New Nikon ES-2 Film Digitizing Adapter

[Ed_Ingold]

It's easy to copy slides with a digital camera and match the color. The same digital camera will produce images quite different from the same scene on its own. It would seem if you want "natural" color, film isn't the best choice.

 

The quality of a scanned or copied negative depends a lot on the negative. Much as slide film does better if underexposed about 1/3 stop, negative film works better if overexposed at least by the same amount if you want good detail in the shadow areas. Highlights take care of themselves, unless you go totally berserk with settings. The opposite is true with slide film and digital.

 

One of the options in Photoshop "Curves" (ctl-M) is "Color Negative". Sometimes it works, other times not so well. Color balance in negative film is highly sensitive to the exposure.

[JDMvW]

If one can find a way to couple one to a digital camera there exists a perfect solution!

 

Coupling is not a problem, most use one or another from of T-mount or specific camera adapters which are cheap.

It's the "perfect solution" that is the problem :p

[PapaTango]

Coupling is not a problem, most use one or another from of T-mount or specific camera adapters which are cheap.

It's the "perfect solution" that is the problem :p

 

Honey, you have just made my DAY! If you were in town, I would hug you and buy you dinner... :eek:

 

A short search on Ham-azon showed me a few adapters. Yes, the T mount needed to put my gnarly old Spirotone back into service--but another gem of inestimable value!

 

Whatsit? Why a Fotodiox Nikon F mount to Canon FD/FL lens with aperture control!

 

https://www.amazon.com/dp/B003EB0H5U/ref=psdc_3347851_t2_B00893PX66

 

I have a boatload of prime lenses in my A-1 kit--all of which I was seriously thinking of selling off. $45 later, I now have both mounts arriving at my secret lair--on or before September 1... :cool:

[JDMvW]

Honey, you have just made my DAY! If you were in town, I would hug you and buy you dinner

 

One of the reasons I went to Canon EOS for digital was the ability to use my old non-AI Nikkors and many other mounts of lenses in stop-down mode.

I have always found the Fotodiox brand to be reliable, although even cheaper and almost always OK adapters are available on eBay and such like. [avoid the "focus confirmation" versions as the plague]

 

Unfortunately, the adapters for the nFD mounts to EOS require a minus-lens (or shooting only macro) and quality is lost on an EOS camera. Like the later EOS, the FD mount will accept a broad range of other mounts if you can find the right adapters.

[Fiodor]

Fiodor - yes, the method is great for slides, not so much for negatives - black and white or colour

 

at least that's what my experience shows

 

and yet again you have to fight to get the right colours afterwards editing

 

otherwise - you can make NEF files using the ES-1/2, but as I already said - the best result I get when using the HDR function in the 810 camera

 

we have to wait to see exactly how this new feature is working in 850

 

So, there is no information…

 

If the camera only reverses, it is not too helpful. And if it does some kind of automatic adjustment, like a scanner would do, it should create a RAW file for further editing.

 

In fact, the ideal option would be that the scanning software does this particular job of properly reversing RAW files gotten from ”DSLR scans”. Is there any application which does this job? Or even you could do the settings before actually shooting the photo, the camera connected to the computer and truly acting as a film scanner, but probably I am asking too much, or most probably what I just said doesn’t make any sense.

[Wouter Willemse]

nd if it does some kind of automatic adjustment, like a scanner would do, it should create a RAW file for further editing.

 

That's not really possible - once a camera starts editing, the data is no longer raw. It could probably export the edited data in a container format like dng, but the actual data would be demosaiced, and no longer real raw data.

 

You could shoot in raw, and use something like ColorPerfect afterwards to handle the inversion with good curves. I never used it but it's reported to deliver good results.

[rodeo_joe1]

Les. You've obviously studied "How to win an internet argument 101"

 

1. Provide no cogent counter argument, but make an unwarranted personal attack on your chosen "opponent".

 

2. Insult and patronise anyone who disagrees with your unsupported opinion to put them off guard.

 

3. Disregard any evidence that contradicts your viewpoint and refuse to acknowledge that anyone other than yourself has any skill or expertise.

 

I've wasted enough of my life doing film v digital comparisons thanks. For some strange reason the quality of result I get from digital is always far superior to that from 35mm film. Despite the person behind the viewfinder remaining the same, and using the same camera techniques.

[Sandy Vongries]

Another strategy would be to make a collar to fit the outside with a step-down ring as an integral part of the plastic collar. We used to do such things for special applications at a place I worked nearly a half-century ago--but this requires some machining skills and a metal lathe.

Actually, since the gadget would be used only occasionally and at home, and wouldn't need to win beauty contests you could follow the collar idea by using an adjustable pipe strap from the hardware store. Just get the right size apply some tape or other gasket to the inside and gently tighten the screw. Since the ES1 is designed for a 60mm, the distance media to lens might even help.

[Ed_Ingold]

There are ways to fix a dented tube which might be uses to repair a filter ring. One method would be to make an expanding swage. Unless there is an existing tool, it would cost more to make than to replace the lens. An alternative would be to salvage the filter ring in a damaged lens and refit the dented one. This too would be an expensive choice, bur probably less than a new lens.

 

If you want to pursue copying slides and negatives in lieu of purchasing an high quality scanner, now of unicorn scarcity, the cost of a new 60 MM macro lens is less than 1/3rd that of an LS-4000 when it was new, 15 years ago or so. Used Nikon 55/2.8 lenses are still relatively inexpensive, and IMO better for this purpose than an AF model. Either way, a good macro lens is very useful for copying artwork and for closeups of stationary objects (e.g., stamps, coins, jewelry). If you prefer a longer lens, which is better for general purposes, the ES-2 is not for you. A copy stand with a light box would work (in lieu of cardboard and gaffer tape0, starting at about $500.

 

Nikon Nikkor 55mm F/2.8 Micro AIS Manual Focus Lens {52}

[Ed_Ingold]

I've wasted enough of my life doing film v digital comparisons thanks. For some strange reason the quality of result I get from digital is always far superior to that from 35mm film. Despite the person behind the viewfinder remaining the same, and using the same camera techniques.

Les does an admirable job scanning film, and has it down to a science. He is no more evangelical about film than most of us are about digital. Just a thought, though. Much as scanners are becoming extinct, so are their repair services. Think of the ES-2 as "Plan B."

[Greg Fight]

I have a copy stand. I could fashion some sort of light source and be off to the races. Since I can't attach the ES-2 to my 60mm, I'll save $139.95 from the start. My D300 will give me enough resolution for my purposes. This is my Plan B. My primary slide/negative scanning is done with my old CanoScan 2710. Again, not optimum but sufficient resolution. Occasionally there is a problem slide - too dense or problematic color - where I want to copy it with the DSLR.

 

Thanks everyone for the advice.

[Ed_Ingold]

You can get LED light tables relatively inexpensively. You might even make one from a light panel and a sheet of opal plastic or glass. I have an old light table with fluorescent lamps, but the spectrum was so uneven that I could't get good color even with slides. LEDs are much better. Light tables made for copy stands are expensive, but probably the best in the long run.

[rodeo_joe1]

Due to the hardness of the Dural that old MF Nikkor lens barrels are made of; I don't think it's feasible to swage a dent out of a filter ring and still retain the thread. Maybe Aralditing the ring of an old filter to the front would work?

 

I still think Nikon's ES-x options can be bettered, and at a lower cost. The ES-x has no way of varying its length that I can see, and so no way of varying the reproduction ratio.

 

I have a Sunagor (was mistaken about it being Aico) tube + slide/filmholder that does exactly the same job as an ES-2 - except the tube has a helicoid and extends.

 

There's no need for a light table. All I do is mount a speedlight on the camera hotshoe, point the whole assembly at a sheet of white card, and let the reflected flash illuminate the slide or film via the opal diffuser. I usually need about 1/4 power on the flash, and fine-tune the exposure with the lens aperture. Easy-peasy.

[Ed_Ingold]

I still think Nikon's ES-x options can be bettered, and at a lower cost. The ES-x has no way of varying its length that I can see, and so no way of varying the reproduction ratio.

What's this again? I actually have an ES-1, and the length can be varied nearly an inch. I also have a set of Nikon K tubes, 1 through 5, which can increase the distance between the lens and the slide copier by over two inches. I said as much in a previous post, but it bears repeating.

 

I'm sure there are other adapters which would work as well, but I was not able to find any still being made, other than cheesy versions which relied on simple lenses rather than a sophisticated macro lens like the 55/2.8. A key advantage of the Nikon ES-1 is that it screws into the lens, making the entire copying assembly rigid and stable, once set up. It is not necessary use a tripod, since long exposures are not subject camera shake relative to the target. Furthermore, the entire assembly is light tight, except for light passing through the film, so there is no loss of contrast due to ambient light.

 

You can use a flash, but it is not necessary. The spectrum of a simple daylight LED light bulb is easily accommodated within the range of white balance of the camera. Using the desk lamp (which I reflect from a tilted white card for uniformity, makes it much easier to align and focus the film. It's easy enough to calibrate the color using an X-Rite Passport, but it doesn't seem to be necessary. Automatic exposure also accommodates variations in the slide's exposure, and is more reliable than AE using a flash.

 

If you use a copy stand, a light table simplifies the operation. The light is continuous, making setup and alignment easier to check, and it provides a flat surface to hold the film. Using an ES-1 or ES-2 eliminates the need to use a copy stand, unless you want to copy film larger than 35 mm.

 

I have a Novoflex focusing rail with a copy attachment, which serves as a copy stand without the stand. I can copy film up to 6x7, using a Sony 90/2.8 Macro or a Nikon 105/2.8 AFD Macro. Focusing and adjustments of the film holder are via a rack and pinion. You must take precautions, as with a copy table, to exclude as much ambient light as possible. I can also use a digital Hasselblad, and use a compendium (bellows) light shade to bridge the gap between the lens and film holder.

 

Next time, please be more judicious in your description of things with which you have never used nor carefully thought out.

[rodeo_joe1]

Ok. I admit that I was mistaken over the ES adapters, but I did qualify my assumption with the words "that I can see".

 

Perhaps if Nikon provided better pictures of the item, or a better description, then such errors might not be made.

 

The ES-2 is still an overly expensive item if the full kit is bought.

[tom_halfhill]

I just posted a message about the ES-1 versus the new ES-2 in the Digital Darkroom forum. I won't repeat it here, but I recently modified an ES-1 to copy film, and buying two ES-1 units (one for slides and one for film) is cheaper than buying the ES-2. Also, some folks are unaware that the ES-1 changed over time. Older models have enough clearance for a film-strip holder without modification, so only one unit is needed to copy slides or film. Recent ES-1 units are tighter and can't accommodate a film-strip holder without modification. Believe me, I tried, before giving up and buying a second one.

 

Regarding quality, my copies of b&w film using my modified ES-1 with a Nikkor 40mm f/2.8 macro lens are plenty sharp and easily resolve the film grain. My scans using a Nikon CoolScan V are no better and take much longer. Also, I have successfully used my ES-1 to copy old negatives that were in such extremely poor condition (actually crumbling) that I wouldn't dare to feed them into my CoolScan for fear of jamming it.

[glen_h]

 

(snip)

 

In fact there's nothing new at all in using a DSLR to copy slides/negatives. It can be done with any camera with a megapixel count in excess of 20 plus a decent macro lens. Since most film is only capable of a resolution around 4000 pixels per inch, a 4000 x 6000 pixel DSLR (either DX or FX) will scrape all the detail off a 35mm frame that it's capable of holding.

 

It is getting harder to find the film data sheets, but here is Portra 400:

 

http://imaging.kodakalaris.com/sites/prod/files/files/products/e4050_portra_400.pdf

 

The MTF curves go out to 80 cycles/mm, or about 4000 pixel/inch.

 

The curve isn't down to zero, but that is the point that it stops, so it seems that is what Kodak (or KodakAlaris) thinks it should be.

 

Usual film scanners use line CCD arrays, either three rows of them, or one row and three light sources.

 

For the DSLR, you do have the Bayer array filter, so you don't get the full resolution at each color.

 

I believe the data sheets for other color films that I looked at also go out to about 80 cycles/mm.

[glen_h]

Tri-X: http://imaging.kodakalaris.com/sites/prod/files/files/resources/f4017_TriX.pdf

 

only goes out to 70 cycles/mm on the curve, which is at about 25% response.

 

Again, I suspect that 4000 pixels/inch is about right.

[Ed_Ingold]

I recently modified an ES-1 to copy film, and buying two ES-1 units (one for slides and one for film)

The only film strip holder which fits my ES-1 is the FH3 which came with an LS-4000 scanner. It is about 50% thicker than a cardboard mount, but only 2" wide. It's a little hard on the leaf springs in the ES-1, but workable. Inexpensive film strip holders for other slide copiers (B&H) are 1/8" thick and 1/4" wider than a mounted slide. They won't fit in the ES-1 even with the springs removed, and place the center of the slide about 1/8" too high. Modifying the ES-1 would be a lot more involved than installing a spacer. In a nutshell, that's why the ES-2 is so welcome.

 

The MTF curves go out to 80 cycles/mm, or about 4000 pixel/inch.

80 lp/mm is for a target with 1000:1 contrast. It's more like 40 lp/mm with a normal scene, which has a contrast ratio of about 6:1

 

For the DSLR, you do have the Bayer array filter, so you don't get the full resolution at each color.

True, but that's for color. You get pixel for pixel resolution for luminance, which affects accutance more than color.

 

Nikon scanners do a very good job with film, in terms of resolution and more important, film handling and negative conversion. In my experience using a 24 MP camera comes very close in resolution, and is better at renewing color. If you dial in the fact that using a camera is p to 10x as fast, the game is over. Furthermore you could by a Nikon D850 or a Sony A7Rii for the cost of a used Coolscan, and have change left over.

 

The downside is I'm to happy with the quality I get converting negatives to positives. I wasn't all that happy with the Coolscan software either, but it's faster and generally better. It's just a matter of time before someone writes a plugin that nails the issue down tight.

[iosif_astrukov]

Fiodor - what you talk about is more like a photoshop plug in, from Nikon, which could convert the negative images using the presets of the Coolscans, for example, or other similar method - indeed such thing would be great

[Fiodor]

That's not really possible - once a camera starts editing, the data is no longer raw. It could probably export the edited data in a container format like dng, but the actual data would be demosaiced, and no longer real raw data.

 

You could shoot in raw, and use something like ColorPerfect afterwards to handle the inversion with good curves. I never used it but it's reported to deliver good results.

 

Thanks. I didn’t know this plug-in, it looks interesting. Have you tried it with dsrl negatives?

 

Yeah, you are right, if there is editing, it is not Raw anymore, unless the changes are Raw variables and it creates also a .Xmp file (but then again, we don’t know what the d850 would be doing when it reverses negatives). Anyway, probably the d850 saves also as Tiff, which is preferable than Jpeg for editing.

[Fiodor]

My doubt was how scanners treat the orange mask, if they do it by hardware or software. Then I read this article through the Nikon F-Mount - How to digitise your slides and negatives which says “Colour negatives have a strong orange mask. Formerly, I thought the scanner would correct it with a special mix of its RGB light components. But I was wrong - it is just software.” So, apparently, it is merely a software job, and we are dealing with an orange tinted image from the start, which if you think about it, is terrible, specially because this orange is not a Raw color temperature variable.

 

(I still don’t know how this orange mask is apparently useful for printing, and how color enlargers and photographic papers work in relation to this mask, but this is another story)

 

The author of the aforementioned article recommends Vuescan for doing the inversion. My question would be: if scanners treat the orange mask only with their scanning software, can we get exactly the same results with a photographed negative than with a scan, using Vuescan in both cases? If not, what is so different in these initial images from which the software works?

 

Then I read this Removing Orange Mask from Colour Negative Scans - Filmshooting . The idea is to find a method to reconstruct the yellow and magenta masks in order to subtract them from the image. It makes sense, because the masks are not flat signals. Below in the comments the author says that he was partially wrong about the theory, but the practice would be still the same. I read until the first page of the comments, so I don’t know if there is something interesting later. What do you think of this? I am sure we can find on the internet more methods like this, but before investing more time exploring, it would be interesting to know your opinion on this.

 

At the end of the first page of the comments, another poster suggests this Home > Resources > Film conversion a more rudimentary method, just using the orange of the film leader to compensate the color cast (so it doesn’t take into account the non-flat characteristics of the masks). The article also suggests Vuescan and Digital ROC, a Kodak plug-in.

[Fiodor]

Using an ES-1 or ES-2 eliminates the need to use a copy stand, unless you want to copy film larger than 35 mm.

 

Ed, so basically you recommend a slide adapter for 35 mm, and a copy stand only for larger formats?

 

Using a slide adapter for 35mm, could we photograph different fragments of the image for stitching them later with software? (I mean, does the adapter allow you to move the position of the frame?)

 

Is it possible to insert somehow a medium format holder in an ES-1?

 

And what about DX cameras, are they as useful as FX ones to do this digitizing work?

[Fiodor]

Fiodor - what you talk about is more like a photoshop plug in, from Nikon, which could convert the negative images using the presets of the Coolscans, for example, or other similar method - indeed such thing would be great

 

Yes, sure. If Nikon doesn’t have the balls to keep producing scanners, the least they could do is developing a decent software to treat dsrl scans in combination with their cameras and slide adapters.

[tom_halfhill]

Ed_Ingold said:

"Modifying the ES-1 would be a lot more involved than installing a spacer. In a nutshell, that's why the ES-2 is so welcome."

 

Actually, modifying the ES-1 was easier than I thought. No spacer was required. I just peeled off the label to expose the two screws holding unit together, removed the screws, bent the two metal clips to make room for my film holder (which came with my Nikon CoolScan V), and then screwed it back together. It took only about 15 minutes, and most of that time was experimentation to fit the film holder. Had I documented the process with photos, I could have made an instruction guide that would enable anyone else to do it in five minutes. The reconfigured clips tightly squeeze the film holder and keep it in place while shooting dupes.

 

When using a DX-format camera and 40mm macro lens, the ES-1 is a better solution than the much more expensive and complicated ES-2. And as I noted before, buying two ES-1 units (one for slides, one for film) is cheaper than buying one ES-2.