Lenses for Film Scanning - article

[ray_parkhurst]

I had an inquiry regarding negative scanning, and was pointed to an interesting article on Pixl-latr about lens tests for scanning done by Richard Karash. I don't know Richard, though it appears he lives and works here in the SF Bay Area. I also don't know if he's a member here. A quick search shows no hits on his name, but that might not mean anything. Anyway, the article links also to his original post on Rangefinder forum. They were interesting enough I thought others might benefit.

https://www.pixl-latr.com/how-good-a-ma ... -scanning/

https://rangefinderforum.com/threads/ho ... g.4791636/

[Sym P. le]

Re: Richard Karesh - for starters, though I haven't dug too deep yet, it looks like he could have some interesting contributions.

https://www.facebook.com/richard.karash/about
https://forums.negativelabpro.com/u/ric ... sh/summary

as ColSebastianMoran on rangefinderforum.com

From Wayback Machine - here and here

Whoopsies, I reposted your links. Sorry.

Regarding "how good a lens do I need", I find it frustrating when I use a "good enough" lens only to see the mush time and again thereafter, always wishing I had a decent lens, and knew how to use it.

[CrispyBee]

It's a bit difficult to understand some of the choices for the lens picks, for example the APO Rodgaon D 2x in normal orientation would only be somewhat OK when digitizing larger formats on a smaller sensor, definitely not for 35mm on fullframe or APSC. And even then it's only a bit of a mediocre performer, requiring to be stopped down to f8.0 to combat CA and spherical aberrations.
Likewise the Sony 90mm 2.8 Macro is not good and suffers from a lot of CA, something that Lightroom and some other RAW editors automatically remove at the cost of resolution. Perhaps that's why it ended up on the list, but that might lead to some unwelcome surprises in (for example) RawTherapee or CaptureOne.
I wouldn't have put those lenses near the Sigma 70mm Macro Art, the Rodagon D 1x or the Nikon CoolScan 8000/9000 lenses - whereas I would have put the 105mm 2.8 Micro-Nikkor AF-D in the Tier1 list instead of Tier2.
And seeing the 105 2.8 Micro-Nikkor AF in the Tier 5 list is especially weird, as far as I know it's the same optics as the Micro-Nikkor AF-D, I think the only difference is that the "D" version transmits focus-distance-information.

So... yeah a strange list.

Regarding "how good a lens do I need", I find it frustrating when I use a "good enough" lens only to see the mush time and again thereafter, always wishing I had a decent lens, and knew how to use it.

You mean for digitizing?

[jmarkus]

Just want to jump in with my opinion on digitizing film via a digital camera & lens - though I also own multiple film scanners as well. I tested a raft of lenses. Dedicated macro/micro, enlarging lenses, tubes, MF and AF - I could get good results out of all of them. One lens was sharper, and resolved more detail than any other lens - the Nikkor 55mm f3.5 ai, or a older 55mm f3.5 Nikkor-P.C. that was factory ai'd. In fact - they were too sharp resolving individual grains. I preferred the look of the Schneider Componon-S 50mm f2.8 enlarging lens. It maintained sharp edges, but had less objectionable grain in the gradients. Due to the volume of film I posses - neither ended up being the main way I scan.. I went with AF, and flatbeds for speed - even then - I will be lucky to live long enough to make it though the pile of film I have left to do.

[Lou Jost]

Just want to jump in with my opinion on digitizing film via a digital camera & lens - though I also own multiple film scanners as well. I tested a raft of lenses. Dedicated macro/micro, enlarging lenses, tubes, MF and AF - I could get good results out of all of them. One lens was sharper, and resolved more detail than any other lens - the Nikkor 55mm f3.5 ai, or a older 55mm f3.5 Nikkor-P.C. that was factory ai'd. In fact - they were too sharp resolving individual grains. I preferred the look of the Schneider Componon-S 50mm f2.8 enlarging lens. It maintained sharp edges, but had less objectionable grain in the gradients. Due to the volume of film I posses - neither ended up being the main way I scan.. I went with AF, and flatbeds for speed - even then - I will be lucky to live long enough to make it though the pile of film I have left to do.

The article highlighted in this post puts the Micro-Nikkor 55mm 2.8 far ahead of the 55mm 3.5. Did you test the 2.8 as well? Did you test any other of his Tier 1 lenses?

[jmarkus]

The article highlighted in this post puts the Micro-Nikkor 55mm 2.8 far ahead of the 55mm 3.5. Did you test the 2.8 as well? Did you test any other of his Tier 1 lenses?

Lou,
I did test the 55mm f2.8 ais as well, and it is also a good lens. In my opinion, comparing my four f3.5 version/copies to my one f2.8 copy - the 3.5's are a frog's hair better image quality and sharpness. The f2.8 has a different helicoid, one extra lens, one extra lens group , and one extra iris blade. It is actually a much more complicated lens than the f3.5, but none of that matters. I'm old and cynical that a handful of frames tested a handful of times on a handful of lenses will actually generate any meaningful data. I can tell you from experience that almost any major manufacturer's macro or enlarging lens will out resolve any film you are likely to throw at it. The reason I responded was that you actually can degrade image quality with accentuated grain due to a lens' resolving power or finding the finest details.
Jim

[ray_parkhurst]

I own an early 55/3.5 and a late one, as well as a 55/2.8. The late 55/3.5 slightly beats the 55/2.8 in the center, but that swaps in the corners. Overall they are close enough that IMO they all should be Tier 1. I find it weird the 55/3.5's are in Tier 3.

One lens I was happy to see in Tier 1 was the 50mm Tominon. I've always liked that lens but it seems not to have a good rep. It's a little short for coin photo systems, but would be great for film copying IMO.

Also IMO the 35mm Tominon should not be in the list at all. Way too short for most copy work. Very good for 3-5x though.

Finally, I'm not sure if Richard put a lot of emphasis on CAs. The 80mm Oly has pretty strong CAs, and IMO should not be as highly rated as Tier 2.

[Lou Jost]

The article highlighted in this post puts the Micro-Nikkor 55mm 2.8 far ahead of the 55mm 3.5. Did you test the 2.8 as well? Did you test any other of his Tier 1 lenses?

Lou,
I did test the 55mm f2.8 ais as well, and it is also a good lens. In my opinion, comparing my four f3.5 version/copies to my one f2.8 copy - the 3.5's are a frog's hair better image quality and sharpness. The f2.8 has a different helicoid, one extra lens, one extra lens group , and one extra iris blade. It is actually a much more complicated lens than the f3.5, but none of that matters. I'm old and cynical that a handful of frames tested a handful of times on a handful of lenses will actually generate any meaningful data. I can tell you from experience that almost any major manufacturer's macro or enlarging lens will out resolve any film you are likely to throw at it. The reason I responded was that you actually can degrade image quality with accentuated grain due to a lens' resolving power or finding the finest details.
Jim

Thanks Jim for your response. To be honest I am concerned more with applications other than slide copying. In most of those applications I strongly prefer the sharpest possible image. And it seems to me that one can always soften an image as much as desired with Guassian noise.

A lens that beats all the ones he listed is the 105mm Printing Nikkor, though the Scanner Nikkor he listed is very close.

[Lou Jost]

I own an early 55/3.5 and a late one, as well as a 55/2.8. The late 55/3.5 slightly beats the 55/2.8 in the center, but that swaps in the corners. Overall they are close enough that IMO they all should be Tier 1. I find it weird the 55/3.5's are in Tier 3.

One lens I was happy to see in Tier 1 was the 50mm Tominon. I've always liked that lens but it seems not to have a good rep. It's a little short for coin photo systems, but would be great for film copying IMO.

Also IMO the 35mm Tominon should not be in the list at all. Way too short for most copy work. Very good for 3-5x though.

Finally, I'm not sure if Richard put a lot of emphasis on CAs. The 80mm Oly has pretty strong CAs, and IMO should not be as highly rated as Tier 2.

I agree. I don't really trust someone who doesn't qualify what he or she means by "best". Robert O'Toole was always careful to give both center and corner evaluations, and to give specific comments on CA. You also were careful about those issues in your tests.

[jmarkus]

Thanks Jim for your response. To be honest I am concerned more with applications other than slide copying. In most of those applications I strongly prefer the sharpest possible image. And it seems to me that one can always soften an image as much as desired with Guassian noise.

Lou,
That is how I ended up with so many 55mm f3.5's. I saw some posted landscapes, and I could tell there was something special about individual leaves being resolved on the horizon at infinity. I asked what lens, and it was a Nikkor-P version of the f3.5 that he had filed off the tab to mount on ai bodies. So my first purchase was the f2.8 ais - thinking "newer is better" - newer coatings, design etc. It did not have the same bite. Every 3.5 I got after that was an older copy and seemed even better - til I got to the factory ai'd Nikkor-P.C. which is still my favorite. One caveat - none of the 55's are focused on infinity at their hard stop. Once I found that little tidbit they became very close in performance. I know - it's weird using a macro lens for landscapes - try it sometime. I just saw and almost EOL serial number on the ai going for $35 last week. Good luck on your search.
Jim

[Lou Jost]

Thanks Jim for your response. To be honest I am concerned more with applications other than slide copying. In most of those applications I strongly prefer the sharpest possible image. And it seems to me that one can always soften an image as much as desired with Guassian noise.

Lou,
That is how I ended up with so many 55mm f3.5's. I saw some posted landscapes, and I could tell there was something special about individual leaves being resolved on the horizon at infinity. I asked what lens, and it was a Nikkor-P version of the f3.5 that he had filed off the tab to mount on ai bodies. So my first purchase was the f2.8 ais - thinking "newer is better" - newer coatings, design etc. It did not have the same bite. Every 3.5 I got after that was an older copy and seemed even better - til I got to the factory ai'd Nikkor-P.C. which is still my favorite. One caveat - none of the 55's are focused on infinity at their hard stop. Once I found that little tidbit they became very close in performance. I know - it's weird using a macro lens for landscapes - try it sometime. I just saw and almost EOL serial number on the ai going for $35 last week. Good luck on your search.
Jim

That's interesting; I used my Micro-Nikkor 60mm as my main landscape lens for decades!

[jmarkus]

Well I guess that just means it is old news to you. I own the 60mm f2.8 afd micro and guess what I used it for - macro shots.