Semi-final verdict on best option for film-out? - Page 3 at DVinfo.net

Go Back   DV Info Net > JVC ProHD & MPEG2 Camera Systems > JVC GY-HD Series Camera Systems

JVC GY-HD Series Camera Systems
GY-HD 100 & 200 series ProHD HDV camcorders & decks.


Reply
 
Thread Tools Search this Thread
Old October 9th, 2006, 10:25 PM   #31
Regular Crew
 
Join Date: Feb 2006
Posts: 64
what is YPbPr?
Hayk Paul is offline   Reply With Quote
Old October 9th, 2006, 10:28 PM   #32
Regular Crew
 
Join Date: Aug 2005
Location: New York
Posts: 123
Quote:
Originally Posted by Keith Gruchala
What I can definitely confirm is that when you switch the JVC into "filmout" and watch a waveform, you will see both the top and bottom of the wave expand to cover a 0.0% to 110% scaled signal. NTSC broadcast is 7.5% to 100%. By stretching this range and expanding the blackstretch, you will definitely get a broader range in which to work final color correction. The danger is the m-peg splotchiness in the low end (for the JVC) as you will need to raise up the midpoint of the gamma and add more black to the blacks to redefine contrast in the shadows.
Hi Keith,
A lot of good comments in your post. As you rightfully note, itís the high compression of this format that undermines the cameraís filmout gamma setting. Although I have not yet tested this setting with uncompressed output (and I should), I have tested it in HDV in both filmout and transfer to uncompressed HD and I can confirm that the results for both are unusable.
Quote:
Originally Posted by Keith Gruchala
If you've ever seen the raw output from the Panavision Genesis, it looks exactly the same- in fact it's locked into the camera's programming. The beauty of the Panasonic monitors is that you can do this at the flick of a switch so everyone can be happy viewing a mid-tone gamma corrected image, but you can sneak over and look at the raw output to check white and black levels and actual perfect exposure. And by the way, this is how you get the most out of your camera for ANY application of the footage, including finishing back to NTSC broadcast, because you are basically dealing with more information spread out over a larger contrast curve- it's just that not a lot of Dp's or producers understand exactly how this works.
I also find the Panasonic monitors very useful. The problem is that they will only tell you the truth if you are recording and uncompressed signal like the monitors are seeing. When you record that filmout gamma to the HDV format the reality is very different. Unlike raw images on the Genesis which have the underlying data to build a pleasing image once curves are applied, images with this type of gamma recorded in HDV do not. All you end up with is noise and lots of it.
__________________
Andy Young
Director/DP www.ArchipelagoFilms.com
VP, Special Projects www.DuArt.com
Andrew Young is offline   Reply With Quote
Old October 12th, 2006, 03:29 PM   #33
Regular Crew
 
Join Date: May 2006
Location: Los Angeles
Posts: 52
film out noise

Quote:
Originally Posted by Andrew Young
Hi Keith,
A lot of good comments in your post. As you rightfully note, itís the high compression of this format that undermines the cameraís filmout gamma setting. Although I have not yet tested this setting with uncompressed output (and I should), I have tested it in HDV in both filmout and transfer to uncompressed HD and I can confirm that the results for both are unusable.

I also find the Panasonic monitors very useful. The problem is that they will only tell you the truth if you are recording and uncompressed signal like the monitors are seeing. When you record that filmout gamma to the HDV format the reality is very different. Unlike raw images on the Genesis which have the underlying data to build a pleasing image once curves are applied, images with this type of gamma recorded in HDV do not. All you end up with is noise and lots of it.
I'd love to see the results of that test- it would be interesting to find a possible solution- I wonder if JVC has thought about adding the noise reduction features that the Canon XL H1 has- it has two differednt types of NR and they are adjustable in steps- could be a great feature to improve the film out in their new product line.....
Keith Gruchala is offline   Reply With Quote
Old October 12th, 2006, 09:20 PM   #34
Regular Crew
 
Join Date: Aug 2005
Location: New York
Posts: 123
Quote:
Originally Posted by Keith Gruchala
I'd love to see the results of that test- it would be interesting to find a possible solution- I wonder if JVC has thought about adding the noise reduction features that the Canon XL H1 has- it has two differednt types of NR and they are adjustable in steps- could be a great feature to improve the film out in their new product line.....
Hi Keith,

Noise reduction is not going to do it, I'm afraid. The filmout gamma setting (which is really no gamma) leaves a distribution of picture information that is heavily skewed towards the shadows. Once you compress such an image, too much of the information is lost in the mids and lows for it to be corrected back into a pleasing image. When you are working with such a limited bandwidth as HDV, it makes more sense to apply a curve that allows you to maintain the most important picture information through the compression process. Unfortunately, high compression is not conducive to wide latitude.
__________________
Andy Young
Director/DP www.ArchipelagoFilms.com
VP, Special Projects www.DuArt.com
Andrew Young is offline   Reply With Quote
Old October 23rd, 2006, 04:17 PM   #35
Inner Circle
 
Join Date: Dec 2005
Location: New York City
Posts: 2,615
There are two "Cinegamma" setting in the HD100 menu. One is listed on it's own and the other is sub-menued under "Gamma". Would it be a good assumption that they should not be used for a film-out?

What setting is ideal?
__________________
William Hohauser - New York City
Edit/Camera/DCP production/Animation
William Hohauser is offline   Reply With Quote
Old October 23rd, 2006, 06:46 PM   #36
Trustee
 
Join Date: Mar 2004
Location: Milwaukee, WI
Posts: 1,719
I have a question about a film out test.

Could somebody, in theory of course, take a still image and print it out or better yet get it printed to a 35mm slide and then use a slide projector to see how check what modes worked well for you for a film out test?

I would think it wouldn't be exactly the same as a true 35mm film output but by using a slide it should work very well. This would allow you to pick a few good images from some camera tests to see what setups would look good when projected on a screen and it wouldn't cost very much.

Also how would a test work out by printing the images to a 35mm negative and getting those developed as photos? You could also just print to a certain size on photo paper to the test very cheap.

Anyways just a goofy thought I had when I was talking to a photographer friend of mine who recently made some slides from digital photos. I thought the same could be done with still images from HDV and the results should be pretty good. I mean if a slide on a 10' screen is going to look like garbage I would think a high quality 35mm print would look even worse.
Thomas Smet is offline   Reply With Quote
Old October 23rd, 2006, 09:04 PM   #37
Regular Crew
 
Join Date: Aug 2005
Location: New York
Posts: 123
Quote:
Originally Posted by Thomas Smet
I have a question about a film out test.

Could somebody, in theory of course, take a still image and print it out or better yet get it printed to a 35mm slide and then use a slide projector to see how check what modes worked well for you for a film out test?

I would think it wouldn't be exactly the same as a true 35mm film output but by using a slide it should work very well. This would allow you to pick a few good images from some camera tests to see what setups would look good when projected on a screen and it wouldn't cost very much.

Also how would a test work out by printing the images to a 35mm negative and getting those developed as photos? You could also just print to a certain size on photo paper to the test very cheap.

Anyways just a goofy thought I had when I was talking to a photographer friend of mine who recently made some slides from digital photos. I thought the same could be done with still images from HDV and the results should be pretty good. I mean if a slide on a 10' screen is going to look like garbage I would think a high quality 35mm print would look even worse.
Hi Thomas,

In order to learn anything valuable about your settings in a filmout test you need to know not just whether or not it looks like garbage, but more important, what kind of garbage it looks like.

Sure, there are technical issues that can be measured within the digital and photographic realms, but to know if you are actually going to like your image there is no substitute for taking some footage through the exact same process it would go though for your final filmout. Slide film and printed photographs both have very different characteristics from the negative and positive stocks used for filmouts, so your results may have no more baring on reality than simply looking at your footage on a poorly calibrated monitor. Believe it or not, motion also has a huge impact on how things look. It you're simply trying to see how your footage holds up big, find a video projector to hook it up to (we do it for free). But if your trying to judge settings, a filmout test is the only reliable way. You'd be surprised how many setups you can squeeze into a minute.

If you're really sold on the economy of still frames that's fine, but do them on an ArriLaser using the same filmout stocks and workflow. Then you can cut them, mount them and project them to your heart's content. Just make sure the footlamberts and color temperature of your slide projector are within SMPTE spec.

Another no cost way to get a feel for the look is to screen filmout tests from different cameras. We screen them every week, if you're ever in New York.
__________________
Andy Young
Director/DP www.ArchipelagoFilms.com
VP, Special Projects www.DuArt.com
Andrew Young is offline   Reply With Quote
Old October 23rd, 2006, 09:15 PM   #38
Regular Crew
 
Join Date: Aug 2005
Location: New York
Posts: 123
Quote:
Originally Posted by William Hohauser
There are two "Cinegamma" setting in the HD100 menu. One is listed on it's own and the other is sub-menued under "Gamma". Would it be a good assumption that they should not be used for a film-out?

What setting is ideal?
You might be mistaking the one of the cinegammas for CINELIKE: ON/OFF. This is a single setting that changes both the color matrix and the gamma to the CINE setting. Until I see definitive test results to the contrary, I recommend leaving this off, the reason being that the cinelike gamma setting seems to compress things a bit. The cinelike color matrix, on the other hand, doesn't seem to have implications for post on way or the other - use it if you life the look.
__________________
Andy Young
Director/DP www.ArchipelagoFilms.com
VP, Special Projects www.DuArt.com
Andrew Young is offline   Reply With Quote
Old October 23rd, 2006, 09:33 PM   #39
Inner Circle
 
Join Date: Dec 2005
Location: New York City
Posts: 2,615
Yup, I got the name confused. Thanks for the advice.

One day there will be a setting called "Videolike".
__________________
William Hohauser - New York City
Edit/Camera/DCP production/Animation
William Hohauser is offline   Reply With Quote
Old October 23rd, 2006, 11:56 PM   #40
Regular Crew
 
Join Date: May 2006
Location: Canada
Posts: 149
Quote:
Originally Posted by Tim Dashwood
Andrew Young at Duart NY and James Tocher at Digital Film Group Vancouver seem to have the most experience with filmout of HD100 material. They are both members here at dvinfo.net, so hopefully one or both of them will notice this thread and give us their opinion.

I have discussed this topic at great length with Andrew and a little with James. Although there seem to be two schools of thought with regards to latitude, there are some commonalities:

Lens: Use the Fuji 13x3.5 lens if possible, double check back focus on a HD monitor all the time.

Detail level: Don't turn it off, but turn it down to at least -7. Some have suggested MIN for a filmout, but personally I think -7 offers a good balance of natural "filmlike" edges without video enhancement. MIN almost seems to blur the pixels.

Frame rate: Obviously 24P is ideal. 25P is usable (with a 4% speed reduction when projected.) NEVER EVER use 30P unless it is for slightly "overcranked" slo-mo. There is no algorythm to convert 30P to 24fps and maintain true speed.

Gain: always on "0"

H Frequency: Middle
V Frequency: Low (this has something to do with video noise being introduced on HIGH setting.)

Color Matrix: Normal


As for gamma, black stretch and knee settings, opinions differ.
James Tocher helped JVC design the linear FILMOUT curve that is now in the camera, but the limitations of 8-bit HDV seem to hinder the use of a linear curve the way we would normally on a Varicam. The latitude of that curve is also very limiting and a monitor with a gamma box would be required for WYSIWYG monitoring.

I am of the mind that since any filmout is going to go through a D.I. timing stage, you might as well capture as much information as possible. For me this means an 80% knee and possibly Black Stretch as high as 3.
However, I have been told that the trade-off for a filmout might be the introduction of noise. I think I would still use black stretch simply to have the option to crush blacks if needed in D.I.

The other thing to keep in mind is to avoid the cine curve when shooting for film. The Cine curve emulates a film curve when presented on a TV, but will not help you with video response when colour correcting for a filmout. Stick with standard curve and normal colour matrix.

Based on the above info, can the scene files posted on this site be used for a successful film out? ....True Color, panamatch?
Brian Ladue is offline   Reply With Quote
Old October 24th, 2006, 07:30 AM   #41
Regular Crew
 
Join Date: Aug 2005
Location: New York
Posts: 123
Quote:
Originally Posted by Brian Ladue
Based on the above info, can the scene files posted on this site be used for a successful film out? ....True Color, panamatch?
Hi Brian,

Adam Wilt and I are involved in testing these and other settings, however, due to the realties of our schedules it has been a slow process. My apologies for that. I can only say that I have seen reasonable filmouts from the camera default setting from well exposed footage and the question we are trying to answer is how much better can we do. Unfortunately, the HDV compression scheme doesn't leave a lot of wiggle room.
__________________
Andy Young
Director/DP www.ArchipelagoFilms.com
VP, Special Projects www.DuArt.com
Andrew Young is offline   Reply With Quote
Old October 24th, 2006, 07:55 AM   #42
Trustee
 
Join Date: Apr 2005
Location: Chicago
Posts: 1,214
Quote:
Originally Posted by Brian Ladue
Based on the above info, can the scene files posted on this site be used for a successful film out? ....True Color, panamatch?
Yes, the scene files can be used with a sucessful film out. I was most impressed with Tim's "Warm" scene file. That scene file was before Panamatch and TC3 existed. Now we are working on a Panamatch to Vision 2 project. The scene files you mention are viable for a filmout project. You really should perform your own film transfer using your expected workflow and your camera settings to really know what you're getting as a resultant film. As an independent film maker on a budget you can be assured that the film transfer will not be grossly off in quality as to detract from the impact of your story.

S.Noe
Stephen L. Noe is offline   Reply With Quote
Old October 24th, 2006, 08:54 AM   #43
Trustee
 
Join Date: Mar 2004
Location: Milwaukee, WI
Posts: 1,719
Quote:
Originally Posted by Andrew Young
Hi Thomas,

In order to learn anything valuable about your settings in a filmout test you need to know not just whether or not it looks like garbage, but more important, what kind of garbage it looks like.

Sure, there are technical issues that can be measured within the digital and photographic realms, but to know if you are actually going to like your image there is no substitute for taking some footage through the exact same process it would go though for your final filmout. Slide film and printed photographs both have very different characteristics from the negative and positive stocks used for filmouts, so your results may have no more baring on reality than simply looking at your footage on a poorly calibrated monitor. Believe it or not, motion also has a huge impact on how things look. It you're simply trying to see how your footage holds up big, find a video projector to hook it up to (we do it for free). But if your trying to judge settings, a filmout test is the only reliable way. You'd be surprised how many setups you can squeeze into a minute.

If you're really sold on the economy of still frames that's fine, but do them on an ArriLaser using the same filmout stocks and workflow. Then you can cut them, mount them and project them to your heart's content. Just make sure the footlamberts and color temperature of your slide projector are within SMPTE spec.

Another no cost way to get a feel for the look is to screen filmout tests from different cameras. We screen them every week, if you're ever in New York.
Ok thanks for the feedback. I wasn't sure if it would work well or not. I knew in graphic design the same things were pretty much true. We were taught to always print cross section samples on the printer and medium of our final destination to check to see how exactly it would look. I kind of thought the same rules would apply here as well but like I said it was just a goofy thought.
Thomas Smet is offline   Reply With Quote
Old October 24th, 2006, 02:50 PM   #44
Trustee
 
Join Date: Sep 2005
Posts: 1,063
It's true!

Quote:
Originally Posted by William Hohauser
One day there will be a setting called "Videolike".

Don't laugh - that day's already here! In fact, I'm preparing to shoot a scene that requires a TV in the background (a news broadcast) - I figured I'd have to shoot this w/another camera, perhaps a HVX, to get that interlaced look. Any thoughts on how to achieve this with the JVC?

It's kind of wierd, but I think someday regular standard-def VHS cameras will be worth a lot of money, sort of like old key boards are now, for that 'classic' look.

This thread is very helpful to us hopefuls (as in, we hope we'll need a film-out!).

john
evilgeniusentertainment.com
John Vincent is offline   Reply With Quote
Old October 24th, 2006, 08:32 PM   #45
Inner Circle
 
Join Date: Dec 2005
Location: New York City
Posts: 2,615
Quote:
Originally Posted by John Vincent
Don't laugh - that day's already here! In fact, I'm preparing to shoot a scene that requires a TV in the background (a news broadcast) - I figured I'd have to shoot this w/another camera, perhaps a HVX, to get that interlaced look. Any thoughts on how to achieve this with the JVC?

It's kind of wierd, but I think someday regular standard-def VHS cameras will be worth a lot of money, sort of like old key boards are now, for that 'classic' look.

This thread is very helpful to us hopefuls (as in, we hope we'll need a film-out!).

john
evilgeniusentertainment.com
Switch to DV. Looks like video to me!
__________________
William Hohauser - New York City
Edit/Camera/DCP production/Animation
William Hohauser is offline   Reply
Reply

DV Info Net refers all where-to-buy and where-to-rent questions exclusively to these trusted full line dealers and rental houses...

Professional Video
(800) 833-4801
Portland, OR

B&H Photo Video
(866) 521-7381
New York, NY

Z.G.C.
(973) 335-4460
Mountain Lakes, NJ

Abel Cine Tech
(888) 700-4416
N.Y. NY & L.A. CA

Precision Camera
(800) 677-1023
Austin, TX

DV Info Net also encourages you to support local businesses and buy from an authorized dealer in your neighborhood.
  You are here: DV Info Net > JVC ProHD & MPEG2 Camera Systems > JVC GY-HD Series Camera Systems

Thread Tools Search this Thread
Search this Thread:

Advanced Search

 



Google
 

All times are GMT -6. The time now is 02:21 PM.


DV Info Net -- Real Names, Real People, Real Info!
1998-2017 The Digital Video Information Network