Starting my focus calibration of V850 using resolution target

[Pere Casals]

The scanner claims resolutions of 4000+dpi

Bernard, EPSON specifies that V850 has 6400 dpi, in fact this is spi, samples per inch, but true optical effective dpi are 2200 to 2800 depending on the axis, 4000 is fairly out of reach for that device.

[faberryman]

Bernard, EPSON specifies that V850 has 6400 dpi, in fact this is spi, samples per inch, but true optical effective dpi are 2200 to 2800 depending on the axis, 4000 is fairly out of reach for that device.

But you still have to scan at maximum specified resolution to achieve maximum actual resolution.

[Jim Andrada]

I always thought you should scan at maximun OPTICAL resolution to get - well, maximum optical resolution. Of course, according to my wife, I'm wrong about almost everything else, so maybe this is no exception...

[Steven Ruttenberg]

Kudos to the OP for doing actual tests, but scratching my head. The scanner claims resolutions of 4000+dpi and you test it with a target that maxes at 500lpi. What am I missing?
And, your focus increments, as in 2700, 3000, 3300dpi, etc... are not hardware settings; they represent the result of some software data massaging; so, even though you perform no explicit sharpening, the perceived quality is more a judgement on the quality of the hidden re-sampling software.

As stated I did not adjust focus for this test. I simply used the focus set I had previously determined. Then ran thru the different dpi settings to see how resolution was affected at this focus setting. Next I will determjne the best focus by making adjustments to the betterscanning holder.

Nothing missing, etc. And 500lpi was just what this standard was. I got it for adjusting focus. This is just a side test for giggles.

[Steven Ruttenberg]

Well,
Let's not get crazy here. I am not trying to benchmark the scanner, but it does appear that 4000 to 6300 ( only went this high. An extra100dpi won't matter) is the sweet spot with quality improving at higher dpi. This is to test and set focus at desired dpi.

I may some day get a real fancy standard and see how it does.

One interesting fact is I can plot dpi vs file storage which is useful as one can readily determine the file size for a given dpi. I can curve fit an equation that can be used with a transfer function to account for a 4x5 negative over this. Or since we know pixel count of image I can plot Tha vs file size which may be more useful without need for transfer function.

[Steven Ruttenberg]

I think I will get a fancy hi rest target. The only issue is they are all on glass and I need it to be on film at 4x5 size. If contact printed there is the question of how well the glass target resolves to the film. Ie, no loss in accuracy of targeting reproducing line pairs. Otherwise I have to devise an adapter to hold target at correct height plus since I wet mount which is my preferred method that would be a difference in using glass target only and coils skew my results for my chosen method of scanning.

[Pere Casals]

I always thought you should scan at maximun OPTICAL resolution to get - well, maximum optical resolution.

Jim, to record all image quality a lens can deliver we need a sensor that fairly outresolves the lens performace.



That table shows that performace increases when scanning at higher dpi than the maximun OPTICAL resolution, to obtain best result we have to oversample.

Of course from a certain point we obtain a larger file with little or no benefit.

[Steven Ruttenberg]

What is the article that yable comes from? I would like to understand tgeir data collection methods, what targeys they use, wet or dry scan, how best focus achieved, type of holder, etc. I also want to understand the data reduction methods used to arrive at reaults. This includes all assumptions.

As a side note I find it hard to believe that a manufacturer like Epson and others would be that misleading. It may be true that it can't resolve beyond a certain number of line pairs, but effective pixel is misnomer. You can have high pixel count and crapy resolution whereas fewer oixels nd extremely high resolution.

Suffice it to say I will do my own tests. I also feel showing a verse without the chapter and book can be extremely misleading. I am not arguing for or agaonst any type of scanner or brand.

Anyway more to come omce I get focus set. Next is 31 step wedge with 0.1 step increments.

[Pere Casals]

What is the article that yable comes from?

https://archivehistory.jeksite.org/c.../appendixc.htm

https://web.archive.org/web/20180407.../appendixc.htm

Next is 31 step wedge with 0.1 step increments.

To find the V850 limits (with multi-exposure active) you may need a 3.5D target

[mitrajoon]

Jim, to record all image quality a lens can deliver we need a sensor that fairly outresolves the lens performace.



That table shows that performace increases when scanning at higher dpi than the maximun OPTICAL resolution, to obtain best result we have to oversample.

Of course from a certain point we obtain a larger file with little or no benefit.

Pere,

Sorry if this is a dumb question but: Does the Table imply that setting DPI at approximately 3200 is the most "efficient" way to maximize the resolution of one's negative for an Epson 750. i.e., most resolution for file size?