Announcement

Collapse
No announcement yet.

Dateigröße ändert sich nach dem Speichern - aus welchem Grund ?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Dateigröße ändert sich nach dem Speichern - aus welchem Grund ?

    Hallo zusammen,

    ich habe eine Frage bezüglich der Bildqualität bei Verwendung von IrfanView, und zwar konkret an folgendem Fall:

    Ich habe mir ein Bild vom Internet heruntergeladen und als Jpeg abgespeichert. Das Bild hat eine Dateigröße von 5.870 kB. Da ich den korrekten Dateinamen noch nicht festgelegt hatte, habe ich das Bild später nochmals geöffnet und unter einem anderen Namen abgespeichert. Jetzt hat das Bild die Dateigröße 877 kB.

    Mein erster Gedanke war "Prima, 80% Speicherplatz gespart".

    Mein zweiter Gedanke war "Oha, hat sich die Qualität des Bildes jetzt um 80 % verschlechtert ?"

    Ich muss dazu sagen, es ist das erste von 250 Bildern, insofern spielt der Speicherplatz / die Dateigröße eine größere Rolle bei mir.

    Der Grund für die Reduzierung der Dateigröße ist, dass beim Speichern die Einstellung "Qualität 80" und kein Haken bei "Farben-Subsampling deaktivieren (dh. 1x1 nutzen)" gesetzt war.

    Eine Suche bei der Hilfe von IrfanView und bei Google brachte die Ergebnisse, dass sich die Dateigröße reduziert bei Anwendung dieser Einstellung. Leider fand ich nirgends die Angabe, was genau die Ursache der Dateigrößenreduzierung ist. Denn interessanterweise sind die Angaben "Auflösung", Originalgröße", "aktuelle Größe", "Druck-Größe (aus DPI)", "Originalfarben", "aktuelle Farben" und "benötigter RAM-Speicher" unter "Bildinformation" (Reiter "Bild", erste Zeile "Information") exakt dieselben. Nur "Gezählte Farben" und "benötigter Plattenplatz" unterscheiden sich.

    Meine Frage also: Was passiert mit der Datei, wenn eine Speicherung mit "Qualität 80" und ohne Haken bei "Farben-Subsampling deaktivieren (dh. 1x1 nutzen)" durchgeführt wird ?

    Gruß Rolf

    #2
    Originally posted by Google Translate
    I have a question regarding the image quality when using IrfanView, specifically in the following case: I downloaded a picture from the Internet and saved as jpeg. The image has a file size of 5.870 kB. Since I had not set the correct filename yet, I opened the picture again later and stored under another name. Now the picture has the file size 877 kB. My first thought was "Prima, 80% saved space". My second thought was "Oha, has the quality of the picture now deteriorated by 80%?" I have to say, it is the first of 250 pictures, so the space / file size plays a bigger role with me. The reason for the reduction of the file size is that the setting "Quality 80" and no "Disable color subsampling" (ie 1x1 use) was set when saving. A search at the help of IrfanView and at Google brought the results that reduces the file size when using this setting. Unfortunately I found nowhere the indication, which is exactly the cause of the file size reduction. Interestingly, the "Resolution", "Original size", "Current size", "Print size (from DPI)", "Original colors", "Current colors" and "Required RAM memory" "The first line" information ") is exactly the same, but the "number of unique colors" and "required disk space "are different, so my question is: What happens to the file when a storage with "quality 80" and no tick in "disable color subsampling (ie, use 1x1)"?


    Yes. JPG compression is very efficient. 80% is pretty high quality. Obviously some JPG artefacts are added, but only a close examination will spot the difference. As you reduce the quality to below 50% the space savings are not so significant, but the quality begins to deteriorate.

    Disabling colour subsampling also save a lot of space, for very little loss of quality.

    I rarely use anything higher than 75%. Can you spot the difference between these three images? The original PNG image is 18.9 Mbytes.

    No chroma subsampling at 75 = 762K
    Allow chroma subsamling at 75 = 569 K
    Allow chroma subsampling at 50 = 349K

    I can spot the difference between the 50% quality JPG (Left) if I zoom in very close and the PNG (right)
    The colour difference betwen the 75% without Chroma subsampling and the PNG is not obviious to me. The artefacts are just visible.

    Bottomline: any quality much above 75% is a waste of bytes unless you're a photographer selling high quality prints.

    Note: I intentionally chose an area of the picture with text, as areas with high contrast shows the most obvious JPG artefacts. Photos that are just landscapes or seascapes will have even fewer artefacts at high compression settings.
    Attached Files
    Last edited by Bhikkhu Pesala; 19.10.2016, 02:00 PM.
    Before you post ... Edit your profile • IrfanView 4.67 • Windows 10 Home 19045.2486

    Irfan PaintIrfan View HelpIrfanPaint HelpRiot.dllMore SkinsFastStone CaptureUploads

    Comment


      #3
      Hello Bhikkhu Pesala,

      thank you very much for your response.

      You wrote that using higher percentage of Quality and using chroma subsampling will result in saving space.

      But how is this (technically) possible when the resolution, current size and so on stays the same ?

      Where does the programm take something away so that the data amount decreases ?

      Best regards,

      Rolf

      Comment


        #4
        See my notes on Compression

        This site: Digital Image File Types Explained explains more about JPG and Chroma Subsampling
        Last edited by Bhikkhu Pesala; 19.10.2016, 07:47 PM.
        Before you post ... Edit your profile • IrfanView 4.67 • Windows 10 Home 19045.2486

        Irfan PaintIrfan View HelpIrfanPaint HelpRiot.dllMore SkinsFastStone CaptureUploads

        Comment


          #5
          Hello Bhikkhu Pesala,

          thank you once again for your time and your help.

          I checked the links you provided and have the following remarks:

          a) Compression

          The site says that with saving at lower quality the file size will be smaller. It doesn't say why the file size gets smaller (how is the smaller file size achieved ?).

          b) Digital Image File Types Explained

          Now I'm beginning to understand - Saving a picture as a jpeg-file is like zipping a file.

          Meaning if I save my 80%/subsampling-file with 100%/subsampling-off, I get my 5870 kB file back.

          Let me try this ...

          ... well, it's now 2401 kB big, but I think I got the meaning.

          c) Chroma Subsampling

          Here I got an explanation what happens when I use Chroma Subsampling: It "breaks the image into 2x2 pixel blocks and only stores the average color information for each 2x2 pixel group"

          This means I only have a quarter of the colors of the original file.

          That could explain the different numbers in "number of unique colors" I have. 154402 in the original file, 64170 in the reduced file.

          O.K., so Chroma Subsampling reduces the amount of colors, therefore the smaller file size.

          One part orf my question is answered.

          Remains the part with the quality:

          I understand that reducing the quality is like zipping the file - only that you are not able to get the original file back.

          I tried this and compared the information data of the files - still the same numbers except for "number of unique colors" and "required disk space".

          I still don't understand where the programm takes what away when using less than 100% quality.

          What is compressed when I use lesser quality ?

          Best regards,

          Rolf

          Comment

          Working...
          X