Epson v750 - optimum resolution

[twise]

I recently acquired an Epson v750 scanner and am trying to find the optimum height of the film holders.

To that end, I have scanned the same image (4 x 5 bw negative) @ each of the 3 levels of the Epson film holder.

Each scan was identical in process, in that I used Epson Scan, set the output levels @ 0 - 255, set the resolution at 2400 dpi and used no sharpening.

I printed each image (an 8 x 10 detail from the scan) in Lightroom on Red River Ultra-pro Satin using Eric Chan's ABW profile for this paper. The print was done on an Epson 3880 @ 360 dpi with no enlargement needed.

I was anticipating differences in the apparent sharpness of the prints, but even using a magnifying glass, I cannot perceive any distinction between the 3 prints.

I am hoping to get input on the following:
1. Should I see obvious differences in the apparent resolution of the 3 prints?
2. If the answer to #1 is 'yes' then what might I be missing in my workflow that prevents me from observing differences in apparent resolution?

Thanks for any info you can provide

[Harley Goldman]

I would look at the files at 50% or 100% rather the prints, scanning at 2400dpi. The side by side should be obvious at that magnification.

[Joanna Carter]

You might like to also consider that, depending on the accuracy of manufacture of the scanner (not always that reliable), the sharpest focus may be above or below the range of the Epson holder; which is why some of us use the Better Scanning (http://www.betterscanning.com/) mounting stations, which are fully adjustable.

[John Rodriguez]

When I tested my 4870 I viewed at 100% on the screen. It would be pretty tough to see the differences at small print sizes. While I didn't print my tests, I'm guessing I wouldn't be able to see the difference in the prints until around 6x or more. That's a moot point for me, as the largest I print with the Epson scans is 16x20, any larger and I get a drum scan.

Mine turned out to be sharpest with no shims, although it may be sharper if I were to go to a negative setting.

[Tony Evans]

+ Joanna. As an example, my optimum height with the Epson holder (using washers) was 3.75mm. Subsequently confirmed with the BSH.

[twise]

Thanks for everyone's reply to my query.

At the risk of belaboring the point, I have attached 4 files that are details of the three scans I made, changing the height of the film holder for each, and a thumbnail of the whole image for context. These are reductions of original files that are 100% views with no interpolation. I still cannot see significant differences between the 3 close ups.

I followed Harley Goldman's advice & viewed the images @ 100%, but could only discern subtle differences.

What I do wonder about at this point is if, as Joanna suggested, the area of sharpest focus might be above or below the range of the Epson film holder

I followed the same process for each scan, thinking that any differences seen would be a result of changing the height of the film holder.

As an aside, the initial photo was a response to an assignment given in an LF class in the local community college where we were supposed to illustrate the theme of 'Greed'.

Thanks for your input
twise

[Nathan Potter]

Optimum height is typically between 3 and 4 mm above the surface of the platen. I made a plot of the resolution vs the image height using a high resolution wedge mask a while back. It is posted here somewhere. Well let me grab it and I'll repost. I gives you an idea of the depth of field that you can expect. Perhaps you did not vary the height substantially enough to yield definitive results.

Nate Potter, Austin TX.

http://farm5.static.flickr.com/4147/4989733373_1d5cc658b1_z.jpg (http://www.flickr.com/photos/argiolus/4989733373/)
V750RESjpg (http://www.flickr.com/photos/argiolus/4989733373/) by hypolimnas (http://www.flickr.com/people/argiolus/), on Flickr

[Brian Ellis]

A computer monitor is a poor tool on which to judge critical sharpness so I think you were right to make prints. But FWIW, when I open your three test files in Photoshop and view them side by side at 450x450 it looks to me like your LoPass is somewhat sharper than the other two.

[Richard M. Coda]

Excuse me for being ignorant... but how do you change the "levels" of the stock Epson holders? Thx.

[Jon Shiu]

Excuse me for being ignorant... but how do you change the "levels" of the stock Epson holders? Thx.
There are feet on the Epson holder and you can adjust to 0, +, or take them off. It is in the pdf manual, buried under what to do if you get blurry scans.

Jon

[Richard M. Coda]

There are feet on the Epson holder and you can adjust to 0, +, or take them off. It is in the pdf manual, buried under what to do if you get blurry scans.

Jon

Damn! Now I'll have to play around with them...

[twise]

Nathan,
Regarding your resolution v distance from platen graph. If I am reading this correctly, the highest resolution is where the film is 3 mm above the platen? The resolution in micrometers means that the smaller the number, the more detail can be resolved?
Thanks
twise

[Nathan Potter]

Nathan,
Regarding your resolution v distance from platen graph. If I am reading this correctly, the highest resolution is where the film is 3 mm above the platen? The resolution in micrometers means that the smaller the number, the more detail can be resolved?
Thanks
twise

Yes you are reading correctly. But the sweet spot for maximum resolution will vary some from one machine to the next. Also note that the high maximum resolution number I measure on that plot was made by adjusting the contrast to the highest possible in Photoshop then examining the line pairs at 1600X. If one preserves the full tonal range while scanning and in PS the resolution will be much lower because, in effect, you will be exchanging a high contrast image for a wide density range image.

If you are a bit non-technical then consider: 1 micron (1um) is 1 millionth of a meter.
There are about 25 um in 1 thousandth of an inch (.001 inch). Most high quality imaging lenses can resolve around 5 to 10 um feature sizes using white light. Using monochromatic light close to 1 um can be resolved.

Nate Potter, Austin TX.

[SeanEsopenko]

I use the epson holders for 4x5 because I'm not printing anything with crazy enlargement when shooting 4x5. I do have betterscanning.com holders for the 6x7 I shoot.

I found with the aftermarket holder a difference of a quarter turn (0.25mm) makes a difference in sharpness. I also scan at 6400 dpi (even 4x5, yes) and down-sample to somewhere between 2000 dpi and 2400 dpi. I see a marked improvement in detail when I do that.

I don't have a large enough inkjet (R3000) but I've printed 8x10 crops of a simulated 20x24 print and for what it's worth the inkjet is the weak link when printing a scan of a 4x5 transparency and observing the sharpness. I can't print optically any larger than 16x20 and honestly I'd never be able to sell anything larger than that for a long while so I don't have much to say regarding large prints.

But I can confidently say that the betterscanning adjustable holder improved my scanning immensely for 6x7 (which is subjected to much more magnification than 4x5 work I do).