DSLR scanning vs Dedicated flatbed scanning.

[robertraymer]

Currently I am using an Epson V850 to scan all of my negatives from 35mm up to 8x10 including wet plates, and while I don't have any specific complaints about it, I always feel that the process takes quite a while and that I am never quite getting what I want out of the negatives. Im sure that some of this has to do with me simply needing more practice, comfortability with the scanner and software, etc, but I have recently been considering other options that may be faster/easier/better. This lead to me considering purchasing a copy stand and using my LED light panel I use as a Lightbox and my D800 to "scan" my images as RAW files and then edit the files in Lightroom, something that has been made even more potentially appealing by a review of NegativeLab Pro I just read.

I was curious as to what other peoples opinions on the matter are? Do you feel there are significant pros/cons to either? If I did make the switch, is there anything specific I should know or consider?

Just some other information that may be useful (or may not, I don't know). Currently I do not batch scan or batch process film scans, and do not "need" that capability, though it would probably be nice if it was easy to do. Most of my scans are usually made so that I have something digital to share on social media/forums/etc, but I do have/would like to keep the ability to make quality prints up to around 16x20 or possibly larger. Also I still shoot a lot of digital work, so RAW editing in LR/PS is actually a process I am more familiar/comfortable with than evaluating/processing scans with the epson/vuescan software.

Thoughts?

[Larry Gebhardt]

Years ago I wrote up my experience doing this for 35mm film compared to a drum scanner. The upshot was the D800E was almost as good resolution wise, equal on tonality, and was much more convenient. I use a Nikon PB4 bellows with the slide/negative holder.

When I have scaled this up to medium and large format some issues were found, almost all related to alignment, film flatness, and moving the film for stitching. Ideally I'd design a jig so the alignment wouldn't drift and there would be a repeatable x-y table for moving the negative for stitching. So far I've just found it easier to use the Epson or the drum scanner than designing that hardware solution or setting up the copystand.

[Pere Casals]

The D800 "scanner" outresolves the V850 for 35mm, with right lens/procedure. For MF, to outresolve the V850 you have to take several images and stitching in Photoshop(or other). For LF...

As the format gets larger the V850 is more suitable. Just make some tests with a tripod before purchasing the copy stand. The V850 is decent for MF, good for 4x5 and superb for 8x10. So IMHO the D800+copy stand would be suitable for 35mm and for jobs requining not much quality.

Just check how it goes before purchasing the stand, use a tripod on a table or place the lightbox on the ground.

With the stand you have to control dust even better than with the scanner, you can use a HEPA air purifier (home models) to get rid of dust.

[robertraymer]

Thanks for the input so far. I will keep it mind. right now it sounds like "right tool for the job" kind of thing. Ill be gone for a few days but am anxious to hear advice/experiences from others as well.

[Peter De Smidt]

There are very extensive threads on this in the DIY section.

[rdeloe]

There are several massive threads on DSLR scanning on the forum that you should check out. Many people have been down this path and are reporting very good results with standard digital cameras like your D800. Some museums seem to be going this route too, albeit with much more specialized equipment. Here's an interesting story about digitizing Autochromes using medium format digital backs to camera scan these very delicate photographic media. They're using stitching when they need higher resolution. (Thanks to Jeffrey G. for putting me on to this story.) https://nglibrary.ngs.org/public_home/filmpreservationblog/Preservation-of-Autochromes

I "camera scan" 4x5 negatives (black and white only so far) and process them in Lightroom. I adamantly reject any notion that camera scanning is for times when you don't need much quality. If you have good equipment and a controlled workflow, you can get excellent results. I digitize my 4x5 negs at 2,667 ppi. This is more than enough for making large, high quality prints. If you only need a good quality "scan" for electronic display, you can easily get away with 3 overlapping frames for one 4x5 negative; on my setup that's still 1,500 ppi. If you math that out, you get a file that is 6,000 x 7,500 ppi -- or 45 MP. Over in digital photography land, nobody who has a clue would ever tell you that 45 MP isn't "enough" to make a good quality print!

This is a 13,221 x 10,554 scan (2,667 ppi) after processing. It's stitched from 12 individual frames.

192412

And here's a 3,294 x 2,626 pixel crop.

192413

I'm currently not using anything fancy. I have a home made scanning template that allows me to quickly make 12 frames from a wet-mounted negative. I use an APS-C camera with a 24 MP sensor, and an Olympus OM macro lens. A copy stand would be better, but a sturdy old tripod is working well in the interim. Details here if you're interested: https://www.largeformatphotography.info/forum/showthread.php?150162-Camera-scanning-on-the-cheap-an-example-approach The only update to this setup I've made is adding additional masking to control extraneous light.

As for processing, I do everything in Lightroom except for spotting, which is excruciating in Lightroom with files this size, but painless in Photoshop. I've tried various add-on tools for inverting the negatives, but (like presets in general) don't see any benefits relative to my own needs. I do the stitching in Lightroom, then invert the negative, export as a TIFF, spot in Photoshop, and re-import the cleaned-up TIFF for further processing. Of course all of this can be done with lots of other tools.

There's only one occasional major downside with camera scanning and stitching, and that's cases where the component images won't stitch. A solution is to use a very precise template when you shoot the frames, and then stitch them in PT GUI with templates. This is discussed at length in the "DSLR scanning" threads; see especially the posts by Peter De Smidt where he talks about the very cool semi-automatic system he and a friend made.

[Peter De Smidt]

One further note: cameras have improved significantly since when I first started to do this. Back then, I had a Nikon D200. It had limited dynamic range and resolution, but it still wasn't hard to beat an Espon scanner. These days there are a number of great cameras that offer high resolution, high dynamic range, and zero-vibration shutters. All of those things matter. What this all means is that one could get very high quality while doing less stitching. If I remember rightly, I used 25 exposures to capture a 4x5 piece of film with my D200. If you're making 16x20s, I bet that 3 exposures with a D800 or better camera could do it, and it'd be very fast.

Dslr scans and film scanner scans don't necessarily line-up pixel-wise. What I mean is that even when the pixel dimensions of the two scans are the same, it doesn't follow that they are equal in quality. I'll limit myself to Epson scanners, in particular I'm not talking about drum scanners, but my D200 outperformed my Epson scanner significantly, even when the pixel dimensions were the same. The dslr produced cleaner and richer files. Yes, those are subjective terms, but they were also supported by step wedge and resolution target scans. Lastly, Lightroom, Capture One, Photoshop, all are significantly more powerful and easier to use than any scanner software I've used, including Nikon Scan, Epson Scan, Vuescan, Colorgenius....

I believe that Lightroom now allows outputting a raw file from stitched raw files. If so, that's huge.

[Pere Casals]

with 3 overlapping frames for one 4x5 negative; ...

The OP's D800 crafts around 18MPix effective in a single frame, not all the 36Mpix are worth, there is a loss in the lens and D800 has a low pass filter (OLPF) (800E and 810 don't have the OLPF, so they are 25Mpix effective able in practice), also we have the bayer mosaic so true resolution may even be well under 18MPix for color resolution, anyway color resolution is less important for accutance tha resolution in the luminance.

APSC cameras are usualy crafting 12MPix effective perhaps.

Then the frames have to overlap for the stitching so a frame yields less than in theory, as a frame overlaps with all neighbour crops then the effetive average recorded pixels can be around 60%, so a D800 perhaps averages around 10MPix per frame. I say "averages" because frames in the corner have a better yield than frames in the center, as there are less overlaps.


A V850 4x5 scan yields around 125MPix effective, so to compete with a V850 a D800 requires from 9 to 16 frames-crops to be stitched in Photoshop, this is a 3x3 or a 4x4 image mosaic. A D800E would require 9 shots.

The question is if it's worth to be stitching a lot of crops or not, as many LF shots have little or no detail beyond what the V850 is able to scan.

A DSLR scanner may even outresolve any drum if the number of crops is large enough and the setup is smart enough...

[rdeloe]

Pere, the bit after the semi-colon in my sentence matters: "If you only need a good quality "scan" for electronic display, you can easily get away with 3 overlapping frames for one 4x5 negative; on my setup that's still 1,500 ppi. If you math that out, you get a file that is 6,000 x 7,500 ppi -- or 45 MP." On my setup, I'm shooting a 24Mp Fuji X-T2. It's also not a Bayer sensor, not that this makes much difference. I get excellent results quickly and easily, so I'm happy. Happy is good -- leaves me more time to make pictures.

The OP wondered about camera scanning. My advice to the OP: don't listen to forum dwellers -- try it for yourself and see if you're happy. If you're not, you already own a good scanner that will produce great results!

[Pere Casals]

Ok, but think that one thing is ppi and another one is effective ppi, those 1500 theoric may end in 1000 effective...

[rdeloe]

This is a very important point Peter. It's noteworthy that DT Cultural Heritage (the company whose equipment is showcased in that post I linked to above) went with a camera-based system for digitizing items for conservation purposes. Have scanners reached the end of the line in terms of development? It rather looks that way. In contrast, no end is in sight for improvements to sensors for digital cameras. In that respect, it makes perfect sense that a company that wants to sell equipment to museums for digitizing flat artefacts would go with a camera-based system.

When I read the article, my first thought was that this was a strange business to get into given how niche film is, especially large format film. But then the light went and I thought about the vast amount of existing historical material in collections around the world that needs to be digitized -- negatives, positives, prints, etc. -- all eventually (hopefully!) to be made available online. And then there's all the new material HABS/HAER photography. The conservation world knows what it's about, so presumably they have no concerns about the quality of the files generated by "camera scanning".

The business model makes a lot more sense in that context... Sadly, these systems are seriously expensive so they won't be making an appearance in my life!



One further note: cameras have improved significantly since when I first started to do this. Back then, I had a Nikon D200. It had limited dynamic range and resolution, but it still wasn't hard to beat an Espon scanner. These days there are a number of great cameras that offer high resolution, high dynamic range, and zero-vibration shutters. All of those things matter. What this all means is that one could get very high quality while doing less stitching. If I remember rightly, I used 25 exposures to capture a 4x5 piece of film with my D200. If you're making 16x20s, I bet that 3 exposures with a D800 or better camera could do it, and it'd be very fast.

Dslr scans and film scanner scans don't necessarily line-up pixel-wise. What I mean is that even is the pixel dimensions of the two scans are the same, it doesn't follow that they are equal in quality. I'll limit myself to Epson scanners, in particular I'm not talking about drum scanners, but my D200 outperformed my Epson scanner significantly, even when the pixel dimensions were the same. The dslr produced cleaner and richer files. Yes, those are subjective terms, but they were also supported by step wedge and resolution target scans. Lastly, Lightroom, Capture One, Photoshop, all are significantly more powerful and easier to use than any scanner software I've used, including Nikon Scan, Epson Scan, Vuescan, Colorgenius....

I believe that Lightroom now allows outputting a raw file from stitched raw files. If so, that's huge.

[rdeloe]

I believe that Lightroom now allows outputting a raw file from stitched raw files. If so, that's huge.

Indeed it does. In my current setup the input is 12 DNG files (I convert my Fuji RAF files to DNG with a third part developer, outside of Lightroom). Inside Lightroom, I stitch them together, and it creates a DNG. Unfortunately, I end up having to work on the file as a TIFF because Lightroom does not have a usable tool for inverting negatives. It's easy to invert them: you simply flip the tone curve. However, all of the sliders are then "backwards", which makes editing the files a huge pain. Plus, as I mentioned above, spotting is awful in Lightroom compared to Photoshop -- so I end up exporting the DNG as a 16 bit TIFF, spotting it in Photoshop, and then bringing it back into Lightroom.

I was initially concerned that this would be a problem, but a DNG is basically a TIFF so it hasn't made any difference that I can detect.

[Peter De Smidt]

As Robert mentions at the start of the thread, there's a new plugin for Lightroom: NegativeLab Pro. It looks promising, especially for scanning color negatives!

[Eric Leppanen]

As Robert mentions at the start of the thread, there's a new plugin for Lightroom: NegativeLab Pro. It looks promising, especially for scanning color negatives!BTW, I received notification today that version 2.0 of Negative Lab Pro is now available, with much improved support for flatbed scanners among other things:

https://forums.negativelabpro.com/t/introducing-negative-lab-pro-v2-0/256

I have been experimenting with a Fuji X-T3 (26 Mpix) and the previous version of Negative Lab Pro for proofing LF color and B&W negs, as well as digitizing old family photos. Because I am not striving for critical sharpness at this point, I have been using an inverted tripod rather than a copy stand and taking a single shot rather than stitching. My results so far with negative materials have been pretty darn good, good enough for web display and small prints. Chrome has been more challenging, for some reason images tend to come out "hot" (high contrast with a tendency to blow out highlights) so I've had to reduce exposure and more aggressively open up shadow areas in Lightroom. I am not sure if my old, uncalibrated monitor is part of the problem, or perhaps I am hitting some limitations in my camera's dynamic range?

Regardless, this rudimentary setup seems to be working well as long as one stays within its crude tolerances and the resolution limits of the camera. On another forum one of the MF digital reps is claiming that the super high-end DT Film Scanning Kit can in some cases outperform drum scans (particularly in terms of shadow detail recovery), so it will be interesting to see how good single shot scans can get once models such as the Fuji GFX100 (102 Mpix) come down in price.

[rdeloe]

Just tried it. The author has done a good job of integrating his plugin with Lightroom. Unfortunately, you still need to make a separate TIFF copy if you want the sliders to go in the right direction. Another catch, for me, is the matter of spotting. Yes, you can send the TIFF to Photoshop from inside Lightroom. It works, but for reasons I haven't been able to figure out, the TIFF that Photoshop saves back into your Lightroom catalogue becomes massive. Anyway, if you want to stay inside Lightroom, this plug-in is worth exploring.

** Important: I'm working in black and white, which is a bit more straightforward when it comes to inverting (in that you don't need special tools). Inverting colour negatives is a whole other thing (and I have no opinion on how well that works with this plug-in!) **


As Robert mentions at the start of the thread, there's a new plugin for Lightroom: NegativeLab Pro. It looks promising, especially for scanning color negatives!

[asnapper]

BTW, I received notification today that version 2.0 of Negative Lab Pro is now available, with much improved support for flatbed scanners among other things:

https://forums.negativelabpro.com/t/introducing-negative-lab-pro-v2-0/256

Regardless, this rudimentary setup seems to be working well as long as one stays within its crude tolerances and the resolution limits of the camera. On another forum one of the MF digital reps is claiming that the super high-end DT Film Scanning Kit can in some cases outperform drum scans (particularly in terms of shadow detail recovery), so it will be interesting to see how good single shot scans can get once models such as the Fuji GFX100 (102 Mpix) come down in price.

Once Fuji add Multishot I think the GFX 100 will be able to produce results rivalling a drum scan

[Pere Casals]

Once Fuji add Multishot I think the GFX 100 will be able to produce results rivalling a drum scan

Sorry, but IMHO Multishot is LOL :)

You can make a test, just download some 100% crops of multishot side by side tests, then sharpen each crop to its best, you will find the same:

192446

192447


Just take this image and sharpen with Ps the "1 Shot" crop and compare:

https://www.dpreview.com/files/p/articles/9807859981/H6D-400c_MS_comparison_Zoom.jpeg
https://www.dpreview.com/news/9807859981/the-hasselblad-h6d-400c-multi-shot-spits-out-insane-400mp-images

You have a lot of side by side posted tests, what I find really funny is that the shots are posted but no optimal sharpening is applied latter, with is the first a Photoshop professional checks after loading a new image.


This suggests that Multishot feature mostly has a better embedded sharpening (or super-resoultion effect from accumulating images) than doing a real job, because the lens normally is a limiting factor.

Usually 100 MPix MF sensors yield around 50 to 70Mpix effective...

Even with a cheap DX DSLR of the Nikon D3400 class you may beat a drum by stitching a number of shots, for example stitching 16 shots (4x4) for a 35mm film frame, (use a reversed lens in this case), if stitching crops then we have no practical limitations.

[StuartR]

I have the Panasonic S1, which has multishot. I have tried it using Leica S lenses, in particular the 120mm APO macro, which should have no problem resolving the claimed 100mp. The mutli shot is not a gimmick. It really works. But it does not give 100mp of pure sharp data. If you zoom to 200-300%, you can occasionally see glitches. In comparison to the Leica S 006 itself, it has a larger file size, but the resolution is not so far apart. In some cases the greater acutance of the CCD in the S made it look sharper, but in others the S1 looked better. What it does do, however, is make a better file than the camera can make without multi-shot. I have no question about that. I intend to try it for scanning, but I have not had the time to dedicate to it yet. I do think the GFX100 will be a superb camera for camera based scanning, multishot or not.

I do think camera based scanning is the future. The biggest issue is not the cameras at all, but the transport. Once someone builds something that can hold the cameras rock solid and plane parallel with the film flat, and provide a good light source, I think film scanners will start to disappear from high end applications. Digital Transitions seems to have done this for institutions, but the price is of course made with unlimited budgets in mind. I looked into it once, but the software license alone was 6000 dollars, so I figured it was not in the price league of a small market and business like I have here in Iceland.

[Pere Casals]

I have tried it using Leica S lenses, in particular the 120mm APO macro, which should have no problem resolving the claimed 100mp.

Stuart, 100MPix (at contrast extintion) in a 24x36mm frame means 115740 efective pix per mm2, which exactly requires 170 lp/mm. This is possible in optimal lab consitions and comparable with other comon FX primes (Nikon, Canon...). Well, a Nikkor EL 50mm enlarger lens resolves remarkably more... and it can be found for $30, used...

Another thing is crafting 170lp/mm on a sensor, pixel size would be 0.003mm, so for a 120mm lens focal angle is 0.0007º.

So if during the taking you have an angular vibration of only 0.001º you will have 1/2 of the lp/mm and 1/4 of the pixels, this is 25MPix at contrast extintion limitation from shake. (sorry for the math)


Just check your own multi-shots vs single shots after you sharpen each crop to the optimum


There is a way to enhance the result, this is using a 1/100,000s or a 1/1,000,000s strobe, a 1/30,000 strobe would be a good start... but also focus and alignment has to be perfect.


One thing is shooting flat targets in a lab and another thing is practical conditions.

[Peter De Smidt]

Here's another company that's been doing this for a long time: http://www.gigamacro.com/gigapixel-macro-imaging-system/
They offer the Nikon MM objectives as an option. They are outstanding for high resolution work. I have one of the 3x MM lenses. It's terrific....for scanning resolution targets. It's not that useful for scanning film, except, maybe, 35mm and smaller

Before we get too carried away, though, note that PMT scanning has in theory an advantage in resistance to flare and signal to noise compared to cmos scanning. A high quality drum scanner in excellent condition can likely scan higher density materials with better contrast than can a cmos-based system. Is that difference important? Well, that depends on you, and what you're doing with the files, and whether you can find a good drum scanner in tip-top condition with a good operator.

[Tin Can]

I once and only once had a Pro Drum Scan of two 35mm slides.

Drop off in Chicago, maybe 2003. High end joint.

$60 wasted on crap.

Now I use the Nikon film copy attachment, Nikon PB5 Bellows + PS4 Copier.

[StuartR]

Hi Pere,
I am sorry, but I just got back from shooting and it is 2:45am in the arctic morning, so I do not follow your math, nor am I frankly all that interested in it, in so far as it relates to the practical side of photography. I know that the Leica 120mm macro renders razor sharp detail across the frame on the 37mp non AA filtered CCD of the Leica S, even at nearly wide open apertures. I also know that Peter Karbe at Leica is one of the better lens designers around, and he has stated that the Leica S lenses are designed to resolve 100mp. Given that it is the best of all of them, I do not doubt that the 120mm can do so. I agree that vibration and stability are totally critical in making high resolution photos. Using the same lens on both bodies, the multi shot has advantages compared to the single shot of the S1, but it does not seem to achieve a true 100mp. I think this is most likely less a stability issue than an issue with the precision of the sensor shift mechanism and the stitching of the resulting images.

[robertraymer]

Thanks for all the great info.

[Pere Casals]

Peter Karbe at Leica .... Leica S lenses are designed to resolve 100mp. Given that it is the best of all of them, I do not doubt that the 120mm can do so.

Looking to datasheets it's a good lens:

MTF at 5, 10, 20, and 40 lp/mm

192479

Page 29: https://www.fotoskoda.cz/images-old/multi/popisobr/LEICA/S_NEW/The%20new%20S%20brochure_en.pdf

The Leica S is just half the surface of the 645 film format. Analyzing this system is quite interesting, it is a good system but nothing far from current Nikon D850 and the like, IMHO.

Anyway we don't requiere expensive cameras and lenses to make a powerful DSLR scanner, we always can stitch more shots.

[sperdynamite]

I'm doing camera scanning with the Pentax K-1 II and using pixel shift (aka multi shot). It's a subtle difference, but once you do post processing on the scan the image is much cleaner and more resolved. I've done quite a lot of scanning in my time and I'm not going back to traditional CCD line scanners, or getting a drum scanner.

These are already enough for my 24" P6000 printer. Drum scanners, flextights, and cool scans are just getting older, while camera scanning rigs get better every year.

[sperdynamite]

I'll also add that there is an IMO misconception that more resolution is always better. I've seen the discussions about resolving the actual grain clumps, etc... I am of the opinion that your desired output is really all that matters.

Example: When it comes to 4x5 I can get a 16x20 print at 300ppi with a single capture using a D810. I have done this actually with Provia 100F, printed with an Epson P6000. The print looks...stunning! Just as you would expect from Provia in that size, there is no visible grain, just detail. Plus I was able to recover significant shadow information due to the latitude of the CMOS capture.

Now as I mentioned I'm using a K-1 II with pixel shift so my results are going to be even better, and this is to say nothing of stitch techniques.

Bigger is bigger and I appreciate that, but in normal print sizes such as 13x19, 17x22, 24", 44", camera scanning techniques are beyond adequate. I'm a big believer in the idea that sometimes you gotta leave the numbers at the door and just use your eyeballs to evaluate what you think is acceptable. Talking about resolving grain clumping just kind of makes most people roll their eyes back into their heads. I'm getting fabulous results using 35mm full frame cameras and conventional macro lenses. Theories and back of the napkin math isn't going to change that.

[Peter De Smidt]

Yes, I agree that we often grossly overestimate the required resolution for a good print, especially if grain won't be visible.

[Pere Casals]

I'll also add that there is an IMO misconception that more resolution is always better.

What is critical is the digital processing in Ps, what user does with software.




I've seen the discussions about resolving the actual grain clumps, etc... I am of the opinion that your desired output is really all that matters.

A refined technique in scanning/processing can be critical to get a sound grain depiction. Sometimes grain is irrelevant (specially in LF), but sometimes grain is very important for the aesthetics and it requires a proficient management.





Example: When it comes to 4x5 I can get a 16x20 print at 300ppi with a single capture using a D810.

In that case your print is suboptimal, a good 4x5" negative has 400MPix effective, the print has 28,800,000 continuous tone pixels (16x20x300x300), but the 810 may take some 25 MPix effective with contrast extintion at pixel level, so you may need 2 pixels in the print to have a tonal change when it could be done with a single one. This can be more or less relevant (or simply not relevant) depending on the scene and on what you want, but you effectively print at around 150 effective resolution instead 300, 150 is not bad, but it's not 300 effective.


For sure, if only taking some 20Mpix from the 4x5 negative you may not need a view camera, you may shot MF roll film (with a fraction of the cost/weight) and you will notice absolutely no image quality change, or you may simply shot the scene with the D810 directly.





I'm getting fabulous results using 35mm full frame cameras and conventional macro lenses. Theories and back of the napkin math isn't going to change that.

For sure... no doubt. A human eye can only see around 60MPix. LF greatness is beyond Image Quality, crazy high IQ comes as a bonus, it's nice having some 800 effective Mpix in a 8x10 shot, a crazy overkill, but a true artist using a view camera knows why he wants the bellows !!!

[sperdynamite]

What is critical is the digital processing in Ps, what user does with software.

A refined technique in scanning/processing can be critical to get a sound grain depiction. Sometimes grain is irrelevant (specially in LF), but sometimes grain is very important for the aesthetics and it requires a proficient management.

In that case your print is suboptimal, a good 4x5" negative has 400MPix effective, the print has 28,800,000 continuous tone pixels (16x20x300x300), but the 810 may take some 25 MPix effective with contrast extintion at pixel level, so you may need 2 pixels in the print to have a tonal change when it could be done with a single one. This can be more or less relevant (or simply not relevant) depending on the scene and on what you want, but you effectively print at around 150 effective resolution instead 300, 150 is not bad, but it's not 300 effective.

For sure, if only taking some 20Mpix from the 4x5 negative you may not need a view camera, you may shot MF roll film (with a fraction of the cost/weight) and you will notice absolutely no image quality change, or you may simply shot the scene with the D810 directly.

For sure... no doubt. A human eye can only see around 60MPix. LF greatness is beyond Image Quality, crazy high IQ comes as a bonus, it's nice having some 800 effective Mpix in a 8x10 shot, a crazy overkill, but a true artist using a view camera knows why he wants the bellows !!!

I think my suboptimal print looks great great. I should know as I have the original chrome for comparison. I think telling someone that their print is "sub-optimal" sight unseen based on some math that 'should' apply (but really only applies to Plato's Scanner) is the very worst example of internet photography forum punditry.

192494192495

[Peter De Smidt]

Amen! Some forum members seem stuck in Plato's cave. Alas, I expect Plato's scanner would only image ideas and not material objects. ;)

[rdeloe]

I am deeply envious of people who are part of a community of photographers who can actually get together over an evening and look at prints of each others' work. I have to think the conversations would be very different than the ones that happen on photography forums. There would certainly be a lot more "show" and a lot less "tell".

[Tin Can]

Like!



I think my suboptimal print looks great great. I should know as I have the original chrome for comparison. I think telling someone that their print is "sub-optimal" sight unseen based on some math that 'should' apply (but really only applies to Plato's Scanner) is the very worst example of internet photography forum punditry.

192494192495

[Pere Casals]

I think my suboptimal print looks great great. I should know as I have the original chrome for comparison. I think telling someone that their print is "sub-optimal" sight unseen based on some math that 'should' apply (but really only applies to Plato's Scanner) is the very worst example of internet photography forum punditry.

192494192495

:) not saying that your prints are not nice...

Let me say it in other words: technically suboptimal, and I reiterate that this can be irrelevant, it can be a fantastic print anyway.

Just I state, and this is completely true, that a single D810 shot cannot take all graphic information that a 300ppi 20x16" print can contain, so Image Quality of the print can be technically improved with a better workflow. No doubt.

The impact of that improvement is debatable, of course... but it is for sure that result from a single D810 shot can be technically improved. Let me reiterate that if that technical improvement makes a difference or not for a particular print/criterion is another question, or if a x2 magnifier has to be used (or not...) to see well the difference.

Also it has to be mentioned that a good Ps processing may mitigate the scanning flaws... or it can introduce new flaws...

Your 4x5 negative may have 400Mpix effective, it's up to you how many of those you take, and at what modulation transfer between pixels: up to you...


Look, you say that a single D810 shot of a 4x5" negative is fantastic, I say no: this is suboptimal, four (2x2) overlaped-stitched shots would be near optimal for a 20x16/300.

[sperdynamite]

Fine, you're right. The image is SUBOPTIMAL.

Next time I'll:

1. Use a 150/5.6 APO Sironar S, the most optimal of lenses. My Schneider APO Symmar is not optimal enough.

2. Anchor my tripod 50' into the ground to ensure vibration free exposures.

3. Use the shortest possible flash duration.

4. People move so I should have a lifelike sculpture of my subject made from a precision 3D scan to maximize stillness.

5. The air in the room probably causes some issues so maybe I could have it all sucked out so we're shooting in vacuum?

6. Have Fuji helicopter in the very freshest batch of Provia and ensure it's at the perfect temperature for exposure, and then have it helicoptered out directly to AGX lab for processing immediately.


Sarcasm aside, I'm not saying that my scan is the best possible scan of this image. I'm sure a Tango would do a better job in some ways. I'm saying that my scan resulted in a great print, and that is the very minimum possibility of camera scanning. I have no use for a 400mp scan of this image, it's just wasted data for my purposes. You may believe that it's prerequisite for...something...but those of us who have to use the equipment we can afford in the reality that exists before us simply make due. It's not a good idea to approach art making this way. Do you run around high school darkrooms telling students about their suboptimal prints because they used maybe-not-the-best enlarger lenses? Do you comment that every home cooked meal is 'sub-optimal' because it was not prepared by Michelin Star chefs?

I'll make sure to put the term 'modulation transfer' in my next artist statement and see how that goes over.

[Pere Casals]

Fine, you're right. The image is SUBOPTIMAL.

Next time I'll: .....




sperdynamite, first is that everyone prints/scans like he wants... do what you want, of course. And of course sharpness is a very overrated concept in photography.


...but we may find useful to understand what is Image Quality, what amount do we require to fulfill what a human eye can see in our print, how IQ is degradated through our workflow, and when that degradation is seen in the print.

It is a personal choice taking care or not about that.

Lately I've no problem, mostly I make contact prints from 8x10 so my prints are crazy sharp even if inspected with an strong magnifier.


I would recommend this book to anyone wanting to understand well the underlying concepts, me... I've learned a lot from it:

192505

$8 used, Ebay. Worth $800+.

[Tin Can]

I think Pere means well and there may be native language interference.

But some do carry on about perfection in every aspect...

I try to enjoy my time here, not always easy!

As we trundle to the next castle to lay siege...


Fine, you're right. The image is SUBOPTIMAL.

Next time I'll:

1. Use a 150/5.6 APO Sironar S, the most optimal of lenses. My Schneider APO Symmar is not optimal enough.

2. Anchor my tripod 50' into the ground to ensure vibration free exposures.

3. Use the shortest possible flash duration.

4. People move so I should have a lifelike sculpture of my subject made from a precision 3D scan to maximize stillness.

5. The air in the room probably causes some issues so maybe I could have it all sucked out so we're shooting in vacuum?

6. Have Fuji helicopter in the very freshest batch of Provia and ensure it's at the perfect temperature for exposure, and then have it helicoptered out directly to AGX lab for processing immediately.


Sarcasm aside, I'm not saying that my scan is the best possible scan of this image. I'm sure a Tango would do a better job in some ways. I'm saying that my scan resulted in a great print, and that is the very minimum possibility of camera scanning. I have no use for a 400mp scan of this image, it's just wasted data for my purposes. You may believe that it's prerequisite for...something...but those of us who have to use the equipment we can afford in the reality that exists before us simply make due. It's not a good idea to approach art making this way. Do you run around high school darkrooms telling students about their suboptimal prints because they used maybe-not-the-best enlarger lenses? Do you comment that every home cooked meal is 'sub-optimal' because it was not prepared by Michelin Star chefs?

I'll make sure to put the term 'modulation transfer' in my next artist statement and see how that goes over.

[interneg]

I'll make sure to put the term 'modulation transfer' in my next artist statement and see how that goes over.

Unlike your interlocutor, you have a far greater practical understanding of it in real world use. Pretty much any decent contemporary DSLR/ ILC will outperform an Epson in real world performance because it's easier to eliminate vibration in copying set-ups & enhance precision of focus - both of which dramatically improve contrast performance.

[Pere Casals]

any decent contemporary DSLR/ ILC will outperform an Epson in real world performance.

:) :) :)

And also a hasselblad x5, that sports only 1800dpi effective for 4x5.

:) :) :)

Interneg, this is LOL !!!!


BTW, a DSLR scanner may easily reach 9000 dpi effective... if the right technique is used, which is an overkill.

[faberryman]

What exactly is dedicated flatbed scanning? How does it differ from regular flatbed scanning?

[interneg]

And also a hasselblad x5, that sports only 1800dpi effective for 4x5.



Feel free to continue deluding yourself. The difference is not small. If you really need to see the difference between the '2400' you claim for the Epson and a high MTF response 2040 I could show you, but it will depend on whether I can be bothered to upload the files to somewhere they can be seen without being severely compressed.

Given that most of the Epson optical path is held together with clips not out of place in cheap picture frames & involves three mirrors before the lens in a structure to which the phrase 'precision' seems an alien concept, it's not surprising it's as poor performing in critical sharpness as it is. And no, no amount of sharpening will solve it. And yes, you'll see the difference even in small prints from 4x5 scans. BTDT on a regular basis.

[Pere Casals]

...The difference is not small...

Interneg... Interneg...

Did you know that the Hasselblad X5 only delivers 8000 points in the scan width ?

Of those 8000 only 6900 are effective, this happens both when scanning 35mm film and when scanning 4x5 sheets, so you only can enlarge to 6900/300 = 23" the short side, for good quality.

Compared to a good darkroom print... a digital print from an X5 image starts being technically worse than a darkroom print from 12", but from 24" (for the short side) this can be better noticed. Still you can make acceptable YMMV 40" (for the short side) prints.


Hmmmm, I was thinking you were aware...

In fact the X5 resolves less than 1800dpi for 4x5", it has contrast extintion by 1725dpi

An V850 resolves 2900dpi in the horizontal axis and 2300 in the vertical one.

[EH21]

I have been scanning with an iQsmart3 and also a fancy set up with a multishot / microstep digital back with schneider 80mm macro symmar with behind the lens leaf shutter attached to my Rolleiflex 6008AF. The microstep produces a non interpolated file 528mb in size and takes about 1 minute (19 frames) and the iqsmart3 can generate a very large file - in the GB that also is non-interpolated. I think when I do everything exactly right, the 6008AF/CF528 can produce a file that has slightly better detail and more information in the shadows/highlights but it's a tiny difference. The iqsmart is much more convenient and does everything right without fuss so mostly I use it. The iQsmart3 gets significantly more convenient with larger negatives since it does the stitching not me.

And I discovered that a lot of my negatives that were sent out for processing on the cheap were not so great. When you pay attention to the scanning you start to realize that what you really should be paying attention to is the shooting and more importantly the processing!

[Pere Casals]

and more importantly the processing!

+1

IMHO processing is critical. It's relatively easy to get good enough images from good LF negatives. A refined Ps technique counts a lot, but this is about learning... what's a bit more complicated is having a sound aesthetic criterion to edit the image. Sometimes it's about not being much intrusive and just balancing some factors, sometimes a deep edition is what yields impressive results. I know some Ps artists that are very good doing that, they are gifted !

[interneg]

Interneg... Interneg...

Did you know that the Hasselblad X5 only delivers 8000 points in the scan width ?

Of those 8000 only 6900 are effective, this happens both when scanning 35mm film and when scanning 4x5 sheets, so you only can enlarge to 6900/300 = 23" the short side, for good quality.

Compared to a good darkroom print... a digital print from an X5 image starts being technically worse than a darkroom print from 12", but from 24" (for the short side) this can be better noticed. Still you can make acceptable YMMV 40" (for the short side) prints.


Hmmmm, I was thinking you were aware...

In fact the X5 resolves less than 1800dpi for 4x5", it has contrast extintion by 1725dpi

An V850 resolves 2900dpi in the horizontal axis and 2300 in the vertical one.

This is an impressive set of completely wrong statements. Much like your claims on multigrade emulsions & dyes.

It's a question of mechanical precision. The difference between 6900 and 8000 is half a micron. Half a micron! That's better than most drum scanners which can have a micron of variance depending on wear etc. Because of the construction, that tolerance is going to be fairly constant, so if you allow for that half micron when designing the scan widths for the larger formats, then you'll deliver the specified resolutions without issue. This is basic maths & the sort of thing that was clearly obvious to the highly competent engineers who designed the machine - but clearly not to you. For the record, the 4x5 width is not 4" but aimed to match the average actual image width between the film rails in Fidelity/ Toyo holders.

As for your bizarre assertions about the quality of an Epson scan, no. Everything you state there is wrong and easily disproved in scans of actual negatives. The output from the Epson falls to bits very rapidly, especially at large print sizes. BTDT. The Flextight output is incredibly clean and natively sharp in comparison. Your almost total lack of real world experience with high end scanners and large exhibition prints is blatantly obvious. Applying the same lazy assumptions to a 3xCCD as you would to a Bayer array is not going to hold up. I could give clear examples of the difference between an Epson and a Flextight for 4x5, but it is clear you are intent on not learning.

[Pere Casals]

What is wrong?

The X5 only resolves 1725dpi effective for 4x5" !!!

The V850 is way better for 4x5, delivering much cleaner and sharp images.

It's impressive you did not know that!

[Corran]

I think my suboptimal print looks great great. I should know as I have the original chrome for comparison. I think telling someone that their print is "sub-optimal" sight unseen based on some math that 'should' apply (but really only applies to Plato's Scanner) is the very worst example of internet photography forum punditry.

Amen! I like the "Plato's Scanner" comment. I'll have to remember that.

[interneg]

Amen! I like the "Plato's Scanner" comment. I'll have to remember that.

Agreed. Especially as there's a repetitively offensive miscreant.

[Audii-Dudii]

What is wrong?

The X5 only resolves 1725dpi effective for 4x5" !!!

The V850 is way better for 4x5, delivering much cleaner and sharp images.

It's impressive you did not know that!

Pere:

No offense intended, but would you mind elaborating on your real-world experience scanning film and then printing from those scans?

I'm asking because by your own admission, you contact-print from 8x10 film most of the time, so it's not clear to me to what extent, if any, your theoretical knowledge is tempered by real-world results.

I don't have a horse in this race, but your numbers-only analyses (which inherently assume all other factors are equal and, of course, they're not) remind me of an argument I recently overhead in the reception area of a doctor's office. Two kids, both of whom were too young to have ever driven any car, let alone a supercar, were arguing about which supercar was "the best" based on the various specs they read in the dog-eared copy of a several years-old Motor Trend magazine they found on a table.

As anybody who has actually driven a supercar at speed [raises hand!] will attest, the claimed (and usually unverified!) specs detailing the car's performance only tell a very small part of the overall story about how it performs in the real world. Drawing a conclusion about its overall performance by extrapolating from those specs is potentially very misleading, because there are many other factors that come into play besides those, including many that are inherently subjective in nature and can't readily be measured and quantified, but absolutely can be sensed by somebody who has accumulated a sufficient amount of experience.

Personally, I found this to also be true with the various film scanners I've owned and used in the past (Polaroid 45i, Polaroid Sprintscan 120, Minolta Dimage, Howtek D4000 and probably a few others that I've forgotten about over the 14 years since I last used one) and have no doubt it remains true today as well. Ditto for modern day inkjet printers.

I'm just sayin', that's all...

[Peter De Smidt]

Very well said, Audii-Dudii!

For years, Pere claimed that an Epson V700 reaches a dmax of 4.0, which is ridiculous, and very easy to check with a stepwedge. Despite being shown evidence, he persisted in continuing to make the false and dis-proven claim. Theories based on false premises are worse than useless. Moreover, dmax, whether with a scanner or with a printer, isn't as important as a uniform response throughout the tonal range. A scanner's response can shoulder off way before dmax is reached.

[interneg]

Audii-Dudii; Peter,

Nothing to disagree with there.
It's the work of minutes for me to run a vertical comparison between an Epson 12000xl and an X5 & indeed I did so for amusement's sake about an hour ago. If Pere's claims were true, then the Epson optical system is worse resolving in reality than whatever number he's plucked from the air for the Imacon/ Hasselblad. And that's with the aid of autofocus.

More to the point, the article he's taken the maximum resolution figures for the X5 from very clearly states that the scanner delivers the specified resolutions for 120 & 4x5 when tested with the same test chart... Thus the only reasonable conclusion that can be drawn is that it's either a lack of reading comprehension or outright active misrepresentation on Mr Casals' part.

[Pere Casals]

Pere: No offense intended, but would you mind elaborating on your real-world experience scanning film and then printing from those scans?

While my own photographic work is of low quality, I scan and edit the work of several artists for their international exhibitions, one of them won the Zoom Photo Festival Saguenay prize, (Quebec), for example.

I use X5 scans mainly for 35mm, it makes an extraordinary work depicting P3200 grain structure. The X5 is fantastic for 35mm, not that good for MF, and suboptimal for 4x5".

Instead the V850 is suboptimal for 35mm, good for MF, very good for 4x5, extraordinary for 5x7(still uses the HR lens) and very good for 8x10".





Howtek D4000

Remarkably, you can download samples in the https://www.largeformatphotography.info/scan-comparison/ and adjust the Howtek 4500 and V700 to their best, I did it and I found that (please try it on your own)

https://live.staticflickr.com/8561/28420386682_d481942db8_o.jpg

A particular V700 dark sample in the collaborative scanner test is flawed, as multi-exposure was not used, IMHO.

The V700 result is not digitally optimized by the scanner, as pro scanners do by default, so V700 samples will always improve in the edition.

[Pere Casals]

Very well said, Audii-Dudii!

For years, Pere claimed that an Epson V700 reaches a dmax of 4.0, which is ridiculous, and very easy to check with a stepwedge. Despite being shown evidence, he persisted in continuing to make the false and dis-proven claim. Theories based on false premises are worse than useless. Moreover, dmax, whether with a scanner or with a printer, isn't as important as a uniform response throughout the tonal range. A scanner's response can shoulder off way before dmax is reached.

Peter, you are in my ignore list, but I'll answer you anyway because I want to reply your slander.

Discrediting people with lies is not nice, Peter: https://www.largeformatphotography.info/forum/showthread.php?149339-Next-step-up-from-Epson-scanners&p=1473043&viewfull=1#post1473043

EPSON claims DMax is 4.0D, I claim that V700 it reaches 3.4D if multiexposure used. For V750 and V850 it may be a bit higher because lenses are multicoated.


I'm very well documented about the SNR vs D of the V850, if you want I may show you the graphs.

[Pere Casals]

Thus the only reasonable conclusion that can be drawn is that it's either a lack of reading comprehension or outright active misrepresentation on Mr Casals' part.

Mr interneg, it happens that X5 has contrast extintion by 1725dpi for 4x5". Not surprising as the X5 has only 8000 pix in the 4" scan width.

I guess that you are not capable to understand that, but it's well proven that a V700 resolves 2900dpi efective in the horizontal axis, the V850 lens cannot be as crappy as you say, because it takes 2900dpi in a 5.9" scanning width, totalling 17000 effective pixels in the 5.9" scanning width, which is amazing for 5x7".

On any doubt, Pali K made a side by side test:

https://www.largeformatphotography.info/forum/showthread.php?150020-Scanner-Comparison-2019-Epson-Flatbed-Eversmart-Flatbed-Drum-Scanners

In practice, the V700 crops are matching those from high end expensive flatbeds, just we need to edit a bit the V700 crops as it does not digitally optimize the scans (red rectangle) automaticly as Pro gear do:
https://www.largeformatphotography.info/forum/showthread.php?150020-Scanner-Comparison-2019-Epson-Flatbed-Eversmart-Flatbed-Drum-Scanners&p=1479178&viewfull=1#post1479178
https://c1.staticflickr.com/5/4857/46755757932_c7010da815_o.jpg


Interneg, if you are brave enough then you may comment on that fact.

[faberryman]

Don't egg him on. He is just going to show us the same stuff he has been showing us for years. Please don't show us the image of the church bell.

[Peter De Smidt]

Maybe our only reply should be, "Plato's scanner."

[Pere Casals]

Don't egg him on. He is just going to show us the same stuff he has been showing us for years. Please don't show us the image of the church bell.

My pleasure:

I reiterate, the bell shows in your monitor like if it was in a 6m print.

Well, the bolts in the wood shows how powerful is the EPSON / 8x10 combination.

https://live.staticflickr.com/669/32535835184_2a10b880f7_o.jpg

[Audii-Dudii]

Don't egg him on. He is just going to show us the same stuff he has been showing us for years. Please don't show us the image of the church bell.

I didn't know ... sorry! <blushes>

[interneg]

The X5 is fantastic for 35mm, not that good for MF, and suboptimal for 4x5".


How many scans of 135, 120 and 4x5 have you ever made on an X5? And if you have, why do you constantly repost the same clickbait, flawed/ warped tests & useless comparisons?

[Corran]

Sharpen
Sharpen More
Unsharp Mask

[Pere Casals]

I didn't know ... sorry! <blushes>

Forget the bell...

Any comment about images in posts 51 and 53 ?

No comment ?

:)

[faberryman]

I just knew he was going to post that bell image.

[Pere Casals]

How many scans of 135, 120 and 4x5 have you ever made on an X5?

I edited hundreds of X5 and Imacon scans, mostly from 35mm Velvia, which is the situation where an X5 shines.

For BW and color negative 4x5" it does not shine at all.

For 4x5" velvia the X5 can be interesting if wanting to recover very deep shadows.



And if you have, why do you constantly repost the same clickbait, flawed/ warped tests & useless comparisons?

IMHO the side by side tests made by Pali are very honest and enlight a lot.

Also editing the crops in the Collaborative Scanner Comparisson https://www.largeformatphotography.info/scan-comparison/ it's really meaningful.

We are talking about reliable information. Pali is not exactly beginner in scanning, while people that made the "Collaborative" are masters of their tools.

To me those examples are Facts, discredit those facts if you are able...

[Pere Casals]

I just knew he was going to post that bell image.

You suggested it... No opinion about the Image Quality of the crop of a 6m elargement?

[faberryman]

You suggested it... No opinion about the Image Quality of the crop of a 6m elargement?

Actually I suggested that you don't show it, but that would have required that you read my post.

[Pere Casals]

Actually I suggested that you don't show it, but that would have required that you read my post.

So no comment about the image quality of the crop...

[interneg]

So Pere, let's be clear, have you scanned anything other than 35mm transparency on an Imacon?

Secondly, I was not referring to Pali's excellent tests (which you have spent a great many posts trying to misrepresent the evidence he gathered) but to your whataboutery whenever you are directly challenged to show personal evidence of your exceptional claims. You have not yet answered the question about why you pick & choose evidence from the filmscanner.info resolution test rather than dealing with the totality. Is it because doing so would destroy the fallacy you have spun?

[faberryman]

So no comment about the image quality of the crop...

I replied a couple of years ago. I don't have anything new to add. And I am not interested in going down that rabbit hole with you again.

[Audii-Dudii]

Excuse me.

I naively contributed to what I thought was a new thread and didn't realize it was actually the continuation of a long-running drama.

I'm relatively new here and don't wish to step on anyone's toes, so I'm just going to show myself out and y'all can carry on...

Cheers! :rolleyes:

[Pere Casals]

So Pere, let's be clear, have you scanned anything other than 35mm transparency on an Imacon?

Secondly, I was not referring to Pali's excellent tests (which you have spent a great many posts trying to misrepresent the evidence he gathered) but to your whataboutery whenever you are directly challenged to show personal evidence of your exceptional claims. You have not yet answered the question about why you pick & choose evidence from the filmscanner.info resolution test rather than dealing with the totality. Is it because doing so would destroy the fallacy you have spun?


interneg, let's be clear:

X5 yields under 2000 dpi effective for 4x5, a shame for a $16,000 machine. A 8000pix sensor for a 4" width will never yield more than 2000dpi effective, but well under.


let's be clear:

a cheap V850 for 4x5" resolves more than 2000dpi effective, 2900 in the Hor and 2300 in the vert. It also makes 8x10".


let's be clear:

Pali's tests discredit you. Also the crops in the Collaborative scanner tests discredit you.


let's be clear:

Some of you go to personal attack and lies where you have no technical arguments:
https://www.largeformatphotography.info/forum/showthread.php?152777-DSLR-scanning-vs-Dedicated-flatbed-scanning&p=1504824&viewfull=1#post1504824
https://www.largeformatphotography.info/forum/showthread.php?152777-DSLR-scanning-vs-Dedicated-flatbed-scanning&p=1505192&viewfull=1#post1505192


let's be clear:

I've no commercial interest, so also no need to say lies to customers. Some here are in a different situation :)

[Pere Casals]

Excuse me.

I naively contributed to what I thought was a new thread and didn't realize it was actually the continuation of a long-running drama.

I'm relatively new here and don't wish to step on anyone's toes, so I'm just going to show myself out and y'all can carry on...

Cheers! :rolleyes:

Don't worry, a scanner riot was my welcome party in this forum, at that time by I was not prepared to deal with such kind of people.

Now it's different, it's pure fun to allow them discredit on their own.

Here interneg has to say that 2000 samples per inch (1725dpi effective, a shame) are pure gold. LOL :)


BTW, the good prints are those crafted with an enlarger by a sound printer, this is what I'm learning to do.

[interneg]

Pere: the question you have to answer is very simple. Have you ever scanned anything other than 135 colour transparency on a Flextight type of scanner?

That is the question you need to answer honestly. It will cause you a great deal less grief if you do, as painful as you might find it.

[Pere Casals]

Pere: the question you have to answer is very simple. Have you ever scanned anything other than 135 colour transparency on a Flextight type of scanner?

That is the question you need to answer honestly. It will cause you a great deal less grief if you do, as painful as you might find it.

Interneg, I don't own an X5 like you, but I've been sitting at the side of the X5 operator and owner to make the 135 and MF scans that later will end in remarkable exhibitions. In some of those sittings I brought my 4x5 sheets and I tested with operator and compared.

That X5 operator is an extraordinary X5 and Ps operator, extraordinary in the artistic criterion, in the technical skills and in the honesty. He rated with a USAF 1951 target his X5 and he found between 1700 and 1800 effective, depending on the pass. I've personally seen those tests.

As you may know each single pass may vary a bit the rating depending on the particular alignment of the strips with the pixels. I gave them ADOX CMS 20 4x5 sheets/rolls and to make contact copies of the USAF glass slide on hires film to make the tests, and I discussed with them the results.

So look, you are speaking with somebody that's pretty well informed about what an X5 does. An extraordinary machine that I love for 35mm, but that I find it's not worth for 4x5" BW.

This is a LF forum, most people shot BW, and I think it's nice to report what scanners really do with sheets. Beacuse of that the Collaborative test it's really useful.


Probably you have not rated your hassie because of lacking a 45 target, the way to make one, I inform you, is making contact copies of the USAF 1951 glass slide on CMS 20 film. If you are interested I'll explain you how to make a good contact copy, it's not that difficult.

[Peter De Smidt]

I have one of Edmund's high resolution versions of the USAF 1951 chrome on glass test targets. It's useful, but by no means the be-all and end-all of scanner performance. So Pere has sat next to a scanner operator....compare that to people who have used the devices under discussion for years....I don't have any X5 experience, and so I don't make claims about it.

[Chester McCheeserton]

Here's the same negative, scanned side by side. first one is an Epson V700 and 2nd one is Imacon Flextight X5. Both at 2000. Pere, do you like apples?

192572192575192573

[interneg]

Pere: because you didn't operate the Flextight, there are critical details that you (and by the sounds of it, the operator) missed. The biggest one is that if you allow Flexcolor to do the inversion, it can subtly but noticeably reduce sharpness & resolution - quite possibly by the amount you describe (which is an extremely minor amount of variance between two very small elements on a target that is already going to be an imperfect copy). If you get an un-inverted .tif file (or force convert a .fff) & do the mask correction, inversion etc in Photoshop, you will get a nasty shock at just how much Flexcolor mangles the sensor output when it's allowed to do the inversion. The number of times I've been sent seemingly soft files with awkward crossed curves & weird saturations from Flexcolor scans is never ending, yet if they were scanned in 3F, or I have the original negative, a fresh un-inverted scan or converting the 3F to tif (literally delete .fff, type .tif on the file name in the folder) and taking it to Photoshop as a positive gets amazingly different results - razor sharp, colours go logically & quickly to where they are supposed to be etc, much less noise too. In other words, it largely matches the other high end scanners that use the same sensor, albeit in xy form, in terms of colour/ resolution. In terms of colour rendering, sharpness, my results line up with the results possible from similar sensors in Pali's tests.

For the record, every test I have done essentially matches Chester's in terms of outright resolution and contrast transfer. I've prepared a number of side by sides of 2040 origination on the X5 and 2400 downsampled to 2040 on the Epson & the results do not look good for the Epson, even after the harsh sharpening you claim makes them match (which it doesn't, it just enhances already unpleasant noise).

[Pere Casals]

I have one of Edmund's high resolution versions of the USAF 1951 chrome on glass test targets. It's useful, but by no means the be-all and end-all of scanner performance. So Pere has sat next to a scanner operator....compare that to people who have used the devices under discussion for years....I don't have any X5 experience, and so I don't make claims about it.

Peter, what you make is claims about lies. https://www.largeformatphotography.info/forum/showthread.php?152777-DSLR-scanning-vs-Dedicated-flatbed-scanning&p=1505192&viewfull=1#post1505192

You are in my ignore list, I'm not to debate with you, I only had to answer one of your your slanders and that's all.

[Pere Casals]

Here's the same negative, scanned side by side. first one is an Epson V700 and 2nd one is Imacon Flextight X5. Both at 2000. Pere, do you like apples?

192572192575192573

Chester, this is flawed side by side.

> this looks 35mm film, at least the aspect matches exactly 3:2. Here we talk about 4x5", were the X5 loses 75% of the linear performance while the V700 holds the same.

> in 35mm, even if you scan 2000dpi the X5 takes advantage of the crazy good 6900dpi effective because of the optic zoom and digital processing.

> With 35mm film the X5 is way, way superior to the V700, no doubt: 6900 vs 2500. For 4x5 the X5 drops to 1725 (zoom out) while the EPSON retains 2500 because the lens is fixed.

> The EPSON requires to scan at higher dpi and then downsampling, a the right edition.



Your test was cooked to discredit the EPSON, while it's true that the X5 is extraordinary for 35mm it's close to the EPSON for MF, and inferior to the EPSON for LF.


This is apples: a honest side by side test for MF, where the EPSON is still slightly inferior.

https://petapixel.com/assets/uploads/2017/05/sidebyside-800x549.jpg
https://petapixel.com/2017/05/01/16000-photo-scanner-vs-500-scanner/

[Pere Casals]

For the record, every test I have done essentially matches Chester's in terms of outright resolution and contrast transfer. I've prepared a number of side by sides of 2040 origination on the X5 and 2400 downsampled to 2040 on the Epson & the results do not look good for the Epson, even after the harsh sharpening you claim makes them match (which it doesn't, it just enhances already unpleasant noise).

Probably you made flawed tests with the EPSON, it's not that easy to obtain optimal results with the EPSON than with an X5.

The EPSON is semi-pro gear, the X5 is pro gear and you have the flextight. With the EPSON you should ensure film flatness and distance and scanning 3200 or 4000 to end in 2500 effective.

Please, just answer me a question:

How it can be that if you download the crops in the Collaborative test you obtain those close matches with the EPSON after the right edition?



https://live.staticflickr.com/8561/28420386682_d481942db8_o.jpg

https://www.largeformatphotography.info/scan-comparison/scanimages/prem-4x5-fullframe-u.jpg

[interneg]

It doesn't matter what format the negative was, as long as it was scanned at 2040ppi. Answer Chester's substantive question (which he backed up with clear evidence) with scans that you have made on an X5 or other Imacon. Otherwise you're just spamming desperately.

Secondly, let's see that whole image after that ugly sharpening then? I see crassly oversharpened, yet fundamentally poor resolving Epson V7/8/9xx scans on a daily basis. It's a high contrast part of the original, so of course it's going to even things out slightly in terms of visual acuity. Let's see a low contrast area from a scan you've made & see how that stacks up? Should not take you long to scan a fresh 4x5 for us, should it?

[Pere Casals]

Otherwise you're just spamming desperately.

Please omit this kind of wording and go to technical reasoning, I'm not going to personal attack, but you'll find me and then I'll force you into technical contradictions that will discredit you.



It doesn't matter what format the negative was, as long as it was scanned at 2040ppi.

It matters a lot !!! because for 35mm the X5 delivers a 2000dpi that's downsampled (with smart digital processing) from crazy good 6900dpi effective, so the "modulation transfer" of detail is crazy good, while the EPSON works closer to extintion.

By 4x5" the X5 has lost that advantage because its contrast extintion is at 1725 dpi, and not at 6900. ...while the EPSON keeps exactly the same performance than with 35mm.

Format matters, this is seen in the Petapixel test, while the X5 is crazy better in 35mm it's almost equal by 6x6cm, and inferior for 4x5. We can discuss why.




Let's see a low contrast area from a scan you've made & see how that stacks up? Should not take you long to scan a fresh 4x5 for us, should it?

You can do it on your own, problem is not in the low contrast areas, but in the low contrast areas of high density. Solution is multi-exposure feature, that yields perfectly clean dense areas.

My guess is that X5 obtains multi-exposure performance in a single pass form two possible factors: probable larger pixels, and probable multi-exposure before the scanner advances a row, while the epson makes two passes.


_____

Please answer the question in post #78, not answered


_____

No doubt that the X5 is very pro gear: it has the flextight, it does not deliver unnecessary pixels, it digitally optimizes the image very well, crazy good DMax in a single pass...

But the cheap Epson also has very strong points, in special for LF. A 5x7 (or 8x10) Epson scan delivers an absolutely pro result, with very acceptable 4x5 jobs that are better than those from an X5 if not crazy high densities there.

[interneg]

It's very easy to scan 35mm at 2040 ppi native (not downsampled) in an Imacon/ Hasselblad. Let's see if you can figure it out. If you'd operated the scanner even the most minimal amount you'd know how.

It is otherwise fundamentally clear you know very little about high end scanning or post-production of any sort at any level other than that of a consumer of the finished scan.

More importantly, was the person who had you scan those transparencies aware that you were not going to be scanning them yourself?

[Pere Casals]

It's very easy to scan 35mm at 2040 ppi native (not downsampled) in an Imacon/ Hasselblad.

Interneg, in a job I had I designed the PCBs and firmware interfacing a linear sensor. Let me explain that you won't know if the image is downlampled in firmaware or not. When interfacing a linear sensor you can do it in two ways:

1) Pixel binning, you read the averaged analog value of several pixels in the linear sensor row feeding a single voltage for the bin to the amplifier and A/D. Later you digitally average several rows to have square pixels.

2) You read all sensor resolution and then you digitally "average" rows and columns.

My guess is that the X5 uses the second way because it has more room to use a refined advanced processing to get a better output, I cannot imagine the X5 designers using the first approach, as the 8k pix sensor they used can be read very fast.

In the V850 I've doubts, because the sensor has +40,000 pix so it's slower to clock out all voltage readings from the sensor.

What I say is that while you won't downsample the image (best guess) the X5 has done it in firmware for you, cooking an optimal result.


While I've never operated an X5, by sitting at operators side I may see things that I guess you have not realized if never have designed and fielded a high end linear sensor camera for demanding industrial applications.





It is otherwise fundamentally clear you know very little about high end scanning or post-production of any sort at any level other than that of a consumer of the finished scan.

Again going to personal attack, and not answering the question in post #78

Please answer the question in post #78, not answered






More importantly, was the person who had you scan those transparencies aware that you were not going to be scanning them yourself?

Me, the artist and the X5 owner are very close friends. I do all the edition of his work for free, we share darkroom and the CPE2. My editions of his work (beyond locally) have been exhibited in several countries, some as distant as Finland vs Canada, and at least nobody complained.

________

Attached, the (area) camera config (made last week) of a system I'm designing know, running at 240 FPS under DCAM Format 7, it has 8 cameras per subsystem totalling 32 in the system, cameras work VIS+IR. Using Intel Image Processing Library to speed up processing. System intelligence uses Genetic Algorithms, Neural Networks and Expert System.

I show that only to point that it would be not easy for you to discredit my technical understanding about we are speaking.

[interneg]

The Epsons use folded optics & likely a similar sized sensor to most other CCD flatbeds - the X5 uses a 72mm sensor. Nowhere in their own materials does Epson claim to use anything other than that system.

Post #78 is a nonsensical question. It is based on a high contrast edge specifically chosen, then oversharpened until horribly noisy. You've chosen the one part of those tests that can be distorted to make the Epson look better. How about showing us the other bits of those images that clearly show the big flaws in the Epson?

You might know about aspects of electronic design, but it is clear that the critical optical and mechanical relationships that high end scanners rely on are far outwith your knowledge. The way the sensor reads out is rather less important than the lens being held in a precise relationship to the film & sensor with no transmitted vibration to kill sharpness & contrast.

If what you state is true about your access to an X5, why have you not been allowed to operate it? They are very popular in higher education because they are easy to teach students to operate & hard to truly break.

[Chester McCheeserton]

Chester, this is flawed side by side.
> this looks 35mm film, at least the aspect matches exactly 3:2. Here we talk about 4x5", were the X5 loses 75% of the linear performance while the V700 holds the same.
Your test was cooked to discredit the EPSON, while it's true that the X5 is extraordinary for 35mm it's close to the EPSON for MF, and inferior to the EPSON for LF.



My example is from my own 5 x 7 inch negative. The imacon is capable of much more precise focus on the film. The Epson, even if you use the plastic holders (which is not an option for 8x10 or 5x7)' or one of the 'betterscanning' holders, or your own wet mount glass sandwich, simply doesn't have the capability to focus on as tight of a plane. (even if you go thru the trouble of shimming your wet mount 'betterscannning' holder with the supplied shims)

This goes back to Peter's much earlier comment about how comparing numbers with different types of machines doesn't necessarily line up. A 2000 dpi drum scan can easily beat a 7000 dpi epson flatbed. Just like a 12 megapixel Nikon D700 can easily beat a 20 megapixel micro 4/3rds camera.

[Pere Casals]

The Epsons use folded optics & likely a similar sized sensor to most other CCD flatbeds - the X5 uses a 72mm sensor. Nowhere in their own materials does Epson claim to use anything other than that system

My own tests and all information you will find around says that V700 resolves 2900dpi in the 5.9" scan width, so it can see 17000 effective separated pixels in the scan width.

Look, this is a remarkable technical achievement with folded or non folded optics. An X5 will never see more than 8000 pixels in a row, and it actually sees 6900 pix in the scan width, the epson sees 17000 effective.

If the EPSON lens covered 3" instead 6" then it would sport 6000dpi effective, even for MF. I don't know if you are able to understand that, but's it's pretty straight.

The V700 has not more resolving power because it scans FOUR 35mm strips in a single pass, plus the holder margins.



You might know about aspects of electronic design, but it is clear that the critical optical and mechanical relationships that high end scanners rely on are far outwith your knowledge.

Again a personal attack. Let me explain you that systems I design and field makes non contact metrology, of course of the subpixel kind. I measure very well system accuracy and I know what comes from optics, from frame vibrations and from product motion blur.



If what you state is true about your access to an X5, why have you not been allowed to operate it? They are very popular in higher education because they are easy to teach students to operate & hard to truly break.

If the V700 is as crappy as you say, how it is possible that the samples posted by Pali and the ones in the Collaborative tests are corroborating what I point.




Post #78 is a nonsensical question...

LOL, interneg, I have you in the point I wanted: negating facts from reliable sources :)




interneg, let's be clear:

X5 yields under 2000 dpi effective for 4x5, a shame for a $16,000 machine. A 8000pix sensor for a 4" width will never yield more than 2000dpi effective, but well under.

Of course 4x5 was not a priority in the X5 design, they could have used a 16K sensor.

_____________________________


The X5 and the V850 are very different machines, "8k sensor with variable magnification" vs "40k sensor with 2 fixed lenses".

Sweet point of the X5 is 35mm, sweet point of the V850 is 5x7".

You cannot say that the X5 is a marvel and the V850 is pure crap, and demonstrating this with personal attacks.


Look, the V850 beats the X5 for 4x5" and up, you should have some respect for a cheap product doing that, in the same way an X5 deserves respect for taking 6900dpi effective from 35mm film.

[interneg]

Pere: how about you show us the results from the Imacon 949 (largely same guts as an X5) in that collaborative scanner test alongside the Epson? Specifically the table edge & the cloth detail. It shows quite a serious difference to what you keep telling us, doesn't it?

And you might also want to consider the results of an actual teardown and MTF test of a V700, which dismantles your assertions about its useable resolution. (http://dasch.rc.fas.harvard.edu/papers/Scannerevaluation1.pdf)

[Pere Casals]

Pere: how about you show us the results from the Imacon 949 (largely same guts as an X5) in that collaborative scanner test alongside the Epson? Specifically the table edge & the cloth detail. It shows quite a serious difference to what you keep telling us, doesn't it?


Once the V750 crop is edited with 2 or 3 clicks it's mostly like all the other crops:

192619

Pro scanners have a very good automatic digital sharpening in firmware or in software, and the delivered image is optimitzed yet to its best, with the EPSON you have to do it on your own later, this is well known, the Petapixel test also explains it.

If we could see that negative with a microscope probably we are sharpening the image in the negative that may have some blur there.




And you might also want to consider the results of an actual teardown and MTF test of a V700, which dismantles your assertions about its useable resolution. (http://dasch.rc.fas.harvard.edu/papers/Scannerevaluation1.pdf)

The MTF of the X5 for 4x5 at 1900dpi is zero, while the V700 has a Transfer of Modulation until 2900dpi (in the Hor axis).

Please see this Pali V700 test: https://www.largeformatphotography.info/forum/showthread.php?150020-Scanner-Comparison-2019-Epson-Flatbed-Eversmart-Flatbed-Drum-Scanners&p=1478033&viewfull=1#post1478033

Elements 5.6 and 5.4 were resolved, which is 2900x2300 effective.

https://www.largeformatphotography.info/forum/showthread.php?150020-Scanner-Comparison-2019-Epson-Flatbed-Eversmart-Flatbed-Drum-Scanners&p=1477892&viewfull=1#post1477892

Remember that the X5 for 4x5 only resolves Elements 5.1 or 5.2. Test you'll hassie and you'll find that.

[interneg]

Once the V750 crop is edited with 2 or 3 clicks it's mostly like all the other crops:

192619

Pro scanners have a very good automatic digital sharpening in firmware or in software, and the delivered image is optimitzed yet to its best, with the EPSON you have to do it on your own later, this is well known, the Petapixel test also explains it.



No. Wrong. It's because professional scanners don't give severely misleading data about their sensors. They are straightforward 3-line CCD - often available on the open market from Mouser etc - not a 6-line attempting to claim better resolution & their useful resolution limits are by design or mechanical precision limitations. The teardown covers why the Epson sensor design has severe limits in useful resolution before going blurry. And because of its design it gets really noisy when sharpened & the double line interpolation design does really odd things to the grain.

That's a horrid, noisy thing you posted there, and the fine details are still wooly. It would be really obvious in a print. I know because it looks like every other Epson scan that people try to pass off as being 'good' in some way.

That you didn't show a side-by-side with the 949's non sharpened results is very telling.
192643
When you have that much more detail & less noise, sharpening has a better effect if you wish to add it.

[Pere Casals]

Interneg, I don't understand why you are such a hater of the Epsons :)

Let's go point by point:



No. Wrong. It's because professional scanners don't have misleading data about their sensors. They are 3-line CCD, not a 6-line attempting to claim better resolution.

> The 6-Line sensors are a great improvement over the 3-Line ones, the second RGB row covers the gap between the microlenses of the first row, this is the perfection. This is basic knowledge. Epson literature shows they are proud of the 6-Lines.

> Anyway if you consider only the first 3 Lines of the EPSON still it has +20,000Pix rgb, 2.5x what the X5 has.

> You should understand that the High Res lens of the EPSON covers 5.9", this is a lot, this is the weak point of the EPSON for 35mm, while the hassie covers 1" !!!

> As format is larger the X5 loses its advantage, by 4x5 the Epson resolves element 5.6 (Hor), the X5 resolves the 5.2. By 5x7 the X5 only resolves the 5.1 (or 4.6) while the Epson still holds the 5.6, as it's a fixed lens the Epson always resolves 5.6 from 35mm to 5x7. The X5 lowers resolution as it zooms out, from 6900 dpi at 35mm to 1500 at 5x7".

> And the Epson also makes 8x10 :), here we cannot compare, but the Epson delivers around +2000dpi effective for 8x10 while the X5 with the 5x7 adapter only performs 1500.




That's a horrid, noisy thing you posted there, and the fine details are still wooly. It would be really obvious in a print. I know because it looks like every other Epson scan that people try to pass off as being 'good' in some way.


https://www.largeformatphotography.info/forum/attachment.php?attachmentid=185109&d=1543958092

The dB SNR vs D of the V700 is sound, it shows pretty right at 3.1D. Add the multiexposure effect that's not included in this graph, IIRC. With multi-exposure you have around the 3.1D reading of the table for 3.4D.



This test was made by Pali, with the V700 sharpened by me in two clicks, honestly, do you thing that result from high end flatbeds are better ?
To me, same noise, same sharpness, if edited also all scanners the same colors:


https://c1.staticflickr.com/5/4857/46755757932_c7010da815_o.jpg

[Pere Casals]

I'd add that another source of performance for the V850 is the fixed magnification of the lens, so the lens is optimized for that exact fixed magnification which is around 1:3 for the HR lens

Instead the X5 has to magnify 2:1 to 1:2.5, which is a 5 times change, so I guess they use the 1x version of the Rodagon D, which is in the center of the range.

The lens in the X5 has a retail price that's higher than a whole V850, I know very well those Linos Rodagon D, I've used them for machine vision projects, very fine glass, no doubt, but they are sold in the 1x and 2x versions, each optimized for a particular magnification. Of course, lenses at 1x or 2x miss the optimal performance when you move from the optimal magnification.


In the case of the V850 they take crazy 5.9" with the lens, which is an insane amount to scan 35mm and even MF, still the optics perform better than expected just because the lens works at the magnification at what it was optimized.

Some in this forum had been saying that the V700-850 had cheap plastic lenses inside. Not true... resolving those 2900 dpi (Hor) in a 5.9" field is not that easy, those 17000 total effective pixels at extition for the covered field is a remarkable number.

So, IMHO, every machine has weak and strong points, and a cost.