[Kobayashi.K]

Staff note: This post may contain affiliate links, which means Pentax Forums may earn a small commission if a visitor clicks through and makes a purchase. If you would like to support the forum directly, you may also make a donation here.

---

Funnily enough, I was trying to find a source for a decent test negative yesterday I've found one seller on eBay that does them in various formats, though they're really designed to aid focusing rather than test resolution. Do you happen to know of a supplier that offers test negatives?

I searched on Fotowand in Germany (fotowand.de) but I could not find one but perhaps I didn't look accurately enough. I also searched a few other shops without success. Perhaps you can also photograph a standard test card and use that negative, but it must be inverted to get back the positive.

Edit 18-01-22 - I found one at ScanDig Germany (www.scandig.de). Select tab Miscellaneous, LaserSoft Imaging USAF-1951 test chart scanner resolution 35mm, in 35mm slide mount (5x5cm). Or search 'LaserSoft'.
There is also a page on ScanDig with an explanation how to use this target (The USAF-1951 test target):
https://www.filmscanner.info/en/Aufloesung.html

---

Last edited by Kobayashi.K; 01-18-2022 at 03:19 PM.

[reh321]

in fact Lightrooms spot removal is adequate in most circumstances, whereas the scanner always required me to head to Photoshop and the clone tool but it was low end and had no ICE.

How does LR remove spots?
Does it handle scratches also?

[mir]

How does LR remove spots?
Does it handle scratches also?

It over writes/fills the spot with the surrounding texture - both for spots scratches.

[3by2]

How does LR remove spots?
Does it handle scratches also?

As Mir says it effectively clones or heals an area by looking for a similar pattern elsewhere and substituting it. You have control over how it chooses that area and it works reasonably well, very well on most spots. Larger areas like scratches can be trickier depending on whether the scratch is confined to one particular area or whether it overlaps differing and distinct areas. It doesn't work if you have say, a lot of fine scratches close together, you'll need Photoshop for that. This quick five minute tutorial is a useful overview.

You Tube

You Tube

[BigMackCam]

How does LR remove spots?
Does it handle scratches also?

It over writes/fills the spot with the surrounding texture - both for spots scratches.

As Mir says it effectively clones or heals an area by looking for a similar pattern elsewhere and substituting it. You have control over how it chooses that area and it works reasonably well, very well on most spots. Larger areas like scratches can be trickier depending on whether the scratch is confined to one particular area or whether it overlaps differing and distinct areas. It doesn't work if you have say, a lot of fine scratches close together, you'll need Photoshop for that. This quick five minute tutorial is a useful overview.

How To Remove People or Objects in Lightroom (Spot Removal in Adobe Lightroom) - YouTube

On the subject of dust and scratch removal...

I've seen results from flatbed scanners using ICE, and they've varied from excellent to not-so-good, in terms of the final image quality. There's a softening effect, but @LesDMess posted some examples for me recently that looked pretty good overall, with minimal softening - and the automatic nature of it is very appealing... but, of course, it only works on colour images.

I really like the ability to easily clone out dust and scratches using Lightroom, Darktable, GIMP and other photo processing / editing tools, but it seems to me that the best approach is to ensure negatives (or transparencies) are as dust-free and in as good condition as possible prior to digitising, and the digitising equipment - copy stand, light panel, film holder etc. - cleaned before use to ensure no (or minimal) dust and fibres can transfer to the negatives. That keeps the amount of cloning needed in post-production to a minimum. Wet wipes and a rocket blower work wonders

[Kobayashi.K]

I have never spots or dust or scratches on my negatives, so I guess I am a neat person. It saves a lot of work. The only things I clone out are power lines, contrails and birds, and that is done by hand.

[LesDMess]

I'm wondering how those scanners do with heavy shadow areas, particularly on any slides such as Kodachrome. My old, lower end scanner, a Dimage Scan Dual would just fill those areas with red noise and banding. In fact I'm getting so much better results now, I've largely given up on the idea of another dedicated scanner. There might be one or two slides I'll consider getting drum scanned rather than trying to find something better than I have.

Most of mine are in good condition and date back 40 years but the older slides have picked up marks and spots I can't get off. What I have found is that camera digitising is a lot cleaner than scanning and doesn't seem to pick up dust and scratches in the same way a scanner does. I'm spending far less time cleaning up than I used to using a film scanner, in fact Lightrooms spot removal is adequate in most circumstances, whereas the scanner always required me to head to Photoshop and the clone tool but it was low end and had no ICE.

If you put an intense light behind your Kodachrome's deep dark shadows and see no information then save your money as an expensive real drum scan won't find any either.

In one scan/dslr capture of a frame of film, you may have to use post in order to bring out detail from shadows/blown highlights if you want/need those. For example in this single frame of Kodak Portra 400 - with deep shadows and blown highlights, I had to use post to bring out detail from those areas. In some cases I may need to adjust scan/dslr exposure to retain the blown areas.



Keep in mind that most color negatives and b&w film have practically endless overexposure range.

OTOH, slides not so much and the shadows can get really dense particularly Kodacrhome and Velvia and perhaps others too. Again you can try post shadows/highlights on a single frame or even an over and under scan mixed together ala HDR in post. Below is a frame of Velvia (RVP50) and I did a normal, +1 & +2 scan and combined in HDR compared to post shadows/highlights on the normal scan. In this case you can see that a lot of shadow detail can be brought out.



I am familiar with the usual clone tool and have gotten adept at using it over the years but I cannot match the results provided by the Coolscan 9000 + Nikonscan in quality and time. The overhead in scan time with ICE is very small. In this example of a perfectly scratched Kodak 160VC, I show the DSLR copy (no ICE) compared to the Coolscan's results without as well as the two levels of ICE.



I would be very interested to find out how good LR's spot removal works on something like this.

---

Last edited by LesDMess; 01-18-2022 at 08:58 PM.

[LesDMess]

Funnily enough, I was trying to find a source for a decent test negative yesterday I've found one seller on eBay that does them in various formats, though they're really designed to aid focusing rather than test resolution. Do you happen to know of a supplier that offers test negatives?

That aside, the fact that it requires a test negative rather than a real working photographic negative to see the effects of diffraction at the aperture I've chosen kind of justifies my compromise to some extent. I know the results are softer at f/10 than f/5.6 or f/8, because the MTF50 chart for the lenses I'm using tells me so... but if I don't see it in my digitised negatives at 1:1 reproduction (let alone smaller repro sizes), it seems like it's not much of a practical issue.

So I bought my used Pentax SMC Macro 50mm f4 lens so cheap on a local CL and noticed how sharp the results were so I wanted to see just how high resolving this was using Kodak Techpan # ISO25 processed in Kodak Technidol. I scanned the various frames captured at different apertures with my Coolscan and it looks like the 4000dpi is the limitation at apertures f4 - f11 and starts loosing detail at f16, f22 and f32.

I then digitized frame 04-27 with my K20D + this macro lens and of course it is sensor limited from f4 to f8 and starts dropping down from there f11 - f32. Assumably due to diffraction.



The large crop to the right is the optical magnification of frame 04-27 at about 4.5X and clearly you can see detail far beyond what the K20D and Coolscan could resolve. I believe this is what you are looking to do?

[stevebrot]

Oh! My! Such a discussion! I will limit my comments to that which I have hands-on or well-researched experience with.

1. RE: the Sigma 50/2.8 EX DG Macro and sweet spot. Yes, f/5.6 - f/8 is where it does best.

2. RE: DOF for copying 35mm negative to APS-C digital. In theory, needed DOF would be the thickness of the negative's emulsion layer (very thin). If there are problems with negative flatness, one may want to reassess the negative carrier. Increased DOF will not solve flatness issues. Added: It seems carriers capable of keeping a negative flat are not capable of reasonable throughput and vice versa.

3. RE: How many pixels in a 35mm negative (or any negative for that matter). The answer is, of course, ZERO! That aside, there are upper limits of usable resolution for a given emulsion/processing combination. A good estimate might be made using MTF data from the film manufacturer. Makers of some recent emulsions have claimed MTF that translate to very high resolutions such that using a 24 Mpx camera for copy work might be entirely appropriate.


Steve

[stevebrot]

and it looks like the 4000dpi is the limitation

By calculation that resolution would render a 21.4 Mpx TIFF from the 5000 ED. Based on what you observed, it appears that use of a 24 Mpx camera with premium optics might be quite appropriate.


Steve

(...only worked with Technical Pan once...amazing stuff...)

[3by2]

I would be very interested to find out how good LR's spot removal works on something like this.

The answer is it wouldn't. That would have me heading to Photoshop and the clone tool for a few hours work. But digitizing by camera, I would expect to be much closer to your Coolscan and ICE straight from the off, this is the difference I've found between the shots I've scanned with my Minolta Dimage and stuff I've re-digitised with the camera.

It's very interesting to see the testing you've done and the results you're getting.


Edit: I would add further, I've seen the conclusion of 4000dpi, which you reached, before. Namely on the Scandig website which is comprehensive on scanning. On a personal note I don't think I've got anything on 35mm film which tests those limits.

[LesDMess]

By calculation that resolution would render a 21.4 Mpx TIFF from the 5000 ED. Based on what you observed, it appears that use of a 24 Mpx camera with premium optics might be quite appropriate.


Steve

(...only worked with Technical Pan once...amazing stuff...)

In page 1 of this thread I posted the results from dslr scanning of this same frame of techpan with a 36MP Nikon D800 and the 4000dpi seems to exceed it by a smidge even though the D800 applies more pixels.

A colleague has the 45MP D850 and I'm hoping to be able to barrow it to try it on this frame of film as well as see how good the built-in color negative conversion works. I do have a 24MP Sony A6000 but I figure the results would still be below the D800.

Speaking of 24MP here's a side by side similar shot of two different cats one taken with a 24MP Sony A900 set at ISO400 and the other with my Canon EOS3 using Fuji Sensia 400 scanned at 4000dpi with ICE.



Of course to "properly" digitize what was achieved on the film, the taking device must exceed it but I am not sure it does in this case.

---------- Post added 01-19-22 at 08:01 AM ----------

The answer is it wouldn't. That would have me heading to Photoshop and the clone tool for a few hours work. But digitizing by camera, I would expect to be much closer to your Coolscan and ICE straight from the off, this is the difference I've found between the shots I've scanned with my Minolta Dimage and stuff I've re-digitised with the camera.

It's very interesting to see the testing you've done and the results you're getting.


Edit: I would add further, I've seen the conclusion of 4000dpi, which you reached, before. Namely on the Scandig website which is comprehensive on scanning. On a personal note I don't think I've got anything on 35mm film which tests those limits.

That's too bad about LR's spot removal though I still wonder what the results would look like. Would it make a mess of it like Canon's FARE - their version of ICE, does with Kodachrome?

I've also conducted this test with Fuji Velvia and Kodak TMAX100 and I believe they exceed Coolscan's 4000dpi.

Here's one on Fuji Velvia and if you look at the same areas - no ICE, it looks similar to the Techpan. I will have to do the same 4.5X magnification to verify if there are more details not achieved with the 4000dpi scan.

[BigMackCam]

OK... so...

Firstly, thanks for all the interesting input and discussion. You good folks never disappoint when I have a query or I'm looking for opinions

I spent some time this morning running a few tests. For the same 35mm Ilford XP2 Super 400 negative, I digitised it first with the K-5 + DFA100/2.8 at f/5.6, f/8 and f/10, then with the K-3II in non-pixel-shift mode, and finally using pixel-shift. I loaded the files into RawTherapee using AMaZE or pixel-shift demosaicing as appropriate, without noise reduction or sharpening of any kind applied, then opened them in GIMP 2.10 as 16-bit TIFF files using RT's external editor feature. I cropped the borders of each image, inverted the tone curve, then automatically adjusted the input levels to attain almost-identical black-point, white-point and gamma for all six files. I then examined each one at 1:1 reproduction. As a final comparison, I scaled the K-3II files to the same pixel dimensions as those from the K-5.

Looking at the K-5 files, there's little noticeable difference between the f/5.6 and f/8 shots... but - and I say this while chowing down on a large slice of fresh humble pie - you folks were quite right... the f/10 files are slightly softer; enough that I will use f/8 by default in future, unless I consider the additional depth-of-field to be essential on a case-by-case basis.

Comparing the K-3II non-pixel-shift files, they benefit from greater acutance than the K-5 files - and without any unpleasant rendering effects in the grain, with this negative at least. The scaled-resolution files also look better, to my eye, than the K-5 files - though both are very good, and when the K-5 images are sharpened just very slightly, it's difficult to tell the difference. Still, the K-3II definitely produces sharper results, IMHO.

Which brings me to the pixel-shift files... and, rather surprisingly, I don't see any obvious improvement over the non-pixel-shift versions. It's possible that the film is the limiting factor here, though, and I'd like to try again with negatives from a low-grain, high-detail film. I've not shot any such film yet, so this is something for the future.

I'm still very happy with the results from the K-5, especially at f/8 - but since I note a definite improvement and nothing undesirable from the K-3II, I may switch to that (or my K-3) going forward. I can see how it could make a difference for larger print sizes.

Thanks again, all

[BigMackCam]

Follow-up to the above:

Regarding pixel-shift vs non-pixel-shift, I've viewed the files in Pentax DCU5, and I actually can see a difference after all... It's as if there's an increase in local contrast with the PS files. I can see how this might be preferable on very detailed, low-grain negatives, but it somehow works against the grainy, high-speed film I've chosen for this test. I'll keep pixel-shift in reserve for now...

[3by2]

OK... so...

Firstly, thanks for all the interesting input and discussion. You good folks never disappoint when I have a query or I'm looking for opinions

I spent some time this morning running a few tests. For the same 35mm Ilford XP2 Super 400 negative, I digitised it first with the K-5 + DFA100/2.8 at f/5.6, f/8 and f/10, then with the K-3II in non-pixel-shift mode, and finally using pixel-shift. I loaded the files into RawTherapee using AMaZE or pixel-shift demosaicing as appropriate, without noise reduction or sharpening of any kind applied, then opened them in GIMP 2.10 as 16-bit TIFF files using RT's external editor feature. I cropped the borders of each image, inverted the tone curve, then automatically adjusted the input levels to attain almost-identical black-point, white-point and gamma for all six files. I then examined each one at 1:1 reproduction. As a final comparison, I scaled the K-3II files to the same pixel dimensions as those from the K-5.

Looking at the K-5 files, there's little noticeable difference between the f/5.6 and f/8 shots... but - and I say this while chowing down on a large slice of fresh humble pie - you folks were quite right... the f/10 files are slightly softer; enough that I will use f/8 by default in future, unless I consider the additional depth-of-field to be essential on a case-by-case basis.

Comparing the K-3II non-pixel-shift files, they benefit from greater acutance than the K-5 files - and without any unpleasant rendering effects in the grain, with this negative at least. The scaled-resolution files also look better, to my eye, than the K-5 files - though both are very good, and when the K-5 images are sharpened just very slightly, it's difficult to tell the difference. Still, the K-3II definitely produces sharper results, IMHO.

Which brings me to the pixel-shift files... and, rather surprisingly, I don't see any obvious improvement over the non-pixel-shift versions. It's possible that the film is the limiting factor here, though, and I'd like to try again with negatives from a low-grain, high-detail film. I've not shot any such film yet, so this is something for the future.

I'm still very happy with the results from the K-5, especially at f/8 - but since I note a definite improvement and nothing undesirable from the K-3II, I may switch to that (or my K-3) going forward. I can see how it could make a difference for larger print sizes.

Thanks again, all

Excellent: My only comment is that XP2 might not be typical with it being a chromogenic film but I don't know as I've only got one XP film on file and that's on 120 and it's XP-4 and somehow I managed to overprocess that by 3 times the recommended time, so I'm not confident I can deduce anything from it, besides the fact I'm an idiot sometimes, maybe somebody else knows.