|
|||||||||
|
Thread Tools | Search this Thread |
November 20th, 2010, 11:00 AM | #1 |
Regular Crew
Join Date: Jul 2010
Location: Parsippany, NJ, USA
Posts: 62
|
Increasing the dynamic range on the 60D
Hi all,
I read some article about increasing the dynamic range on a 7d to achieve a better film-look of the video by defining a custom Picture Style for video and I would like to do the same for the 60d. May I ask what specific settings that works for you. Thanks... |
November 22nd, 2010, 07:43 PM | #2 |
Regular Crew
Join Date: May 2010
Location: Los Angeles, CA
Posts: 25
|
Hi there,
I'm just using the dialed-down neutral setting that many of the superflat early adopters seem to have conceded is the best general approach. There are probably quite a few threads here that have already covered this topic, but I can't seem to locate them at the moment. I don't recall any staggering finds in truly difficult lighting conditions ever. Often the supporting examples have been captured under very flat lighting conditions to begin with, such as overcast pastoral scenes. |
December 13th, 2010, 03:12 PM | #3 |
Regular Crew
Join Date: Jul 2010
Location: Parsippany, NJ, USA
Posts: 62
|
Thx for the post!!!
|
December 13th, 2010, 03:48 PM | #4 |
Inner Circle
Join Date: Nov 2006
Location: Tallahassee, FL
Posts: 4,100
|
I finally settled on the Marvel Cine picture profile, and found I got about a stop to a stop and a half on the top end compared to shooting the Neutral profile. I don't shoot with anything else any more.
__________________
DVX100, PMW-EX1, Canon 550D, FigRig, Dell Octocore, Avid MC4/5, MB Looks, RedCineX, Matrox MX02 mini, GTech RAID, Edirol R-4, Senn. G2 Evo, Countryman, Moles and Lowels. |
December 13th, 2010, 04:46 PM | #5 |
Trustee
Join Date: Dec 2003
Location: Malvern UK
Posts: 1,931
|
I've been playing around with this one;
SagePictures.com |
December 13th, 2010, 09:15 PM | #6 |
Inner Circle
Join Date: Nov 2006
Location: Tallahassee, FL
Posts: 4,100
|
Don't have a Mac and don't use FCP. I have heard from people who've used this program that it does eek out a slightly superior picture, but takes it's sweet time doing it.
__________________
DVX100, PMW-EX1, Canon 550D, FigRig, Dell Octocore, Avid MC4/5, MB Looks, RedCineX, Matrox MX02 mini, GTech RAID, Edirol R-4, Senn. G2 Evo, Countryman, Moles and Lowels. |
December 14th, 2010, 01:10 PM | #7 |
Major Player
Join Date: Nov 2010
Location: Santa Ana, CA
Posts: 499
|
Whats the most recent consensus on this topic, all that info is from a year ago? I've heard conflicting opinions, some say get it right in camera, others shoot flat and grade it in post. Correct me if I'm wrong, contrast decreases dynamic range, so you loose information? But it also increases color saturation, so it would seem to preserve color information? I'm thinking of cloudy days. You can correct color in an overcast day in post, but does it compare to getting the light correct and retaining the color information?
|
December 14th, 2010, 10:53 PM | #8 |
Regular Crew
Join Date: May 2010
Location: Los Angeles, CA
Posts: 25
|
@Greg
Well yes it is quite a bit from a year ago, but then Canon dslr video specs haven't really changed in that year. I think it comes down to what you visually prefer. In my opinion, testing is the only way to really tell. From my own personal tests with flat picture styles I would say that yes you can preserve tonalities that might otherwise be lost. But at the same time fine details seem to suffer and colors get, well, chalky. I don't believe that any picture style can give you more dynamic range, but it can change the subtleties in which that range is captured. Thus higher contrast doesn't imply lower dynamic range, just reduced finesse of tonalities within the dynamic range. I'm surprised that Perrone claims a stop and a half on the high end with marvel, but keep in mind that he is a professional and I am certainly not. I have done my own personal testing after reading many posts by others, and this is my opinion for the moment based on my own experimentation. |
December 15th, 2010, 12:44 AM | #9 | |
Inner Circle
Join Date: Nov 2006
Location: Tallahassee, FL
Posts: 4,100
|
Quote:
When doing this the RIGHT way, with LOG signals, we don't introduce luma or chroma noise during this expansion. But doing this the way it must be done with these consumer cameras, we are most certainly introducing noise. To combat this, I try to never record above 200 or 400 ISO. And I know that if I have to shoot beyond that, I am going to have a lot of work to do later in post. In fact, if I have to shoot beyond 400 ISO I am likely going to be shooting a different profile like neutral. But this is quite rare for me. The idea of getting it right in camera is wonderful. But you're walking a tightrope, and these cameras record signals that are VERY inflexible in post should you change your mind. If they were recording RAW signals, things would be different. So for people looking to get more dynamic range from these cameras, it can be done to an extent, but the trade-off is more noise in the signal. Sometimes that is worth it. Other times not so much. On these camera with no histograms or zebras, and the potential for being overexposed half a stop leading to unrecoverable footage, I'll take all the help I can get, and worry about the noise later.
__________________
DVX100, PMW-EX1, Canon 550D, FigRig, Dell Octocore, Avid MC4/5, MB Looks, RedCineX, Matrox MX02 mini, GTech RAID, Edirol R-4, Senn. G2 Evo, Countryman, Moles and Lowels. |
|
December 15th, 2010, 01:24 AM | #10 |
Trustee
Join Date: Apr 2006
Location: United States
Posts: 1,158
|
shooting flat isn't doing you any favors. you have 8 bits, thats it. if you shoot in a flatter setting, recording 7 or even 6 bits worth of range, you've lost, not gained anything. this is especially true with saturation where you are reducing gradation. with contrast, reducing it tends to let you also lower your exposure to recover highlights better in camera. so you might get another stop or so compared to stock in the middle settings. OTH there is nothing wrong with shooting it in camera, if you are confident about what you are doing. I really get tired about having a millions options in post, really you only need one, the right one. if you get it right in camera, you're 90% done and just tweaking... but then again I'm old school like that when you couldn't do it in post, or it cost a fortune to do.
|
December 15th, 2010, 03:45 AM | #11 |
Trustee
Join Date: Dec 2003
Location: Malvern UK
Posts: 1,931
|
There's no real right or wrong answer to this. Personally I'm not a fan of reducing saturation too much in camera. I can always reduce saturation later, but it is harder to bring more saturation in if it is reduced too far to begin with. I like saturated colourful images anyway.
Regarding the tonal range, the flat styles could be a hinderance or an advantage depending on how things are seen. The type of S-curve that styles like Marvel Film introduce appear to me to be adding more tonal range at the highlight and lowlight range of things (in other words their S-curve is the inverse of the type of s-curve most people use in their NLE's to get a 'film-look') therefore producing a less contrasty image. The big question is, is it worth the trade off of tonal range in the mids, and is there anything further to be gained by reducing the contrast setting in the camera? My feeling on the latter would be perhaps no. But without knowing exactly what the camera is doing when the contrast is adjusted it wouldn't be fair to comment. I think though that a subtle s-curve as produced by the Marvels Film and the Digital Cinema picture style I recently found could be an advantage and would produce much the same gains as using the film gamma in Panasonic cameras and the hyper gammas in HDCAM and XDCAM cameras, since adjusting the gamma curves is exactly how these work too. |
December 15th, 2010, 04:06 AM | #12 |
Inner Circle
Join Date: Feb 2004
Location: switzerland
Posts: 2,133
|
well what all of you are trying, is in fact to produce an HDR picture with only one shot.
this trick is possible with some trade off, deciding if you want to bet on dark or bright side. trying to get the most details in bright area is great because usually burned pictures looks ugly. the downside of it is that trying to recover the dark side brings noise. (but we know how to cure it). if you choose to give the better chance to the dark side , then it is a lot more tricky to get back the bright portion of the picture, because all get burned/over exposed easily and it is a lot more visible/disturbing than crushed blacks. You also have to work with codecs that are supporting extra bits for the calculation or you get banding very easily during the process. The flat profile found on the canon is very popular, because usually , people are shooting low light and the need to restore from the dark side is bigger than the one found one well exposed pictures. |
December 15th, 2010, 12:58 PM | #13 | |
Regular Crew
Join Date: May 2010
Location: Los Angeles, CA
Posts: 25
|
Quote:
|
|
December 15th, 2010, 02:10 PM | #14 | |
Inner Circle
Join Date: Nov 2006
Location: Tallahassee, FL
Posts: 4,100
|
Quote:
"...to try and compress information that the camera would normally [use] lose, into a space that it can actually record." But yes, you are right. And I suspect that if you are defining the "dynamic range" of a camera as extending the darkest darks and brightest brights it can record, then I am unaware of any camera that offers this. Various methods of doing what I described are available by shooting LOG, by using curves, etc. But the physical properties of the sensor cannot be changed.
__________________
DVX100, PMW-EX1, Canon 550D, FigRig, Dell Octocore, Avid MC4/5, MB Looks, RedCineX, Matrox MX02 mini, GTech RAID, Edirol R-4, Senn. G2 Evo, Countryman, Moles and Lowels. |
|
January 2nd, 2011, 03:48 PM | #15 |
Major Player
Join Date: Dec 2005
Location: Los Angeles
Posts: 663
|
If you expose properly for your scene, then no amount of flattening or pretending to shoot log will make an improvement. Yes, the footage will look different on the LCD because you've raised the lows and lowered the highs. But what you've effectively done is reduce the dynamic range to what the LCD or monitor can show, which is less than what the camera can capture.
For the most amount of dynamic range, expose properly, which can mean underexposing certain situations in order to maintain highlights. Shooting "flat" is simply underexposing or overexposing digitally, not optically, and will introduce noise and reduce latitude. People preach "log" and "flat" because they have no idea how to operate a monitor; they assume that if they can't see it, it's not there, so they force it into the 6 bit range of the cheap LCD.
__________________
software engineer |
| ||||||
|
|