Announcement

Collapse
No announcement yet.

Resolution issues in cinemascope

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

  • Resolution issues in cinemascope

    On http://vimeo.com/79671820 I’ve found a tutorial about film look (shooting, grading,…). After ~4min it’s about the 1:2.35 Aspect Ratio. The rather non classic way to get this cinemascope view resulted in a significant loss of vertical resolution when applied to my interlaced HD video’s. I would appreciate if some forum members are willing to check this issue. I used my test movie on https://www.dropbox.com/s/wfssckmhce...ter%2050i.m2ts
    but any interlaced 1920x1080 video clearly shows the problem I think.

  • #2
    Originally posted by dree View Post
    On http://vimeo.com/79671820 I’ve found a tutorial about film look (shooting, grading,…). After ~4min it’s about the 1:2.35 Aspect Ratio. The rather non classic way to get this cinemascope view resulted in a significant loss of vertical resolution when applied to my interlaced HD video’s. I would appreciate if some forum members are willing to check this issue. I used my test movie on https://www.dropbox.com/s/wfssckmhce...ter%2050i.m2ts
    but any interlaced 1920x1080 video clearly shows the problem I think.
    Hi Dree.

    The vast majority of content that would use that SAR are sourced as progressive, as this is more typical of feature film production. While it does not stop the use of interlaced, from at least an aesthetic point of view, it is not really designed for it. You shouldn't really see any difference in resolution within the picture area, as all you are doing is effectively blanking the top and bottom, and keeping the actual content area 1:1 with the source pixels. Any change in resolution would mean that you are changing the PAR or resizing.

    For any target SAR, regardless of the source, you may want to try this. Create a project that is 1920xXXX, the XXX being the vertical pixel count for the desired output SAR. When you apply your source 1920x1080 clips to the timeline, just make sure they are set to original size. This will fit the width of your source to the width of your new SAR output timeline. You now have space top and bottom in your source clips that you can move vertically to get your framing correct.

    When you output this, you will now have a master that is in the exact correct ratio for your chosen SAR and with no loss of resolution compared to your source resolution. You can then feed this to many encoders that will make a 1:1 encode without black bars top and bottom. If you want a master compatible for legal Blu-ray compatibility, as most encodes won't add the necessary black bars to refill the necessary 1920x1080 requirement for Blu-ray. Create a project that is 1920x1080 and place your new SAR master so that it fits original width, and then run off a new 1920x1080 master.

    "There's only one thing more powerful than knowledge. The free sharing of it"


    If you don't know the difference between Azimuth and Asimov, then either your tapes sound bad and your Robot is very dangerous. Kill all humans...... Or your tape deck won't harm a human, and your Robot's tracking and stereo imagining is spot on.

    Is your Robot three laws safe?

    Comment


    • #3
      Hi Dree.

      Here is a link to an Edius 7 example project. It may help you better visualise the process I have described. If you don't use Edius 7, don't worry, there are media files inside the project that you can import into an earlier version or play through an appropriate media player.

      Here are some descriptions of the files and the process, the project contains sequences that reflect the descriptions, as does the media.

      The master project is 1920x1080P 25FPS, for the original SAR. I have used 1920x800 for the new SAR, this is 2.4:1

      1920x1080.m4v

      This is the output from the original media file.

      1920x800.m4v

      This is the output from a project that was set to 1920x800. The original 1920x1080 media file was placed in this project as per my original description, and exported to 1920x800.

      1920X800 BACK TO 1920X1080.m4v

      This is the 1920x800 media placed back inside the original 1920x1080 project and then outputted as 1920x1080. This then adds black bars as per my original description.

      Difference Between The Two In Original Frame, Showing No Change In Resolution Or Internal Ratio.m4v

      This is both the original 1920x1080 media and the new SAR 1920x800, shown one overlaying the other in a split screen. As I did not make any vertical adjustments to the new SAR media. You can see that the actual visible pictures between the two are exactly in the same place. This shows that the process does not alter the PAR or size of the active video area between the two media clips. There is also a sequence in the project that shows this live across two video tracks.

      It has to be said that between Edius and its layouter tool, this process is a lot easier than on any other NLE that I have used.

      Hope this helps.



      "There's only one thing more powerful than knowledge. The free sharing of it"


      If you don't know the difference between Azimuth and Asimov, then either your tapes sound bad and your Robot is very dangerous. Kill all humans...... Or your tape deck won't harm a human, and your Robot's tracking and stereo imagining is spot on.

      Is your Robot three laws safe?

      Comment


      • #4
        Hi,
        Thanks for your detailed information. I only have time today to view your splitscreen proofing, and it looks perfect. My problem however is to let other people use/test the aproach presented in the tutorial because some friends use it and are not fully convinced it is not correct. So I am trying to get other people's results on this..

        Comment


        • #5
          Be careful on what you use to do this. Edius really does make the whole process very easy. Edius will also let you change the project frame size live. This can be really usefull to test things quickly from one SAR to another, although you should stay in one project size for the main master output SAR. You will find that the layouter is one of the best standard editing tools in any NLE. If you want, you can also use it to keyframe tilt movements. This will allow you to track a persons vertical movement in the new SAR, so that you can chose the compromise on how much of their head to cut off :)

          The whole process is basically a Pan and Scan edit, but without the pan.

          "There's only one thing more powerful than knowledge. The free sharing of it"


          If you don't know the difference between Azimuth and Asimov, then either your tapes sound bad and your Robot is very dangerous. Kill all humans...... Or your tape deck won't harm a human, and your Robot's tracking and stereo imagining is spot on.

          Is your Robot three laws safe?

          Comment


          • #6
            Thanks again. I am a bit confused on your SAR definition. I feel as it is used as screen aspect ratio, while I thought that in the PAR/DAR/SAR story SAR is used for storage aspect ratio. Of course SAR and DAR are the same with square pixels.

            Comment


            • #7
              Originally posted by dree View Post
              Thanks again. I am a bit confused on your SAR definition. I feel as it is used as screen aspect ratio, while I thought that in the PAR/DAR/SAR story SAR is used for storage aspect ratio. Of course SAR and DAR are the same with square pixels.
              For what you are wanting to achieve, and to avoid any geometry distortion in between the start picture and end picture. You only need to concentrate on Screen Aspect Ratio, picture resolution and Pixel Aspect Ratio. Just remember that both your input and output are square pixels, going by your use of standard HD as your starting point.

              My example takes all this into consideration, so hopefully visualising my description with either the full Edius project or the standalone media, will help.

              You can also do this test with another NLE using my media clips. Although as I said earlier, Edius in my experience has the best tool set to do this. If you are testing with other people on different platforms, just make sure that they understand the base concept of what you are trying to achieve. Although most people using Edius will probably better understand the process, due to the fact that Edius is super intuitive and is excellent for film type work.

              Post back if you have any issues with the example project or media clips.

              "There's only one thing more powerful than knowledge. The free sharing of it"


              If you don't know the difference between Azimuth and Asimov, then either your tapes sound bad and your Robot is very dangerous. Kill all humans...... Or your tape deck won't harm a human, and your Robot's tracking and stereo imagining is spot on.

              Is your Robot three laws safe?

              Comment

              Working...
              X