4x5: LF Print quality from epson V750 scans

[Kburg928]

I've been kicking around the idea of picking up an Epson V750 scanner and I was wondering how good the scans are off of this scanner? I print using an Epson 7800 with normal print sizes of 16x20 and 24x30 but I plan on picking up the Epson 9900 and would be making prints up to 40x50.

I guess my question is, what is the largest gallery quality print I can make from an epson V750 scanner?

Thanks for the help
Keith

[Brian Ellis]

I have the Epson 4990, the predecessor to the 700/750 scanners. Unless the 750 greatly exceeds the quality of the 4990, and I don't think it does based on the tests I read when the 700/750s first came on the market, it wouldn't do the job to my satisfaction at least with a 40x50 print.

The largest print size that's acceptable to me with the 4990 and an Epson 3800 printer is about 17x25 and that's pushing it for exhibit where most people won't be putting their noses on the print. For my own use I stop at about 13x18. But different people have different standards. If your large prints are going to be viewed from a distance commensurate with the size of the prints things might be different but I kind of doubt it. I just don't think these Epson prosumer scanners are made with 40x50 prints in mind.

[Richard M. Coda]

I had some (color) 16x20s made from 4x5 scans (1800 dpi on V750 PRO) and they were not as sharp as the 16x20s I had made from drum scans. I would be hesitant to go that large from a desktop scanner. There are plenty of places to get reasonably priced drum scans, but then again, if you want the best quality you will have to pay for it (I would recommend West Coast Imaging (http://westcoastimaging.com)).

[Ken Lee]

Most reports here, convey that the recent crop of Epson scanners deliver somewhere around 2100 spi at best. (Even when you set them for 2400 spi and beyond, all you get is bigger files - but no more actual detail).

If you intend to print at 300 dpi, this suggests you can make a 7X enlargement, and get whatever data was present. If you plan to print at 360 dpi, then you can make a 5.8X enlargement.

If you are among those who rate their scanner more rigorously, then perhaps you get 1800 spi from it, and can make correspondingly smaller enlargements, like 5X or 6X maximum. This has been my experience. Besides, Large Format Lenses, after that much enlargement, start to lose their "luster" you might say.

Numbers aside, all of this is ultimately a matter of personal taste, so you should compare real results, to see what *you* want.

[kaiyen]

Richard - did you just seriously compare the V750 to a drum scan?

They are not the same, no question. When I want a scan for exhibits - which...isn't that often :-( - I get a drum scan. But for $500 (or whatever the 750 sells for now), which is a lot cheaper than, say the Howtek 3500 for sale here recently, I'm pretty darn happy with it.

[sanking]

Most reports here, convey that the recent crop of Epson scanners deliver somewhere around 2100 spi at best. (Even when you set them for 2400 spi and beyond, all you get is bigger files - but no more actual detail).

If you intend to print at 300 dpi, this suggests you can make a 7X enlargement, and get whatever data was present. If you plan to print at 360 dpi, then you can make a 5.8X enlargement.



Another consideration is that it is very unlikely that your LF film will have more than 2400 dpi of real information. So if you assume the very best resolution of 2400 dpi from the Epson V750 that will only get you a print of 32X40" at 300 dpi.

Anything beyond that and you are in the realm of genuine fractals.

Also, bear in mind that whether you scan with a drum scanner or with an Epson V750 a lot of the ultimate print quality depneds on the skill of the scanner operator and on post processing of the image file. Some people make beautiufl prints with real detail as low as 200 dpi that others can not equal with two or three times that much detail.

Sandy King

[Ron Marshall]

With most of the images I scanned on the Epson 4990 I was only happy with 13x16. I did several comparisons to drum scans.

[nonuniform]

Perhaps the question of whether the v750 produces an image that can be printed larger than 17x25 is subjective. I've used the v750 wet mount option and produced prints from those scans at 24x30 that I really love. Then again, the images were shot handheld with a press camera, so the expectation of sharpness is very different!

[rdenney]

I own a 20x40 color print that the photographer made from a 6x12 negative on an Epson 4990. The print quality is excellent--comparable in resolution to a 16x20 Cibachrome print from 4x5 Velvia made by another well-known photographer. Both stand up to close scrutiny.

That doesn't mean that it would compare favorably to a similarly sized print from a drum scan. There is always something better out there that I can't afford, or that makes life too difficult. Thus, every choice we make presents a compromise.

If the 20x40" print shows compromise, it's not in the resolution. It may be in a subtle loss of tonality.

I have only made one scan from my V750 so far (just having set it up last week). I made that scan at 2400 spi, and when viewed actual pixels in Photohop, it did show a bit of softness. Basic corrective sharpening required a radius of 1 pixel, which indicates quite good detail in the scan. The actual-pixel display is not overwhelmingly unlike the actual-pixel display from my Canon 5D with images made using very good prime lenses. And, of course, there are many, many, many more of those actual pixels. (Ten times as many.) Again, if there is a difference, I would expect it to be in tonality rather than resolution.

The actual-pixel display on my monitor represents an image of 96x124 inches. Maybe a third of that would sample out that bit of softness I see on the monitor, and yield a 32x40" print. That 20x40" print from 6x12 seems to bear that out. So, I'm expecting decent results at 8x.

I haven't compared it to my Nikon film scanner as of yet. That will tell a different tale.

So, 40x50 will really be stretching it, but 32x40 will be good enough for many applications.

I know people who make prints that big from digital SLR's and think them good enough. I've already two planets away from them just starting with large-format film.

In this image, I can just about read the label on the beer bottle floating in the water in the background in the original scan, which was made at 2400 on a V750. That bottle is soft at full view, but it's about the softness I would expect from the lens (an Ilex Paragon at f/22) as much as anything. Again, actual pixels on the display represents about a 24x enlargement.

http://www.rickdenney.com/images/Espada-Aq-032793-13-lores.jpg

I haven't attempted a print yet.

Rick "not expecting it to look like [fill-in-the-blank] more expensive alternative" Denney

[Lenny Eiger]

I didn't do these, but they are credible examples, IMO, of the difference.

750 is on the left, 4500 on the right.

Lenny

[Darin Boville]

I didn't do these, but they are credible examples, IMO, of the difference.

750 is on the left, 4500 on the right.

Lenny

And yet when you sharpen the left one (the 750) it looks awfully close to the 4500...

--Darin

[rdenney]

And yet when you sharpen the left one (the 750) it looks awfully close to the 4500...

--Darin

And printed at an 8x enlargement would still look darn good.

Rick "who can definitely see the difference at 24x!" Denney

[PenGun]

I didn't do these, but they are credible examples, IMO, of the difference.

750 is on the left, 4500 on the right.

Lenny

I don't know what the particulars are for those shots. I do know the 700/750 is hard to focus.

[mrpengun]

http://www.largeformatphotography.info/scan-comparison/

that site has lots of good comparisons of various scanners.

I never made a print from it, but i did scan a 35mm slide with the maximum resolution, and it was precise enough to see drops of water from the slide.

<edit> just noticed our names... weird.
I should confess though, mine was a mistyped "penguin"...
</edit>

[Lenny Eiger]

And yet when you sharpen the left one (the 750) it looks awfully close to the 4500...

--Darin

No it doesn't. It's a tiny little file. You can't really see anything. It is true that one can sharpen and get a lot from a consumer flatbed. One starts very blurry and adds the sharpening effect and if you're careful you can eke out some quality.

With a drum scan you start out with a very sharp file, and usually plenty of pixels. There is no comparison, especially when you want to make a print larger than 13 x 19. There is also more dynamic range, more sensitivity, etc. Not everyone can properly take advantage of it all, but it makes a huge difference.

Lenny

[percepts]

Fuji Provia gives resolution figures of:

Chart Contrast 1.6 : 1 .......................... 80 lines/mm
Chart Contrast 1000 : 1 ........................ 160 lines/mm

What that means is that at a contrast ratio of 1000:1 it is possible to get 160 lines per millimeter. But what most people blindly ignore is that means alternating black and white lines with an 11 stop contrast ratio between them which in the real world will very very rarely exist over such minute distances on film. But fuji also give the real world contrast ratio of 1.6 : 1 where you only get 80 lines per millimeter. 80 times the number of millimeters in an inch is 80 x 25.4 = 2032. The really funny thing is that people persist in blaming the cheap scanners for not being able to get more resolution out of film when most images never had it in them to start with. Thats what happens when amateurs start trying to make a name for themselves on the web. People start believing them. Drum scans will give better quality scans with less noise and better colour but even a drum scan can't get resolution out that isn't in the film.

So assuming you are only going to get 80 lines per millimeter, then you are only going to get 2032 lines per inch and that means a max print width of 26.6 inches or so (360 ppi) before you start stretching out the gaps beween samples on an epson printer.

I am no expert of this stuff and Fuji say lines per mm. They don't say lines pairs per millimeter. If they had said line pairs per mm then a flatbed scanner might be able to get them or it might not. If it could then you can double that print size to 53 inches. But the last time I tried to get any sense out of anyone on that question, I was given spurious answers because nobody actually seemed to know if fuji means lines per mm or line pairs per mm. I don't know but is it pure coincidence that what most people seem to be able to get out of the film is the same as fuji say you are likely to get in the film.

p.s. :

Fuji numbers for acros which only give 60 lines per mm for 1.6:1 contrast:

Chart Contrast 1.6 : 1 ....................... 60 lines/mm
Chart Contrast 1000 : 1 ..................... 200 lines/mm

[Jim Michael]

Another consideration is that it is very unlikely that your LF film will have more than 2400 dpi of real information. So if you assume the very best resolution of 2400 dpi from the Epson V750 that will only get you a print of 32X40" at 300 dpi.

Anything beyond that and you are in the realm of genuine fractals.



Sandy, could you elaborate on the realm of genuine fractals? There was a fellow here in Atlanta who was doing some interesting stuff with fractal algorithms and image processing, are you referring to that?

[Darin Boville]

It may be a tiny little file, as you say--but it was your sample image, not mine! :)

I'd be happy to look at any giant, un-little files you'd care to share...

--Darin


No it doesn't. It's a tiny little file. You can't really see anything. It is true that one can sharpen and get a lot from a consumer flatbed. One starts very blurry and adds the sharpening effect and if you're careful you can eke out some quality.

With a drum scan you start out with a very sharp file, and usually plenty of pixels. There is no comparison, especially when you want to make a print larger than 13 x 19. There is also more dynamic range, more sensitivity, etc. Not everyone can properly take advantage of it all, but it makes a huge difference.

Lenny

[Lenny Eiger]

I'd be happy to look at any giant, un-little files you'd care to share...
--Darin

We are very close geographically - why not come by and look at some prints? I'll show you the difference in real terms.... where it matters.

Pt Reyes National Seashore is always worth a visit....

Lenny

[Bruce Watson]

I've been kicking around the idea of picking up an Epson V750 scanner and I was wondering how good the scans are off of this scanner? I print using an Epson 7800 with normal print sizes of 16x20 and 24x30 but I plan on picking up the Epson 9900 and would be making prints up to 40x50.

First, you realize I hope that you aren't the first person to ask this question, yes? If you use the search feature you'll pull lots of threads on this. Vast quantities of opinions. Lots of smoke. Little fire.

The bottom line is, you can do what you want. If you want to make 40x50 inch prints from 35mm scanned on an Epson, go for it. The only valid judge of whether it meets your standards or not, is you. Yours is the only opinion that counts when it's your money you are spending.

For my money (literally), I tried quite a while to coax quality out of an Epson consumer flatbed scanner. I found for my needs I couldn't go more that 4x enlargement with it.

I fought this for a year or so. I didn't want to buy a drum scanner. I didn't. I didn't want to have to climb all the learning curves, time, mess, expense, etc. But I finally had to have the scan quality. And holy cats, the difference is huge. I could see it at 2x enlargements in my first scans. And I've gotten a lot better at scanning since then. A lot better. I finally quit playing with the consumer flatbeds altogether because no matter how much I learned about scanning I just couldn't make them do what I needed them to do.

I should point out here, because these threads always latch hold of resolution and debate it endlessly, that there's a lot more to a high quality scan than just resolution. I'm just sayin'.

The quality of the final print depends on everything that happens before the printer. You have to ask yourself why you want scanning to be your weakest link.

[Paul Kierstead]

At what point does v700 questions become trolling?

[percepts]

At what point does v700 questions become trolling?

When V900 comes out next month.

[PenGun]

At what point does v700 questions become trolling?

Yes there is a large body of opinion and even a few facts.

I'd say Bruce has it about right. About 4X is good from a 700/750. I'm surprised how good my V700 is but even a sharp MF negative at full res is a bit thin at 16x20.

My 4X5 is a very nice fit for my V700 and Pro 3800. 16X20 is pretty sweet and I think it's a great toolchain for me to learn with.

Focus is an issue and you have to experiment to see where the hi res system actually focuses. Mine for instance is sharpest with no feet on the standard 4X5 mounts.

[Kburg928]

Thanks for all the replies,

So basically the V700/750 will work out well for scanning images to be viewed on my website but for high quality prints above 16x20 it's best to go with a scanner like the Imacon's or a drum scanner.

Thanks again
Keith

[Ken Lee]

You should also have a look here, for threads that deal with focusing these scanners.

[Michael Gordon]

I highly recommend the BetterScanning.com (http://betterscanning.com/) Variable Height Mounting Station. It makes a huge difference in the quality of the scans coming off my V700.

[Lenny Eiger]

Thanks for all the replies,

So basically the V700/750 will work out well for scanning images to be viewed on my website but for high quality prints above 16x20 it's best to go with a scanner like the Imacon's or a drum scanner.

Thanks again
Keith

Keith,
I would basically agree with you on this one, but that's me. I would rather have a drum even for 11x14's, that's partially because I already have one (not particularly fair). However, you are now in the realm of religion vs science (or opinion).

It has to do with what different people see as "quality" in a print. I am looking for every bit of graytone separation, all the 3-dimensional quality, etc., to create a longer tonal range than a platinum print and generate all the atmosphere that I can. Not everyone is trying to do that - some people like a lot more contrast. Everyone has a different aesthetic - and they should.

If you want 3-dimensional midtones as I described, a drum scanner is a better bet. You have a much higher dynamic range, the scanner can really separate things out as it is very accurate, sharp and all the rest.

I hope that helps.

Lenny

[Frank Doering]

In a pinch, the 750 can be pushed surprisingly far. I recently had to use it to prepare files for 8x10' (that's feet) prints. A writeup of the experience with sample crops is here: http://doeringphoto.com/projects/powerplant/info2.html.

[Juergen Sattler]

Frank, really enjoyed your write-up of the process you went through. I can only imagine what you went through to get these prints done in time with all the throw backs you experienced - very well done. Would love to see your exhibit in person.

[D. Bryant]

Yes there is a large body of opinion and even a few facts.

I'd say Bruce has it about right. About 4X is good from a 700/750. I'm surprised how good my V700 is but even a sharp MF negative at full res is a bit thin at 16x20.

My 4X5 is a very nice fit for my V700 and Pro 3800. 16X20 is pretty sweet and I think it's a great toolchain for me to learn with.

Focus is an issue and you have to experiment to see where the hi res system actually focuses. Mine for instance is sharpest with no feet on the standard 4X5 mounts.

You use the tools you can afford, but you owe it to yourself to get at least one drum scan made to understand the significant difference a drum scan makes in quality.

Compare your 16x20s then and see what you think. :)

You may also discover that your lenses and view camera technique aren't as good as you thought.

That's what happened to me after I did the comparison.

BTW, Do you also use the login: mrpengun?

Don Bryant

[PenGun]

You use the tools you can afford, but you owe it to yourself to get at least one drum scan made to understand the significant difference a drum scan makes in quality.

Compare your 16x20s then and see what you think. :)

You may also discover that your lenses and view camera technique aren't as good as you thought.

That's what happened to me after I did the comparison.

BTW, Do you also use the login: mrpengun?

Don Bryant

Not me.

I have yet to see a drum/flatbed comparison with the V700/V750 where the flatbed was in focus. I can appreciate the ability of PMTs to see further into the black and suspect a better tonal range for similar reasons but I'm not convinced the V700/V750 is as bad as the drum scan crowd claims.

I did photography long ago and made some money even. Now it's entirely to please myself. I am a brutal critic so I am confident I'll get as close to as good as I can. I have decided that what I am doing now is art. It may be bad art but it is an expression of my broken vision. ;)

[Bruce Watson]

I have yet to see a drum/flatbed comparison with the V700/V750 where the flatbed was in focus. I can appreciate the ability of PMTs to see further into the black and suspect a better tonal range for similar reasons but I'm not convinced the V700/V750 is as bad as the drum scan crowd claims.

There's no reason for you to be convinced. The only person who can convince you is you yourself. So... what's stopping you?

Go off and find a scan from a V7xx that satisfies you. Then get a drum scan made from the same film that satisfies you (same number of pixels, same bit depth, etc. of course). If you give a capture sharpen to the flatbed, be fair and give an appropriate capture sharpen to the drum; IOW make both scans as good as they can be. Then make same size prints from both scans (or sections from what would have been full size prints), put them up on the wall side-by-side under the same lighting. If you want a fair reading, make this as much an apples-to-apples comparison as you can.

Then stand there and evaluate the prints; decide which print you like best. Show the prints to your friends and neighbors (without telling them anything about them), ask them just a single question: Which print do you prefer? Listen to what they say (no prompting from you or anyone who knows which print is which of course).

Sometimes the only way to convince yourself is to do the work. After all, why guess when you can know?

[PenGun]

There's no reason for you to be convinced. The only person who can convince you is you yourself. So... what's stopping you?

Go off and find a scan from a V7xx that satisfies you. Then get a drum scan made from the same film that satisfies you (same number of pixels, same bit depth, etc. of course). If you give a capture sharpen to the flatbed, be fair and give an appropriate capture sharpen to the drum; IOW make both scans as good as they can be. Then make same size prints from both scans (or sections from what would have been full size prints), put them up on the wall side-by-side under the same lighting. If you want a fair reading, make this as much an apples-to-apples comparison as you can.

Then stand there and evaluate the prints; decide which print you like best. Show the prints to your friends and neighbors (without telling them anything about them), ask them just a single question: Which print do you prefer? Listen to what they say (no prompting from you or anyone who knows which print is which of course).

Sometimes the only way to convince yourself is to do the work. After all, why guess when you can know?

I'm a poor person. One of Lenny's scans is about 25% of a V700 price.

As I said, I'm calling it art. I'm not all that concerned with the ultimate details, the picture is what I'm after. If my art is so poor that the scanner res is important, I've failed.

I used to roll up 3 or 4 35mm carts from a 100' roll every day and blow that off, process it, and even get some prints done that night. Mostly E6, all my own work. Used to make masks so I could print cibachromes. I've spent many hours sweating the details.

I don't do that anymore. I try to see well and take only what I need.

I may pay for a drum scan someday, but it's not a priority.

<vulture>It's also likely your economy will collapse sooner and worse than ours and drum scanners may become very cheap in Can $s in the next while.</vulture>

[Frank Petronio]

In my ideal world I'd have a top quality Creo scanner (along with two spares for parts) networked in with double Mac workstations loaded with memory and multiple 30" monitors, all tied to the latest greatest Epson printing on the most expensive paper, mounting the prints on Plexi and sending copies of my 600-page self-published monograph to galleries around the world.

In the real world I do everything with a two-year old MacBook Pro, an older Epson 4990, and an ancient Epson 2200 and some external drives for back-up, which all cost less than $4k including software. Also, my total camera investment -- 4x5, 35mm, digital -- is only about $1200.

I've worked as a commercial photographer and boughten plenty of drum scans "back in the day" and I love them. But even using my very basic gear, I can produce satisfying work, at least most of you guys seem to compliment me on it. If I were primarily concerned with detail and having the "ultimate" I wouldn't be able to function, much less sleep at night... I'd be like one of those gigapixel anal retentive freaks. Life is too short... As it is now, if I had enough money to buy all those Creos, Epsons, and Macs I wouldn't... I would spend the money on experiences worth photographing. The damn curators who ignore me now can rescan my stuff after I'm dead if it bothers them so much.

So just buy what you can afford and learn how to use it to its full potential.

[D. Bryant]

but I'm not convinced the V700/V750 is as bad as the drum scan crowd claims.



Sorry but it's not a claim. Instead of small objects being seen as blobs you can actually see the details of the object. This isn't an exaggeration nor is the V700/750 almost as good. There really is a major difference. That's why the drum scanning crowd keeps pointing that fact out. They are simply trying to refute the mis-information about the invalid comparison.

I'm sorry if these posts sound pedantic.

Don Bryant

[PenGun]

Sorry but it's not a claim. Instead of small objects being seen as blobs you can actually see the details of the object. This isn't an exaggeration nor is the V700/750 almost as good. There really is a major difference. That's why the drum scanning crowd keeps pointing that fact out. They are simply trying to refute the mis-information about the invalid comparison.

I'm sorry if these posts sound pedantic.

Don Bryant

The main problem I have is that the drum scanning crowd seems unable to focus a V700/V750. That does not lead me to have much confidence in their tests. If you don't have focus you have blobs.

It is pedantic actually. The response from the people who do drum scanning is almost a chorus now. The Ravens outside, who are quite loud this time of year, are almost starting to sound better. ;)

The light is nicely sideways, I have wild roses to hunt.

[Lenny Eiger]

In the real world I do everything with a two-year old MacBook Pro, an older Epson 4990, and an ancient Epson 2200 and some external drives for back-up, which all cost less than $4k including software. Also, my total camera investment -- 4x5, 35mm, digital -- is only about $1200.

I've worked as a commercial photographer and boughten plenty of drum scans "back in the day" and I love them. But even using my very basic gear, I can produce satisfying work, at least most of you guys seem to compliment me on it.

I see nothing wrong with your approach to your own work. It sounds sane.

For me it isn't about detail, it's about 3-dimensional quality. When you see spray from the sea, I want you to feel the cold water on your arm. I can't get it with a 4x5 and I can't get it with a flatbed scanner. There are a lot of extra tones in my scanner and other materials, and the scanner gets me everything the neg can give.

If someone is satisfied with the prints they have, then the least expensive, easiest way to go is great.

I think the only problem comes when someone suggests that med format can deliver as much as an 8x10, or that a flatbed is equal to a drum. If a smaller neg or a flatbed gives them all they want, it is. But if someone wants more, there is another level up to go. It isn't for everyone. Everyone has a different area of the sport that they focus on - we all get to choose.

Lenny

[D. Bryant]

The main problem I have is that the drum scanning crowd seems unable to focus a V700/V750. That does not lead me to have much confidence in their tests. If you don't have focus you have blobs.

You misunderstand what I mean. I've got an Epson 4990 and have the Betterscanning wet scanning adjustable height film holder. Even after calibrating the focus height porperly there are details that cannot be sharply resolved. I've sanned the same negative on the 4990 and a drum scanner and the details that are little blobs from the 4990 scan are detailed objects in the drum scanned image file. The same holds true for the V700/V750 scanners. That's all I'm saying.

If you wish to refute these facts then you are doing so based on assumptions, not on facts.

I'm not trying to convince you that the Epson scanners are a piece of junk. For what they cost and what they do they are a fair value (although some Microtek users will say that their machinces are better performers.)

But don't continue to say that the Epson scanners have almost as good output as the high end flatbeds and drum scanners because they don't, not even close. If they do please prove it.

Don Bryant

[percepts]

The following thread gives a clue to why drum scanners produce better results than ccd scanners.

http://www.largeformatphotography.info/forum/showthread.php?t=20817&highlight=drum

ccd scanners have fixed size samples with fixed spacing. They simply cannot physically compete with a drum scanner even though they have several overlapping scanning lines. The sample size is still fixed and that forces software interpolation in most cases. And the optic path and sensitivity is never as good as a drum scan can be.

[PenGun]

The following thread gives a clue to why drum scanners produce better results than ccd scanners.

http://www.largeformatphotography.info/forum/showthread.php?t=20817&highlight=drum

ccd scanners have fixed size samples with fixed spacing. They simply cannot physically compete with a drum scanner even though they have several overlapping scanning lines. The sample size is still fixed and that forces software interpolation in most cases. And the optic path and sensitivity is never as good as a drum scan can be.

You can read. Well done.

This is an old discussion. I suspect the drum boys actually know how good the Epsons can be and it scares them. The chorus is consistent and has the same choir.


You misunderstand what I mean. I've got an Epson 4990 and have the Betterscanning wet scanning adjustable height film holder. Even after calibrating the focus height porperly there are details that cannot be sharply resolved. I've sanned the same negative on the 4990 and a drum scanner and the details that are little blobs from the 4990 scan are detailed objects in the drum scanned image file. The same holds true for the V700/V750 scanners. That's all I'm saying.

If you wish to refute these facts then you are doing so based on assumptions, not on facts.

I'm not trying to convince you that the Epson scanners are a piece of junk. For what they cost and what they do they are a fair value (although some Microtek users will say that their machinces are better performers.)

But don't continue to say that the Epson scanners have almost as good output as the high end flatbeds and drum scanners because they don't, not even close. If they do please prove it.

Don Bryant

How did you focus your 4990? The mount make no difference really, just makes it easy to adjust. I know what I did to get focus on my V700. The difference from stock setting is profound.

I have claimed nothing except that you people cannot achieve focus.

[sanking]

I have claimed nothing except that you people cannot achieve focus.

So why would Don Byrant and the rest of "we people" not be abel to achieve focus?

It is really a pretty simple matter, IMO. You take resolution target or a fairly high contrast negative, scan it at the highest optical resolution of the scanner at varying distance from the glass of the scanner bed, and then visually compare the scans.

I have owned an Epson 2540, a 4870 and a 4990 and it never took me more than about 15 minutes to find the plane of best focus and it sure does not take a lot of skill.

Sandy King

[PenGun]

So why would Don Byrant and the rest of "we people" not be abel to achieve focus?

It is really a pretty simple matter, IMO. You take resolution target or a fairly high contrast negative, scan it at the highest optical resolution of the scanner at varying distance from the glass of the scanner bed, and then visually compare the scans.

I have owned an Epson 2540, a 4870 and a 4990 and it never took me more than about 15 minutes to find the plane of best focus and it sure does not take a lot of skill.

Sandy King

We have been through this before. Your example from the last time was admitted to be out of focus.

All those scanners focus on the glass, or should. The V700/V750 is DIFFERENT.

The high res scan system on the V700/V750 focuses somewhere above the glass. Where is up to you to find. I do not think many have found it.

[D. Bryant]

PG,


I have claimed nothing except that you people cannot achieve focus

I'm sorry but that just isn't correct in my case. I use a pair of high contrast, high resolution negs that were originally created for use to focus enlargers and automated printers in the photofinishing industry. I used these same type of negatives to adjust the focus of printers and enlargers when I worked in the professional photofinishing industry. They allow one to judge focus very easily and precisely.

Additionally my monitors are extremely sharp (NEC Multisync 2090UXi) so I'm confident what I'm viewing on screen represents reality. I wonder if you can say the same.

So in short you don't know what you are talking about. Your ridiculous position simply puts your judgment in question. I'm through discussing this with you.

Don Bryant

[neil poulsen]

. . . I should point out here, because these threads always latch hold of resolution and debate it endlessly, that there's a lot more to a high quality scan than just resolution. I'm just sayin'. . .

Bruce,

You make an interesting point here. Could you please elaborate on this.

For example, I'm thinking about dynamic range, texture in shadows, in highlights, etc. What else would you include?

[sanking]

We have been through this before. Your example from the last time was admitted to be out of focus.

All those scanners focus on the glass, or should. The V700/V750 is DIFFERENT.

The high res scan system on the V700/V750 focuses somewhere above the glass. Where is up to you to find. I do not think many have found it.


I think you have me confused with some other body. To the best of my recollection I have never shown any example that was "admitted to be out of focus." Course, I will stand corrected if you can point me to the "out of focus" example you claim I presented but I am pretty sure you are blowing smoke.

It is factually incorrect that the scanners I mentioned focus best on the glass. In fact, none of the scanners I mentioned focus best on the glass. They vary from best focus at about 0.5mm to as high as 2.5mm, and that is pretty much the experience of others. I tested a V700 and it focused at about 2mm above the glass.

As I previously stated, finding the point of best focus takes virtually no skill or time, and it is not magic. You are making this seem a lot more complicated than it is. Most likely because, as Don remarks, you really don't know what you are talking about.

Sandy

[PenGun]

I think you have me confused with some other body. To the best of my recollection I have never shown any example that was "admitted to be out of focus." Course, I will stand corrected if you can point me to the "out of focus" example you claim I presented but I am pretty sure you are blowing smoke.

It is factually incorrect that the scanners I mentioned focus best on the glass. In fact, none of the scanners I mentioned focus best on the glass. They vary from best focus at about 0.5mm to as high as 2.5mm, and that is pretty much the experience of others. I tested a V700 and it focused at about 2mm above the glass.

As I previously stated, finding the point of best focus takes virtually no skill or time, and it is not magic. You are making this seem a lot more complicated than it is. Most likely because, as Don remarks, you really don't know what you are talking about.

Sandy

Have it your way. I really don't care.

[sanking]

Have it your way. I really don't care.

OK, I will read your response this way. You made several incorrect statements, got called on them, and could not back any of them up.



Sandy

[PenGun]

OK, I will read your response this way. You made several incorrect statements, got called on them, and could not back any of them up.

Sandy

Not at all. Try reading what I said. There are reasons I do not want to argue with you, you are not good at it.

[PenGun]

PG,



I'm sorry but that just isn't correct in my case. I use a pair of high contrast, high resolution negs that were originally created for use to focus enlargers and automated printers in the photofinishing industry. I used these same type of negatives to adjust the focus of printers and enlargers when I worked in the professional photofinishing industry. They allow one to judge focus very easily and precisely.

Additionally my monitors are extremely sharp (NEC Multisync 2090UXi) so I'm confident what I'm viewing on screen represents reality. I wonder if you can say the same.

So in short you don't know what you are talking about. Your ridiculous position simply puts your judgment in question. I'm through discussing this with you.

Don Bryant

Please yourself. That was probably harsh and overly personal.

What I mean is, and I have said this too many times, I have never seen a comparison between the V700/V750 and a drum scanner where the flatbed was in focus.

[PenGun]

I think you have me confused with some other body. To the best of my recollection I have never shown any example that was "admitted to be out of focus." Course, I will stand corrected if you can point me to the "out of focus" example you claim I presented but I am pretty sure you are blowing smoke.

You are right it was Jim Kitchen. It was in a thread where this was discussed.



It is factually incorrect that the scanners I mentioned focus best on the glass. In fact, none of the scanners I mentioned focus best on the glass. They vary from best focus at about 0.5mm to as high as 2.5mm, and that is pretty much the experience of others. I tested a V700 and it focused at about 2mm above the glass.


All those scanners have one system. They should focus on the glass or else everything on the glass, 8x10s for instance, are out of focus. The cheap flatbeds are like that.

You tested a V750 with the low res system on the glass and in a holder.

http://www.largeformatphotography.info/forum/showpost.php?p=438099&postcount=30

You were not that sure what you tested actually. You did not test the high res system AFAIKT.



As I previously stated, finding the point of best focus takes virtually no skill or time, and it is not magic. You are making this seem a lot more complicated than it is. Most likely because, as Don remarks, you really don't know what you are talking about.

Sandy

Well yes that is nearly always true.

[Sevo]

All those scanners have one system. They should focus on the glass or else everything on the glass, 8x10s for instance, are out of focus. The cheap flatbeds are like that.

"Those" Epson scanners are all dual mode, with low-res paper mode on the glass and high-res transparency in holders. Period.

Sevo

[Bruce Watson]

Bruce,

You make an interesting point here. Could you please elaborate on this.

For example, I'm thinking about dynamic range, texture in shadows, in highlights, etc. What else would you include?

All parts of tonality, from separation of tones to proper rendition of textures, particularly smoothness. Much of this lacks vocabulary to adequately describe it, which is perhaps another reason to latch hold of resolution. But it's all part of getting the scanner to make an accurate representation of the film.

There's also something I call clarity for lack of a better word. It's one of the things I find when I scan the same piece of film on different scanners. When you compare prints from the different scans, sometimes you can see that one print has more clarity than the others. It's like the others have a thin veil between your eyes and the print. I suspect it's a local contrast effect, but I don't really know. A fluid mount tends to increase clarity, but it won't cure a veiled scanner completely in my experience.

Edit: I see below that Frank says this quality is "presence". I'll buy that. And there's a depth to a better scan that you don't see in a lesser scan. I don't know how to adequately describe it.

[Frank Petronio]

From my experience, the difference between the flatbeds and the high-end scans are kind of like the difference between 120 versus 4x5 (scanned on your flatbed). You can get a pleasing full range images from the 120 but you get that much more depth and presence from the larger format. And in my experience, I've seen properly scanned 120 on a good scanner blow away my 4x5 scans from my Epson.

You still have to manage with what you can afford. If you only shoot a dozen decent images per year, then by all means, you can rationalize spending the time and money on drum scans. But I think that quantity equals quality myself. Nothing beats shooting and while the post work is important, it is still secondary to finding/making good photos to start with.

Probably half the images I see people do seem to have "errors" to my eye. A lot of people fail to even hit a black and white point and put the mid-tones in some pleasing area... so just try to keep things in perspective, you can do respectable work with a flatbed.

[Ken Lee]

By analogy...

You don't have to go to an expensive drum scanner to improve on consumer flatbed scans. When scanning medium format, the difference between a dedicated film scanner and a consumer flatbed is plain to see. As others have said, what looks like a blob in one scan, is an object in the other. I presume that drum scanners, on Large Format, would show an analogous result. (Correct me if I'm wrong)

My old Microtek 2500f scanned without any glass, and had autofocus. Nevertheless, it paled in comparison to my old Minolta Dimage Multi Pro. If I were using Medium Format on a regular basis, I would go out and get a Nikon scanner immediately - because any enlargement greater than 3 or 4x, would demonstrate a distinct difference in quality.

Here is a small section of a 6x7 color slide:


http://www.kenleegallery.com/images/tech/Microtek2500ppiSilverfast.jpg
Microtek Scan: 2500 spi


http://www.kenleegallery.com/images/tech/Minolta2400ppi.jpg
Minolta Scan: 2400 spi

[Jim Michael]

I thought Keith Walklet did a great job of demonstrating its capability in this thread (http://www.largeformatphotography.info/forum/showthread.php?t=46503&highlight=mylar).


Please yourself. That was probably harsh and overly personal.

What I mean is, and I have said this too many times, I have never seen a comparison between the V700/V750 and a drum scanner where the flatbed was in focus.

[sanking]

There's also something I call clarity for lack of a better word. It's one of the things I find when I scan the same piece of film on different scanners. When you compare prints from the different scans, sometimes you can see that one print has more clarity than the others. It's like the others have a thin veil between your eyes and the print. I suspect it's a local contrast effect, but I don't really know. A fluid mount tends to increase clarity, but it won't cure a veiled scanner completely in my experience.


I think the veil you describe is due to the fact that the lenses in consumer type flatbeds cover a wide area of film and are used at apertures that are diffraction limited, or close to it it. What this means is that in practice there is an area of 2-3mm above and below the plane of best focus that looks reasonably sharp, and that even when you locate the plane of best focus its clarity is still reduced by the results of diffraction and having to cover such a large area of film. Recovery sharpening regains some of the clarity but introduces sharpening artifcacts that degrade smoothness.

Sandy King

[Bruce Watson]

I think the veil you describe is due to the fact that the lenses in consumer type flatbeds cover a wide area of film and are used at apertures that are diffraction limited, or close to it it. What this means is that in practice there is an area of 2-3mm above and below the plane of best focus that looks reasonably sharp, and that even when you locate the plane of best focus its clarity is still reduced by the results of diffraction and having to cover such a large area of film. Recovery sharpening regains some of the clarity but introduces sharpening artifcacts that degrade smoothness.

Sandy King

That might be some of it. Some of it might come from the different ways scanners handle light -- and people have been arguing the best way to do that since the old diffusion vs. condenser (and more recently, vs. cold light) enlarger arguments.

And some of it may come from the electronics and firmware/software that handles the A/D.

There are many potential sources. I don't know how to adequately describe the effects, let alone propose any solutions.

[rdenney]

All this discussion has been enlightening, and my own experience comparing sensel-level results from my V750 with results from my Nikon 8000ED show a similar difference to what Ken Lee showed above.

But we have to keep things in perspective, too. One person described that only 8x10 negatives scanned with a drum scanner met his quality standards. That, to me, does not result from a process intended to provide good or even excellent image quality, but rather from a process intended to obtain the best possible image quality at any price. Of course, that motivation drives people who chase larger and larger formats.

And the use of the term "presence" is also enlightening. It exactly describes the difference between small and large formats even with low-power enlargements that should be well in hand for either format. I'm not sure it's micro-contrast as much as the smoothness of the tonal gradient at small as well as large scales within the print.

The print that I used as an example (a 20x40" print from a 6x12 negative) does not have all of that smoothness. But it does have the ability to draw the viewer into the image, and having done so, does not disappoint the viewer with unreadable mush at close range. So, yes, it is possible to do better, but in this case it met the photographer's expectations, and it met mine, too.

The improvement from 35mm to 4x5 is vast, and much greater than the improvement between 4x5 and 8x10. Of course both improvements are absolutely not subtle, but the first overcomes so many limitations and represents such a leap in quality and "presence" that people need to get used to it a bit before even contemplating the next step. Jumping from 35mm to 8x10 in one step will likely result in sensory overload, and seeking the appearance of quality (by buying a high-end capability) rather than finding quality in the images themselves. As the old amateur telescope maker's aphorism goes, if you want to grind a 12" mirror, it's faster in the long run to grind a 6" mirror first.

4x5 to Epson V700 or 750 to Epson printer is such a quality leap from 35mm or small-format digital (no matter how it's printed) that it represents a good place for a new large-format photographer to pause a while and catch up with the required technique. That will create a foundation for further improvements, so that the results drive the process and not the process driving itself. And the cost is commensurate with that level of development.

For 6x9 and smaller, though, it's not much more costly to use a Nikon film scanner (used 8000's are still completely usable now and sell for little more than a new 750). That film scanner is a better match for the medium and small format images. And the price is still within a range that is reasonable for someone starting out on the adventure.

Those of you who are insisting on the difference between a flatbed and a drum scan must know that if the image fails because of the flatbed scan it will also fail with a drum scan. That extra bit that the drum scan adds is not subtle, but it's also not what defines a compelling image. It may do no more than further expose the limitations in one's technique. Starting small and working up to one's own level of technique and desired result seems ultimately the cheapest path to that result.

Rick "quoting another aphorism: the great can be the enemy of the good" Denney