Make a note of what you think “this would be better.”

  • Output Video:Project” is n:1, not 1:1

    • Using one long video as material, cut it out to make multiple videos, or use part of the previous video in one video.
  • I’m talking about a large project to create multiple videos from multiple long material videos.

    • Put a short prefix on the video material.
      • Sort by name to get it all together.
      • X_, etc. is good.
        • At first, it was prefixed with an explanatory

          • Descriptive prefixes, shaking.
            • Rush Video,” “Assault Video,” etc.
            • It is not possible to give an appropriate short summary at the work stage
        • The “ones I’m working on right now” are capped at a few at most, so the alphabet is sufficient.
        • Alphabets come first in the list.
        • A descriptive name can be freely added after the prefix
      • When you’re done working on it, you can make a bin and put it away there.
        • bin with an appropriate short summary
  • At the time I took the video, I should have written in Scrapbox what video was taken when and with what file name.

    • I can’t tell from the sequentially numbered video file names alone.
    • You have to look back to find out → inefficiency
    • Best to write it down on the day it was taken.
  • You can write down ideas for videos you want to do. - List of things you would like to do to create a video

  • It’s hard to edit if you move right after you finish speaking.

    • Leave about 3 seconds of space before and after.
  • Your eyes are swimming.

    • I’d rather have a campa by the camera at this level.
    • I put the camphlet (PC) near the camera so that my gaze no longer wanders in the void.
      • But obviously the eyes go down and so on.
      • Maybe “look, look back at the camera, look back at the camera, talk.”
        • It’s easier to edit if it’s separated by a silent section, and
    • Because you’re facing the camera, it’s a “metaphor of two people talking”.
      • In the game play style of P in P video shot from an angle, the eyes are not looking at the camera in the first place.
      • This is “the metaphor of listening to someone talking about games on the side.”
    • Phenomenon that the chassis version is not seen.
    • A bad title or thumbnail can make a noticeable difference.
    • Three-week review
    • Six-week review
  • Trying to make one perfect explainer video raises the psychological hurdle too high.

    • Let’s make short instructional videos and release them sequentially.
    • I made one video, then realized after a while that I forgot to say something, so I made two videos and put them together into a playlist.
    • Then I realized I forgot to say it again.
    • Facebook
  • I don’t trust UNDO.

    • It would be better to put the project files on Dropbox or something and have them automatically backed up.
  • Disk space is tight

  • I used to use a lot of shortcuts to cut with Cmd+K, but now I can only cut one track when there are multiple tracks.

    • Started editing two videos on top of each other in a game commentary video and got into trouble.
    • I also had video and audio that would cut out on one side, but I could “link” them together.
    • I’m more likely to cut everything together in the first place.
    • Cmd+Shift+K “Add edit point to all tracks” replaced with Cmd+K
  • Ripple trimming of previous edit point to playback head” Convenient

    • In short, “delete and pack up here.”
    • You can use one key to fill in the blanks that you don’t need in your speech.
  • Dragging the edge with the selection tool “trims” it

    • Intuitive in the direction of contraction, but “what was gone is back” in the direction of extension.
    • Ripple, rolling, rate adjustment
  • live game playing

    • When recording with Win+G in Windows, the microphone should be muted.

      • Game volume is too loud and audio is difficult to hear
      • I wish there was a setting that would record it on a separate channel.
      • Peak normalization -10db due to loud sound on screen recording side
    • About compositing facial images (mixed opinions)

      • If the face image is subsequenced first, audio peaks are not visible.
        • Is there a setting to display it? I can’t find it.
        • I think it should be done with a regular clip.
        • Bundle and replace in sub-sequences after editing is complete.
          • To finish the mask editing in one session
      • Overlaying a face image with a mask results in a heavy preview.
        • Because it would require synthesis.
        • Just finish editing the audio and the ticker.
      • When overlaying a face image with a circle mask, it is difficult to apply a common mask and scale after cutting, so it should be done first
        • In the meantime, you can create and apply a sub-sequence
      • based on the above
        • First, track separation of video and audio
        • Masking the video images together after grouping them into sub-sequences (A)
        • Can you link (a) and audio?
        • Hide (A) by track
        • Horizontal movement to match the screen recording video
          • Find the voice that is unintentionally mouthing something like “Oh,” and adjust it so that you don’t feel out of sync with the screen.
    • Location of the channel registration button 350, 325 with a font size of 100 is OK.

  • Put keywords that you want to be found in searches in the description.

    • No hits when I put it in the tags.
    • Considering that searches are done by voice input, enter “Nishio Games” instead of “NishioGames”.
    • How it’s done these days
      • 1: Write a good description in the description section of the video.
      • 2: Reprint it on Facebook
      • 3: Post it on Twitter and cut it short because it’s over the word count.

This page is auto-translated from /nishio/ć‹•ç”»äœœæˆăźă‚łăƒ„ using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.