First fluid mounted scans...question

[David Luttmann]

It doesn't matter what you are scanning. If the scanners highest native resolution is 1600, going higher is just going to interpolate, like upsizing in Photoshop.

Did you read the whole thread? Tim Parkin only does 4x5 and found the same thing.

The Epsons are rated around the 2000 to 2400ppi mark in terms of real resolution. I'm not sure where this 1600 figure comes from, but enough testing has been posted here and published in View Camera that we know 1600 is below the capabilities of the scanner.

[sanking]

I tested the 1600 dpi theory using 8x10 and MF 6x7 on my scanner myself.

Check out the test of the Epson V700 at http://www.filmscanner.info/en/Films...tberichte.html

Results show *real* resolution of around 2300 spi - 2400 spi, which is consistent with my own testing of this scanner using a resolution target.

Sandy King

[mrladewig]

And back tot he last question from the OP...

Most of us scan with the output set to "Actual Size" and then select the resolution.

I scan on a 4990 and usually scan at 2400ppi. I haven't got a target to scan, but 2400 seems to be the sweet spot on this scanner too. Occasionally if I want to print a bigger size, I'll scan at 3200ppi fluid mounted on my betterscanning holder. It usually doesn't give much if any additional detail, but color is smoother and I prefer to downrez from an oversized scan into a final print size than to upres from a smaller file size.

[John NYC]

Check out the test of the Epson V700 at http://www.filmscanner.info/en/Films...tberichte.html

Results show *real* resolution of around 2300 spi - 2400 spi, which is consistent with my own testing of this scanner using a resolution target.

Sandy King

From this it sounds like they are scanning at 6400 dpi, but seeing a real resolution at 2300 dpi? It doesn't sound like they are saying scanning at 2300 dpi will get you an effective 2300 dpi. Is that your interpretation?

[John NYC]

The Epsons are rated around the 2000 to 2400ppi mark in terms of real resolution. I'm not sure where this 1600 figure comes from, but enough testing has been posted here and published in View Camera that we know 1600 is below the capabilities of the scanner.

I don't know where it came from, but I'd love for someone to repeat what I did and see if they get the same results:

1. Scan a neg of your choice at 1600 and 2400
2. Make a copy of the 1600 one and upsize in photoshop to 2400
3. Compare the upsized one to the native 2400 scan

When I did this, the upsized one was marginally (and I mean marginally) sharper than the native 2400 dpi scan.

[sanking]

From this it sounds like they are scanning at 6400 dpi, but seeing a real resolution at 2300 dpi? It doesn't sound like they are saying scanning at 2300 dpi will get you an effective 2300 dpi. Is that your interpretation?

In my tests the highest real resolution I got was around 2300 dpi, and I got very close to that scanning at about 3200 dpi. You can see some slight increase in real resolution at 4800 dpi and at 6400 dpi but not enough to justify the huge file size for LF film, IMO.

If you scan at 2300 dpi you will get effective resolution of around 2200 dpi, so in practical terms you might want to do your scans in the 2300 dpi to 3200 dpi range which will give you from about 95% to 98% of the potential effective resolution of the V700/750.

Some people have claimed higher resolution up to 3200 dpi, but that is beyond the ability of my V700.

Sandy

[coops]

Most of us scan with the output set to "Actual Size" and then select the resolution.

I scan on a 4990 and usually scan at 2400ppi. I haven't got a target to scan, but 2400 seems to be the sweet spot on this scanner too. Occasionally if I want to print a bigger size, I'll scan at 3200ppi fluid mounted on my betterscanning holder. It usually doesn't give much if any additional detail, but color is smoother and I prefer to downrez from an oversized scan into a final print size than to upres from a smaller file size.

Makes sense. Thanks

[JC Kuba]

I don't know where it came from, but I'd love for someone to repeat what I did and see if they get the same results:

1. Scan a neg of your choice at 1600 and 2400
2. Make a copy of the 1600 one and upsize in photoshop to 2400
3. Compare the upsized one to the native 2400 scan

When I did this, the upsized one was marginally (and I mean marginally) sharper than the native 2400 dpi scan.

When you upsize aren't you doing some sort of software sharpening? If so, and you compare it to an unsharpened 2400 scan, how can you say an unsharpened 1600 scan is better?

[John NYC]

When you upsize aren't you doing some sort of software sharpening? If so, and you compare it to an unsharpened 2400 scan, how can you say an unsharpened 1600 scan is better?

I am not sharpening on purpose, for certain. I am just using the image size adjustment in photoshop to enlarge. Obviously, the software does some sort of interpolation to do that and might be introducing sharpening. Don't know.

Prior to reading that other thread, I had taken all the well-worn advice on this forum to scan at 2400. Then that thread popped up, and I tried the experiment I described. And then did some reading and found Tim Parkin's similar investigations. I was very surprised at the result I got.

I also downsized a 2400 image and compared to an unsharpened 1600, and the differences were negligible, with the 1600 being slightly sharper.

So, whatever the reasons, I can't see any reason to scan (on my particular sample of the V750 Pro scanner) at the higher resolution when I can't see the results in real world images that contain very fine details. Would love to be able to get the higher resolution going and understand why this is case.

[sanking]

[QUOTE=John NYC;617378
So, whatever the reasons, I can't see any reason to scan (on my particular sample of the V750 Pro scanner) at the higher resolution when I can't see the results in real world images that contain very fine details. Would love to be able to get the higher resolution going and understand why this is case.[/QUOTE]

When you did your tests did you, 1) make sure that film holder was selected in order to engage the highest resolutions lens, and 2) did you determine the best plane of focus for that lens? If you were to use the lens that is engaged when film area guideis chosen you would not get better than about 1600 spi - 1800 spi.

Of course, the difference between 1600 spi and 2300 spis is not all that great if you are comparing real negatives. To observde a difference that small might require testing with a high resolution target.

Sandy King