Canon slide copying set-up

[M.Chapman]

2 hours ago, Chuck Nacke said:

I am afraid that I do not wish to post a full sized file on this thread, I have my reasons.

 

Ok, let's try a different tack. Can you direct me to an image in your Alamy portfolio that you feel demonstrates the best quality you have from scanning a 35mm chrome, that is 5000 px or more which has passed through regular Alamy QC? I can then buy a copy as I'm genuinely intrigued to see. You'll get a Presentation use sale and my curiosity will be satisfied.

 

Mark

Edited February 11, 2020 by M.Chapman

[MDM]

3 hours ago, Chuck Nacke said:

MDM,

 

She has a P.h.D. in Solid State Physics and your posts have made her laugh hard..... There are digital images of her, working out outside on Alamy (shot with a DSLR) 

 

To everyone else,

 

I am afraid that I do not wish to post a full sized file on this thread, I have my reasons.  I have been very careful with everything I have written and expressed my own opinion.  I do happen to know and have worked with professional pre-press people, very high-end printers, and I can assure you that they have extensive training and experience working on images with their sophisticated and expensive imaging equipment.  I have still not seen a 35mm chrome copied with a DSLR that meets my standards using my simple desktop scanner, just my opinion.

 

Chuck

 

Well please give her my best wishes 😂. I am glad I made her laugh. My PhD is in volcanology and geochemistry. In fact some of those  pics I have been posting of Sollipulli are from my PhD research. 
 

In relation to DSLR copying, I am making all my judgements from my own experience, not from hearsay and I stand firmly by my assertions. 

[MDM]

1 hour ago, M.Chapman said:

 

Ok, let's try a different tack. Can you direct me to an image in your Alamy portfolio that you feel demonstrates the best quality you have from scanning a 35mm chrome, that is 5000 px or more which has passed through regular Alamy QC? I can then buy a copy as I'm genuinely intrigued to see. You'll get a Presentation use sale and my curiosity will be satisfied.

 

Mark

 
This is actually meaningless really in resolving this particular debate  as it is not about one good image of Chuck’s. It is about which process produces consistently better results and it would require comparison of numerous scans and camera pics of various slides to make a sensible judgement. These would have to have exactly the same treatment in post as well. 

[Harry Harrison]

I've just watched another video by Nick Carver about scanning on Epson V7xx, V8xx scanners using the standard Epson medium format holder. It would be good to watch for anyone with one of those scanners, and it is very good on how to use Silverfast 8 and Negafix for scanning colour negative.

 

https://www.youtube.com/watch?v=qtpmlEeJodw

 

It seems with Negafix that you actually adjust the settings to get the colour and tone of the image that you prefer on screen, it seems pretty powerful but it's no surprise that at the end of it you get a scan that pleases you which isn't possible with an outsourced scan. He also goes into the dust and scratches removal which seems to work well (probably not for Kodachrome though). He glosses over its suitability for 35mm but uses 2400 ppi as his standard resolution, coincidentally this is what reviews suggest is the maximum it is capable of.

 

At about 21m 15s he talks about Smart Sharpen in Photoshop, he obviously rates it and uses it before he takes the scans into Lightroom. I don't have it but I've not noticed it mentioned on here with respect to scanning, is it good? He used it in combination with standard Unsharp Mask when doing his giant medium format print scans, e.g.

 

DSLR / Fuji Velvia

 

Unresized (larger) image

Unsharp mask in Photoshop - Amount 400%, Radius 1 px, Threshold 2 px

Resize down to 72" x 24" @300 ppi (bicubic automatic)

Smart Sharpen in Photoshop - Amount 150%, Radius 1.3 px, Reduce Noise: 0, Remove: Gaussian Blur

then

Smart Sharpen in Photoshop - Amount 150%, Radius 1 px, Reduce Noise: 0, Remove: Gaussian Blur

 

 

Edited February 11, 2020 by Harry Harrison

[Inchiquin]

21 hours ago, geogphotos said:

 

Using a camera is clearly a lot faster than a film scanner, but a film scanner has dust removal

 

 

 

Unless the bulk of your collection is, like mine, Kodachrome in which case the scanner's dust removal facility is useless.

 

Alan

[M.Chapman]

15 hours ago, MDM said:

I am not sure I can see any difference worth mentioning between your Topaz images and the previous versions but I have been on my computer most of the day so my eyesight is a little strained right now for distinguishing very fine detail. Unless I am missing something, I think the effort and time taken is not worth it. If you are going to do some Photoshop work, then there are a some better options I think.

 

1. Select the sky and apply a gaussian blur as I have described by selecting the sky. Sky selection in those two images would be trivial in Photoshop.

 

 or

 

2. Process two images in Lightroom, one with strong noise reduction and no sharpening and the other with no noise reduction but some sharpening for the ground area. Open as a single file with two layers in Photoshop and do as you describe or better still use layer masks which allow non-destructive editing. Ultimately this might be better than 1 as LR noise reduction is probably better than a gaussian blur in Photoshop. For real accuracy select the sky in Photoshop so you don't paint or erase over the boundaries. This whole process would be even better with a pressure-sensitive tablet than a mouse.

 

Or process the entire image in Lightroom with local noise reduction. I find Photoshop much more accurate though for detailed selections. 

 

After much experimentation I have to agree.  Although Topaz seemed like a "quick fix", after a lot of trial and error I find I can achieve a better result by making selections in PS and applying smart sharpening and noise reduction (not Gaussian blur) filters where needed. Very close inspection of Topaz results always reveals over-sharpening artefacts, even when their sharpening is set to 0, and even after trying to reduce by blending back with the original image.

 

Thanks so much for your help, and patience! I think I'm now ready to start working my way through some of my old slides and hopefully passing Alamy QC.

 

Mark

[MDM]

1 minute ago, M.Chapman said:

 

After much experimentation I have to agree.  Although Topaz seemed like a "quick fix", after a lot of trial and error I find I can achieve a better result by making selections in PS and applying smart sharpening and noise reduction (not Gaussian blur) filters where needed. Very close inspection of Topaz results always reveals over-sharpening artefacts, even when their sharpening is set to 0, and even after trying to reduce by blending back with the original image.

 

Thanks so much for your help, and patience! I think I'm now ready to start working my way through some of my old slides and hopefully passing Alamy QC.

 

Mark

 

I ran a series of Topaz Denoise with various settings last night while doing other things such as making dinner, watching a movie (BlackKKlansman - well worth the watch) as it hangs up my computer when it is running and I can't do much else. I was waiting until today to examine the results as my eyesight is a lot better early in the day when I haven't been staring at screens for hours. Anyway I have not found anything new that would convince me to use it in my slide copying workflow. As you say it produces artifacts and that combined with the slowness make it an easy decision. I guess if one had a state-of-the-art computer then it would run at reasonable speed but not on my Mac.

 

I may welll change my workflow a little in that I will do all the sharpening and noise reduction on the raw files (much better noise reduction than doing it in Photoshop), do two separate files and combine them in Photoshoo rather than the gaussian blur on the sky that I have been using. I use a Wacom tablet so the pressure sensitive pen makes the job of combining easier. Having said that I am not too concerned if the skies are slightly  blurred and virtually noise free as they are generally just background and nobody is really going to examine them in detail as long as the effect doesn't affect the main subject.

 

It is a matter of weighing up what degree of perfection to aim for. I am a pragmatic perfectionist (just made that up but it sounds good). If I am producing images that are good enough to pass Alamy QC and will print to my satisfaction at A4 at least and maybe even A3 viewed at sensible viewing distances and not through a loupe, then that is good enough for me.

 

 

 

 

[MDM]

1 hour ago, Harry Harrison said:

I've just watched another video by Nick Carver about scanning on Epson V7xx, V8xx scanners using the standard Epson medium format holder. It would be good to watch for anyone with one of those scanners, and it is very good on how to use Silverfast 8 and Negafix for scanning colour negative.

 

https://www.youtube.com/watch?v=qtpmlEeJodw

 

It seems with Negafix that you actually adjust the settings to get the colour and tone of the image that you prefer on screen, it seems pretty powerful but it's no surprise that at the end of it you get a scan that pleases you which isn't possible with an outsourced scan. He also goes into the dust and scratches removal which seems to work well (probably not for Kodachrome though). He glosses over its suitability for 35mm but uses 2400 ppi as his standard resolution, coincidentally this is what reviews suggest is the maximum it is capable of.

 

At about 21m 15s he talks about Smart Sharpen in Photoshop, he obviously rates it and uses it before he takes the scans into Lightroom. I don't have it but I've not noticed it mentioned on here with respect to scanning, is it good? He used it in combination with standard Unsharp Mask when doing his giant medium format print scans, e.g.

 

DSLR / Fuji Velvia

 

Unresized (larger) image

Unsharp mask in Photoshop - Amount 400%, Radius 1 px, Threshold 2 px

Resize down to 72" x 24" @300 ppi (bicubic automatic)

Smart Sharpen in Photoshop - Amount 150%, Radius 1.3 px, Reduce Noise: 0, Remove: Gaussian Blur

then

Smart Sharpen in Photoshop - Amount 150%, Radius 1 px, Reduce Noise: 0, Remove: Gaussian Blur

 

 

 

I can't take time and nor do I have the inclination to watch any more of his videos. I watched enough of the first one to understand what he was doing. I didn't enjoy his style or consider his assessment of DSLR copying of medium format negs to be of any real relevance to my work. It would not make sense to take a raw image into Photoshop and apply what are massive levels of sharpening if what you have quoted there is accurate.

 

My philosophy is to always to as much on the raws as possible and I do my sharpening in Lightroom whether it is an image from a digital camera or a slide copy. If I was to do sharpening in Photoshop, I would use the ACR filter rather than any of the traditional Photoshop sharpen tools.

 

Martin Evening's assessment of Smart Sharpen in his Adobe Photoshop CC for Photographers (I have the 2016 edition) is worth reading. Basically he says don't be taken in by the fact that it's called smart and that he considers it to be a tool for corrective rather than general sharpening. Far too much to summarise here but I do take Martin Evening's opinions very seriously.

 

 

 

[Harry Harrison]

2 minutes ago, MDM said:

Basically he says don't be taken in by the fact that it's called smart and that he considers it to be a tool for corrective rather than general sharpenin

Thanks, just wondered as I'd not come across it. Yes, he was using very high levels, fairly high even on the drum scans.

 

Unresized (larger) image

Smart Sharpen in Photoshop - Amount 200%, Radius 3 px, Reduce Noise: 0, Remove: Gaussian Blur

Resize down to 72" x 24" @300 ppi (bicubic automatic)

Unsharp mask in Photoshop - Amount 120%, Radius 1.5 px, Threshold 14 px

Smart Sharpen in Photoshop - Amount 120%, Radius 1 px, Reduce Noise: 0, Remove: Gaussian Blur

[Betty LaRue]

17 hours ago, MDM said:

 

I only watched a few snippets to see what he was doing and it's not really relevant to the current discussion Betty in relation to using a camera to digitise slides.  He is digitising a panoramic 6x17 negative, using a Canon 6D camera which required him to take a load of shots and then try to stitch them afterwards. This is nothing like the sort of procedures we have been talking about here but it illustrates the extreme difficulty of getting a piece of film aligned with a camera. This is completely different to using a ES-1 adapter on a quality camera as I have described in terms of quality and difficulty.

 

His comparison of his actually scanning methods for panoramic medium format film is probably a lot more valid. The drum scan turns out the best quality unsurprisingly but it is incredibly expensive. Check out the Michael Strickland website he used for prices. 

That’s the first thing I did. At $35 a pop that is far and away out of my budget. 
The weird thing is that when you find someone that supposedly does superior work, you feel that someone more affordable is dog meat! 😁 I’m afraid I tend to be guilty of that. That’s the story of my life. Finding what I absolutely consider the best and having to have it.

So...when I can’t afford it, I just start saving for it. Many times that means a long wait, but in the end, I don’t have buyer’s regret.

 

Sometimes what I consider the best, though, doesn’t mean it’s the most expensive. 
While I was watching the video, I realized taking one shot as I would be doing, not a series, would be much easier.  And I trust you that the ES-1 would be easier yet.

Betty

[MDM]

41 minutes ago, Betty LaRue said:

That’s the first thing I did. At $35 a pop that is far and away out of my budget. 
The weird thing is that when you find someone that supposedly does superior work, you feel that someone more affordable is dog meat! 😁 I’m afraid I tend to be guilty of that. That’s the story of my life. Finding what I absolutely consider the best and having to have it.

So...when I can’t afford it, I just start saving for it. Many times that means a long wait, but in the end, I don’t have buyer’s regret.

 

Sometimes what I consider the best, though, doesn’t mean it’s the most expensive. 
While I was watching the video, I realized taking one shot as I would be doing, not a series, would be much easier.  And I trust you that the ES-1 would be easier yet.

Betty

 

Human nature Betty. People think that by paying more they are getting the best which is not necessarily true at all. It can be hard to balance it as well when selling services. I live with this dilemma. 
 

If you do decide that you want a setup for copying then I think your best bet would be to look at something like Phil Crean posted on page 1 of this thread. Getting an ES-1 type setup to work with a Fuji system and macro lens might take some experimentation whereas the system Phil mentions will work with any camera and macro lens. 

[Betty LaRue]

7 minutes ago, MDM said:

 

Human nature Betty. People think that by paying more they are getting the best which is not necessarily true at all. It can be hard to balance it as well when selling services. I live with this dilemma. 
 

If you do decide that you want a setup for copying then I think your best bet would be to look at something like Phil Crean posted on page 1 of this thread. Getting an ES-1 type setup to work with a Fuji system and macro lens might take some experimentation whereas the system Phil mentions will work with any camera and macro lens. 

Duly noted. Thanks.

[Harry Harrison]

11 hours ago, Betty LaRue said:

Sometimes what I consider the best, though, doesn’t mean it’s the most expensive. 

Betty, you might be interested in this earlier thread:

 

https://discussion.alamy.com/topic/10923-cheap-slide-scanner/

 

ReeRay has a beautifully elegant solution with an old (but beautifully engineered) Leica copy stand, an enlarger lens and a Fuji T?, there's a nice picture of his setup. He says in the thread that he owns a very high end Minolta Multi Pro scanner and that the quality that he gets from his Fuji matches the scans from the Minolta scanner.

 

Because the stand was designed for a small, light Leica it probably would be unbalanced with a big DSLR but the Fuji seems to fit it perfectly. Those stands aren't easily available though and you probably pay a premium Leica price but it shows the quality that can be achieved with extension tubes and an enlarger lens provided that you've got a means of aligning everything correctly.

Edited February 12, 2020 by Harry Harrison

[M.Chapman]

On 10/02/2020 at 21:01, M.Chapman said:

 

I took a look at the site as well. Very interesting. I found the bit on light source collimation interesting. See Section 4 on this page. The author claims a collimated light sources give better sharpness and contrast. I might try increasing the separation between my light source and my slide to see what happens. Although, with my less than perfect slides, I suspect the limit on the sharpness I can get lies way back at the time I took the picture (lens + film + technique).

 

Mark

 

I just tried taking several shots with my light source at different distances and the results with the light further away do indeed appear to be noticeably sharper. My setup was a bit "Heath Robinson" so I need to improve it and retest. Then, if the result is confirmed, I'll post the results. Not sure if anyone else here has a setup where they can easily change the separation? (NB. If using a diffuser then it has to be moved further away from the slide). Negative Lab Pro site says the following about light collimation.

 

Collimated light has two effects when DSLR scanning a film negative:

1. Collimated light produces sharper looking scans (whereas diffuse light will produce softer scans)
2. Collimated light produces more contrast in scans

Which is better? Well, it’s really a matter of opinion and the debate has been going on for a while (in a darkroom setting, you can choose between using a condenser head (for collimated light) or a diffusion head 90.)

But I would say this:

• If you like the sharpness and detail you see in drum scanners or dedicated film scanner (like Nikon Coolscans), try to get a more collimated light source.
• If you prefer the “softness” of a traditional fine-art print, go for a more diffused light source

 

Mark

Edited February 12, 2020 by M.Chapman

[Harry Harrison]

8 minutes ago, M.Chapman said:

Not sure if anyone else here has a setup where they can easily change the separation?

Not with the Illumitran anyway, the slide/film holder is fixed to the 'light box'.

 

I remember that in the Leitz Valoy condenser enlarger it was the actual glass condenser that holds the negative flat from above and is treated to avoid Newton rings, the condenser focused the light from the opal bulb above. Interesting that you should see a difference through simply moving your light source further away. I've seen setups where they shine a slide projector from the other side, presumably through some sort of diffuser. I thought it looked crazy but perhaps that's why they do it.

[MDM]

7 hours ago, M.Chapman said:

 

I just tried taking several shots with my light source at different distances and the results with the light further away do indeed appear to be noticeably sharper. My setup was a bit "Heath Robinson" so I need to improve it and retest. Then, if the result is confirmed, I'll post the results. Not sure if anyone else here has a setup where they can easily change the separation? (NB. If using a diffuser then it has to be moved further away from the slide). Negative Lab Pro site says the following about light collimation.

 

Collimated light has two effects when DSLR scanning a film negative:

1. Collimated light produces sharper looking scans (whereas diffuse light will produce softer scans)
2. Collimated light produces more contrast in scans

Which is better? Well, it’s really a matter of opinion and the debate has been going on for a while (in a darkroom setting, you can choose between using a condenser head (for collimated light) or a diffusion head 90.)

But I would say this:

• If you like the sharpness and detail you see in drum scanners or dedicated film scanner (like Nikon Coolscans), try to get a more collimated light source.
• If you prefer the “softness” of a traditional fine-art print, go for a more diffused light source

 

Mark

 

 

I suspect that any differences between consensed and diffused light sources are illusory and related to contrast rather than true sharpness (actual rendering of detail). The most important factors by far are surely a properly focused and exposed original in good condition as well as a high quality camera and lens combo used for copying. The ES-1 has an immovable diffuser directly behind the slide which helps greatly in providing even light across the image (as measured by spot meter). I can not think of any reason why moving the light source would change rendering of detail except perhaps that moving it away necessitates longer exposures which if anything would likely to lead to less sharp images due to camera shake. However, using the camera on a tripod with mirror up (if DSLR) should negate any such problems.

[Betty LaRue]

11 hours ago, Harry Harrison said:

Betty, you might be interested in this earlier thread:

 

https://discussion.alamy.com/topic/10923-cheap-slide-scanner/

 

ReeRay has a beautifully elegant solution with an old (but beautifully engineered) Leica copy stand, an enlarger lens and a Fuji T?, there's a nice picture of his setup. He says in the thread that he owns a very high end Minolta Multi Pro scanner and that the quality that he gets from his Fuji matches the scans from the Minolta scanner.

 

Because the stand was designed for a small, light Leica it probably would be unbalanced with a big DSLR but the Fuji seems to fit it perfectly. Those stands aren't easily available though and you probably pay a premium Leica price but it shows the quality that can be achieved with extension tubes and an enlarger lens provided that you've got a means of aligning everything correctly.

Thanks, Harry, very interesting.

[Chuck Nacke]

On ‎11‎/‎02‎/‎2020 at 01:43, M.Chapman said:

 

Ok, let's try a different tack. Can you direct me to an image in your Alamy portfolio that you feel demonstrates the best quality you have from scanning a 35mm chrome, that is 5000 px or more which has passed through regular Alamy QC? I can then buy a copy as I'm genuinely intrigued to see. You'll get a Presentation use sale and my curiosity will be satisfied.

 

Mark

Mark,

 

No I am sorry, but will not do what you suggest.  The bottom line is that I work with the equipment that I have and many of those scans have been licensed by clients and are in a number of agencies, I am happy with them, clients are happy with them, what else is there to say?.  I am also chasing down the Illegal use of one of my very exclusive images, it was not stolen from Alamy,  I do believe that you can get a good enough idea of the quality of my scans from the images that I have on Alamy.

 

Best,

 

Chuck 

[M.Chapman]

4 hours ago, Chuck Nacke said:

Mark,

 

No I am sorry, but will not do what you suggest.  The bottom line is that I work with the equipment that I have and many of those scans have been licensed by clients and are in a number of agencies, I am happy with them, clients are happy with them, what else is there to say?.  I am also chasing down the Illegal use of one of my very exclusive images, it was not stolen from Alamy,  I do believe that you can get a good enough idea of the quality of my scans from the images that I have on Alamy.

 

Best,

 

Chuck 

 

Fair enough, your decision. 

 

I've made further improvement to my workflow and have decided that Topaz Denoise can be a useful tool if used with care on the right pictures (sorry I've changed my mind again). Here's another example from 35mm Velvia 50ASA using DSLR copying technique downsized to meet Alamy minimum requirements (6MP).

 

Whole image

 

100% crop before Topaz

 

100% crop after Topaz Denoise (15, 15, 0, 0.00) with 30% original image blend + 0.2 radius pixel Gaussian blur

 

I find it takes about 1 minute to apply Topaz to a 6MP image + 30 seconds to blend with original to retain a little noise/grain and reduce the edge sharpness slightly (in my opinion Topaz "overcooks" things a bit). No selections were used or necessary.

 

Here are links to the 6MP jpg files for anyone that wants to download to inspect more closely.

 

Falcon before Topaz denoise 

Falcon after Topaz denoise 

 

Mark

[M.Chapman]

17 hours ago, MDM said:

I suspect that any differences between condensed and diffused light sources are illusory and related to contrast rather than true sharpness (actual rendering of detail). The most important factors by far are surely a properly focused and exposed original in good condition as well as a high quality camera and lens combo used for copying. The ES-1 has an immovable diffuser directly behind the slide which helps greatly in providing even light across the image (as measured by spot meter). I can not think of any reason why moving the light source would change rendering of detail except perhaps that moving it away necessitates longer exposures which if anything would likely to lead to less sharp images due to camera shake. However, using the camera on a tripod with mirror up (if DSLR) should negate any such problems.

 

You may well be right. I improved my setup and am struggling to recreate what I saw yesterday (although I'm now not able to move the light source as far away). One thing that I'm pretty certain happens is that dust on the slide is rendered much more noticeably (sharper and/or higher contrast) when the light is less diffuse. I think that more diffuse light must illuminate the sides of the (3D) dust more and reduce the contrast. But the difference in the slide itself is less noticeable. So I won't be increasing the separation beyond my existing 10mm which ensures that the texture of my diffuser (and any dust on it) is defocussed

 

Mark

[Harry Harrison]

35 minutes ago, M.Chapman said:

Here are links to the 6MP jpg files for anyone that wants to download to inspect more closely.

For me that's a particularly impressive result, very noticeable in the even green background which would also translate into clear blue skies I imagine. I'm surprised that the detail in the falcon's face seems to have been enhanced also.

[M.Chapman]

9 minutes ago, Harry Harrison said:

I'm surprised that the detail in the falcon's face seems to have been enhanced also.

 

Clever isn't it, especially as no selection was used (not even available in Topaz). Another thing that I noticed was that noise (grain) in blue skies are smoothed, but the similar sized graininess of sand on a sandy beach isn't. It's a shame that, in my opinion, that Topaz overdoes things (hence my blend back with original) and doesn't seem to be available as a simple filter in PS (on my Mac I don't see the plugin at all, and all the online help relates to the Windows plugin.)

 

Free 30 day trial of Topaz Denoise AI can be downloaded from their website here. 

 

Mark

Edited February 13, 2020 by M.Chapman

[Harry Harrison]

6 minutes ago, M.Chapman said:

Free 30 day trial of Topaz Denoise AI can be downloaded from their website here. 

Thanks, I'll take a look, I must say I was scared off by the blue spilling over on to MDM's mountains but it certainly looks as if it is worth exploring for scans.

 

Might you be tempted not to downsize quite so much? I know we're all running scared of QC failures but it does seem rather a shame. I've seen quite a few scans pass through this forum that are more in the 4500 > 5000 px range though there is a possibility that they were uploaded in more lenient times.

Edited February 13, 2020 by Harry Harrison

[MDM]

1 hour ago, M.Chapman said:

 

Clever isn't it, especially as no selection was used (not even available in Topaz). Another thing that I noticed was that noise (grain) in blue skies are smoothed, but the similar sized graininess of sand on a sandy beach isn't. It's a shame that, in my opinion, that Topaz overdoes things (hence my blend back with original) and doesn't seem to be available as a simple filter in PS (on my Mac I don't see the plugin at all, and all the online help relates to the Windows plugin.)

 

Free 30 day trial of Topaz Denoise AI can be downloaded from their website here. 

 

Mark

 

It's good you are finding Topaz DeNoise (let's abbreviate as TDN) useful. Would you care to elaborate a bit on how you have blended the images as this is a vital part of the process and it is not clear what you did.  By the way, there is a Lightroom plugin installed on my machine (maybe not available for the trial version or maybe you didn't install it?) but it is limited as it can't work on the raw files. There is also a Photoshop plugin.

 

I am not saying it is no good - it certainly does an excellent job on removing grain and noise - but the benefits for me come nowhere near the drawbacks in terms of a reasonably speedy workflow. I have been using it on the raw images (36MP, 16-bit Nefs) and it is very slow (a few minutes per image after the initial preview in TDN has been generated). I have been opening the raw files directly into TDN and there is no way of opening a cropped version of the raw so it is all or nothing.  I save it as a DNG and import into Lightroom. 

 

To use the TDN-processed file in my workflow I would then need to open it into Photoshop and open a sharpened version of the original NEF, after making similar adjustments to the white balance and exposure-related adjustments to both files. As noted before TDN can do some weird stuff to the white balance which is also a negative. I would then blend the two files using layers and layer masking for non-destructive editing so that the areas that need to be sharp and where noise is not a problem (the detail areas such as the ground in a landscape) are from the sharpened Nef and the rest is from the TDN-processed DNG.

 

Now compare that to the much faster speed of producing two different versions of the same Nef in Lightroom, one sharpened and one with a reasonably strong LR noise reduction applied, opening and blending in Photoshop. Lightroom noise reduction is actually very good I find and can be made as clean as Topaz.  There is also the benefit that there is no introduction of artifacts at boundaries which happens in TDN (I have noticed that the slide mount can cause problems at the margins of the slide but artifacts can also be introduced at boundaries within the image).

 

Whether one is marginally better than the other in quality of the resultant image is up for debate but doing the noise reduction in Lightroom is much faster because of not having the very slow step in TDN. Again keep in mind that I am talking all the time about working on raw images. So while TDN could be very useful for some images, the drawbacks in terms of speed leave me in no doubt as to how I will proceed from here. 

 

Edited February 13, 2020 by MDM

[MDM]

56 minutes ago, Harry Harrison said:

Thanks, I'll take a look, I must say I was scared off by the blue spilling over on to MDM's mountains but it certainly looks as if it is worth exploring for scans.

 

Might you be tempted not to downsize quite so much? I know we're all running scared of QC failures but it does seem rather a shame. I've seen quite a few scans pass through this forum that are more in the 4500 > 5000 px range though there is a possibility that they were uploaded in more lenient times.

 

I think anything above 4000 pixels in my processing is starting to look a bit noisy. Fine for my purposes if I want to make a print but why risk a QC failure when 3000 pixels will easily clear the bar and it is unlikely that a larger file would be required by a buyer. Whether QC is harder or easier now is moot.