Upscaling to 4K
Over the next couple of years, therefore, and assuming that 4K sets take fire in the marketplace, the smart money will be on upconverting 2K sources to 4K. No form of upconverting can add real resolution; genuine Ultra HD starts and ends with 4K resolution. Nevertheless, we expect plenty of action on the 2K to 4K upconversion front. Since consumer 2K is largely (though not entirely) 1920 x 1080 pixels, and consumer 4K is 3840 x 2160, it would appear that such upconversion might simply involve taking the content of each 2K pixel and quadrupling it (with no added enhancement) to fill a 2 x 2 pixel area on the 4K display. But that will gain nothing in subjective resolution, and may actually reduce image quality due to the added processing required. Most upconversion, therefore, will likely include enhancement and/or other digital manipulation, designed to both eliminate possible upconversion losses and better simulate the look of true 4K.
At this past week’s CES Unveiled in New York, Technicolor planned to demonstrate a promising 4K upconverter from Silicon Valley chip developer Marseilles. Located as I am on the Left Coast, I was not at CES Unveileda sort of Mini-me CES designed to fill in the January-to-January gap between the big, annual CESs in Las Vegas. Others, from Home Theater’s New York office, however, attended the New York event.
But in late May I was invited to Marseille’s California headquarters to view a preview of its technology. Their demo used two Sony 55-inch 4K flat screen Ultra HDTVs. The first comparison had native 4K on one display and Marseille’s upconversion of a 2K original on the other. There were essentially no visible differences, though if pressed I’d say that some scenes looked marginally sharper on the upconverted version. When they later showed upconverted 2K material on one set (using the Sony’s on-board upconversion) and their own upconversion on the other, the result was a visible win for Marseille as well.
I don’t think 55-inch 4K sets are the most revealing choice for such a face-off, but it did show that Marseille’s processing could be a serious player. I suspect they’ll have a lot of competition in this game, but the partnership with Technicolor is a smart move for both companies. Marseille is not a household name in video upconversion, but then again there are no household names in video upconversion. Videophiles alone may be familiar with Marvell, DVDO, Lumagen, Faroudja, and others, but everyone has heard of Technicolorunless they’ve recently arrived from Krypton. Technicolor hopes to use its name recognition to promote Marseille’s technology. If this works out, expect to see the Technicolor name on a variety of future products that offer 4K upconversion. Likely candidates include Blu-ray players, A/V receivers, and perhaps stand-alone processors. I’d add televisions as well, but many HDTV manufacturers have a not-invented-here mindset.
The point I made above about the upconverted 4K looking a shade sharper than true 4K is an experience I’ve also had in Sony demonstrations. It may result in a slightly too ambitious use of enhancement to overcome inherent upconversion losses. I hope that when Technicolor/Marseille upconversion (or anyone elses) is included in products, or at least in products likely to appeal to serious videophiles, it offers a choice of several enhancement modes, from virtually none (or at least only enough to overcome upconversion losses, perhaps erring a bit on the conservative side) up to whatever the intended market, or marketers, require. The Marseilles reps commented that consumers in some markets actually like over-cooked sharpness enhancement in their HD (?) images! Offering a variety of options would allow the end user to tune out those over-eager designers at Edgy-R-Us Video.