More Aspect Ratio Discrepancies
Dear Gary: I’m a bit confused by some of your DVD reviews. You often note that the DVD is 2.35:1 in letterbox and 2.25:1 in anamorphic versions. First of all, I don’t see how this can be, since both are derived from the same bits, one by deleting every 4th line or some more complicated interpolation (ala Sony), one by using them all. Moreover, I don’t observe it on my TV. I have a Sony 32-inch XBR direct view set, which can properly display anamorphic DVDs by accessing the service menu and reducing the VSIZ parameter from 42 to 0. When I do this with Amadeus (one of the DVDs that you say in Issue 27 has a smaller aspect ratio when viewed anamorphically), I get exactly the same picture area as in plain letterbox—i.e., nothing is cropped from the sides compared to the non-anamorphic version. Could the varying aspect ratios be an effect of your display devise? I could imagine the settings on a projection TV screwing things up in this way. If so, I hope it will not be endemic to all 16:9 sets when they start appearing. Surely widescreen TVs can be designed to show all that is on the DVD. Any illumination on this issue would be welcome.
David S. Johnson
Research Editor Michael Coate Comments:
Beginning with Issue 31 of Widescreen Review (the one you are presently holding in your hands) you will notice that all anamorphically encoded DVDs are reported in our reviews and discographies as having identical aspect ratio measurements for both the anamorphic AND letterbox versions. In short, this is due to us utilizing a different technique to measure the ratios as compared to the method used in past issues. The long answer is a little more complicated, but here goes. Our review facility is equipped with a professional cursor generator (the Pandora PBX3001A), the same unit installed in many telecine bays. The cursor generator (which is 100 percent accurate) is connected to two different 1.78:1 widescreen monitors, one for LaserDiscs, the other for DVDs. Both monitors have been calibrated for zero overscan. One catch is that anamorphic program material cannot be measured while displayed in anamorphic mode. Due to the connection, measurements must be taken in 1.33:1 mode with the anamorphic squeeze still intact. This shouldn’t pose any problems as all that needs to be done is for us to apply a formula to the measurement to arrive at the eventual display ratio. When we began reviewing DVDs two years ago, it was decided to continue measuring aspect ratios to be consistent with the reviews established with LaserDiscs five years earlier. The crew that installed the DVD players and related equipment, informed Gary that anamorphic measurements would have to be made with the squeeze intact but that simply adding “.45” to the measurement would provide the correct measurement. This principle was based on 1.33 + .45 = 1.78. At first, this seems to make sense… until one thinks about why some DVD discs have two different measurements. The ratios should be identical since only one widescreen transfer is utilized in the creation of a DVD and the DVD player electronically converts the anamorphic image to “normal” proportions and reduces it in size to fit on a standard 1.33:1 TV for those viewers who do not have 1.78:1 (16:9) monitors. The anamorphic technology that is utilized in motion picture photography and exhibition is all based on multiplication factors—i.e., CinemaScope’s 2.35 ratio is based on a 1.18 aperture multiplied by an anamorphic squeeze factor of 2 (or more specifically, 1.2 x 2 = 2.4). One could add 1.2 to 1.2 and also arrive at 2.4, but that would be a coincidence. If one added 1.2 to any other number, a different result would be achieved. Ultra Panavision’s 2.75:1 aspect ratio could not be achieved by adding 1.2 to its 2.2 aperture. It is achieved by MULTIPLYING 2.2 by 1.25 (the anamorphic squeeze factor of the lenses employed by that process). Because of this, I theorized that a multiplication formula would need to be utilized for our measurements rather than one based on addition. After some thought, 1.33 was determined and tested to be the correct multiplication factor. A perfect 2.35:1 transfer of a SCOPE movie would measure 1.77:1 with its electronic squeeze intact. Multiplying 1.77 by 1.33 equals…you guessed it, 2.35:1, whereas adding .45 to 1.77 falls short of 2.35:1. I hope this clears up any confusion this matter may have generated. Our website database and future Special Edition Discographies will be updated to reflect the current and correct information.
You can E-mail Widescreen Review @ mailto:editorgary@widescreenreview.com