Photographing Emulsion side and what resolution?

[Lewipix]

I am embarking on a big project photographing (camera scanning) color negatives, B&W negatives and slide positive transparencies using Sony A7r4 / FE90 Macro / backlight diffused CRI 97 light source. Tethered setup, remote release, and full manual focus and exposure, ISO 100. I am also photographing some older prints (negatives no longer available)

Is film emulsion side photographing always recommended for sharpness?.... If photographing at F8 does it make any difference (DoF) ?

Is 'texture' or grain an issue depending on photographing emulsion vs non-emulsion side? With a 60 megapixels sensor, can there be too much detail of the photographed emulsion surface, or for that matter the paper surface of a photographed print? Lets assume you do want the option of larger print sizes from your 'scanned' media.

Thanks in advance for your experience

David

 

[McDiesel]

I have always digitized with emulsion side up. Never bothered to try and compare doing it the other way around.

However, as an owner of this equipment, I see a couple of other potential issues with your approach:

• FE90 Sony Macro does not have a flat focus field, which leads to softer corners: grain in the corners will look different. You will have to zoom out a bit to keep the negative in the central portion of the frame, losing some resolution.
• F8 is not the best aperture for this lens. f/4 or f/5.6 are better.

In other words, this lens is not a good match for the fantastic 60MP sensor in this application. It can't provide enough detail even for a standard acquisition, and it won't allow you to exploit pixel-shifting. So perhaps you're optimizing the wrong variable here?

Sorry if you already knew all that, just pointing out to the lessons I've learned trying to scan using exactly the same hardware.

 

[Lewipix]

No I did not know that about the FE90 Macro. I thought all macro lenses had a flat focus field ( I was told) but obviously some lenses are softer at the edges than others (is that another way of saying the same thing, soft edges is not having a flat focus field?).

The recommendations I got for the FE90 were pretty good as a macro and it can also work as a regular portrait lens. Nonethelss, thanks for your tip, I will move a little further from the negs. I am not capturing x 1.0 as was leaving some room for the border (to get some orange mask etc)

I can easily stop up to f5.6 so no problem there

The issue with 60mp resolution was that I read "too much resolution" can exaggerate film grain, which prima facie sort of makes sense.

Thanks

 

[JerseyDoug]

In theory, results with the emulsion side up might be very slightly better because of internal reflections in the film base. In practice, which side of the film should be up is much more dependent on other issues such as film flatness and avoiding Newton's rings if there is film to glass contact.

Concerning scan resolution, the scans that McDiesel has posted in other threads here put an end to worries about too much resolution as far as I am concerned.

 

[Alan9940]

I camera scan all my film with the emulsion up to avoid reflections. IMO, you can't have "too much resolution." One of the YouTube photographers that I follow uses a Fuji GFX 100 to scan 35mm and MF; never heard a peep regarding resolution (other than that the results were fantastic!)

 

[gone]

The recommendations I got for the FE90 were pretty good as a macro and it can also work as a regular portrait lens.

That would have concerned me, those are nearly always 2 different beasts. Although I do see how a lens that's soft in the corners might be good for portraits. It's certainly possible, just depends on the lens.

 

[Lewipix]

In other words, this lens is not a good match for the fantastic 60MP sensor in this application.

Just wondering what lens you ended up using with your A7R4 for copying film? If not too much trouble is there a link to a thread as alluded to by @JerseyDoug that might show the differences?

 

[McDiesel]

@Lewipix I've been following https://www.closeuphotography.com/ for a long time, and based on their recommendation and testing I ended up with Sigma 105mm f/2.8 DG DN Macro Art. I compared it side by side with the Sony 90mm macro and it was clearly better in the corners.

I want to be clear here: the Sony is an excellent macro lens, it's just its focus field is not exactly flat at 1:1 distance, which frankly is rarely something macro shooters care about. The blog I linked to doesn't have the best navigation, but he tested all of these (and a bunch of other) lenses at 1:1.

Also, I am a bit puzzled by @JerseyDoug reference, as I haven't shared my work (it's not good and I'm shy the only thing I can think of is my Ilfosol 3 testing in this thread?

 

[Huss]

I tested to see if there was any difference photo scanning (using a Nikon Z7) film emulsion side up or down. I saw zero difference in the results.

I highly recommend using a 1:1 AF lens as opposed to manual focus. It is instantaneous and way more accurate than manual focus. Also stop down to F10-F11 to compensate for any deficiencies in film flatness, no matter how slight it may be.

 

[Moose22]

I tested to see if there was any difference photo scanning (using a Nikon Z7) film emulsion side up or down. I saw zero difference in the results.

I highly recommend using a 1:1 AF lens as opposed to manual focus. It is instantaneous and way more accurate than manual focus. Also stop down to F10-F11 to compensate for any deficiencies in film flatness, no matter how slight it may be.

Exactly my experience. I scan shiny side up now so images face the right way when I review them in lightroom.

I generally go f/10 or f/11. There's no way I can tell the difference between f/5.6 or 8 and f/11, though I don't go too nuts. I think f/5.6 to f/8 is the "best" range for my lens, but a touch more dof solves more problems than the theoretical optimum of my lens.

I'm using a Tokina, which was a phenominal lens for the price, but doesn't autofocus with the FTZ. Tempted to get a native macro, or at least one that works with the FTZ. What are you using?

 

[JerseyDoug]

@Lewipix ...Also, I am a bit puzzled by @JerseyDoug reference, as I haven't shared my work (it's not good and I'm shy the only thing I can think of is my Ilfosol 3 testing in this thread?

Yes, that's the thread I was thinking of. I did not remember the context when I posted.

 

[Huss]

Exactly my experience. I scan shiny side up now so images face the right way when I review them in lightroom.

I generally go f/10 or f/11. There's no way I can tell the difference between f/5.6 or 8 and f/11, though I don't go too nuts. I think f/5.6 to f/8 is the "best" range for my lens, but a touch more dof solves more problems than the theoretical optimum of my lens.

I'm using a Tokina, which was a phenominal lens for the price, but doesn't autofocus with the FTZ. Tempted to get a native macro, or at least one that works with the FTZ. What are you using?

Nikon 60 2.8G macro with the FTZ. You need a G lens to have AF w the FTZ. It focuses on the grain in an instant at your working aperture.

 

[JerseyDoug]

I am using a 50/2.8 Schneider Componon-S enlarging lens with a Fuji X-T20 for digitizing 35mm negatives. Coinimaging.com says it works well at magnifications from 1X to 4X but is a little soft in the corners. Using it with an APS-C sensor can help in that regard.

 

[Lewipix]

@Lewipix I've been following https://www.closeuphotography.com/ for a long time, and based on their recommendation and testing I ended up with Sigma 105mm f/2.8 DG DN Macro Art. I compared it side by side with the Sony 90mm macro and it was clearly better in the corners.

I want to be clear here: the Sony is an excellent macro lens, it's just its focus field is not exactly flat at 1:1 distance, which frankly is rarely something macro shooters care about. The blog I linked to doesn't have the best navigation, but he tested all of these (and a bunch of other) lenses at 1:1.

Also, I am a bit puzzled by @JerseyDoug reference, as I haven't shared my work (it's not good and I'm shy the only thing I can think of is my Ilfosol 3 testing in this thread?

I looked back at the reviews and threads and actually, I did know about the better optical performance of the Sigma 105 DN in the corners. I think the "not flat focal field" thing threw me. For whatever technical reason the Sigma performs a bit better optically by looking at the chart tests etc. Dustin Abbott similarly acknowledges the Sigma's better optical performance but OTOH describes the Sigma's optical performance as fantastic and the slight issues with "distortion and vignetting" in the corners "clean up" well".

The Sigma 105DN macro would appear to be a no brainer for the A7r4 for dedicated copy work. Also a bit cheaper.

The Sigma was also said to be a good portrait and general Macro lens. The Sony FE90 just seemed to win as a better overall performer with things like better autofocus, better manual focus and added image stabilisation, none of which are important for copy work except arguably the clutch method and precision of manual focus.

When I added it all up, and not having the luxury of buying too many dedicated lenses, I guess I went with the 'package' that was the Sony.I think I will get better use from it once finished with my film digitizing project. Never say never but I doubt I will go back to shooting in film/slides. If I did, the Sigma would be the better choice.

Again thanks for reminding me of the limitations of the Sony FE90. I hope to be able to compensate by using the sweet spot aperture and pulling away a bit from the negative to avoid filling the edges/corners of the frame. I suppose I might still get 40 or so megapixels (from the 61 megapixel sensor). I print at 16 x 20" and for example have used less than 40 megapixels in the past with good results

 

[Lewipix]

That would have concerned me, those are nearly always 2 different beasts. Although I do see how a lens that's soft in the corners might be good for portraits. It's certainly possible, just depends on the lens.

Thats interesting but I guess i would question why. is it due to mutually exclusive and competing engineering requirements or simply a practical cost issue. My impression is its the latter. Both the Sony FE90 and Sigma 105 DN lenses are said to be great macros and can double as good portrait lenses

 

[MattKing]

FWIW, traditionally macro lenses were evaluated primarily on their flat field performance - are the corners as well resolved with good contrast as the centres of flat, two dimensional subjects?
And I'm sure that some still evaluate those lenses that way.
But I think there are also now people who evaluate modern "macro" lenses on how well they perform at close distances with non-flat subjects, as well as how well they perform at portrait and farther distances.
For film digitization, one needs really good 1:1 flat and near field lens performance at close distances. And one is quite happy if the lens is absolutely crummy at longer distances.

 

[Lewipix]

For film digitization, one needs really good 1:1 flat and near field lens performance at close distances. And one is quite happy if the lens is absolutely crummy at longer distances.

From all accounts the sigma 105 DN does a good job at both flat field 1:1 as well as portrait distances. I suppose (rightly or wrongly) every lens would ideally have a good "flat focus field" if that translates to good edge to edge, corner to corner sharpness. I am just guessing it is more important in macros where the field of focus is slither thin.

If a lens nails it at close 1:1 distances with sliver thin DoF I would have assumed the great optics involved would also benefit portrait or further distances, or at least not be a disadvantage in any way.

Where the Sony macro loses a bit with IQ in the corners up close for copy work on a stand/tripod, as compared to the Sigma macro, it arguably is a better or as good performer in general macro work (with 3D objects centred in the frame) and has additional and costly features for better portrait and overall performance eg better focusing and stabilization. If that means you get a better focused image and less image shake, it likely trumps any small corner sharpness issue, to the extent it manifests.

Maybe a Sony FE macro G Master is the answer but if it comes it will likely be x2 or x3 price.

Thank you Matt, I appreciate your reply.

I guess we must all learn to use what is available to us. TBH I am blown away by the performance of the FE90 Macro in the very brief experience thus far.
David

 

[MattKing]

If a lens nails it at close 1:1 distances with sliver thin DoF I would have assumed the great optics involved would also benefit portrait or further distances, or at least not be a disadvantage in any way.

The best lenses I've ever worked with were designed for reproduction work in a graphic arts environment. Absolutely flat field response at 1:1 and similar magnifications - perfect for the requirements of magazine, book and newspaper publication. Designed to image to film the size of a newspaper broadsheet.
In most cases though, those lenses were slow, extremely large and heavy, and potentially not particularly resistant to off-axis flare. They also were likely to be mediocre at portrait and longer distances (there are exceptions).
And those lenses would have been much, much more expensive than the Sony or Sigma lenses you refer to.
When you design a macro lens to also perform well at other distances, you are forced to accept compromises - some of which are substantial. The characteristics that ensure flat field performance with even illumination at high magnifications tend to reduce performance at farther distances.
High end enlarging lenses are a much better candidate for film digitization.
Here is an example of a lens that I would like to use for digitization:
OM Macro bellows 80mm f/4 lens
You can't even focus that lens if it is mounted on a camera body only.

 

[Lewipix]

The characteristics that ensure flat field performance with even illumination at high magnifications tend to reduce performance at farther distances.

Thanks Matt. hopefully my approach mentioned previously, will mitigate those trade-offs and when my digitization work is done I will be left with a good macro lens for real world (non flat) objects. the other option would be to sell the Sony and buy a Sigma.

 

[250swb]

I use a Nikon Z7 with a 60mm D as I'm not bothered about auto focus, and whether the scans are shiny side or emulsion up it doesn't make any difference to quality, except if you have a curling negative use ANR glass to flatten it. In fact manual focus is useful to warn of a curly negative or if it's not sitting flat for some reason because you always know if the focus has changed, the little box will go from green to red whereas with AF it will always be green.

 

[Adrian Bacon]

I'll throw my 2 cents in here. I digitize with a Canon EOS R5 and Sigma 70mm ART Macro lens. It's an excellent combination. I always digitize emulsion side up. Emulsion side down tends to be ever so slightly softer, but you have to zoom in to see it. If you're outputting at smaller file sizes than the native resolution of your camera, you'll never see that difference. Other than that, the only other guidance I'd give is open up your aperture to f/5.6. I can't say anything about your particular lens, but if you want to go with a different lens, the Sigma 70mm Macro ART lens (the newer one) is among the better lenses out there. I've used several macro lenses over the years and this one so far has been the best.