Wonderful example

Jeff Donald · December 12th, 2002, 09:21 AM

The days of editing and color correcting for just broadcast are over. I have clients that have me correct (color, gamma, etc) a piece different ways for different uses. If the project isn't going to broadcast we stretch the black level, gamma, white clip etc to fit the medium. The 7.5 IRE black level is only for broadcast. Why lose part of your range if the piece will never be broadcast? I correct one way for web use and another for broadcast and even another for DVD. It's like giving a painter only one size brush.

I don't do much work for transfer to film. But I would correct DV for film differently than DV for web.

Jeff

Jay Gladwell · December 12th, 2002, 10:48 AM

Jeff, you bring up and excellent point. I hadn't thought of it, but you're right. It depends on the final form in which the piece is being exhibited.

Robert Knecht Schmidt · December 12th, 2002, 12:39 PM

Rob Lohman had me right, and I'm sorry if I wasn't clear. To restate and rephrase, For digital video, viewing the video signal on a computer monitor is sufficient to make the assessment that the material originated on DV (not film) simply by observing contrast flatness.

Of course, I am also of the school that a properly calibrated high-resolution production monitor (NTSC or PAL, depending on where you hail from) is an essential tool for shooting and editing. On my editing desk I have twin 17" Viewsonic computer monitors and twin 14" SONY NTSC video monitors for my work.

When it comes to digital video, however, what you see on your computer screen (given that your system is gamma-calibrated and the material you're working with has a dynamic range of 24 bits per pixel as DV does) is the truest representation of the video signal--truer that the DA converted signal you'll see on NTSC output, which, in systems math terms, just adds one (or several) more transform matrix (or matrices) to the signal pathway chain.

It should be pointed out that most video production houses, those facilities riddled with big expensive NTSC monitors--TV stations comprise the largest segment of this class of facilities--still work in analog video.

Post facilities that work in high definition, however--e.g., CG work for film or 24P video--tend not have lots of NTSC or even HD TV monitors sitting around. These folks work off computer screens.

If it's in the digital domain, and the material is of standard dynamic range (e.g., 24 bits per pixel), your computer monitor is plenty sufficient to observe the whole true signal.

Jeff Donald · December 12th, 2002, 04:14 PM

If your work is intended for broadcast, an NTSC Production monitor is used (along with a waveform monitor and vectorscope) for quality checks and final approval. Computer RGB differs from DV NTSC in color space and the monitors require different phosphors. Different phosphors reproduce different colors. I don't know of any post house (digital or analog) that uses computer monitors for final approval of work intended for broadcast.

Work destined for film output may be a different matter. Film recorders and their associated support equipment are not my area of expertise.

Jeff

Henrik Bengtsson · December 18th, 2002, 03:35 PM

Robert. My point was more in the fact that you are watching a compressed (lossy compression at that) stream than what monitor you were looking at (sorry if this didn't come through clearly). Compressing video involves a lot of messing about with the pixels, and colourcorrection and foremost, judging the colours compared to a video or film media usually goes right out the window.

I'm definately not good enough to judge if the image is flatter on a highly compressed quicktime stream. Hell, im not sure im good enough to judge it on a "proper" media =)

/Henrik

Jeff Donald · December 18th, 2002, 04:34 PM

The rotoscoping questions and answers got split out to here http://www.dvinfo.net/conf/showthrea...&threadid=5564

Jeff

Adrian van der Park · December 18th, 2002, 05:40 PM

Not entirely... when creating computer gen'd images for output to film or HD, 24bit per pixel (8bit flat) colour space is NOT correct for matching to a film output. What you need is to work in a higher colour space (i.e. 10bit log, 12bit flat, 16bit flat or HDR), and to simulate what it would look like when imaged to film, you must use a LUT (or Look-Up-Table) to alter your 8bit flat display to mimick that of a 12bit flat or 10bit log of tiff or cineon out to be laser imaged to film or dumped to HD.

/Adrian

<<<-- Originally posted by Robert Knecht Schmidt : Rob Lohman
Post facilities that work in high definition, however--e.g., CG work for film or 24P video--tend not have lots of NTSC or even HD TV monitors sitting around. These folks work off computer screens.

If it's in the digital domain, and the material is of standard dynamic range (e.g., 24 bits per pixel), your computer monitor is plenty sufficient to observe the whole true signal. -->>>

Robert Knecht Schmidt · December 18th, 2002, 07:00 PM

I don't disagree, Adrian, and I stand by my statement that a computer monitor is sufficient to evaluate the dynamic range of DV footage.

Victor Muh · January 16th, 2003, 08:13 PM

I dunno about MB. It wouldn't put those tacky, ramdom fake scratches on the image like you get with the stock Quicktime effect.

<<<-- Originally posted by Jeff Donald : In the credits, Final Cut Pro is listed. My best guess is it's Magic Bullet http://www.theorphanage.com/ It may have been done with another AE plugin, but it looks like MB to me.

Jeff -->>>

Davi Dortas · January 27th, 2003, 09:07 AM

Well just finished watching the movie. Great piece, although the one thing that really bothered me about the movie, technically was how the filmmaker gave the video a "film look". Whatever process he used, it gave the film a too blurry look, like as if he shot it with a 360 degree shutter. It looks similiar to 1/30 sec. shutter which annoys the hell out of me.

It definitely does not look like film. It reminds me of some cheapie DV commercials that air here in Vancouver, most noticeably those annoying Shaw Cable commercials. They employ the same smeary look and overblown highlights that always look bad. Im just commenting on the technical aspect of the piece. It was well lit and all, but the annoying film look really bugged the hell out of me. It would have looked alot crisper had the filmmakers used a straight de-interlace, without mucking around with the frame rate or whatever.

It was a good piece overall so no negative comments here. Anyone else notice the smeary look of the piece?

December 12th, 2002, 12:39 PM	#33
Robert Knecht Schmidt Wrangler Join Date: Feb 2002 Location: Cleveland, Ohio Posts: 1,933	Rob Lohman had me right, and I'm sorry if I wasn't clear. To restate and rephrase, For digital video, viewing the video signal on a computer monitor is sufficient to make the assessment that the material originated on DV (not film) simply by observing contrast flatness. Of course, I am also of the school that a properly calibrated high-resolution production monitor (NTSC or PAL, depending on where you hail from) is an essential tool for shooting and editing. On my editing desk I have twin 17" Viewsonic computer monitors and twin 14" SONY NTSC video monitors for my work. When it comes to digital video, however, what you see on your computer screen (given that your system is gamma-calibrated and the material you're working with has a dynamic range of 24 bits per pixel as DV does) is the truest representation of the video signal--truer that the DA converted signal you'll see on NTSC output, which, in systems math terms, just adds one (or several) more transform matrix (or matrices) to the signal pathway chain. It should be pointed out that most video production houses, those facilities riddled with big expensive NTSC monitors--TV stations comprise the largest segment of this class of facilities--still work in analog video. Post facilities that work in high definition, however--e.g., CG work for film or 24P video--tend not have lots of NTSC or even HD TV monitors sitting around. These folks work off computer screens. If it's in the digital domain, and the material is of standard dynamic range (e.g., 24 bits per pixel), your computer monitor is plenty sufficient to observe the whole true signal. __________________ All the best, Robert K S Search DVinfo.net for quick answers \| The best in the business: DVinfo.net sponsors

December 18th, 2002, 03:35 PM	#35
Henrik Bengtsson Regular Crew Join Date: Aug 2002 Location: Borås, Sweden Posts: 167	Robert. My point was more in the fact that you are watching a compressed (lossy compression at that) stream than what monitor you were looking at (sorry if this didn't come through clearly). Compressing video involves a lot of messing about with the pixels, and colourcorrection and foremost, judging the colours compared to a video or film media usually goes right out the window. I'm definately not good enough to judge if the image is flatter on a highly compressed quicktime stream. Hell, im not sure im good enough to judge it on a "proper" media =) /Henrik __________________ -------------------------------------------------------------------------------- Henrik "HuBBa" Bengtsson, Imaginara Fotographia,http://www.imaginara.se

December 18th, 2002, 05:40 PM	#37
Adrian van der Park Regular Crew Join Date: Sep 2002 Location: London, England Posts: 87	Not entirely... when creating computer gen'd images for output to film or HD, 24bit per pixel (8bit flat) colour space is NOT correct for matching to a film output. What you need is to work in a higher colour space (i.e. 10bit log, 12bit flat, 16bit flat or HDR), and to simulate what it would look like when imaged to film, you must use a LUT (or Look-Up-Table) to alter your 8bit flat display to mimick that of a 12bit flat or 10bit log of tiff or cineon out to be laser imaged to film or dumped to HD. /Adrian <<<-- Originally posted by Robert Knecht Schmidt : Rob Lohman Post facilities that work in high definition, however--e.g., CG work for film or 24P video--tend not have lots of NTSC or even HD TV monitors sitting around. These folks work off computer screens. If it's in the digital domain, and the material is of standard dynamic range (e.g., 24 bits per pixel), your computer monitor is plenty sufficient to observe the whole true signal. -->>> __________________ Adrian van der Park VFX modeler London, England

December 18th, 2002, 07:00 PM	#38
Robert Knecht Schmidt Wrangler Join Date: Feb 2002 Location: Cleveland, Ohio Posts: 1,933	I don't disagree, Adrian, and I stand by my statement that a computer monitor is sufficient to evaluate the dynamic range of DV footage. __________________ All the best, Robert K S Search DVinfo.net for quick answers \| The best in the business: DVinfo.net sponsors

January 16th, 2003, 08:13 PM	#39
Victor Muh Regular Crew Join Date: Aug 2002 Location: Paris, France Posts: 126	I dunno about MB. It wouldn't put those tacky, ramdom fake scratches on the image like you get with the stock Quicktime effect. <<<-- Originally posted by Jeff Donald : In the credits, Final Cut Pro is listed. My best guess is it's Magic Bullet http://www.theorphanage.com/ It may have been done with another AE plugin, but it looks like MB to me. Jeff -->>> __________________ Strive not to be a success, but rather to be of value. - Albert Einstein http://advaloreminternational.com

December 12th, 2002, 09:21 AM	#31
Jeff Donald Warden Join Date: Mar 2002 Location: Clearwater, FL Posts: 8,287	The days of editing and color correcting for just broadcast are over. I have clients that have me correct (color, gamma, etc) a piece different ways for different uses. If the project isn't going to broadcast we stretch the black level, gamma, white clip etc to fit the medium. The 7.5 IRE black level is only for broadcast. Why lose part of your range if the piece will never be broadcast? I correct one way for web use and another for broadcast and even another for DVD. It's like giving a painter only one size brush. I don't do much work for transfer to film. But I would correct DV for film differently than DV for web. Jeff

December 12th, 2002, 10:48 AM	#32
Jay Gladwell Join Date: Jul 2002 Posts: 1,244	Jeff, you bring up and excellent point. I hadn't thought of it, but you're right. It depends on the final form in which the piece is being exhibited.

December 12th, 2002, 04:14 PM	#34
Jeff Donald Warden Join Date: Mar 2002 Location: Clearwater, FL Posts: 8,287	If your work is intended for broadcast, an NTSC Production monitor is used (along with a waveform monitor and vectorscope) for quality checks and final approval. Computer RGB differs from DV NTSC in color space and the monitors require different phosphors. Different phosphors reproduce different colors. I don't know of any post house (digital or analog) that uses computer monitors for final approval of work intended for broadcast. Work destined for film output may be a different matter. Film recorders and their associated support equipment are not my area of expertise. Jeff

December 18th, 2002, 04:34 PM	#36
Jeff Donald Warden Join Date: Mar 2002 Location: Clearwater, FL Posts: 8,287	The rotoscoping questions and answers got split out to here http://www.dvinfo.net/conf/showthrea...&threadid=5564 Jeff

January 27th, 2003, 09:07 AM	#40
Davi Dortas Membership Suspended Join Date: Oct 2002 Location: Brazil Posts: 78	Well just finished watching the movie. Great piece, although the one thing that really bothered me about the movie, technically was how the filmmaker gave the video a "film look". Whatever process he used, it gave the film a too blurry look, like as if he shot it with a 360 degree shutter. It looks similiar to 1/30 sec. shutter which annoys the hell out of me. It definitely does not look like film. It reminds me of some cheapie DV commercials that air here in Vancouver, most noticeably those annoying Shaw Cable commercials. They employ the same smeary look and overblown highlights that always look bad. Im just commenting on the technical aspect of the piece. It was well lit and all, but the annoying film look really bugged the hell out of me. It would have looked alot crisper had the filmmakers used a straight de-interlace, without mucking around with the frame rate or whatever. It was a good piece overall so no negative comments here. Anyone else notice the smeary look of the piece?