Video Asylum

Posted by David Aiken (A) on June 21, 2007 at 15:03:37

In Reply to: RE: "Unlike claimed audio improvements claimed video improvements can be captured by cameras, know what I mean? " posted by Tom Brennan on June 21, 2007 at 06:01:22

I accept that "many video improvements can be captured by a still camera and demonstrated and studied at length and at leisure", or at least that some certainly can, however your initial statement left out the "many" and didn't make any qualifications about the universality of your camera test.

I disagree with "This is difficult, if not impossible, with audio as with audio you cannot capture a moment and study it. Well you can with measurements I suppose but not the sound itself."

You can capture sound with a recording and one could set up a system that allowed you to 'freeze' the tape at a given moment and play that moment for a continuing period. What you would get is, essentially, coloured noise which gives you little idea of how that moment sounds 'in situ' within the flow of sound.

But that may not be too different to capturing a moment of video with a camera. When I pause a DVD, I find that the frozen image with many DVDs appears very grainy while the picture while the disc is playing looks anything but grainy. The still does not capture a particular aspect of the playing video and some things which appear to produce a quite noticeable improvement in sharpness and other aspects of picture quality simply don't show on the still shot. I had this experience when I got a new equipment rack for my HT system and set the components up in it with the centre speaker on a separate solid stand. Previously the centre speaker had shared a wide AV rack with the components. My centre speaker is large and crossed over to the sub at 40 Hz, and the shelves of the old AV rack had been glass. Not only the picture but the sound impmroved with the change.

While the still image did not show any of the things that I regarded as visual improvments, I tried recalibrating my picture settings using Digital Video Essentials. Another impression I had of the changes in picture quality was that the colours in overall image looked brighter. The black levels didn't appear to have changed when I used DVE but I had to reduce the TV's contrast setting by 1 step in order to restore my peak white levels. An interesting confirmation.

I'm wary of claiming that any one test is the be all and end all of testing for difference. Still photos of video don't reveal everything. Any single audio test doesn't reveal everything. I suspect that there's little difference in how testable claims in both areas are. Too many people run a limited range of tests because those are the only tests they have, see no difference in results, and claim that as a proof of no difference. It isn't. It says nothing about the things left untested. We also don't always have the tests we need at a given point in time. Tests for jitter were not available when CD players were introduced and quite a few people ridiculed claims that there were audible differences between players that tested similarly on the tests available. When tests for jitter became available differences were revealed. It's simply unscientific to assume that we have all the tests we need. We may have, but we can never be certain of that and any time someone invents something which offers an improvment there's going to be the possibility that some new test comes along with the improvment. That won't happen in every case but it will certainly happen in some.

I don't think your comparison of video and audio in relation to tests holds up. There are grey areas for both when it comes to deciding whether or not something makes an improvement and there probably always will be. We have no way of accurately deciding which 'improvements' in the grey area are real and which are not. We occasionally get to move something from the grey area to the proven area but, when it comes to those things left in the grey area, we're each left to make our own judgement. We can say some are more probably genuine improvements than others and we each may draw the line between probable and improbably at different points, but wherever we draw the line the only sure bet is that we're going to make mistakes. Draw the line too conservatively and you'll ignore more genuine improvements than you will accept things that don't work. Draw the line too loosely and you'll pick up a lot of things that don't work along with the ones that do. Essentially you choose the sort of error you're prepared to make and the amount of time and money you're prepared to spend at the fringes. This is going to be equally true for video as it is for audio.





David Aiken