I have also aired content on Discovery, and nationally on several other U.S cable and broadcast networks. I regularly use almost every format on Discovery's list of "approved" and "unapproved" formats. As you know, some formats and specific cameras lend themselves best to certain types of field acquisition - and some don't. Do the DV25 formats get used extensively on SD programs of Discovery? Yeah. Some of the highest-rated programs on Discovery USA (American Chopper, etc.) are reality-based mobile ENG style programs in confined spaces that simply make the use of full-sized cameras impractical. Just this week, while I watched an American Chopper episode, I counted two times that one of the PD170's the camera crew was using was in the frame. I don't think Discovery is in the dark about one of their highest rated programs using a "not permitted" format for acquisition. In this case the format and the camera is perfectly suited to the "look and feel" of the American Chopper programs.

I understand Discovery's need to have airing masters delivered on Digital Betacam for SD and HDCAM (1080i60) for HD. Those are good, solid delivery formats. It's their list of "approved" acquisition formats that is ludicrous. They provide no latitude for some cameras within a given format that have great chipsets and optics, and conversely allow other questionable cameras just because they are within the "approved" format list. At a glance through the approved list of formats for acquisition, the first obvious common trait for the approved SD formats is 4:2:2 color space. 4:2:2 is obviously superior to 4:1:1 if a lot of compositing and motion graphic work will be done on the programs. But how many of the reality-based shows that dominate Discovery's programming have a lot of compositing and motion graphic work done to them? Not many. Their list of "not approved" formats is largely 4:1:1 formats, including all of the DV25 formats. And yet the cameras of those formats are generally the most effective, and cost-effective, tools to use for many of the Discovery programs! Tens of thousands of DV25 format originated programs have been broadcast worldwide by TV stations and networks, but they're not good enough "officially" for Discovery Networks - and yet, as noted above, DV25 formats are used widely on certain Discovery SD broadcasts.

Even though I use and respect all of the formats on Discovery's approved for acquisition list, I find it very questionable that Beta SP is included and the DV25 formats are not. I've been seamlessly intercutting well-shot DV25 format footage with Beta SP footage since the mid-1990's! Scores of producers have. Even such respected video engineers as Adam Wilt (www.adamwilt.com) acknowledge that well-shot DV25 formats are on a par with Beta SP. Discovery's position on this is unreasonable, costly to producers, and as illustrated above - it is often ignored.

Discovery's approved HD acquisition format list is also out of step with economic and field production realities. Discovery HD broadcasts in 1080i60. To exclude DVCProHD and HDV acquired programs, that are conformed to Discovery's Producer's Guidelines spec sheet for mastering, is baffling to me. If top-level shooters and editors are using DVCProHD and HDV for mobile, cost-effective acquisition, and creating excellent programs that are mastered in 1080i60 HDCAM for delivery, what is the problem? The entire program is then going to be compressed down to at least 19.2 mbps MPEG2 for HDTV delivery. In the hands of seasoned professionals, the Z1, HD100, and HVC200 will be very effective acquisition tools. Add in expert editing, NLE-based color correction, etc. and the end product is truly worthy of airing on Discovery HD.

I feel that Discovery and other networks that exclude DV25 formats for SD acquisition, and DVCProHD and HDV for HD acquisition are out of touch with the technology, economics, and dynamics of the rapidly evolving television marketplace. Fox, ABC, and ESPN broadcast in 720p60. Any DVCProHD or HDV camera that can acquire or output a 720p60 image should be considered "approved" by them. Discovery, HDNet, NBC, CBS, and others broadcast in 1080i60. Any small DVCProHD or HDV camera that can acquire or output a 1080i60 image should be "approved" by them. With a proliferation of HD channels and networks that WILL accept DVCProHD and HDV acquired programs on the horizon, the networks like Discovery that exclude those cost-effective programs will begin to lose serious money to the networks and channels that will accept the formats for acquisition. When Discovery has lost enough money they will somehow include DVCProHD and HDV on their "approved for acquisition" list. Economic viabilities drive our entire business. That said, there will always be a solid niche for high-budget HD format productions. Overlooking the equally solid niche for low-budget HD format programming is a very perilous course - even for a company as large as Discovery...

[John M Burkhart]

Hi Steve,

Well I for one completely agree with you. I also think that Discovery's a bit out of touch with modern production realities.

The O.P. asked about if broadcasters like Discovery had approved HDV as a format for broadcast yet. My answer to him was yes they had looked at it, and no they did not approve it, and to give a list of formats they did approve.

Purely an informative post on my part , and should not be construed as an endorsement by me of Discovery's policies and procedures, nor those of any third party, now or in the future.

(Sorry, going through contracts all morning, and lawyer speak starts to pervade everything you do.) :)

A few caveats though (gah! can't stop it!), these are the the specs from 2004, when the only HDV camera available was the JVC GR-HD1U (the jvc one chip camera). So maybe once these new cameras show their quality, Discovery will re-consider.

I also think the relatively high guidelines are there to make sure their budget ends up on the screen and not in the producer's pocket. They don't want to give you a Digi-Beta budget, and then have you turn around and shoot it on DVCAM, and pocket the difference.

The decision to approve BetaSP I think is there as more a concession to reality not to quality. In so much as it's dominance as the ONE TRUE BROADCAST FORMAT for so many years, means that there is a huge amount of stock footage out there (and probably hundreds of thousands of hours in discovery's own extensive archives) and to dismiss all of that out of hand is cutting your nose off to spite your face.

They really really want you to shoot DigiBeta for SD (which I did) or HDCAM for HD. And I mean fair enough, DigiBeta gives you great pictures no doubt, and renting them is not completely out of the question.

They are very hung up on bandwidth (data-rate) as a measure of quality. HDCAM, DigiBeta etc. One of the engineers was talking about a show they did for DiscoveryHD about world capitals that was shot on HDCAMSR, and how beautiful it looked etc. I was thinking in the back of my mind about the feasibility of sneaking through the jungle with a full crew, generator, hard disk array etc.

So there does obviously have to be a middle ground, and like most things these days the middle is coming down in price.

Eventually the engineers will lose this fight for supreme quality only, simply due to the economics of it.

Tell the Discovery producer that your show you're co-producing with them (they hardly just buy shows outright these days) will cost this much if I shoot it on HDCAM, and this much if I shoot it on HDV, and see if they don't come around and have a word or two with the engineers.