Skip to main content
ednortonuk
Participating Frequently
June 5, 2018
Question

Exif history

  • June 5, 2018
  • 3 replies
  • 3804 views

Hey all,

I've been dealing with a lab getting my 120mm MF negs scanned. I just want to know if by looking at the exif raw data (pasted below), can you see if this was a 72dpi scan that has been resampled to 300dpi and saved as a jpg or if it was 300dpi scan that has had the pixel dimensions changed only.

My thinking is that rather than scanning at high res, they're scanning low res and resampling it in photoshop.

Any help would be appreciated.

Here's the data.

<x:xmpmeta xmlns:x="adobe:ns:meta/" x:xmptk="Adobe XMP Core 5.6-c140 79.160451, 2017/05/06-01:08:21        ">

   <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">

      <rdf:Description rdf:about=""

            xmlns:xmp="http://ns.adobe.com/xap/1.0/"

            xmlns:photoshop="http://ns.adobe.com/photoshop/1.0/"

            xmlns:dc="http://purl.org/dc/elements/1.1/"

            xmlns:xmpMM="http://ns.adobe.com/xap/1.0/mm/"

            xmlns:stEvt="http://ns.adobe.com/xap/1.0/sType/ResourceEvent#"

            xmlns:tiff="http://ns.adobe.com/tiff/1.0/"

            xmlns:exif="http://ns.adobe.com/exif/1.0/">

         <xmp:CreatorTool>Adobe Photoshop CS5 Windows</xmp:CreatorTool>

         <xmp:CreateDate>2018-06-05T17:59:59+01:00</xmp:CreateDate>

         <xmp:ModifyDate>2018-06-05T18:24:03+01:00</xmp:ModifyDate>

         <xmp:MetadataDate>2018-06-05T18:24:03+01:00</xmp:MetadataDate>

         <photoshop:LegacyIPTCDigest>CDCFFA7DA8C7BE09057076AEAF05C34E</photoshop:LegacyIPTCDigest>

         <photoshop:ColorMode>3</photoshop:ColorMode>

         <photoshop:ICCProfile>sRGB IEC61966-2.1</photoshop:ICCProfile>

         <dc:format>image/jpeg</dc:format>

         <xmpMM:InstanceID>xmp.iid:D8D91F3EE568E811852FBF74E8439D24</xmpMM:InstanceID>

         <xmpMM:DocumentID>xmp.did:D8D91F3EE568E811852FBF74E8439D24</xmpMM:DocumentID>

         <xmpMM:OriginalDocumentID>xmp.did:D8D91F3EE568E811852FBF74E8439D24</xmpMM:OriginalDocumentID>

         <xmpMM:History>

            <rdf:Seq>

               <rdf:li rdf:parseType="Resource">

                  <stEvt:action>saved</stEvt:action>

                  <stEvt:instanceID>xmp.iid:D8D91F3EE568E811852FBF74E8439D24</stEvt:instanceID>

                  <stEvt:when>2018-06-05T18:24:03+01:00</stEvt:when>

                  <stEvt:softwareAgent>Adobe Photoshop CS5 Windows</stEvt:softwareAgent>

                  <stEvt:changed>/</stEvt:changed>

               </rdf:li>

            </rdf:Seq>

         </xmpMM:History>

         <tiff:ImageWidth>2079</tiff:ImageWidth>

         <tiff:ImageLength>2048</tiff:ImageLength>

         <tiff:BitsPerSample>

            <rdf:Seq>

               <rdf:li>8</rdf:li>

               <rdf:li>8</rdf:li>

               <rdf:li>8</rdf:li>

            </rdf:Seq>

         </tiff:BitsPerSample>

         <tiff:PhotometricInterpretation>2</tiff:PhotometricInterpretation>

         <tiff:Orientation>1</tiff:Orientation>

         <tiff:SamplesPerPixel>3</tiff:SamplesPerPixel>

         <tiff:YCbCrPositioning>1</tiff:YCbCrPositioning>

         <tiff:XResolution>3000000/10000</tiff:XResolution>

         <tiff:YResolution>3000000/10000</tiff:YResolution>

         <tiff:ResolutionUnit>2</tiff:ResolutionUnit>

         <tiff:Make>NORITSU KOKI</tiff:Make>

         <tiff:Model>QSS-31</tiff:Model>

         <exif:ExifVersion>0220</exif:ExifVersion>

         <exif:FlashpixVersion>0100</exif:FlashpixVersion>

         <exif:ColorSpace>1</exif:ColorSpace>

         <exif:PixelXDimension>3000</exif:PixelXDimension>

         <exif:PixelYDimension>2955</exif:PixelYDimension>

         <exif:FocalPlaneResolutionUnit>3</exif:FocalPlaneResolutionUnit>

      </rdf:Description>

   </rdf:RDF>

</x:xmpmeta>

This topic has been closed for replies.

3 replies

Stephen Marsh
Community Expert
Community Expert
June 6, 2018

Consider the physical size of the originals.

Consider the maximum optical resolution (not interpolated) of the scanner.

This will give an indication of the pixel width/height of the scan.

A 35mm neg scanned at 600ppi should yield a file approx. 850 x 567 pixels.

D Fosse
Community Expert
Community Expert
June 6, 2018

Pixels per inch. Stop and consider what it means. Have pixels, will print. How thinly can I spread those pixels out on paper? Small pixels, densely packed = small print. Large pixels, spread out = large print.

Pixels per inch is an equation. Like any equation, it can be inverted and turned upside down and inside out, so that you can start with any two values and derive the third.

ednortonuk
Participating Frequently
June 6, 2018

Honestly, I've been doing this for 10 years and I've never really got my head around ppi/dpi/lpi/spi fully. It was just expected that everything needed to be given to clients at 300dpi and nobody seemed to be able to explain why. I've watched countless tutorials and can see it comes from the older printing press days. I've seen countless people argue for and against the 300dpi myth.

The thing that confused me is when it comes to outputting to, for example, super sizing prints. Let's say you had three files opened. All are 6000x4000px. 24mp images. The first was 72ppi, second 300ppi, third a 3000x2000 72ppi file changed to 300ppi. You need to print at a bus stop 120x80inch print. Enlarging all 3 to print at 120x80 inches.. does that mean the first two images will appear the same in print, and the third suffers in quality as it has been changed from a 3000x2000 to a 6000x4000 image already. Or, will a 120inch... or a 400inch print look the same if you supply the file in 72 or 300 PIXELS per inch. If it's something that can be changed so freely, I'm confused why I'm always asked for files in the magic 300ppi.

Printers have always specified they need the files at 300 or 240 PIXELS per inch. But as I understood it, output resolution should take into consideration desired print size and viewing distance of the print. Billboards can be printed at 15dpi for example.

So my confusion was in having supersized prints - having higher PPI on images would allow the viewing distance to be reduced...

And going back to the original scanning negatives issue. A 3000px scan at 72 or 300 is pretty low res right? But if it was 300 or higher, I could achieve a bigger print size at least?

If this doesn't make sense, don't worry, my brain seems to have always struggled to understand this.. it hurts right now just trying to write all this out and make sense of it!

D Fosse
Community Expert
Community Expert
June 6, 2018

It's a lot simpler than you think, you just need to clear out what you think you know

If you have a certain number of pixels, and you want to print it at a certain physical size, then the ppi becomes a resulting number. It's not given, it's a result of those two premises. That's it. It's an equation, x=z/y.

And inversely - a given number of pixels, printed at a certain ppi, becomes a resulting physical print size.

Yes, you're absolutely right: a billboard can be printed at 10 or 15 ppi. The same file - the very same! - can be printed in a book at 300ppi.

Mylenium
Legend
June 6, 2018

I just want to know if by looking at the exif raw data (pasted below), can you see if this was a 72dpi scan that has been resampled to 300dpi and saved as a jpg or if it was 300dpi scan that has had the pixel dimensions changed only.

You can't. Metadata is arbitrary. Even if they followed correct procedures, that wouldn't affect this data and if it did, they could manipulate it to hide any tampering. If you have doubts about your image quality, only providing a full res image for visual inspection and possible forensic analysis can offer any clues, but even that wouldn't be a definite thing. Resampling an image may produce specific patterns, but so does an out-of-focus scanner or the scanner's arrangement of its photo-sensitive cell.

Mylenium

ednortonuk
Participating Frequently
June 6, 2018

Thanks for getting back to me. Just to follow up as I might be a little confused here..

So I can see from the data that they used a NORITSU KOKI scanner and then have used Photoshop CS5 on a Windows machine. The current file size I have is 3000x2955px at 300dpi which I can see (at least the dimensions) in the above data. Higher up above that it says :

<tiff:ImageWidth>2079</tiff:ImageWidth>

<tiff:ImageLength>2048</tiff:ImageLength>

Does that not mean it was brought into PS CS5 at 2079x2048 and then enlarged to 3000x2955px?

I only say this because prior to these high res scans, they sent me the low res versions which were at 72dpi and 2079x2048px....

Well actually to be clear and provide a full picture: I was told when processing my 35mm/120mm negs that scanning would be done at 300dpi. In actual fact I received the first scans back at 72dpi but originally at 1039x1024px wide. I'm not too familiar with negative scanners but surely they output jpgs in a more usable way. At the least around 2000px for low res scans..?

I then said I was told in store they'd be at 300dpi but apparently that incurs a further £8 surcharge per roll! So I paid up and went ahead with one roll to check it first as they seemed a bit scatty. 4 days later, I get through the files at 2079x2048px but STILL 72dpi. I told them they're still at 72dpi to which they replied 'Very sorry about that, we will link the correct files to you asap.'

That to me seemed odd because these were seemingly new scans with increased length/width pixel dimensions so they didn't link me to the wrong files. Then a few hours later, new scans come back at 300dpi. That's when I looked at the data and saw the old 72dpi pixel width/length dimensions that now looked to be changed to the larger 3000x2955px. Which to me, looked like they've scanned at 72dpi and resample to 300dpi in photoshop. You can't see at dpi change in the data but it all seems a bit suspect to me...

Does all that add up?

D Fosse
Community Expert
Community Expert
June 6, 2018

ednortonuk  wrote

<tiff:ImageWidth>2079</tiff:ImageWidth>

<tiff:ImageLength>2048</tiff:ImageLength>

Does that not mean it was brought into PS CS5 at 2079x2048 and then enlarged to 3000x2955px?

That's what I would suspect too, just from the way it looks. I have no idea how to interpret metadata, but the numbers don't look good.

I googled Noritsu, and this has to be the HS-1800, which is the model that can take 120 film. The funny thing is that it should be capable of considerably higher resolution than this.

Don't read anything into the ppi figures. It doesn't mean anything in this context, it's completely irrelevant. This is all about pixels. 2000 pixels at 72ppi is exactly the same thing as 2000 pixels at 300ppi, 1000ppi, a million ppi, one ppi. Doesn't matter.

Anyway. What I would recommend here is to not scan, but photograph these negatives. You get much better resolution, much better detail, and a raw file is much easier to control than an RGB file from a scanner. All it takes is a 24-50 megapixel camera with a good flat-field macro lens. For Nikon, the Micro-Nikkor 60mm/2.8 is an absolutely stellar performer for these things.

An old repurposed enlarger does splendidly for repro stand, and then you need a piece of white translucent plastic to use as diffuser over whatever light source you choose.