Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

skintone

Adrian Jebef said:
I use a 2000 series 24" Cine-tal Cinemage on-set as my main reference monitor but I also have a 24" NEC MultiSync PA241W that I use as a cross-reference display. Both can be internally calibrated with simple probes. I run my camera signal to both and can double check subtle differences between the Class 1 and the prosumer display. Basically the Cine-tal shows true gamma ranges and beter color resolution (which really helps when trying to neutralize images with a lot of different color temperatures) but I find it invaluable to reference the same image looking at the NEC (as that picture will ultimately be a better representation of what is seen at home). I've used different displays (Cine-tal, Panasonic, Sony E250, Sony L231, Flanders, etc, etc.) throughout the years, I've even used different combos of different displays, and even up to 4 displays all on one cart on-set. I've also gone through high-end color probe calibrations again almost all of these displays attempting to get them to a technically exact match. 90% of the time it all ended in a mess. I've found my current setup the best balance between technical precision and the pragmatically flawed human eye.
Click to expand...
I was just looking into ordering the NEC PA241W as I've been a bit frustrated with some of my computer monitors lately and calibrations being off. This has been a highly praised monitor in my research.
 
Steve Sherrick said:
I was just looking into ordering the NEC PA241W as I've been a bit frustrated with some of my computer monitors lately and calibrations being off. This has been a highly praised monitor in my research.
Click to expand...

Steve, I have only good things to say about the NEC. For the price point it's hard to beat. Plus it allows you to switch between different calibrations, has a multitude of manual color/gamma/luma adjustments, and can be directly calibrated with their SpectraView II software + probe. Overall I find it more accurate than the ubiquitous 17" on-set Panasonic. As all gear that's used on set, it will get beat up over time. I recently replaced the entire LCD panel in one of my Cine-tals after that display was beat up over the course of a few years. Now she's as good as new. I've had my NEC for about a year now and it's gotten a slight color shift due to being bounced around on my cart all day. But for the price point I'm planning on just replacing it rather than attempting any sort of hardware fix.

http://www.necdisplay.com/p/desktop-monitors/pa241w-bk-sv
 
I had always thought R3D to be 12bit ? Is the format now 16 bit capable ?
 
The R3D itself uses 16bit precision. The SDK for development uses float and integer math where appropriate.

Graeme
 
Blair S. Paulsen said:
I have had better results with a 1/4 CTB filter rather than the 80C or 80D. I believe it is because the unique MIRED shift that best suited crossing over film response essentially warps the relatively linear shift required to more equitably excite silicon. Suggest that anyone going down this road try both. YMMV.

Cheers - #19
Click to expand...

Hey Blair/Noel (or anyone), would you guys happen to have CTB/80C/80D filter before and after test shots (in tungsten, obviously.) Preferably sample .r3ds stills would be best, but even just to web compressed video showing the difference would be adequate. I was under the impression that an 80D would suffice, but it seems you guys think it won't. Also do you put it on the lens or do you filter the lights?

I fear that putting the filtration on the lens would just destroy mixed lighting shots, but filtering tungsten-only lights would be a major pain in the ass on small gigs. Ideally I'd like a "set it and forget it" solution (aka on the lens), but I'm worried that if I go with anything more than an subtlety of the 80D, it'll be too much for daylight balanced (and yes, I almost always prefer warmer tones).
 
Last edited:
Sorry Mike, can't post anything today as I away from my system but I can pass along some observations:

1) I can't imagine using anything more than a 1/4 CTB on camera. In my experience a little blue shift is enough to compensate for the inherent properties of silicon. Ever since RC2 came out the color science is pretty good at handling high CRI tungsten sources like 3200 Kinos. Its the 2500-3100 Kelvin range (typical of practicals), that, IMO, still needs help. I would also note that a more subtle blue shift that has less stop loss keeps the cure from being worse than the disease. If you have tons of tungsten light then I suppose giving away a full stop or more doesn't matter, but the more typical situation I run into is both low Kelvin and lack of foot-candles on location so I prefer the lower "price" of a lighter filter.

2) I agree that mixed lighting can prove problematic in some scenes even when grading from RAW. For that reason, I would much rather gel lights than drop glass in front of the lens to ameliorate mixed lighting challenges. It may take more time but in most cases its worth the time you save in post. If you have a lot of incandescent practicals, low quality instruments, red/orange reflective surfaces near lighter complexion talent, very limited time, etc then the on camera filter is your friend.

Cheers - #19
 
Mike P. said:
Hey Blair/Noel (or anyone), would you guys happen to have CTB/80C/80D filter before and after test shots (in tungsten, obviously.) Preferably sample .r3ds stills would be best, but even just to web compressed video showing the difference would be adequate. I was under the impression that an 80A would suffice, but it seems you guys think it won't. Also do you put it on the lens or do you filter the lights?

I fear that putting the filtration on the lens would just destroy mixed lighting shots, but filtering tungsten-only lights would be a major pain in the ass on small gigs. Ideally I'd like a "set it and forget it" solution (aka on the lens), but I'm worried that if I go with anything more than an subtlety of the 80A, it'll be too much for daylight balanced (and yes, I almost always prefer warmer tones).
Click to expand...

I recommend shooting cold and grading warm later. The blue channel sensitivity of CMOS with many CFA's can be over 6db below the red. That can be a real problem in tungsten, where the blue channel strength is also far below that of the red. The 80A/B/C are designed to affect both red and blue. The 80C and D actually apply less correction than an 80A, and happily absorb less light.

In lieu of the 80's, I don't recall a little blue being a bad idea in tungsten, but there's no substitute for testing.

Cheers.
 
I did a couple of tutorials for the creative cow a while back I don't want to send people over there and away from this thread but I do talk about this in two different articles (a bit outdated sort of) but Ted had some great things to say in the audio interview.

the link is here

but come back here to talk.

i tend to use at least an eighth blue on all tungsten just to keep the noise out of the blue channel.

david
 
Thanks Blair, Noel, and David. Yeah, I definitely need to do some filter testing... I wanted to do some LowCon/DigiCon testing as well (they claim that DigiCons actually help skin tones on digital sensors, which might be true if it adds a bit of blue).

I've found, under tungsten, I need to blast the scene with tons more light over any other temp sources. With daylight (or even non-professional ~3800-4400 florescent sources), I can be underexposed badly, but I can massage a pretty great image/look in my one-light out of RCX. Not the case with tungsten. The difference in the histogram is not only how weak the blue channel is, but also how blurry/soft the blue channel looks compared to red and green. The benefit is two-fold too, as it also allows the image to be pushed to 1600-2000 and still be useable (especially when downscaled to 1080p.)

So, yeah, I'm pretty much turned off tungsten unless adding blanket 1/8~1/4 CTB (on-cam filter) can give the image at least a *strong* 3200k. Even if it brings brightness down a bit, if the blue channel is resolved well, I'll be able to push the ISO higher than I would with straight tungsten anyway so I wouldn't worry about light loss too much. Of course, if it's mixed lighting I'd have to gel/swap the tungstens out.
 
Last edited:
Mike, I have those ultra-cons now if you want to test them.


Mike P. said:
Thanks Blair, Noel, and David. Yeah, I definitely need to do some filter testing... I wanted to do some LowCon/DigiCon testing as well (they claim that DigiCons actually help skin tones on digital sensors, which might be true if it adds a bit of blue).

I've found, under tungsten, I need to blast the scene with tons more light over any other temp sources. With daylight (or even non-professional ~3800-4400 florescent sources), I can be underexposed badly, but I can massage a pretty great image/look in my one-light out of RCX. Not the case with tungsten. The difference in the histogram is not only how weak the blue channel is, but also how blurry/soft the blue channel looks compared to red and green. The benefit is two-fold too, as it also allows the image to be pushed to 1600-2000 and still be useable (especially when downscaled to 1080p.)

So, yeah, I'm pretty much turned off tungsten unless adding blanket 1/8~1/4 CTB (on-cam filter) can give the image at least a *strong* 3200k. Even if it brings the entire image down a bit, if the blue channel is resolved well, I'll be able to push the ISO higher without worrying to much. Of course, if it's mixed lighting I'd have to gel/swap the tungstens out.
Click to expand...
 
Bob Gundu said:
Mike, I have those ultra-cons now if you want to test them.
Click to expand...

HA, awesome, thanks for the offer! When did you get those? UltraCons are the "cleanest" low contrast filter, right? Like they don't bloom the highlights like LowCons/DigiCons/Mist Filters and they try to remain as image neutral as possible? Needless to say, I'll let you know when I'm around (hopefully soon).
 
Mike P. said:
Thanks Blair, Noel, and David. Yeah, I definitely need to do some filter testing... I wanted to do some LowCon/DigiCon testing as well (they claim that DigiCons actually help skin tones on digital sensors, which might be true if it adds a bit of blue)..
Click to expand...

the only thing I have added is lo cons and digicon both at the weakest 1/8 do a very nice job bringing out detail in the blacks and rendering skin. The locon is much different to the digicon side by side and the locon does not like "contro Luce" hard backlight needs to be better controlled or it can flare out and go milky with direct light sources.

the cool thing about the dig icon is that the have zero affect on resolution because of the way it affects light. I'm using a 1/4 black promist on a film I am doing right now and I light what it is doing to the backs and the highlight roll off. Can't post any stills of that yet unfortunately.

David
 
Mike P. said:
HA, awesome, thanks for the offer! When did you get those? UltraCons are the "cleanest" low contrast filter, right? Like they don't bloom the highlights like LowCons/DigiCons/Mist Filters and they try to remain as image neutral as possible? Needless to say, I'll let you know when I'm around (hopefully soon).
Click to expand...

Digicons should be the equivalent of UltraCons, but they're more recent so they're probably even better than the UltraCons. I don't have 2 full sets to compare...sorry Mike!
 
Noel Sterrett said:
I recommend shooting cold and grading warm later. The blue channel sensitivity of CMOS with many CFA's can be over 6db below the red. That can be a real problem in tungsten, where the blue channel strength is also far below that of the red. The 80A/B/C are designed to affect both red and blue. The 80C and D actually apply less correction than an 80A, and happily absorb less light.

In lieu of the 80's, I don't recall a little blue being a bad idea in tungsten, but there's no substitute for testing.

Cheers.
Click to expand...

I have some 80C filters for sale (6.6x6.6 and 138mm), LIKE NEW, 75% off discount... surprised that no one jumped on them yet at these very low prices. I thought no one shoots tungsten on RED ONE/SCARLET/EPIC without 80C/80D/CTB filters...

http://www.reduser.net/forum/showth...m-Series-9-3x3)-75-OFF!-Thanksgiving-discount
 
I tend to view lens filtering as a sledgehammer where a scalpel is the tool of choice, but that's just me. I'd rather make the effort to gel light sources if I can help it. :-)
 
Scott Crawley said:
I tend to view lens filtering as a sledgehammer where a scalpel is the tool of choice, but that's just me. I'd rather make the effort to gel light sources if I can help it. :-)
Click to expand...

Agreed. If you have the time/crew to gel lights it is the superior modality, particularly since it allows you to vary the strengths on each instrument to best achieve desired rendition on key elements.

Cheers - #19
 
I also agree and the are good reasons.

1. Not all 3200k tungsten bulbs come in at 3200k most are warmer and dimmed lights require more correction as they are even warmer
2. Gelling lights gives you more control
3. The filters affect the entire scene which is a bit more film style plus usually you lose stops which then forces you to light a bit differently.

David
 
Everyone does this on this forum, so I'm not picking on David AT ALL. But for the record it is "lose stops"...
"Loose" is primarily an adjective (though can be used as a verb). "Lose" is a verb.

"If a man is loose with his wallet, he will lose his money."
"I will loosen this tie so I don't lose oxygen."

"Lose" is such a strange word because it is a long U sound with a single O. It rhymes with "choose" and "snooze" which all have the double O.

Stepping off of soap box now. We now resume normal programming. ;)
 
You must log in or register to reply here.
Back
Top