Page 1 of 1

False Color Standards

PostPosted: Sun Sep 18, 2016 3:01 pm
by Dave Perry
I've never used false color on set, only dabbled with it to start learning it. It seems that there is not a standard for what IRE values correspond to the designated colors. Is this true, or am I just not getting it? Does anyone have a source for a standardized false color chart they could share?

Thanks!

Re: False Color Standards

PostPosted: Sun Sep 18, 2016 4:27 pm
by Timothy Cook
I was curious about this same thing a few months ago and ran across a post on the http://www.bmcuser.com forum.
I can't find the exact post, but JB and some others were discussing how it's different for almost each manufacturer. I can't find the post at this moment.

What was discussed was, what colors are used and what levels of IRE that determine the switch to the next color can vary form company to company.
I believe they commented on the fact that some manufacturers are trying to get on the same sheet of music and standardize, but at that point I can see each company saying that their format is the correct one.

It would be cool if it took on a format like ISO or ASA meaning international.

Re: False Color Standards

PostPosted: Sun Sep 18, 2016 4:43 pm
by Denny Smith
Correct. I have three devices that display false color, two. Knitters and a EVF, and each one is different. Two of them use red for the same IRE, the third purple, cut off levels are also different. Bottom line, I do not use them, except for the clipping false color option on the Cineroid EVF and the PixE, which both use red.

Each monitor/EVF I have that use false color, has a description of how it is implemented in their instructions. Of the three, I like the Pix E implementation the best, seems more logical in color choice and cut off values.
Cheers

Re: False Color Standards

PostPosted: Sun Sep 18, 2016 7:18 pm
by Dave Perry
Thanks guys. That's what I though. It's unfortunate though.

Re: False Color Standards

PostPosted: Sun Sep 18, 2016 7:40 pm
by Kyle Gordon
I think different cameras expose differently, so where you might want to record the sweet spot for skin on one, might be different for another. Same with overall exposure.

Monitors that offer false color for exposure, like the Flanders, have adjustable settings for false color for this reason. There is no standard, because cameras arent all the same, and exposure scenarios vary.

The false color for the Ursa Mini seems to be calibrated for the exposure that BM engineers think is best for this sensor, which is middle grey somewhere between 38 IRE and 42 IRE, (I cant remember the exact number) and typical well exposed light skin a stop over that.

Im not sure how the false color works now when you shoot in video mode tho. It should still show the same values based on the sensor, but I think the change in gamma to "video" happens before the false color calculation, so it relocates proper grey and skin tones off into new areas of the false color display from their equivalent values coming into the sensor, meaning that youd change your exposure to compensate for that LUT to set exposure that reads right on the false color display, and that seems wrong. Thats the way the viewfinder did a few months ago anyway. I dont know the implementation in the camera or latest firmware.

I see a lot of very good ideas coming with each new Firmware update. They are making a really great camera. I hope they've noticed this, and have it working well now too :)

Re: False Color Standards

PostPosted: Sun Sep 18, 2016 8:36 pm
by John Brawley
There's no standard.

The Convergent Design Odyssey 7Q even let's you select what IRE values go with what colour. I think from memory so did the amazing Si2K (remember that one ?)

Generally I think of greens as 50%, pinks as 70%, blues as just barely there blacks that should give you a little detail and blues is unrecoverable blacks, yellows you're nearly screwed on highlights and red means gone.

That's my understanding of the international she'll be right mate standard or ISBRM.

Maybe Hook is listening and will share the IRE values ? That would be handy for everyone to know :-)

Also important to know where it's reading those values from. If you're outputting a REC 709 to a monitor and THEN using the false colour to look at the REC 709, that probably won't be very useful.

False colour should be pre LUT, (and maybe ISO) in my view.

JB

False Color Standards

PostPosted: Mon Sep 19, 2016 1:07 am
by rick.lang
The description in the URSA Camera Manual should be helpful. I believe the middle grey is 38.4 IRE. John's description may be a little too extreme. Red means you may be in danger of clipping the highlights depending upon your exposure. I've had red in my frame and the recorded image was fine. Purple means you may not have sufficient detail in your black, blue is a warning you are close to losing the shadows but I often have blue and the image is fine, yellow is a warming you are close to losing highlights. But with zebras set at 90% the red shows up earlier so there's some margin of safety there.

I'll check this tomorrow and report back if my understanding is flawed. I only expose with false colour now, paying the most attention to pink and green; but I do need to take a peak at zebras more often. Zebras are definitely sensitive to the % zebra setting, so again, I use that as a guide that I might have a blown highlight, but usually it's a false alarm (thank you 15 stops).

Edit
Zebras settings do not affect false colour values on the monitor. So false colour may be a more reliable exposure guide for ETTR but of course false colour is most used in keeping your pink and green on faces consistent across clips while you adjust exposure as appropriate.

Sent from my iPad using Tapatalk

Re: False Color Standards

PostPosted: Mon Sep 19, 2016 2:01 pm
by Leon Benzakein
I have noticed that most false colour displays come with a key of what colour represents what IRE value for that particular manufacturer.

Re: False Color Standards

PostPosted: Mon Sep 19, 2016 3:34 pm
by Dave Perry
Seems like we all have the same experience, no standard. That's too bad because it makes me less likely to lear it when I have to re-learn depending on what gear I'm using. Zebras and waveform it is for now at least.

Thanks for the feedback.

False Color Standards

PostPosted: Mon Sep 19, 2016 3:53 pm
by rick.lang
rick.lang wrote:The description in the URSA Camera Manual should be helpful... But with zebras set at 90% the red shows up earlier so there's some margin of safety there.

I'll check this tomorrow and report back if my understanding is flawed...


Sent from my iPad using Tapatalk


Just checked and John Brawley is correct that the false colour value is set before anything else. I was wrong that a 75% zebra setting would result in false colour red appearing earlier. Thought I had observed that but checking this morning when there was no variation in lighting, while the zebras that may appear on the monitor at 75% and are gone with a 100% zebra setting, the quality of false colour yellow, orange, and red did not alter one iota with the change in zebra settings. That's a very good thing to know for people who expose to the right with zebras. A very quick peak in the false colour display will tell you if those zebras are very bad news or within the safe area anyway. Thanks, John, once again.


Sent from my iPad using Tapatalk

Re: False Color Standards

PostPosted: Mon Sep 19, 2016 4:05 pm
by Dave Perry
rick.lang wrote:
rick.lang wrote:The description in the URSA Camera Manual should be helpful... But with zebras set at 90% the red shows up earlier so there's some margin of safety there.

I'll check this tomorrow and report back if my understanding is flawed...


Sent from my iPad using Tapatalk


Just checked and John Brawley is correct that the false colour value is set before anything else. I was wrong that a 75% zebra setting would result in false colour red appearing earlier. Thought I had observed that but checking this morning when there was no variation in lighting, while the zebras that may appear on the monitor at 75% and are gone with a 100% zebra setting, the quality of false colour yellow, orange, and red did not alter one iota with the change in zebra settings. That's a very good thing to know for people who expose to the right with zebras. A very quick peak in the false colour display will tell you if those zebras are very bad news or within the safe area anyway. Thanks, John, once again.


Sent from my iPad using Tapatalk


So in essence, calibrating your zebras to turn on based on what suites you from false color would be the way to go, don't you think?

False Color Standards

PostPosted: Mon Sep 19, 2016 4:39 pm
by rick.lang
Yes, hadn't thought of that. A solid red is a serious warning, but yellow/orange/red transitions could be a good way to set zebras! For example if you are very conservative, set the zebra to show when false colour shows yellow. Less conservative, set zebras to show when false colour is showing the transition from yellow to red, a risk taker can set zebras to show in the area when red is beginning to show. I'll try that approach instead of using zebras at 90% all the time.

Edit
Okay, I tested this on a light source where I had a large red area with a narrow outline of yellow before going to light grey. With zebras set at 100%, the zebras appeared only in the red area. At 95%, I did not see much difference. At 90% and lower, the zebras showed in the yellow area. Even at 75% zebra, I never saw zebras in the light grey area but the zebras were more solidly over all yellow.

So I guess what I've started doing (set zebra at 90%) isn't that bad a choice. It's an early warning system for those not using false colour. I still recommend using false colour though.

Sent from my iPad using Tapatalk

Re: False Color Standards

PostPosted: Mon Sep 19, 2016 5:42 pm
by Denny Smith
Dave Perry wrote:Seems like we all have the same experience, no standard. That's too bad because it makes me less likely to lear it when I have to re-learn depending on what gear I'm using. Zebras and waveform it is for now at least.
Thanks for the feedback.


I am with you Dave, I learned to set exposure using Waveform monitor, and color balance with Vector scope, most accurate system for me at lest, with Zebras for a quick check. In a ENG camera, you,could,set two Zebra values, one for clipping and the other for skin IRE value, to get people exposure correct. Worked very well in most situations, but not all. You still need to mix in experience with these tools, which provide a starting point.
Cheers

Re: False Color Standards

PostPosted: Mon Dec 03, 2018 4:39 pm
by Robert Baker
<Reviving a thread from long ago>

Has there been any conversation on having BMD put the color chart on the screen (on any of the cameras that support it as well as the Video Assist monitors) so that there is no guess work involved. All the off camera monitors that I used, outside of the Video Assists, have the chart available in False Color mode. It would make setting up exposure easier.

Re: False Color Standards

PostPosted: Tue Jan 15, 2019 8:14 pm
by kuunami
Robert Baker wrote:<Reviving a thread from long ago>

Has there been any conversation on having BMD put the color chart on the screen (on any of the cameras that support it as well as the Video Assist monitors) so that there is no guess work involved. All the off camera monitors that I used, outside of the Video Assists, have the chart available in False Color mode. It would make setting up exposure easier.
That would be great.

Sent from my SM-J700M using Tapatalk

Re: False Color Standards

PostPosted: Thu Feb 21, 2019 4:33 am
by Peter Crawley-Boevey
With all the different standards of False Colour across the industry it seems imperative that every false colour display has a legend that applies to that particular display. I can't for the life of me think why BM cameras and the video assist don't have it?

Re: False Color Standards

PostPosted: Thu Feb 21, 2019 8:40 pm
by Rakesh Malik
John Brawley wrote:There's no standard.

The Convergent Design Odyssey 7Q even let's you select what IRE values go with what colour. I think from memory so did the amazing Si2K (remember that one ?)



My camera has a tool that lets me turn various exposure zones on or off, so I figured out where I wanted my skin tones and turned those + the high/low clip ones on, and use those as a reference. The ones that are turned off render as shades of grey on the monitor, the rest light with the color indicated in the legend.

I can't change what color the camera uses, but just being able to turn them on/off is very nice. I've wondered often why that's not more common.

Re: False Color Standards

PostPosted: Fri Feb 22, 2019 3:11 am
by Uli Plank
You can load a LUT into the camera to get any kind of false colors you want (I was used to Red).

Here is a great tool which supports about any 'standard': https://timeinpixels.com/false-color-plugin/