Marantz VP-12S4 DLP Projector Manufacturers Comment
We at Marantz wish to thank both Tom and Shane for their obviously thorough review of the VP-12S4. In this era of higher resolution (on paper anyway) devices, some people have questioned the viability of offering such a premium priced device that is "older" technology. But Shane's comments answer that question as well as we could have.
We have fine-tuned and tweaked an already state of the art platform for the previous three generations, focusing (pun intended) on performance, both measured and visual, taking into consideration some of the areas that other manufacturers miss completely. Gamma for example. While a linear curve of 2.2 is the "ideal" measurement, we know that subjectively in a dark environment, that curve will feel like it has too much contrast, due to the extremely high intra-field contrast ratio of the projector. Therefore, the curve designed in to the VP-12S4 is not linear, and does not lose any detail in the blacks as a result, but the picture is extremely easy to watch in a dark room. Rainbow effect is another. To minimize this to the level that we have requires extremely precise color wheel timing, which in turn requires a very stable chassis. These are things that go way beyond "how many pixels does it have?" and have contributed to our best-in-class representation over the years.
I would like to address a couple of small points though. The first has to do with the black level "crushing" that Shane noted. When testing the projector, we use a special test signal that generates windowed patterns in 0.5 IRE increments (actually slightly less, with 220 increments between black and white), and we know the projector can resolve this difference. In other words, in a dark room, you can see a 0.5 IRE window against a black background. This is higher color depth than most DVD players can deliver and is a result of the 10 bit Gennum processing. I believe that what Shane may have been seeing is a result of an error in the source device.
The second point has to do with the deinterlacing issues that he noticed. Proper deinterlcaing is a very difficult process to perfect, and some believe that it can never be done perfectly. During the development at Gennum we worked with absolute mathematical geniuses trying different algorithms in real time. Right as you make one artifact go away, another one shows up in a completely different realm. It will always be a compromise between feathering, jaggies, and sharpness. The HQV test disc, while one of the first DVDs in general release that enables critical processor evaluation, is primarily designed to make the HQV process from Silicon Optix look as good as possible. Makes sense. We have material in our possession that can make the HQV processor look bad and make the Gennum processor sparkle. And I'm sure that the people at Faroudja probably have material that can make their process look good too.
I want to stress that what really counts is, if there are distractions when watching real world source material And if that source material hasn't been processed already (a difficult task where cable boxes and satellite receivers are involved), then the VP-12S4 does as good a job as anything on the market. Overall though, we are glad for the opportunity to have the VP-12S4 reviewed by as thorough and accurate a team as you two. Hopefully your readers will understand why we say that it's the picture that counts.
Product Manager, Video