I’m doing some troubleshooting to figure out the source of my pale (inverted) corners and edges that are very obvious in images that have dark backgrounds. I want to try to narrow down the causes as well as figure out what is actually normal procedure, since I’m new to camera scanning.
I was getting uneven density, especially at the sides and corners of darker images (think flash at night, dark background), which I initially attributed to my Nikon ES-2 being perhaps too narrow. Today I received a Valoi Easy35, which supposedly has the unevenness bugs worked out, and I’m getting similar edge/corner issues.
Equipment used: Sony A7Rii and Sigma 70mm Macro Art, usually shot at 5.6 or 8
Light source: initially Viltrox 116 with the Nikon ES-2, now Valoi Easy35
Questions:
Is it normal to have to apply a lens correction in LR for this lens? If I use the lens correction option and set it to 200 for vignetting, it’s mostly fixed.
If the lens is the issue, is there a lens that maintains the impressive sharpness and general “wow” factor of this Sigma, but without the vignetting?
I noticed that if I correct in NLP vs NegSets/Negmaster/PS, it is less noticeable. I then discovered that NLP enables LR’s lens correction automatically. That seems to tell me that this is normal behavior and I’m being too picky (not the first time).
Any insight as to where I should place my expectations? Is there a real problem here, or is this standard procedure? I don’t mind making these corrections part of my workflow, but if it is an abnormal equipment issue, I’d rather fix at the source.
If the lens of your “camera scanner” is the cause…you’d need an other lens - or simply correct the effect as far as it makes sense. Alternatively, you could check out DxO PureRaw and test it with the free trial in order to see if it does what you need.
Lens aberrations are expected, unless you invest in a really good (and hence expensive) gear.
I think the surprise is that the Sigma 70mm f2.8 ART Macro is widely accepted as one of the best lenses that can be used for ‘camera scanning’ on full frame, it’s in Richard Karash’s top tier for example as well as being rated very favourably on closeuphotography.com. It is widely used on this and other forums and no-one has mentioned getting problems like this to my knowledge. I’m baffled by these issues because if they were caused by the lens and others were getting them they would surely have reported them on here and that lens wouldn’t have this excellent reputation. I’d like to think that it’s something else that’s causing it but it’s hard to think what that might be given that it is apparently the same on both the ES-2 and the Valoi easy35.
I’m considering the possibility that everyone else just knows that you need to apply a lens correction profile, or since this is the NLP forum and NLP handles it automatically, they don’t see the issue. I only noticed it when I started trying out Negmaster.
I found a lens review somewhere after posting this that had a vignette test of this lens, and the pattern matched exactly what I’m seeing, so I guess that’s that.
I’ve implemented a default lens correction into my Lightroom preferences, so it’s gets applied before I do any cropping, etc. So far it has solved the problem, so I think this is just normal (right?).
Thanks. I’ve looked into that before, but I wasn’t sure if you need to do it every time or not. The instructions seem to indicate that you need a reference frame for each actual image, which sounds rough.Also, the canned profile seems to fix the problem.
I don’t mind upgrading (eventually) if there is something available that beats this lens. The only thing I want to avoid is spending a lot of time on rigging a process lens that wasn’t intended to mount on a camera. Unless there is a simple adapter to do that out there.
No, you can take just one reference image for an entire setup (camera, lens, light, and so on) and use it as long as you don’t change your setup or every time you start a scan session.
With Valoi easy35 it’s very simple, just remove the film guide and take a picture. That’s all !
Here’s a result with a negative from 1985 (Kodacolor GA 100) scanned yesterday with a Canon EOS 6D (gen 1), Sigma 50mm f/2.8 DG Macro EX and the Valoi easy35 with Flat Field.
If you find that the vignetting is circular and more intense in the corners, then it is due to lens vignetting. Shooting at 5.6, I would definitely expect some vignetting. Remember too that the conversion process adds a good amount of contrast, so I would expect any inherent lens vignetting to basically be doubled during the conversion. This lens correction in Lightroom is really useful, and that’s why it is automatically applied by Negative Lab Pro.
If you find that the edges are more uniformly pale on some or all sides, then it is due to your film holder setup. Basically, if your light source is being diffused too far away from the underside of the film carrier, it will cause a very faint shadow on the edges, which will be magnified by the contrast added during conversion.
There are two things you can do to at least narrow down the problem.
First off, look at the negatives! I was going crazy trying to figure out where I was getting pale edges from, and then I looked at the physical negatives and realized they were in fact slightly darkened right at the edges. Not in a circular way, but right along the border in straight lines. I am not sure why that is the case (my negatives were developed by a lab) but it was quite clearly an issue with the negative itself and not my film holder / scanning setup.
Second, shoot a blank white wall at f/5.6 (or whatever your preference is) with your Sigma, and take a look at lens corrections on/off in your photo editing software (LR or otherwise). That will show you exactly what vignetting is being added by the lens, if any.
I always use lens corrections in Lightroom, for lenses with profiles (which IIRC the Sigma does have, I own the same lens). It’s basically free optical improvement in the digital era, and one reason why lens manufacturers nowadays tend to not prioritize vignetting and distortion in their designs, because of how easy it is to correct in post (unlike in the film days). Optical design like all engineering is a collection of tradeoffs, and eliminating vignetting entirely would probably require significantly bigger, heavier, and more expensive glass.
All that being said, I’d honestly be surprised if the Sigma had very serious vignetting at f/5.6, even before correction (and certainly not after). I’ll probably take it out of the closet later to test for myself.
