DSLR vs Plustek vs Epson.

[Picturenaut]

I'm just using Lightroom and simply invert the curve. It's a bit of a pain after to do the retouching due to the inverted curve but there's not much to do besides dust cleaning and some basic adjustment.

Thanks again. Hm, I am in fact very pleased with the many film presets of Silverfast and wouldn't like to miss them. I guess in LR you need to work out some profiles as standard settings...

Here are some images scanned with my Braun FS120+Silverfast I have already in my little DPR gallery:


Kodak Portra 160, 120 film, New Mamiya 6 (75mm lens)

(actually the green skin of this lady is exactly how she looked like )


Ilford HP5, 120 film, New Mamiya 6 (50 mm lens)


Kodak T-Max 100, 120 film, New Mamiya 6 (75mm lens)

--
Picturenaut

 

[ochastan]

I'm just using Lightroom and simply invert the curve. It's a bit of a pain after to do the retouching due to the inverted curve but there's not much to do besides dust cleaning and some basic adjustment.

Thanks again. Hm, I am in fact very pleased with the many film presets of Silverfast and wouldn't like to miss them. I guess in LR you need to work out some profiles as standard settings...

Here are some images scanned with my Braun FS120+Silverfast I have already in my little DPR gallery:


Kodak Portra 160, 120 film, New Mamiya 6 (75mm lens)

(actually the green skin of this lady is exactly how she looked like )


Ilford HP5, 120 film, New Mamiya 6 (50 mm lens)


Kodak T-Max 100, 120 film, New Mamiya 6 (75mm lens)

--
Picturenaut

I made a quick profile in LR. It takes no time. The actual editing takes a little bit more time but again, nothing dramatic.
I’m going to start scanning MF in a few weeks and will report back. I have a Mamiya 7 and used a 500c/m for most of my past work.

--
www.olivierchastan.com

 

[ochastan]

I'm just using Lightroom and simply invert the curve. It's a bit of a pain after to do the retouching due to the inverted curve but there's not much to do besides dust cleaning and some basic adjustment.

Thanks again. Hm, I am in fact very pleased with the many film presets of Silverfast and wouldn't like to miss them. I guess in LR you need to work out some profiles as standard settings...

Here are some images scanned with my Braun FS120+Silverfast I have already in my little DPR gallery:


Kodak Portra 160, 120 film, New Mamiya 6 (75mm lens)

(actually the green skin of this lady is exactly how she looked like )


Ilford HP5, 120 film, New Mamiya 6 (50 mm lens)


Kodak T-Max 100, 120 film, New Mamiya 6 (75mm lens)

--
Picturenaut

It looks good. How fast is the scanner?
--
www.olivierchastan.com

 

[Overrank]

Here are the scans I had promised. I did not bother doing a scan with the Epson 850 as it’s not even in the same realm as the other two.

Film stock is 1969 Kodak Tri-X.

Leica M10M with Nikon 60/2.8 AF at f8 and ES-2 below. The photo was simply inverted and adjusted black and white points - nothing else. Funny enough the Leica recorded the lens as the Tri-Elmar at 16mm ;-)


Leica M10M scan

Plustek 8300 with no settings except for Negafix Kodak Tri-X setting.

View attachment 310a5be9310444cfb19c13a0db3b1601.jpg
Plustek 8300i Scan

I think the images speak for themselves.

You might want to explain what you think they’re saying …..

You can just go back to the very first original post that I started this thread with which got buried over the past 5 days . The DSLR scan is way better in all aspects: grain structure, details, depth, dynamic range, even how it resolves scratches on old film.

Sorry, I thought they were saying £7,500 specialist digital camera produces digitisations very similar to £350 film scanner

I really don’t see that much of a difference, particularly since you’re using the 7200ppi scan without halving it (as is well know, it only gives 3600dpi of data) and have used a straight inversion for one, which has then been hand optimised, but for the Plustek scan rather than taking a raw scan and hand inverting it, the the Negafix Tri-X setting has been used.



I’m two scanners away from a Plustek, so I’ve no axe to grind, but I think it’s better for people to make up their own mind as they are perfectly good scanners (available new, with a warranty and support).

 

[ochastan]

Here are the scans I had promised. I did not bother doing a scan with the Epson 850 as it’s not even in the same realm as the other two.

Film stock is 1969 Kodak Tri-X.

Leica M10M with Nikon 60/2.8 AF at f8 and ES-2 below. The photo was simply inverted and adjusted black and white points - nothing else. Funny enough the Leica recorded the lens as the Tri-Elmar at 16mm ;-)


Leica M10M scan

Plustek 8300 with no settings except for Negafix Kodak Tri-X setting.

View attachment 310a5be9310444cfb19c13a0db3b1601.jpg
Plustek 8300i Scan

I think the images speak for themselves.

You might want to explain what you think they’re saying …..

You can just go back to the very first original post that I started this thread with which got buried over the past 5 days . The DSLR scan is way better in all aspects: grain structure, details, depth, dynamic range, even how it resolves scratches on old film.

Sorry, I thought they were saying £7,500 specialist digital camera produces digitisations very similar to £350 film scanner

I really don’t see that much of a difference, particularly since you’re using the 7200ppi scan without halving it (as is well know, it only gives 3600dpi of data) and have used a straight inversion for one, which has then been hand optimised, but for the Plustek scan rather than taking a raw scan and hand inverting it, the the Negafix Tri-X setting has been used.

I’m two scanners away from a Plustek, so I’ve no axe to grind, but I think it’s better for people to make up their own mind as they are perfectly good scanners (available new, with a warranty and support).

I obviously did not buy the camera to scan so it’s an irrelevant argument in that case. I also do not “recommend” anything to anyone just reporting my findings. I’ll do a slightly more pronounced close-up (not even pixel peaking) and the differences become pretty obvious. Finally, the decision is as always based on one’s needs and uses. I print large and for exhibits. If it’s to scan for home or other similar usage then yeah, no one should give a crap about this thread which is clearly aimed at people who care more than the average user.

--
www.olivierchastan.com

 

[Overrank]

I obviously did not buy the camera to scan so it’s an irrelevant argument in that case. I also do not “recommend” anything to anyone just reporting my findings.

I don’t want to get into an “Open Talk” discussion of semantics, but your original post certainly reads that way, without using the word “recommend”

I’ll do a slightly more pronounced close-up (not even pixel peaking) and the differences become pretty obvious.

My issue is that you’re not comparing like with like, nothing to do with the resolution. You’ve used the SilverFast presets whereas hand optimizing the Leica image. Also you’re using the SilverFast image in a way that’s non-optimal by scanning so large and then not shrinking it back to the limits of its resolution.



Plus, tbh I can’t actually see any significant difference in the displayed images.

 

[smithim]

Plus, tbh I can’t actually see any significant difference in the displayed images.

Odd, if I pull them up side by side, I see very significant differences.

 

[Overrank]

Plus, tbh I can’t actually see any significant difference in the displayed images.

Odd, if I pull them up side by side, I see very significant differences.

There are contrast differences, but one reason for that is the Negafix profile which I’ve found can be rather harsh with B+W.

 

[ochastan]

Here are the two scans zoomed in.

The Plustek scan was redone at 3600ppi since it seemed to be important to Overrank. I also used the "other" setting in Negafix with monochrome curve which is pretty flat. The Leica scan is exactly the same as before just zoomed in. No I did not downsample the M10M scan.



And for the record, there is zero manipulation with the M10 except for the inversion and black and white points.




M10M Scan Zoom




Plustek 8300 Scan Zoom



--
www.olivierchastan.com

 

[smithim]

Plus, tbh I can’t actually see any significant difference in the displayed images.

Odd, if I pull them up side by side, I see very significant differences.

There are contrast differences, but one reason for that is the Negafix profile which I’ve found can be rather harsh with B+W.

The grain and detail is finer in the camera version, giving a sharper result overall - the eye lashes being a case in point.

There may be valid reasons for the differences, but to say that they don't exist is misleading, IMHO

 

[Overrank]

Plus, tbh I can’t actually see any significant difference in the displayed images.

Odd, if I pull them up side by side, I see very significant differences.

There are contrast differences, but one reason for that is the Negafix profile which I’ve found can be rather harsh with B+W.

The grain and detail is finer in the camera version, giving a sharper result overall - the eye lashes being a case in point.

There may be valid reasons for the differences, but to say that they don't exist is misleading, IMHO

The problem is that almost all of the difference (which is slight at the displayed resolution) can be explained by the Negafix profiles. With some black and white films in SilverFast (e.g. Kentmere) I now scan without using the profiles for just that reason. You’ll see that the rescanned images have a smaller difference, even close up.

 

[Overrank]

Here are the two scans zoomed in.

The Plustek scan was redone at 3600ppi since it seemed to be important to Overrank.

Thanks for rescanning - it wasn’t the 3600ppi setting itself, it was that the 7200ppi setting produces a enormous file with only 3600ppi of data in it, so it can be resized to 3600 without much of a loss in quality. The 3600ppi setting produces images with a smaller resolution (say 2800dpi) but they are not as bloated.

I also used the "other" setting in Negafix with monochrome curve which is pretty flat.

I would have scanned as a slide in raw and then inverted in the same way as the Leica image, but the Negafix “other” settings are much more neutral (as can be seen in the rescan), and what I use for some B+W films for precisely that reason.

I’m surprised that the Plustek picks up the scratches in the film when the camera capture does not. In the OOF area to the top right I prefer the Plustek’s rendering to the camera, apart from the scratches. But then I don’t particularly like the current trend for “grain” in B+W film as a differentiator to digital (Kodak, Fujifilm, Ilford etc spend many millions of pounds trying to minimise grain because it wasn’t wanted pre-digital).

The Leica scan is exactly the same as before just zoomed in. No I did not downsample the M10M scan.

And for the record, there is zero manipulation with the M10 except for the inversion and black and white points.


M10M Scan Zoom


Plustek 8300 Scan Zoom

 

[ochastan]

Here are the two scans zoomed in.

The Plustek scan was redone at 3600ppi since it seemed to be important to Overrank.

Thanks for rescanning - it wasn’t the 3600ppi setting itself, it was that the 7200ppi setting produces a enormous file with only 3600ppi of data in it, so it can be resized to 3600 without much of a loss in quality. The 3600ppi setting produces images with a smaller resolution (say 2800dpi) but they are not as bloated.

I also used the "other" setting in Negafix with monochrome curve which is pretty flat.

I would have scanned as a slide in raw and then inverted in the same way as the Leica image, but the Negafix “other” settings are much more neutral (as can be seen in the rescan), and what I use for some B+W films for precisely that reason.

I’m surprised that the Plustek picks up the scratches in the film when the camera capture does not. In the OOF area to the top right I prefer the Plustek’s rendering to the camera, apart from the scratches. But then I don’t particularly like the current trend for “grain” in B+W film as a differentiator to digital (Kodak, Fujifilm, Ilford etc spend many millions of pounds trying to minimise grain because it wasn’t wanted pre-digital).

The Leica scan is exactly the same as before just zoomed in. No I did not downsample the M10M scan.

And for the record, there is zero manipulation with the M10 except for the inversion and black and white points.


M10M Scan Zoom


Plustek 8300 Scan Zoom

Lucky for you my Plustek 8300i is now for sale. DM me to get your dream scanner

--
www.olivierchastan.com

 

[Overrank]

Lucky for you my Plustek 8300i is now for sale. DM me to get your dream scanner

I spent years scanning many thousands of negatives with a Plustek, but it’s now in storage and I’m using a CoolScan, so unfortunately I’ll have to pass. I imagine it will sell quickly though. Good luck !

 

[sybersitizen]

I’m surprised that the Plustek picks up the scratches in the film when the camera capture does not.

That's typically been the case with my Pacific Image scanners compared to camera capture. I don't know why it's the case, but I assumed it was a common experience.

 

[Picturenaut]

It looks good. How fast is the scanner?
--
www.olivierchastan.com

Slow, as I wrote, at least if you want higher resolution, enough to see film grain (depends on the film, of course). 3600 dpi scanning of 6x6 medium format images in "print quality" setting needs about 5 min minimum - 6400 dpi (highest resolution) goes about 10-15 min per image. I need to measure that more exactly next time I scan images. That's the reason why I am considering to switch to DSLR/ML scanning. But the quality of the scanner with Silverfast 9 is really good...

 

[Overrank]

I’m surprised that the Plustek picks up the scratches in the film when the camera capture does not.

That's typically been the case with my Pacific Image scanners compared to camera capture. I don't know why it's the case, but I assumed it was a common experience.

I’ve not noticed this with the CoolScan, and looking back a few years my Plustek scans look clean in the OOF areas in B+W. I do mostly scan fresh film though so it doesn’t get much chance to get scratches. Most of the older films I’ve scanned have been colour (either negative or slide) so I’ve had ICE on, which removes scratches as effectively as it removes dust.

 

[Svein Eriksen]

IMO the light source influence how visible dust and scratches are. If I scan old BW negatives where I can't use ICE I find that an Epson flatbed show dust and scratches less than a Nikon Coolscan. Some of it probably is because of lower (real) resolution, but I believe the light source also contribute.

 

[Overrank]

IMO the light source influence how visible dust and scratches are. If I scan old BW negatives where I can't use ICE I find that an Epson flatbed show dust and scratches less than a Nikon Coolscan. Some of it probably is because of lower (real) resolution, but I believe the light source also contribute.

My experience with dust on flatbeds was one of the things that moved me over to dedicated film scanners in the first place. But at the time I was mostly scanning old colour negatives and so getting ICE was a big plus point. I still mostly use colour but with the price of colour film at the moment I’m using more B+W as test rolls.

The light source certainly changes the look of the scan - there’s a company that makes a product called Scanhancer for some Minolta scanners which diffuses the light source to smooth off the grain ( http://www.scanhancer.com ). Apparently something similar (“grain dissolver”) is in the later Minolta scanners ?

 

[ochastan]

IMO the light source influence how visible dust and scratches are. If I scan old BW negatives where I can't use ICE I find that an Epson flatbed show dust and scratches less than a Nikon Coolscan. Some of it probably is because of lower (real) resolution, but I believe the light source also contribute.

I found ICE and SRD to be rather unsatisfactory. It’s oK for low level of dust but it does affect the picture at some point. Printing for gallery at 16x20 or 30x40 and you start noticing these artefacts from the correction. Still need to correct manually in the end in my experience.