Nikon ES-2 finally launched

[Akira]

The long awaited (?) film digitizing adapter will finally hit the shelves this month:

https://www.dpreview.com/news/6668188440/nikon-s-es-2-film-digitizing-adapter-for-the-d850-will-finally-ship-in-march

[Jakov Minić]

That is good news Akira, thank you for sharing.

[richardHaw]

too expensive for me 

still going to use my modified ES-1 

maybe one day, i will get one from the auction sites for a nice price.

[Ilkka Nissilä]

ES-1 only includes a mounted slide holder (?), whereas ES-2 comes with film strip holder and other accessories.

[Jan-Petter Midtgård]

Review by Jürgen Becker: http://www.throughthefmount.com/articles_rev_es-2.html

There seems to be  at least one serious limitation with the ES-2. You can not scan a mounted slide without putting it in the FH-5 2-frame slide holder, and the FH-5 can only take frames up to 2mm thick!

[Frank Fremerey]

The long awaited (?) film digitizing adapter will finally hit the shelves this month:
https://www.dpreview.com/news/6668188440/nikon-s-es-2-film-digitizing-adapter-for-the-d850-will-finally-ship-in-march

I bought mine the day before I left for Granada. Had it one pre order since July 2017. Paid 169€. A perfect fit for my 2.8/60G Micro on the D850. Will show some slides, as soon as I finished my current editing job. Only a few hundred pictures to go, then DVDs to burn. 30 of the on three drives in parallel.

[Akira]

I bought mine the day before I left for Granada. Had it one pre order since July 2017. Paid 169€. A perfect fit for my 2.8/60G Micro on the D850. Will show some slides, as soon as I finished my current editing job. Only a few hundred pictures to go, then DVDs to burn. 30 of the on three drives in parallel.

Frank, I would like to know if the 42MP of D850 would make difference compared to 24MP or even 16MP, in terms of copying 135 films (slides or negatives).

[Frank Fremerey]

Frank, I would like to know if the 42MP of D850 would make difference compared to 24MP or even 16MP, in terms of copying 135 films (slides or negatives).

At 46MP I guess I will be able to picture the film corn even better resolutionwise ;-)

My hope is more about the color fidelity and differentiation at base ISO. There are some Fuji slides in my collection that show green differentiation in unbelievable fidelity. I am quite positive the Sony 24MP at ISO100 cannot reproduce that.

Do not hold your breath though. Tomorrow afternoon earliest.

PS: I sold my last D600, will get another one as soon as barely used ones hit the market at next to nothing...

[Akira]

At 46MP I guess I will be able to picture the film corn even better resolutionwise ;-)

My hope is more about the color fidelity and differentiation at base ISO. There are some Fuji slides in my collection that show green differentiation in unbelievable fidelity. I am quite positive the Sony 24MP at ISO100 cannot reproduce that.

Do not hold your breath though. Tomorrow afternoon earliest.

PS: I sold my last D600, will get another one as soon as barely used ones hit the market at next to nothing...

Frank, I won't hold my breath, but would appreciate your contribution.

The reason for my question is that BR's old review of D1X suggested that even a 2.7MP D1 already outperformed Fuji Velvia scanned at 2700dpi (roughly equivalent of an 10MP image), and a 6MP D1X image seemed to outresolved Velvia by a comfortable margin.

http://www.naturfotograf.com/D1X_review.html

[Ilkka Nissilä]

If I recall correctly in this comparison the same area of film and sensor was compared but the film was FX size and sensor DX size so if a longer lens had been used to compensate for the different sensor sizes, the result could have been different. If we extend the D1X's resolution to FX size then I wouldn't be surprised to find comparable results. However, a higher resolution scanner provides significantly sharper images. I recall printed demonstrations of Minolta's 5400 ppi scanner and the resulting detail was substantially better than Nikon's 4000 ppi models but depth of field could be an issue. Film has a lot of resolution but it is hard to digitize it perfectly. The Heidelberg Tango scanned at 11000 ppi. Why do you think they'd make that if there was nothing to be gained but large files?

I looked at some Kodachrome 25 slides under a high grade microscope and could be stunningly sharp spider webs in a flower. Those webs were disappointingly blurred when scanned at 4000 ppi. Furthermore the scan looked much more grainy than the image on the microscope, illustrating that the desktop scanner technology was not all that good.

[Birna Rørslett]

I'm not all that convinced film (24x36mm = 135 format or 'FX' in today's parlance) had a higher resolving power than today's top digital systems. I've been studying a lot of slides under high magnification (actually, a significant part of my work routines in the former days of being a scientist) and even Kodachrome 25 wasn't that impressive in terms of detail rendition. Sometimes we used technical (microscopy) film that improved on Kodachrome, but even these would be no match for the better digital cameras today.

Oversampling would be required mainly to reduce film image graininess but the gain in perceived detail sharpness apparently was slight.

The larger film formats had usually lower contrast than 135 and required high-quality film scanners to do them justice.

[Akira]

If I recall correctly in this comparison the same area of film and sensor was compared but the film was FX size and sensor DX size so if a longer lens had been used to compensate for the different sensor sizes, the result could have been different. If we extend the D1X's resolution to FX size then I wouldn't be surprised to find comparable results. However, a higher resolution scanner provides significantly sharper images. I recall printed demonstrations of Minolta's 5400 ppi scanner and the resulting detail was substantially better than Nikon's 4000 ppi models but depth of field could be an issue. Film has a lot of resolution but it is hard to digitize it perfectly. The Heidelberg Tango scanned at 11000 ppi. Why do you think they'd make that if there was nothing to be gained but large files?

I looked at some Kodachrome 25 slides under a high grade microscope and could be stunningly sharp spider webs in a flower. Those webs were disappointingly blurred when scanned at 4000 ppi. Furthermore the scan looked much more grainy than the image on the microscope, illustrating that the desktop scanner technology was not all that good.

Ilkka, thank you for sharing your observations.  I didn't know about The Heidelberg Tango which seems to have offered amazing quality scan images:

https://www.fineartdrumscanning.de/bilder/

An 11,000dpi scan of 135 film corresponds to nearly 200MP FX sensor.  If the DOF problem is non-existent in the drum scanning, and if that bettered image quality significantly, an ingenious film holder should offer the great improvement for the digitizing with DSLRs.

I wonder how the drum scanner could avoid the Newton ring that could occur between the film and the acrylic cylinder?

[Les Olson]

Good B&W negative films resolve between 100 and 150 lp/mm, roughly the same as four micron sensel cameras - the D500 and D850. Colour negative film resolves about 60 lp/mm. I have seen a wide range of figures for Kodachrome, but you would expect it to be between B&W film and colour print film, so 80 lp/mm may be about right. 

So, yes, if you are looking at a Kodachrome slide under a microscope its resolution is not as high as modern digital cameras. But if you are looking at prints, it ain't the film, it's the format. Unaided human vision, under optimal conditions, can resolve about 15 lp/mm. So a 35mm B&W film, or a D850, can make an 8 x 10 print that just puts on the paper all the detail a viewer can resolve.  Medium format film could put 20 lp/mm on an 8 x 10 print. But an 8 x 10 negative can put the whole 100 lp/mm on an 8 x 10 print - vastly more detail than any digital camera.

[Frank Fremerey]

Les: given best lighting, clean air, a high res lens and proper shooting technique...

[Øivind Tøien]

...
I wonder how the drum scanner could avoid the Newton ring that could occur between the film and the acrylic cylinder?

I have no specific knowledge of that specific scanner, but I believe old professional drum scans involved mounting the film with some sort of oil between the film and the drum. So there would be no glass-air-film interface to create newton rings.

Regarding duplication with an adapter (also applies to a scanner), I wonder to what degree the light source (directional vs. diffused) would have on the perceived sharpness?

There is at least one advantage when using a digital body with processing in for example CNX2 vs. editing the bitmapped Nefs/importing TIFFs from a scanner: There is no exposure compensation available in the latter case. (Other raw converters may have that capability, for instance in RawTherapee, but the exposure slider works differently then.) I discovered this advantage when I digitized a number of family albums (by photographing the pages/individual prints) a few years back. It was very nice then to have the full editing capability in CNX2 from the D5100 files. And I was surprised how well some of them came out, even from not so large prints.