HD DVD Goes Pop

The August 2006 issue of Popular Mechanics devotes one entire page (!) to HD DVD. The main feature of the article is a comparison between HD DVD and the standard disc played back on a much less expensive, upconverting DVD player (an $80 Philips).

The article first states that "…many new DVD players can "upconvert" regular discs to a higher resolution." Not!. As we've written before, in the strictest sense an upconverting player does not increase the resolution of the source. It merely manipulates the pixels in the source by repeating them, averaging them, interpolating them, and performing other clever tricks to enable the 720x480 pixels in a DVD source to fill up all of the pixels in a high definition display, such as 1280x720 or 1920x1080. If it didn't do this, the set itself would. Without any upconversion, the 720x480 source would appear as a small rectangle in the middle of the screen, surrounded by black bars on all sides! The upconverting player may do this pixel manipulation better than the display, but that will depend on the specific player and set.

Yet the popular press continues to write about this upconversion as if it were an actual increase in resolution. It's a semantic trap into which even the enthusiast press can fall all too easily, even when they know better. And the poor consumer is mislead into leaving the store with that cheaper, "high definition," upconverting DVD player" when he or she came in looking for HD DVD or Blu-ray.

Put briefly, high definition in today's consumer market always means not only 1920x1080 or 1280x720 (or some comparable display resolution, like the 1365x768 common to many plasmas) but also a source that begins with that resolution at its video creation in a camera or film telecine and maintains it at every step from there to the final display.

You can upconvert 480i or 480p to higher pixel counts, but there is nothing you can do to a standard definition source to add real resolution. Not now, not ever. That isn't to say you can't manipulate it to look good, even perhaps subjectively better than it did before. But it can never be true high definition. If the detail is not in the signal to begin with, you can't generate it from nothing.

But the Popular Mechanics article doesn't stop there. They sat three viewers in front of a 42" (diagonal), 720p plasma to compare clips from Serenity, played back from both an HD DVD and standard definition DVD, the latter on the Philips upconverting DVD player. They did, at least, take the precaution of using two different seating differences to account for the relatively small screen.

Serenity is a good HD DVD transfer, though not one of the sharpest. But that was the least of the problems with the test. The article states that they watched the two versions "at 720p resolution." I take that to mean that they set both players to 720p. That of course would be an obvious choice for someone to make if they are using a 720p display and hadn't done any prior viewing, or hadn't researched the HD DVD chatter on the web both here and elsewhere. If you've read any reviews of the Toshiba HD DVD players, including ours, you'll know that current HD DVDs barely look better than standard DVDs when the Toshiba is set to 720p.

While the statement that they used 720p to compare the players is subject to interpretation, I'll stick with my reading of it. If those who ran the test actually did set the Toshiba player to 1080i, as they should have, that fact would certainly deserve a mention in the article. It's about the most critical single piece of information you can provide to a potential buyer of an HD DVD player. (It also would have been nice if the testers had used a 1080 display. They are becoming widely available at prices little higher than the 720p plasma used in the test, though the model and brand of the latter were not identified.)

The viewers did pick out the HD version consistently. But the reported differences were small, and confined to a crisper image on the HD DVD. That's not far from what I would expect, given the conditions of the test. The clearly implied conclusion, though it isn't precisely stated, is that HD DVD is still too expensive for the marginal improvement it offers.

One last concern. The article notes that the blacks on both players used in the test were considerably darker than from an ordinary DVD player. There is nothing in a properly functioning HD DVD or upconverting player that would account for this other than an incorrect setting of the players' black level (or an incorrect setting of whatever "ordinary" DVD player the testers were referring to). It also suggests that the display may not have been set up as well as it should have been.