Jump to content

Canon slide copying set-up


Recommended Posts

21 hours ago, MDM said:

 

Imagine holding a toilet roll tube directly in front of your lens which has a front element a lot wider than the toilet roll - that sort of thing. According to Harry's post, that lens will work with a step down ring and extenders in front of the lens. I have a Tamron 90 macro lens with 58mm thread which I will be checking out. At the moment it won't focus close enough. I will know a lot more when the boat arrives from China with the extenders that go in front of the lens and will report here.

 

Thanks got it, like our Avatars here. 😎

 

I use postimage aka postimg all the time, the ads I see are for McDonald's and Fat blockers, must be something about reading the previous sites viewed. But for the people who asked, that hosting is free and we have no control over the ads. Something on your system is deciding what to show you.

 

Free hosting = they pay for the free with ad views.

 

canon-slide-duplicator.jpg

 

Used to post images to forums there are no ads. See above.

 

 

 

Edited by Klinger
Link to comment
Share on other sites

31 minutes ago, MDM said:

Please feel free to download, examine and comment. 

Thanks for doing that, Dropbox works really well, though no Russian ladies sadly (!Joke!).

 

I brought them into Lightroom and compared them side by side at 1:1 and I deliberately didn't read your conclusions first. However, I agree with them completely noticing exactly the effects that you describe. I hadn't heard of Topaz DeNoise but clearly it does have to be used with care.

 

Next time I do some scans I'll try and upload some of mine the same way to see if I can show the differences between DSLR, Microtek film scanner and the Flextight. At least I might learn something anyway.

Link to comment
Share on other sites

9 hours ago, geogphotos said:

 

 

That's very interesting and useful.

 

Have you used this lens for slide copying using a tripod - camera pointing down to slide illuminated below by lightbox?

No, Ian, I’m still in the process of gathering information, so I haven’t tried anything yet.  I only used it similarly by shooting the plants, but that was larger objects than a slide. The results of that were good.

While this bit of news isn’t relevant, it explains a bit. I’m preparing for a huge garage sale in April, and have hundreds of things to go through and price. I find since I moved from Oklahoma to Kansas, then my husband died, it took me a while to know what I actually need and use of the belongings we moved with us. The plus is that I have a huge storage room in the basement. I never had a basement before. So I can contain the mess while working on it. I don’t need a lot now while living alone.  
So it could be May before I have the time to test slide copying. 
I do want to test all of my tripods first, though. The ones that aren’t steady while shooting down might go in the garage sale!

 

Perhaps you are like me. I try to absorb all of the technical information in this thread but find my eyes start glazing over, because the terms are absolutely foreign to me. I can’t form mental images which is how I learn. If I can’t picture it in my mind, that’s when my eyes glaze over.  
I’m amazed at how technically sharp our fellow contributors to this thread are.

Betty

 

Link to comment
Share on other sites

On 07/02/2020 at 10:40, MDM said:

 

Imagine holding a toilet roll tube directly in front of your lens which has a front element a lot wider than the toilet roll - that sort of thing. According to Harry's post, that lens will work with a step down ring and extenders in front of the lens. I have a Tamron 90 macro lens with 58mm thread which I will be checking out. At the moment it won't focus close enough. I will know a lot more when the boat arrives from China with the extenders that go in front of the lens and will report here.

Great description “ I can actually “picture” that! :D

Link to comment
Share on other sites

1 hour ago, Betty LaRue said:

Great description “ I can actually “picture” that! :D

 

Great Betty. Now imagine putting this on the end of the toilet roll and you are almost there.

 

Seriously though if you decide you really want to invest in what you need to copy slides with your existing camera and macro lens, it is not a big step and I am sure we can advise in a simple way that will be guaranteed to work. I would guess about $400-500 and you would have a top class system. 

Link to comment
Share on other sites

6 hours ago, M.Chapman said:

 

Take 10 shots

Import into LR

Make all tonal adjustments (except crop) on 1st image

Synchronise adjustments and export 10 images as 16 bit TIFF or PSD

Load all 10 images into stack with auto align enabled. (there's a single command for this)

Then set layer opacities to (bottom to top) 100%, 50%, 33%, 25%, 20%, 17%, 15%, 13%, 11%, 10% (sequence is 1/2, 1/3, 1/4 etc.

Then flatten image

Dust spot in PS (not easy in LR if images weren't precisely aligned)

Then crop to remove any edge problems caused by image realignment.

Done.

 

Only worth the hassle if you have a really valuable slide to salvage I would suggest.

 

Ads - Oh... I don't see any Ads at all, it must be my ad-blocker saving me. I never realised that. I'll go back and alter the links. Terribly sorry.

 

Update ... turned off my Adblocker and I still don't see any Ads on the links I used. That's weird... Maybe cookies are involved and it knows I posted the images, so doesn't show me the Ads?? I've edited my earlier posting to use direct image links, which hopefully has stopped the Ads?

 

I'll investigate using Dropbox too.

 

Mark

 

 

 

Any reason why you don't convert your layers to smart object > layers > smart objects > stack mode > median ?

Or mean in stead of median in some cases.

 

wim

  • Upvote 1
Link to comment
Share on other sites

3 hours ago, Harry Harrison said:

Thanks for doing that, Dropbox works really well, though no Russian ladies sadly (!Joke!).

 

I brought them into Lightroom and compared them side by side at 1:1 and I deliberately didn't read your conclusions first. However, I agree with them completely noticing exactly the effects that you describe. I hadn't heard of Topaz DeNoise but clearly it does have to be used with care.

 

Next time I do some scans I'll try and upload some of mine the same way to see if I can show the differences between DSLR, Microtek film scanner and the Flextight. At least I might learn something anyway.

 

Interesting to hear that. Yes please do upload some examples as well. Looking forward to that.

Link to comment
Share on other sites

14 minutes ago, wiskerke said:

 

Any reason why you don't convert your layers to smart object > layers > smart objects > stack mode > median ?

Or mean in stead of median in some cases.

 

wim

 

Thanks Wim :)

 

No reason other than I didn't know about that option and I was following the instructions in the article I mentioned. I'll give it a go and see what happens...

 

Mark

Link to comment
Share on other sites

50 minutes ago, MDM said:

 

Great Betty. Now imagine putting this on the end of the toilet roll and you are almost there.

 

Seriously though if you decide you really want to invest in what you need to copy slides with your existing camera and macro lens, it is not a big step and I am sure we can advise in a simple way that will be guaranteed to work. I would guess about $400-500 and you would have a top class system. 

 

I'm just going to mention this, because I did way back in the early pages. For 35 cents a slide, not including postage, you can get 3000dpi 10MP scans of your slides, done by a professional service.

 

That's over 1,000 slides. Maybe the break even point (I used $400) might be something to consider? Over 1,000 slides, start getting all the equipment, lenses, accessories and spend the time, or mail them in?

 

I already have a camera, extension tubes, macro lens, and now a slide duplicator. I'm thinking maybe send in the top 40 slides that I might want to try to upload for stock, spend the $20 which includes shipping, and be free of all the complications and details. Maybe?

 

Link to comment
Share on other sites

55 minutes ago, Klinger said:

 

I'm just going to mention this, because I did way back in the early pages. For 35 cents a slide, not including postage, you can get 3000dpi 10MP scans of your slides, done by a professional service.

 

That's over 1,000 slides. Maybe the break even point (I used $400) might be something to consider? Over 1,000 slides, start getting all the equipment, lenses, accessories and spend the time, or mail them in?

 

I already have a camera, extension tubes, macro lens, and now a slide duplicator. I'm thinking maybe send in the top 40 slides that I might want to try to upload for stock, spend the $20 which includes shipping, and be free of all the complications and details. Maybe?

 

Years ago I sent a few hundred slides to India (via a company in California). My hope was to get lots online ASAP. The scans were ok, mostly, but some not usable. Slides were returned to me all jumbled up, but in the pages, at least. Also tried medium format, and that was a disaster. The people scanning had no idea about film flatness, or even which side to put facing up. After speaking to someone in California, they redid that batch. Second try not much better, so I bought a Coolscan 8000ED and learned to use it. Another company would maybe have done a better job, but I decided to cut my losses. 

 

Despite this long thread having some interest, I'm reminded of something Klinger said early on. After looking at digital files for years, the files from slides won't look that great. I think some can pass QC, but the effort might not reap rewards. I can't seem to stay away from experimenting with some cool gear, but I don't expect to get much back in sales.  

Edited by KevinS
Link to comment
Share on other sites

52 minutes ago, KevinS said:

 

Despite this long thread having some interest, I'm reminded of something said early on. After looking at digital files for years, the files from slides won't look that great. I think some can pass QC, but the effort might not reap rewards. I can't seem to stay away from experimenting with some cool gear, but I don't expect to get much back in sales.  


Same here. If I didn’t say that and I probably did, then it was what I was thinking. Having said that I have been printing some at A4 size and getting very decent results. It’s different looking at them at full size on screen where the quality differences compared to my digital camera images are very apparent. 

 

I am doing this mainly because I am enjoying a winter lull in my photographic activity and taking some time to do some experimentation which I also love. I have a big archive of negs and slides, many of which I can’t even recall until I pull them out. There are family shots, professional stuff, pics related to my other former  profession (geology) and loads of other stuff going way back. I am currently just picking out slides at random with no idea what I will find and it is exciting.  
 

One thing is for sure - the pics are precious to me and I want to get the very best quality I can out of them. Otherwise why bother at all? If I find stuff that I think might be saleable then I will probably upload them to Alamy but that is not my main objective. 

Edited by MDM
  • Upvote 1
Link to comment
Share on other sites

8 hours ago, Harry Harrison said:

Would logging in with a private window allow you to see them?

They've always been there for me, strangely they all live in Wargrave, who'd have thought it. That's where Google seems to think I live also, which is handy.

 

No, I never see any Ads, even when opening an incognito/private window, or when using a different browser, or even a different device. Maybe my ISP is doing some content blocking?

I just tried DropBox, but I couldn't find how to create a gallery/album function in the free (basic) Dropbox and it's less friendly. So I've now uploaded those crops to Google Drive and hopefully set the sharing up correctly.

 

Can you see these (hopefully without ads and Russian women!)

 

Mark

Edited by M.Chapman
Link to comment
Share on other sites

23 minutes ago, M.Chapman said:

 

No, I never see any Ads, even when opening an incognito/private window, or when using a different browser, or even a different device. Maybe my ISP is doing some content blocking?

I just tried DropBox, but I couldn't find how to create a gallery/album function in the free (basic) Dropbox and it's less friendly. So I've now uploaded those crops to Google Drive and hopefully set the sharing up correctly.

 

Can you see these (hopefully without ads and Russian women!)

 

Mark

 

Yes visible and ad-free. Dropbox is a bit unintuitive but you just create or upload a folder, hit the share button and you should get an option to create a link. I have to remember how to do it each time as it is not entirely obvious. I am on the free version. A big advantage is that you can upload raw images. Not sure if that is possible with the other free hosts. 

Link to comment
Share on other sites

28 minutes ago, MDM said:

 

Yes visible and ad-free. Dropbox is a bit unintuitive but you just create or upload a folder, hit the share button and you should get an option to create a link. I have to remember how to do it each time as it is not entirely obvious. I am on the free version. A big advantage is that you can upload raw images. Not sure if that is possible with the other free hosts. 

I didn’t have any trouble uploading and sharing, but there’s not a photos folder so the gallery view didn’t seem to be available. I saw somewhere the photos folder has been deprecated. When did you take out your account?

 

Mark

Link to comment
Share on other sites

3 hours ago, M.Chapman said:

 

Thanks Wim :)

 

No reason other than I didn't know about that option and I was following the instructions in the article I mentioned. I'll give it a go and see what happens...

 

Mark

 

The article probably refers to the situation before CC. Either in CS5 or 6 this was the way to do it. And yes there were some actions floating around (-by Fred Miranda or someone on that board?).

 

wim

Link to comment
Share on other sites

20 hours ago, MDM said:

 

Mine all live in Cambridgeshire funny enough. These Russian ladies do get around it seems. 

 

 

 

Thanks for posting that Mark. It would probably be possible to set up an action to do a lot of it but it still seems like an awful lot of work. 

 

I worry a bit about downloading images from a site that carries ads for dubious services. My Russian ladies all live in Cambridgeshire. I do recommend Dropbox as there are no ads as far as I know and you can post raw images as well which is a big advantage. It is free up to 2GB I think. It was wim who pointed me there as an image sharing site. 

 

Anyway if anyone is interested I have just uploaded some pics to Dropbox which illustrate a few things related to grain and noise. All on Nikon D810 with 55mm Micro-Nikkor, extension ring, ES-1, tripod and Rotolight Neo 2 LED as light source. 

 

The pictures of the little boy (my son about 17-18 years ago) were taken on Ilford XP-2 I think (chromogenic BW negative film). I was using that for a while in the early 2000s as it was easy to get developed and I would then scan the negs on my LS4000.

 

Kid1 is just a straightforward shot, converted to BW in LR to remove a pink-purple cast and inverted to a positive in PS. I have cleaned and adjusted it a bit but no extra noise reduction in PS as the grain this is real film grain and I think it suits the picture.

 

KidTDN1 is the same picture processed from the raw in Topaz DeNoise. At first it looks really clean and almost free of grain or noise but closer inspection makes it look like the mad retoucher has been at work - the skin looks like plastic. And there are some weird noise artifacts. I should say I have only used Topaz DeNoise a few times and it might be possible to get better results as there are a few settings to play with.

 

Sollipulli922DS is a typical example of what I have been doing with my old slides. This is probably Velvia 50 (I would have to break the slide mount to check). Most of the processing, including sharpening and noise reduction, was done in Lightroom and then opened into Photoshop. I selected the sky and applied a gaussian blur of 1.5 followed by cleaning of spots. My slides are remarkably clean as they have lived mostly in the Fuji boxes in plastic bags for years so I don't touch them apart from a little blow with a jumbo blower. I then downsized to its present size to sharpen and decrease noise. I think this would have no problem with Alamy QC.

 

Sollipulli922TDNDS is the same slide processed from the raw in Topaz DeNoise (coffee time) and then some adjustments made on the resulting DNG file in Lightroom before opening and cleaning in PS. The sky is very clean but there is a very noticeable artifact on the boundary between the rocks and the sky which ruins the picture. I could have worked on this more but I don't think it is worth the effort as Sollipulli922DS is fine as it is.

 

Stromboli921DS and Stromboli921TDNDS have had similar treatment except I didn't apply any gaussian blurs in Photoshop. Stromboli921DS is noisy whereas Stromboli921TDNDS is very clean and there are no artifacts that I can see from running it through Topaz DeNoise so that one wins here I think. 

 

Finally Sollipulli923 is an example of the amazing dynamic range obtainable with the system I am using. I was able to restore detail in the snow as well as in black areas on the glacier. Most of the work was done on the raw nef in Lightroom apart from blurring the sky as above and spotting in Photoshop.

 

Please feel free to download, examine and comment. 

 

Very interesting thanks. Following your post I downloaded a free trial of Topaz DeNoise and gave it a whirl. It certainly removes grain and seems to do a pretty good job of preserving detail elsewhere. Even on the default settings it's quite severe though and causes some "watercolour" effect and haloing on high contrast edges, but on lower settings I like it, although in my opinion even on a setting of 0 for sharpening, it over-sharpens. It cleans up the skies so much I almost feel like it needs a little noise added back in! It might be best to create 2 versions of an image one with denoise (at default 15, 15), and one without and blend in PS? I just tried a 50/50 blend and like the result, it looks more natural.

 

Topaz recommends using it as early in the workflow as possible, so I tried it on a RAW file but it really messed up the colour balance and contrast and I couldn't recover it. So I found it best to create a 16 bit TIFF in LR at full resolution, then apply Topaz (it's quite slow, but faster than taking 10 shots and averaging in PS), then retouch and downsize in PS using ordinary bilinear. Downsizing at the end reduces any haloing.

 

I might well buy a copy. I especially like how easy it is to use, and the simple user interface with the minimum number of sliders.

 

Mark

Edited by M.Chapman
Link to comment
Share on other sites

17 hours ago, wiskerke said:

 

Any reason why you don't convert your layers to smart object > layers > smart objects > stack mode > median ?

Or mean in stead of median in some cases.

 

wim

 

Thanks, I just tried that and it works fine, and easier than setting the layer opacities.

 

I loaded the 10 images into PS CC using:-

    File>Scripts>Load files into Stack...

         with Attempt Automatically Align and Create smart object ticked

Then changed the stack mode

    Layer>Smart Objects>Stack Mode>Mean 

 

Mark

Link to comment
Share on other sites

I decided this morning to give these stacking methods a try to see if they have any practical value. I followed the instructions in this more recent article on the procedure which is essentially what wim was suggesting using smart objects as this is a lot faster than manually changing the layer opacities. I tried it on sets of 7 images copied from two different slides at ISO64, 1/2s, f11on my D810 and imported the raw nefs into Lightroom.

 

 For each set of images I did two different stacks - one with no noise reduction or sharpening whatsoever and a second with minimal colour noise reduction but no luminance noise reduction. I used the  Lightroom command on each set - Open as Layers in Photoshop - as suggested in the article (I reckon Open as Smart Object in Photoshop would do the same but I followed the procedure exactly). I did not bother with the layer alignment as the images are already aligned, identical in their crop and taken a few seconds apart with no movement of anything. I then selected the layers, converted to Smart Object and then stacked the images as median and then mean before rasterising. I then examined the resulting images and found nothing. I can honestly say I can see no difference between any of the resultant images in terms of noise or grain reduction.

 

So what is happening? Given that I am shooting at ISO64 I suspect that the noise generated by the copying process is extremely minimal so the "noise" in the images is probably film grain as the D810 is virtually noise free at low ISO. Now the theory behind this multiple image noise reduction is that the noise in each image in the stack is random but that cannot be the case for the film grain which will be fixed for each slide. So I am concluding that if there is any value to the multiple image noise reduction technique (and presumably there must be), then it does not work for slide copying as the grain is not random.

 

 

 

 

 

 

Link to comment
Share on other sites

2 hours ago, M.Chapman said:

 

Very interesting thanks. Following your post I downloaded a free trial of Topaz DeNoise and gave it a whirl. It certainly removes grain and seems to do a pretty good job of preserving detail elsewhere. Even on the default settings it's quite severe though and causes some "watercolour" effect and haloing on high contrast edges, but on lower settings I like it, although in my opinion even on a setting of 0 for sharpening, it over-sharpens. It cleans up the skies so much I almost feel like it needs a little noise added back in! It might be best to create 2 versions of an image one with denoise (at default 15, 15), and one without and blend in PS? I just tried a 50/50 blend and like the result, it looks more natural.

 

Topaz recommends using it as early in the workflow as possible, so I tried it on a RAW file but it really messed up the colour balance and contrast and I couldn't recover it. So I found it best to create a 16 bit TIFF in LR at full resolution, then apply Topaz (it's quite slow, but faster than taking 10 shots and averaging in PS), then retouch and downsize in PS using ordinary bilinear. Downsizing at the end reduces any haloing.

 

I might well buy a copy. I especially like how easy it is to use, and the simple user interface with the minimum number of sliders.

 

Mark

 

As far as I am concerned, the jury is out for me on Topaz DeNoise. I found exactly the same thing with the colour - it is very difficult or impossible in some cases to recover the original colour and I am pretty good with colur balancing in Lightroom in general. It might work well on some images but it is not something I will be using in general. It came as part of a package for me and the recent upgrade was free. It is also very slow on my computer. 

 

My approach to landscape images (copies of slides that is) is to process in Lightroom with a healthy sharpening and some noise reduction. If the sky is easily selectable, I apply a gaussian blur to the sky only at about 1.5 and downsize to around 2000 pixels short side which gives acceptable sharpness overall and would be acceptable for Alamy QC. They print well also with no real visible noise or grain in the sky.

 

You asked about Dropbox as well. I started there in 2017 and am on the free package. Basically I upload a folder of images which keeps them all together and create a link. I only use it for fun stuff like this. As I said a big advantage is that you can upload raw images.

 

 

 

Edited by MDM
Link to comment
Share on other sites

1 hour ago, MDM said:

I decided this morning to give these stacking methods a try to see if they have any practical value. I followed the instructions in this more recent article on the procedure which is essentially what wim was suggesting using smart objects as this is a lot faster than manually changing the layer opacities. I tried it on sets of 7 images copied from two different slides at ISO64, 1/2s, f11on my D810 and imported the raw nefs into Lightroom.

 

 For each set of images I did two different stacks - one with no noise reduction or sharpening whatsoever and a second with minimal colour noise reduction but no luminance noise reduction. I used the  Lightroom command on each set - Open as Layers in Photoshop - as suggested in the article (I reckon Open as Smart Object in Photoshop would do the same but I followed the procedure exactly). I did not bother with the layer alignment as the images are already aligned, identical in their crop and taken a few seconds apart with no movement of anything. I then selected the layers, converted to Smart Object and then stacked the images as median and then mean before rasterising. I then examined the resulting images and found nothing. I can honestly say I can see no difference between any of the resultant images in terms of noise or grain reduction.

 

So what is happening? Given that I am shooting at ISO64 I suspect that the noise generated by the copying process is extremely minimal so the "noise" in the images is probably film grain as the D810 is virtually noise free at low ISO. Now the theory behind this multiple image noise reduction is that the noise in each image in the stack is random but that cannot be the case for the film grain which will be fixed for each slide. So I am concluding that if there is any value to the multiple image noise reduction technique (and presumably there must be), then it does not work for slide copying as the grain is not random.

 

 

You may have missed a critical step. There has to be movement between each shot. Changing the alignment causes a change in the way the grain distribution interferes with the camera's sensor array. A variation on the Moire effect. Imagine trying photograph a piece of fabric which shows moire fringes in the shot. If the camera isn't moved, the moire pattern stays fixed and won't be removed by averaging multiple shots. I know that moire fringes are only seen when two regular arrays interfere with one another, and film grain isn't regular, so what are we talking about here? Well, this is more like moire "speckle" that occurs when a random pattern (film grain) interferes with the fixed sensor array. If the images you took didn't need alignment, then you didn't have enough movement to disturb the speckle.  The ES-1 is a very stable way of aligning everything :).

 

This effect  is explained in the papers referenced in a post Harry made earlier.

https://www.rangefinderforum.com/forums/archive/index.php/t-29937.html

http://www.photoscientia.co.uk/Grain.htm#problem

 

At first I didn't believe it (I'd only heard of moire fringes), but if I take 2 shots of the same slide with a small movement, then import into PS and align as layers, then toggle between at 200% them I see the way the film gain is rendered does indeed change. It's this variation that the stacking technique averages. The effect isn't huge, but it does seem to help IMHO and as illustrated by the shots I posted earlier. It's not removing the true grain in the film (which is fixed), but it does reduce the moire speckle effect that can make the grain look worse. But I do agree that Gaussian blur of the sky is faster and more effective (it can remove the grain itself) for skies. I'm currently thinking Topaz might be better still but have yet to try it on a wider range of images. 

 

Mark 

Edited by M.Chapman
Link to comment
Share on other sites

1 hour ago, M.Chapman said:

 

You may have missed a critical step. There has to be movement between each shot. Changing the alignment causes a change in the way the grain distribution interferes with the camera's sensor array. A variation on the Moire effect. Imagine trying photograph a piece of fabric which shows moire fringes in the shot. If the camera isn't moved, the moire pattern stays fixed and won't be removed by averaging multiple shots. I know that moire fringes are only seen when two regular arrays interfere with one another, and film grain isn't regular, so what are we talking about here? Well, this is more like moire "speckle" that occurs when a random pattern (film grain) interferes with the fixed sensor array. If the images you took didn't need alignment, then you didn't have enough movement to disturb the speckle.  The ES-1 is a very stable way of aligning everything :).

 

This effect  is explained in the papers referenced in a post Harry made earlier.

https://www.rangefinderforum.com/forums/archive/index.php/t-29937.html

http://www.photoscientia.co.uk/Grain.htm#problem

 

At first I didn't believe it (I'd only heard of moire fringes), but if I take 2 shots of the same slide with a small movement, then import into PS and align as layers, then toggle between at 200% them I see the way the film gain is rendered does indeed change. It's this variation that the stacking technique averages. The effect isn't huge, but it does seem to help IMHO and as illustrated by the shots I posted earlier. It's not removing the true grain in the film (which is fixed), but it does reduce the moire speckle effect that can make the grain look worse. But I do agree that Gaussian blur of the sky is faster and more effective (it can remove the grain itself) for skies. I'm currently thinking Topaz might be better still but have yet to try it on a wider range of images. 

 

Mark 

 

At which point I give up. If true, then it would involve far too much extra work moving the slide around, refocusing and reshooting and then aligning them in Photoshop. I can't see any real difference between the images you posted either which would convince me that there is any practical value in the technique. It's all very interesting although I have to say I don't understand how it could apply to film grain and I can't see anything in those links that refers directly to this layer stacking and aligning technique (I am probably missing it and the fact that I don't understand it means only that and not that it is not true).

 

In other words, the effects are so minor if they are indeed visible that it is not worth the time taken in my opinion. Doing some global noise reduction and balancing of sharpness on the single  raw in Lightroom followed by local noise reduction (whether by blurring or additional noise reduction) on areas where noise is apparent is a far more productive approach for me and gets me to a quality that would print well at A4 at least and pass Alamy QC. 

 

The problem I am finding with Topaz is that it is putting artifacts between various boundaries in the images (land and sky, person and background) and I am happier with my tried and tested techniques for local noise reduction. But I have not used Topaz on enough images to make a really definitive judgment. 

 

 

 

 

 

Edited by MDM
Link to comment
Share on other sites

22 minutes ago, MDM said:

At which point I give up. If true, then it would involve far too much extra work moving the slide around, refocusing and reshooting and then aligning them in Photoshop.

 

I just left my AF and IS on, that did the trick, no need to move the slide by hand, it only needs a tiny movement. Image alignment happens automatically upon loading by PS into the stack, although it takes a little time. The second reference is all about film grain. The stacking technique came from this paper on super-resolution. I just put the two techniques together. But I agree the gains are small compared to the other techniques (Gaussian Blur and Topaz Denoise), but there's no reduction in IQ, so I still think it's potentially useful. It might also be useful for reducing sensor noise when opening up those deep Velvia shadows too, although if you have a bright light source, are shooting at low ISO and have a decent DSLR it shouldn't be needed.

 

Mark

Link to comment
Share on other sites

The lens I am using is an old manual focus lens so has no AF or stabilisation. I would not use AF for such close up work anyway as it is quite probably not accurate enough and using stabilisation on a tripod is ill-advised. There is a risk of having poorly focused images I would think. 

 

In any case I have seen enough to rule this out as a technique I would use in practice. I am running some tests with different settings in Topaz DeNoise to see if it is usable for my purposes. Setting sharpness to zero helps but does not completely remove artefacts it seems. My original technique is likely to remain my preferred method I am thinking. 

Link to comment
Share on other sites

42 minutes ago, MDM said:

The lens I am using is an old manual focus lens so has no AF or stabilisation.

OIC

42 minutes ago, MDM said:

I would not use AF for such close up work anyway as it is quite probably not accurate enough and using stabilisation on a tripod is ill-advised. There is a risk of having poorly focused images I would think. 

 

Fortunately I'm using a mirrorless system where I find the contrast based AF is much faster, more accurate and reliable than I can achieve manually (even using magnified live view and focus peaking). Wide area AF also deals with variation from slide to slide due to slight variations in mount thickness and transparency flatness. My IS doesn't hunt on a tripod, maybe it turns off, but it does re-centre the lens, when I half press the shutter before setting off the self timer. I think I get very good shot to shot focus accuracy, indeed if I didn't my 10 shot stack would have been a right mess. The exposure I'm using is also quite short at 1/250th (at ISO200 f/8) as the LED light source is pretty bright.

 

All that being said, I'm very impressed with the detail in your image Sollipulli922DS.jpg. Out of curiosity, do you recall the lens and film stock? I don't think I've got many transparencies that are that good from corner to corner.

 

I'll be trying some more tests with Topaz Denoise tonight too.

 

Mark

 

 

Edited by M.Chapman
Link to comment
Share on other sites


 

24 minutes ago, M.Chapman said:

 

 

All that being said, I'm very impressed with the detail in your image Sollipulli922DS.jpg. Out of curiosity, do you recall the lens and film stock? I don't think I've got many transparencies that are that good from corner to corner.

 

I'll be trying some more tests with Topaz Denoise tonight too.

 

Mark

 

 

 

 

It was either Velvia 50 or Fujichrome 100. I would have to break the mount to check which I don’t really want to do. The camera would have been an OM1 or 2 and probably a Zuiko 50 f1.8 or maybe a Zuiko 85 f2. I changed to Nikon in 2000 I think. 

 


 

 

Edited by MDM
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.