Jump to content


Photo

16mm to 10bit SD


  • Please log in to reply
16 replies to this topic

#1 Adam Thompson

Adam Thompson
  • Sustaining Members
  • 161 posts
  • Cinematographer

Posted 13 September 2007 - 08:33 PM

Coworker got some 16mm tele'd to Hard drive, 10bit uncmp. SD and I noticed the blacks or shadows looked like they had a layer of blue noise over them. I doubt our film recorded that so I'm wondering how we could address this with the lab.

Was this maybe an adjustment issue on their machine or could it be some other issue? We saw it on his LCD in the time line where it wasn't really noticable (as usual on LCD) but when it went out to a bigger CRT monitor, you could really see it then. Made the fresh S16mm shots look a little old and unprofessional. Any ideas?

Also curious about experience with quality differences via an S16 to HD transfer and having that sent out as 10bit SD. I haven't had the chance to do this yet but was reading where someone went this route and noticed huge improvements.
  • 0

#2 Alessandro Machi

Alessandro Machi
  • Sustaining Members
  • 3318 posts
  • Other
  • California

Posted 13 September 2007 - 09:26 PM

I'll just put my vote in for good old betacam sp, or digital betacam if you can afford that. Then when you encode for editing, just make sure to use the component out signal and go to a higher quality codec than DV-25.
  • 0

#3 Adam Thompson

Adam Thompson
  • Sustaining Members
  • 161 posts
  • Cinematographer

Posted 13 September 2007 - 10:52 PM

I'll just put my vote in for good old betacam sp, or digital betacam if you can afford that. Then when you encode for editing, just make sure to use the component out signal and go to a higher quality codec than DV-25.


Why would we do that? I'm confused. The point of using the hard drive was to bypass the crazy costs of Digi deck rental and trouble. Working straight from drives seems to be the way things are going now. The file was 10bit uncompressed in quicktime for direct placement into our timeline on Final Cut studio. How would a tape be any better? Also, why would a tape not have the blueish cast over the blacks and the drive would?
  • 0

#4 John Brawley

John Brawley
  • Sustaining Members
  • 834 posts
  • Cinematographer
  • Atlanta Georgia

Posted 14 September 2007 - 01:41 AM

Coworker got some 16mm tele'd to Hard drive, 10bit uncmp. SD and I noticed the blacks or shadows looked like they had a layer of blue noise over them. I doubt our film recorded that so I'm wondering how we could address this with the lab.

Was this maybe an adjustment issue on their machine or could it be some other issue? We saw it on his LCD in the time line where it wasn't really noticable (as usual on LCD) but when it went out to a bigger CRT monitor, you could really see it then. Made the fresh S16mm shots look a little old and unprofessional. Any ideas?

Also curious about experience with quality differences via an S16 to HD transfer and having that sent out as 10bit SD. I haven't had the chance to do this yet but was reading where someone went this route and noticed huge improvements.




Hi Adam.

It's a bit hard to diagnose. Is it noisy in the blacks ? Or there a blue cast over the blacks which also have noise....or is the noise itself blue amoungst black pixels...in the blacks ?

Have you got an example you can upload ?

It sounds right off to me that it might have been telecine'd this way, especially if it is simply blue in the blacks. Did you attend the grade ?

JB
  • 0

#5 Thomas Worth

Thomas Worth
  • Basic Members
  • PipPipPipPip
  • 372 posts
  • Director
  • Los Angeles

Posted 14 September 2007 - 02:06 AM

Coworker got some 16mm tele'd to Hard drive, 10bit uncmp. SD and I noticed the blacks or shadows looked like they had a layer of blue noise over them. I doubt our film recorded that so I'm wondering how we could address this with the lab.

Who performed the transfer?
  • 0

#6 Adam Thompson

Adam Thompson
  • Sustaining Members
  • 161 posts
  • Cinematographer

Posted 14 September 2007 - 03:19 AM

Hi Adam.

It's a bit hard to diagnose. Is it noisy in the blacks ? Or there a blue cast over the blacks which also have noise....or is the noise itself blue amoungst black pixels...in the blacks ?

Have you got an example you can upload ?

It sounds right off to me that it might have been telecine'd this way, especially if it is simply blue in the blacks. Did you attend the grade ?

JB


John,

Maybe I shouldn't have said noise, sorry about that. It's more of a blue cast over/in the blacks. And like I said, it's hard to see on the Mac's cine LCD screen. The CRT mon. and another test to DVD on a well tuned 31" TV showed the same problem.

No, it was sent off so we weren't there. This was also a best light but that shouldn't matter.

Cinelab did it I believe. I'd have to ask his permission to upload a grab as I didn't shoot it. Still makes me mad though, you'd think that would be a very basic thing to catch before taking someone's money.
  • 0

#7 Adam Thompson

Adam Thompson
  • Sustaining Members
  • 161 posts
  • Cinematographer

Posted 14 September 2007 - 04:02 AM

Ok I got a .jpg showing the extreme of the example. The back of the girl and the dark part of the bookcase on the left is where it really shows. Though on my laptop right now I can't really see it. CRT it's very clear. Also on this PC it looks a little underexposed further into the room but on the Mac it looks a stop or two brighter. ? He said there was only the sunlight coming into the room through the window you see, no other lights. There is a little motion blur as the camera was moving right behind her, both going toward the light.

Attached Images

  • 16mmtest.jpg

  • 0

#8 Thomas Worth

Thomas Worth
  • Basic Members
  • PipPipPipPip
  • 372 posts
  • Director
  • Los Angeles

Posted 14 September 2007 - 04:46 AM

I doesn't look that bad to me. However, if you pump up the dark areas, you can see some big blue grains. Is this tungsten balanced stock? Your blue channel is the grainiest:

post_19925_1189760378_1.jpg

If you've got the footage as 10 bit uncompressed data, you should be able to remove the excess blue easily:

post_19925_1189760378_2.jpg
  • 0

#9 Robert Houllahan

Robert Houllahan
  • Sustaining Members
  • 1585 posts
  • Industry Rep
  • Providence R.I.

Posted 14 September 2007 - 09:43 AM

I doesn't look that bad to me. However, if you pump up the dark areas, you can see some big blue grains. Is this tungsten balanced stock? Your blue channel is the grainiest:

post_19925_1189760378_1.jpg

If you've got the footage as 10 bit uncompressed data, you should be able to remove the excess blue easily:

post_19925_1189760378_2.jpg



I remember transfering this the other day, it's tungsten stock under mixed lighting and some of it is underexposed. The girl moves from fluro to tungsten to daylight rapidly. I assumed that because it's a test you would want to see the effects of the changing lighting and I found a setting that seemed to balance all the different lighting settings relatively well. If you go from tungsten light to daylight without a 85 the blues are going to get a bit funky and add a bit of under exposure to that and you will get some noise.

As someone else said with a 10bit disk recording you can easily work in fcp to do additional color correction, also considering our ridiculously low minimum (I think the xfer cost you 82.50 with the disk recording) I don't think you could get as good a transfer that you could manipulate anyplace else for that.

-Rob-
  • 0

#10 Phil Rhodes

Phil Rhodes
  • Sustaining Members
  • 11943 posts
  • Other

Posted 14 September 2007 - 09:56 AM

What did you transfer it on?

Phil
  • 0

#11 Robert Houllahan

Robert Houllahan
  • Sustaining Members
  • 1585 posts
  • Industry Rep
  • Providence R.I.

Posted 14 September 2007 - 11:10 AM

What did you transfer it on?

Phil



We have 2 Cintel MK3 Turbo telecine's both have DAV afterglow and video sections , Metaspeed servos and both have Quattroscan 10bit 444 SDI framestores, the machine this was xfered on has a Digital Deflection and is controlled by a Copernicus Color Corrector (probably not many in the UK) which was a competetorto the DaVinci 888 DUI with 10bit 444 I/O and 16 bit color processing it also has windows, etc. Our second machine has a DaVinci 888 and a standard cintel scan generator. We record SDI to a G5 with a Decklink SD card....

The copernicus room just got a complete tuneup and checkout by Dave Corbitt for a 35mm dailies job we were doing...


-Rob-
  • 0

#12 John Brawley

John Brawley
  • Sustaining Members
  • 834 posts
  • Cinematographer
  • Atlanta Georgia

Posted 14 September 2007 - 07:10 PM

Ok I got a .jpg showing the extreme of the example. The back of the girl and the dark part of the bookcase on the left is where it really shows. Though on my laptop right now I can't really see it. CRT it's very clear. Also on this PC it looks a little underexposed further into the room but on the Mac it looks a stop or two brighter. ? He said there was only the sunlight coming into the room through the window you see, no other lights. There is a little motion blur as the camera was moving right behind her, both going toward the light.


Hi Adam.

I have to say, for an unsupervised transfer in what looks to be mixed lighting, it seems to be OK. In the image you present I can BARELY make out a bias in the blacks. It's only when you lift it as demonstrated you can start to see it. But are you planning to grade it that way ?

I actually think it looks pretty good given the conditions you describe...

jb
  • 0

#13 Adam Thompson

Adam Thompson
  • Sustaining Members
  • 161 posts
  • Cinematographer

Posted 15 September 2007 - 01:42 PM

Hey guys. Just relayed the comments via phone (he's shooting a short for someone that may end up at Cinelab again). Said this particular shot had only the sun coming in that window, no mixed or other lights, though there were mixed lights in other shots of this test. Also it was perfectly exposed as the progress of this shot shows though this still may not illustrate that well, but what would any of that have to do with the blue anyway? Some other shots in the test were purposely underexposed and some purposely had flares, etc.. Rob, were you reffering to these others shots when you said "....it's tungsten stock under mixed lighting and some of it is underexposed."? Just wanted to clear that up.

The stock was 200T Kodak (7217) and this particular shot had an 85 on it, the one or two before it did not. He waited till about 6pm to do this shot as the sun starts coming into the apartment through some buildings and a tree or two outside. It fills the room to about a 2.8 via bounce from the bright walls.

This was a best light transfer. If it would have been scene to scene corrected, would this issue still come up do you think?

Cinelab did a transfer for me on a short a couple years ago and it was fine so I'm not trying to make anyone look bad, we just want to clear up how to avoid this issue in the future. And correcting it for the blue at this point would also change the color of the girl's denem skirt, and other things, wouldn't it? I'm not into CC work that much but was wondering how, in FCP, would be the best way to do it. The G5 has Final Cut Studio and Magic Bullet on it, if that helps.

I'll try to get a well exposed still from another shot in the same room if you guys want.

It's cool that a frame can be posted here and the guy that did the transfer would end up responding to it! Thanks Rob.
  • 0

#14 Adam Thompson

Adam Thompson
  • Sustaining Members
  • 161 posts
  • Cinematographer

Posted 15 September 2007 - 02:00 PM

Also, does anyone know why pics from the Mac appear darker on a PC? I've run into this before using other PC's too but have never figured it out.
  • 0

#15 John Brawley

John Brawley
  • Sustaining Members
  • 834 posts
  • Cinematographer
  • Atlanta Georgia

Posted 15 September 2007 - 05:59 PM

Also, does anyone know why pics from the Mac appear darker on a PC? I've run into this before using other PC's too but have never figured it out.



Mac's and PC's have a different gamma standard for their respective monitor's. There are little apps that can allows you to switch bewteen the two on Mac. ( don't know about PC) I think it's 1.8 for mac and 2.2 For PC ???

It's pretty standard for web designers to check their designs by looking at both....

jb
  • 0

#16 Robert Houllahan

Robert Houllahan
  • Sustaining Members
  • 1585 posts
  • Industry Rep
  • Providence R.I.

Posted 16 September 2007 - 11:33 PM

This was a best light transfer. If it would have been scene to scene corrected, would this issue still come up do you think?

I'll try to get a well exposed still from another shot in the same room if you guys want.

It's cool that a frame can be posted here and the guy that did the transfer would end up responding to it! Thanks Rob.



I usually run tests like this with minimal changes, I think there were 3 notches on the CC for this, as I would think that seeing an overall look with minimal correction will give you a better idea of the performance of the stock under the lighting conditions you are working under.

Any time you are working with a colorist a discussion about the look you are going for is a must, I could have timed the blue out as was done with the stills in photoshop or made keyframed changes as she moves from the dark into the brightly lit room. There is so much room to work the picture in color correction that a understanding of the artistic intent really should be conveyed to the colorist as anybody doing this work will want to get you what you intend and more.

-Rob-
  • 0

#17 Phil Rhodes

Phil Rhodes
  • Sustaining Members
  • 11943 posts
  • Other

Posted 17 September 2007 - 08:17 AM

I don't see that it's horribly, horribly wrong. Maybe a bit pink in the highlights, maybe a bit blue in the shadows, but not smack-in-the-face incorrect, certainly.

A Rank 3 with every upgrade in the book and properly aligned (which Mr. Houllahan's comments indicate this one is) should be able to produce at least fairly reasonable results.

Phil
  • 0


FJS International, LLC

Gamma Ray Digital Inc

Broadcast Solutions Inc

Visual Products

Opal

Metropolis Post

The Slider

CineTape

Paralinx LLC

Abel Cine

Aerial Filmworks

Ritter Battery

Tai Audio

CineLab

Rig Wheels Passport

Willys Widgets

Wooden Camera

rebotnix Technologies

Glidecam

Media Blackout - Custom Cables and AKS

Technodolly

Willys Widgets

Glidecam

Wooden Camera

Technodolly

CineTape

Abel Cine

Opal

Broadcast Solutions Inc

Paralinx LLC

Metropolis Post

rebotnix Technologies

Gamma Ray Digital Inc

The Slider

Visual Products

Tai Audio

Media Blackout - Custom Cables and AKS

Aerial Filmworks

Ritter Battery

CineLab

Rig Wheels Passport

FJS International, LLC