Scanning Technique for real Resolutions

Love Shack

Love Shack

  • 0
  • 0
  • 17
Matthew

A
Matthew

  • 4
  • 3
  • 809
Sonatas XII-54 (Life)

A
Sonatas XII-54 (Life)

  • 3
  • 3
  • 881
Zakynthos Town

H
Zakynthos Town

  • 1
  • 1
  • 2K
Driftwood

A
Driftwood

  • 13
  • 2
  • 2K

Forum statistics

Threads
199,796
Messages
2,796,764
Members
100,038
Latest member
Pabla786
Recent bookmarks
0

jd callow

Moderator
Moderator
Joined
Jan 31, 2003
Messages
8,466
Location
Milan
Format
Multi Format
Sandy King and others have stated that the real resolution of some of the consumer grade flatbed is considerably less than the stated optical. In my experience I noticed, what I thought were focusing issues with the epsons, which tended to limit their resolution (800 being no better than 1600 on the epson perfection 1600, although this may have to do with pther factors outside of the fixed focus nature of epson's design ), but I was never able to test real resolution as I don't have a target.

My questions are:
How do you ascertain the resolution without a target?
What are the common settings for max res? If the epson V700 maxes out at 2400ppi do you scan at 2400ppi or is that resolution only achieved when scanning at 6400ppi or some other res?
 

keithwms

Member
Joined
Oct 14, 2006
Messages
6,220
Location
Charlottesvi
Format
Multi Format
JD, I look forward to Sandy's repsonse, and in the meantime I'll just state that I think you do indeed need to 'overscan' significantly to get the highest res from a flatbed like a 4990 or 700. Even though the stated max res is not anywhere near what you actually get.

At some point I scanned the heck out of a piece of low speed film at several different resolutions and determined that there was decreasing improvement at progressively higher dpi, but that this improvement was there, and it was really only seen after downsizing the scan.

In other words, I allege that if you flatbed scan to the max and then downsize it to something more reasonable like 2400, you will obtain a better, cleaner result than by scanning straight to 2400. Obviously that is a pain because it involves long scan times and enormous files and then the downsizing.

I haven't thought about why that is so; perhaps it's merely an issue of averaging out the detector noise that is introduced by the scanner itself. Flatbeds use CCDs or similar, I assume, whereas drummers use PMTs.

This is not to say that 4800 or 9600 are true resolutions that you can obtain with a flatbed- they're apparently not anywhere close. Nevertheless, scanning at these resolutions and then downsizing should give some benefit- it's up to you to decide if it's worth it.

As far as a resolution target, I suppose that I could microprint something to allow us to do a flatbed res check. But surely some test samples exist already.
 
Last edited by a moderator:

Doug Fisher

Member
Joined
Oct 19, 2006
Messages
126
A negative that is good and sharp with lots of detail in the image is enough to help you calibrate. If you are wanting to be able to state lpm numbers, you would need to get one of the high quality commercial test slides ($$) and scan that.

Since you mentioned the V7xx and 6400 ppi, a couple of other things that often need to be mentioned/emphasized are:

1. You switch between the lenses by changing the "film with film holder" or "film with film area guide" software setting.

2. You have to use a film holder if you are going to use "film with film holder" or the film will not be at the right height for best focus. If you want to see a classic example of someone not getting this right, look at Aztek's examples for their V7xx fluid mounting supplies kit.

3. The "high resolution lens" can scan at other resolutions other than 6400 ppi.

Between Epson's incorrectly written marketing materials and "reviewers" who never really tested/understood these scanners (and thus perpetuated the misinformation), there have been quite a few people who never realized that they were not correctly setting up or getting the most out of their scanners.

Scanning at more than 2400 ppi can have its advantage. One is that you can downsample and thus see a reduction in noise. Another thing is that many people are taking XX ppi as gospel based on what they have heard other people claim. If they did a valid test, I think they might find the V7xx scanners can do better than many of the XX ppi numbers that are often stated on the forums. I am not saying these scanners are Nikon-killers but a lot of people are producing very nice work with them.

Doug
 

sanking

Member
Joined
Mar 26, 2003
Messages
5,437
Location
Greenville,
Format
Large Format
My questions are:
How do you ascertain the resolution without a target?
What are the common settings for max res? If the epson V700 maxes out at 2400ppi do you scan at 2400ppi or is that resolution only achieved when scanning at 6400ppi or some other res?

You can compare sharpness of two scanners by scanning with each a sharp negative of slide and comparing results.

I don't know of any way to ascertain true resolution in spi or lines/mm without a target. And even with a target you must first determine the point of best focus with scanners that do not auto-focus.

I have never tested the Epson V750, but I have tested the 4870, 4990 and V700. My tests indicate that the maximum effective resolution of the 4870 and 4990, even scanning at 4880 spi, is about 2000 spi. I know some people who have reported up to 2400 spi, but I am skeptical. The V700 tested at about 2200 spi in my case.

Noise can be reduced if one scans at double the needed resolution and then down sizes. But file size may make that impractical, depending on the speed and RAM of the computer.

Sandy King
 
Photrio.com contains affiliate links to products. We may receive a commission for purchases made through these links.
To read our full affiliate disclosure statement please click Here.

PHOTRIO PARTNERS EQUALLY FUNDING OUR COMMUNITY:



Ilford ADOX Freestyle Photographic Stearman Press Weldon Color Lab Blue Moon Camera & Machine
Top Bottom