RayPA
Ignore It (It'll go away)
+1 on the color casts. I usually get the orange-ish/blue cross. I'm not sure why that happens. Sometimes it looks as though it's the result of some 'outside' variable, such as heat or expiration. Very frustrating. It is exposure critical like slide film. I try to not to underexpose it. I have to say that when it's good, it's very, very good. It has to be one of the sharpest color films I've ever used.
DCSANG said:
Dave, when you shoot with Extar and underexpose (iso 125), what color cast do you pick up?
/
bensyverson
Well-known
If you look at the charts for 800 and 400, you'll see that their spectral responses are basically the same. Certainly not different enough to make any visible difference in the rendering of fluorescents.
Portra 400 has a more magenta-tinged orange mask, so if you're seeing differences between 400 and 800 in the greens, odds are your scanning workflow is not completely neutralizing the mask.
Bottom line is that these films exist as a family for a reason... There is just not a appreciable difference between any of them unless you're printing photochemically (and even then, it will be subtle).
If you see a weird color cast with any modern color negative film, it's not the film. It's you. You can live with it or adopt a more careful scanning workflow, but do not blame the film. Not one of these films has any color cast.
Last edited:
Tim Gray
Well-known
Sorry to go off on a tangent like this. I do think some of is relevant though with the discussion of Ektar.
I'd really have to argue with this. I have shots taken at the same time with both films. The Portra 800 has a huge green cast that the 400 doesn't show. It's not me blaming the film or anything else. The color balance looks good in areas where the fluorescents don't fall. Note, I'm not saying the whole image has a cast - that is easily corrected out. Even in shots taken in single types of lighting, most of this can be corrected out. But when lighting is mixed between tungsten and fluorescents, spectral sensitivities can be a real pain to correct out.
I actually talked to a guy who helped work on Portra 800 about this. He did not deny that there were some differences in the spectral sensitivities between these films. He said that Portra 160 and 400 have sensitivities closer to that of the eye than 800 does, resulting in more accurate color. However, it makes it harder to get good color saturation. And that these issues are magnified at higher speeds (he didn't clarify this bit).
I actually have a careful comparison shots on two rolls but they haven't been developed. Otherwise I'd post them. In lieu of that, here's a shot of Portra 800 with a wicked green cast. Notice that the light is in the room in the back and is not affecting the color balance in the foreground.
. by ezwal, on Flickr
The best I can do in a comparison (albeit an unintended one) can be seen at the following link. The first photo is if Portra 400, the second is 800. The light above the soup bowl is halogen - the fluorescent light is on the other side of the room at the top of the picture. You can see a clear difference where the light reflects on the stove top. The third picture is again Portra 400. The fluorescent light is the one above the window. No green cast. The light coming from the left is halogen. The fourth picture is taken in the same lighting. Again, the foreground is lit by the halogen, while the window sill is crazy green from fluorescent light above it. The last picture is duplicated in this post - it's also Portra 800.
http://www.dropbox.com/gallery/100367/1/green?h=e7d868
dcsang
Canadian & Not A Dentist
Ray... I usually pick up a wee bit of green - but I've picked up that green whether I shot at 100 or 125.. You can't see it in this shot I don't think because I've corrected the color via LightRoom (and RAW).
Cheers,
Dave
Cheers,
Dave
Lilserenity
Well-known
I have to say I have never had the problems people have encountered with Ektar, admittedly I have never shot it in medium format (need a decent meter really) but in my OMs it's my favourite film for landscape and general street scenes (Portra always for people!)
Jamie123
Veteran
You can correct for these color casts in "post" production but, well, part of the appeal of using film should be that one shouldn't (necessarily) have to correct for color casts in "post" - if you're going to correct for color cast you're going to do it before shooting (with filters or such).
That said, to correct it in post, when you scan your negative, hopefully, you can scan to a RAW image format. For example, Nikon Scan 4 (and I believe Vuescan) will allow you to scan to RAW formats (.NEF in the case of Nikon Scan). Import said RAW format into your RAW image converter (I use LightRoom) and correct the cast within the conversion program.
That's usually how I do it.
I rarely, if ever, have to color correct Portra - it is, well, immaculate
Cheers,
Dave
You're always correcting in post. The question is just whether the software gets it right or whether you have to do it yourself. As far as I understand it the negative scans in NikonScan are always an interpretation (i.e. auto white balance) by the software. Saving as .Nef will give you Raw editing capabilities in post but it's not really a Raw file. No matter what white balance settings you set in NikonScan the .Nef file WB values will always be set to 0/0. At least that's my experience with my Nikon 9000.
RayPA
Ignore It (It'll go away)
Thanks, Dave. Yeah, I don't see a cast in the shot. I usually pick up a slight blue with underexposed Ektar (still, it's the cooler end of the spectrum).
/
bensyverson
Well-known
I agree, since people always seem to announce "problems" with Ektar or Portra, when the real culprit is scanning technique.
I'm looking at your images now, and they show Noritsu in the EXIF. If they're minilab scans, that's the root of the problem. Not only are these machines fully automated, they use "image optimization" software that screws with your colors in truly unknowable ways. On Noritsu, the software is called AccuSmart, and it's neither accurate nor smart.
Again, I understand how convenient minilab scans are, and they can help you edit your images... But judging a film based on a minilab scan is like judging paint color based on a cellphone picture.
bensyverson
Well-known
Tim, I hope you don't mind, but I'm attaching modified versions of a couple of your images. Just let me know if you want me to take em down.
I hope this illustrates just how "wrong" the images straight out of the Noritsu can be. They're quite difficult to color correct, because the Noritsu takes every opportunity to blow out highlights and fog the blue channel.
I hope this illustrates just how "wrong" the images straight out of the Noritsu can be. They're quite difficult to color correct, because the Noritsu takes every opportunity to blow out highlights and fog the blue channel.
Attachments
Tim Gray
Well-known
I've rescanned this at home on my Coolscan and seen the same thing. In fact, I've spent quite a lot of time learning to manually color balance and correct scans from linear files, converted files, etc., you name it. I've got a pretty good handle on how to do it. I've got no problems with the cyan/blue color cast that people always say Ektar has; that's an easy fix.
If you really think you could correct out the green color of those fluorescents in some of the shots I've posted and not screw up the rest of the color without resorting to masking, you are more than welcome to demonstrate your scanning prowess.
I've tested a bunch of films, both still and motion, and some films are better at suppressing the green spike from fluorescents and others aren't. Portra 400 seems better at it than Portra 800. I've not tested Ektar specifically for this, but I would not be surprised if it was weaker in this area as well since it is more saturated.
If you want to chalk up everything to 'bad scans' and 'technique', you are more than welcome to. You are also more than welcome to shoot some Portra 400 and 800 in mixed fluorescent and tungsten lighting and show us how they react to fluorescent lighting exactly the same.
Sorry again for the interruption. I'm done now.
bensyverson
Well-known
Tim Gray
Well-known
Not at all. What about the electric green in the background in the other room from the fluorescent light?
bensyverson
Well-known
...you have mixed light in this shot. If you correct for the face, the background will go green. That would be the same if you had shot it in any film, or digital... It's just a fact of life.
Tim Gray
Well-known
That's my point. Portra 400 handles the mixed light a lot better than Portra 800. Just like the Fuji film with the 4th color layer handles mixed light of fluorescent + something else.
bensyverson
Well-known
I just haven't seen any evidence of this... All color negative films handle mixed light very well. Fuji's "4th layer" is just marketing hype. That "4th layer" does not alter the spectral sensitivity curves in any meaningful way, and just winds up as plain-old magenta dye. Fuji negatives are CMY just like Kodak.
Last edited:
dcsang
Canadian & Not A Dentist
While I agree with that statement for the most part - I still wonder why the cast in Ektar scans vs Portra... Portra, as I have said (and have been told) is "made for scanning" - is it that NikonScan seems to "understand" Portra better than Ektar?
Cheers,
Dave
Tim Gray
Well-known
In my experience, it does make a difference. I posted some samples and have many more.
bensyverson
Well-known
I think what's needed is a controlled test... One that utilizes the same exact setup and lighting, and doesn't involve drugstore scans. I've been meaning to do this for a couple months, but this is giving me a push to finally get it done!
Tim Gray
Well-known
Yeah, I have a test like that shot on both films, I just haven't sent them off. I included a color checker in some mixed lighting. By the way, the scans I showed were all in the same lighting (same room) and were developed and scanned by Precision Camera. I can provide you a raw scan from my Coolscan if you want.
On the other hand, Portra 400 doesn't seem to have much of an issue with mixed fluorescent/tungsten lighting in my experience. 800 has more of an issue. I've seen examples from Fuji films that convinced me the 4th layer works. I've also shot tests myself on Kodak Vision3 500T which picked up a green cast in mixed lighting, while Fuji Reala 500 (with the 4th layer) didn't.
As it's been explained to me, you can get the same result without a special 4th color layer if you tailor the spectral sensitivities of the layers to better match what the human eye sees. Makes sense to me.
On the other hand, Portra 400 doesn't seem to have much of an issue with mixed fluorescent/tungsten lighting in my experience. 800 has more of an issue. I've seen examples from Fuji films that convinced me the 4th layer works. I've also shot tests myself on Kodak Vision3 500T which picked up a green cast in mixed lighting, while Fuji Reala 500 (with the 4th layer) didn't.
As it's been explained to me, you can get the same result without a special 4th color layer if you tailor the spectral sensitivities of the layers to better match what the human eye sees. Makes sense to me.
bensyverson
Well-known
I'm not arguing with what you're seeing firsthand, but on a technical level, all the "4th layer" does is provide a slight boost in the sensitivity of the magenta-forming layer to a very specific band of "cyan" frequencies.
It's worth noting that Kodak achieves a very similar spectral response curve in their magenta-forming layer without the use of "4th layer" technology.
It's also worth noting that extra sensitivity in the cyan portion of the spectrum does not necessarily mean one film will record a cyan-colored object better than another. The spectra of "cyan" objects may be made up of spikes entirely outside the cyan area of the spectrum. Color is additive after all.
It's worth noting that Kodak achieves a very similar spectral response curve in their magenta-forming layer without the use of "4th layer" technology.
It's also worth noting that extra sensitivity in the cyan portion of the spectrum does not necessarily mean one film will record a cyan-colored object better than another. The spectra of "cyan" objects may be made up of spikes entirely outside the cyan area of the spectrum. Color is additive after all.