4 Replies Latest reply on Nov 13, 2011 9:07 AM by the_wine_snob

    Minimum time for a frame ( milliseconds )


      Greetings. My middle-school daughter is doing a science lab that requires her to create a subliminal message in a video. Using the trial version of Premiere Elements 20 (and having consulted help files, forums, even Lynda.com), we are having difficulty reducing the length of the frame containing the subliminal message to a point that it's impercerptible to the naked eye, yet perceptible to the minds eye (apparently somewhere around 12 miliseconds). We've tried the frame as a jpg, and also rendering it as an MPEG video. Any suggestions? Thank you!

        • 1. Re: Minimum time for a frame ( milliseconds )
          the_wine_snob Level 9

          In Video, the smallest increment is one Frame, which translates to 1/30th sec. for NTSC, or 1/25th sec. for PAL. You cannot have a partial Frame.


          Good luck,



          1 person found this helpful
          • 2. Re: Minimum time for a frame ( milliseconds )
            the_wine_snob Level 9

            One thought that I had, for a subliminal message would be to go with that one Frame Duration, and then alter the Opacity down a bit. Some experimentation might be needed, but that would diminish the audience's awareness of it, but still should "imprint" on their brains.


            Here is a look at a one Frame message, done at an Transparency (PS's term for Opacity, and in terms of settings, it's just the opposite - you could adjust the Opacity to ~ 50% in PrE too), and at 30 FPS, goes by too quickly to read, though there is a tiny "flicker".



            Good luck,



            • 3. Re: Minimum time for a frame ( milliseconds )
              Ted Smith Level 3

              The text should be white letters on black.

              The video level of the text must be quite low.

              It is the slow repetition over a long period that makes it effective.

              Just a few virtually invisible messages will have no effect as the brain.

              If it normally take a second to read then it will take at least 30 frames for the brain to just take it in even if the eyes and the subconscious mind sees every frame.

              Eg 30 frames of one every 10 seconds equals 300 seconds to just get the message to the subconscious.


              Subliminal messages depend hugely on the type of display you will be seeing it on.

              Most LCD TV screens have a lag so that one 1/30 second 'frame' will decay somewhat over at least a few frames because of the lag even though you are hitting it with one frame. It is the lag that you are seeing.


              It was only reliable with the old black and white CRT TV screens that had a very fast decay or when projected by a film movie projector that will definitely show only one fame - then nothing else.


              I worked in TV in the 1960's and carried out quite a few illegal experiments on this subject and found one frame of 10% video level of white letters on a black background was invisible yet gave measurable audience results. I built a keyed image switch that allowed a video signal to get through from one camera for one frame every 10 seconds (out of valves!)


              On one occasion we had a TV Telethon to raise money for Spastics and donations had slowed to a trickle. I put up a message flashing every 10 seconds that said "You can give more!" and within fifteen minutes the switchboard jammed with more donations!


              By adjusting the video level of the text until a person "not in the know" can just see it then reducing it a bit, it you might get a result as long as you show the final result in the same monitor brand and model of screen. A laptop is usually the worst for lag

              1 person found this helpful
              • 4. Re: Minimum time for a frame ( milliseconds )
                the_wine_snob Level 9



                I had not even considered the latency of the display, with newer monitors/TV's. Good to know.


                I would also think that Progressive vs Internaced display, might play a role in the impact too, but that is speculation on my part.


                With film projection, the shutter actually blanks out the image, while the claw pulls the next frame into place. Normally, the human eye never catches that blanking out of the image, due to "presistance of vision" in humans (though some scientists will state that it is not really "presistance of vision," but something much more complex at a neurological level - that is for them to debate, and I'll just stick with "presistance of vision.) If one rigs up a second shutter (this can even be by using ones fingers flicked before the eyes), with a bit of experimentation on the rate, one can then see the blank image on the screen. It's sort of like creating that "finger" shutter, while looking at a ceiling fan. Get the rate just right, and the fan blade appears to stop.


                For "presistance of vision," I encountered it very often, when shooting in a totally dark studio (usually to burn in a CRT, or LED display on an image of computer gear for an ad). When we'd fire the stobes, the entire image of the set (if one was looking at it, when the strobes were fired) remains, and for a good, long time. Turning one's head had zero effect, as that image presisted. It was easy to walk into an object, as the only "view" that one had, was of where they were looking, when the strobes fired - no matter where the head turned, that image was still there, until it had faded completely.


                OK, getting way OT here, so I will wrap up by saying, "thanks for the mention of the latency, and display types. Never considered that."



                1 person found this helpful