The revolutionary technology of pixel-shifting has injected a lot of controversy and confusion into the world of 4K projectors. People are concerned about what is true or genuine 4K and what is fake or "faux-K." What is the difference, and is it anything to worry about? In the end, as we go about categorizing and defining these products, the practical question is what makes a 4K projector 4K?
A Bit of History
Back in the day, the native resolution of a projector was a big deal, and rightly so. There was a massive visible difference in picture quality between SVGA and XGA, and between WXGA and Full HD 1080p. When comparing these projectors there were obvious differences in detail and image sharpness. Visible pixelation (the screendoor effect) on lower resolution projectors was supremely annoying and desperately to be avoided. Anyone in the projector biz could tell at a glance just from looking at pictures on the screen what the native resolution of the projector was. So we all grew up KNOWING that the "native" resolution of the chips was directly related to ultimate picture quality on the screen.
That was then. This is now. In our new reality of 4K resolution and pixel shifting technologies, there is ZERO correlation between native resolution (the number of physical pixels on the chip) and the actual resolution of the picture you see on the screen. The chip's physical resolution, at least in the home theater world, has become irrelevant.
The revolution began with JVC's introduction of pixel shifting in 2013. This was a radical new approach to getting substantially higher picture resolution out of native 1920 x 1080 D-ILA chips. Epson followed using pixel shifting on 1920 x 1080 3LCD chips. Then Texas Instruments released an 0.66" 4K UHD chip with a physical mirror matrix of 2716 x 1528 which uses two-phase pixel shifting to create 8.3 million pixels, which is 4K UHD. They followed that with an 0.47" DLP 4K UHD chip which has a physical mirror matrix of 1920 x 1080. It uses four-phase pixel shifting to create 8.3 million pixels.
To no surprise, this has fueled a lot of controversy over what is "true 4K" and what is often disparagingly referred to as "faux-4K" or simply "faux-K." This term is widely used by industry personnel, reviewers, and consumers alike. It often connotes that there is something inauthentic and undesirable about the pixel shifting technologies and the 4K projectors that use them. In reality, pixel shifting has produced rapid advances in picture resolution at prices far lower than you must pay for projectors that have native resolution 4K chips. It is a terrific technology that is hard not to love once you see it.
And the simple fact is this: we can no longer glance at an image on the screen and know the physical resolution the chips in the projector like we could before. Today there are 4K projectors using chips in various physical resolution formats -- 1920 x 1080, 2716 x 1528, 3840 x 2160, and 4096 x 2160. And they are all capable of producing impressive 4K resolution pictures. Even the current JVC and Epson models that use two-phase pixel shifting on 1080p chips can produce pictures that come a lot closer to replicating a full 4K picture than one might imagine possible--sometimes they can even beat the 4K UHD DLP chips in the subjective impression of image sharpness.
To illustrate the misleading nature of the term faux-K, let's do some side by side comparisons of six different 4K projectors, five of which are using a variation of pixel-shifting, and one using native 4K chips...
- Optoma HD60 vs. Optoma UHD50
- ViewSonic PX727-4K vs. Epson HC 4000
- Optoma UHZ65 vs. Sony VPL-VW285ES
In our first test, we will compare projectors using the two new 4K UHD DLP chips. We can do this by setting up the Optoma UHD60 with its 0.66" chip sporting 2716 x 1528 mirrors, and the Optoma UHD50 with its 0.47" chip, having just 1920 x 1080 mirrors. The UHD60 has two-phase pixel-shifting to double the number of pixels on the screen while the UHD50 has four-phase shifting to quadruple the number of pixels. Let's put these two projectors side by side and take a close look.
Viewing the pictures close up, at a distance of 12 inches from a 5-foot wide image, you can see an obvious difference in pixel structure. The UHD60 with its 0.66" chip and two-phase shifting produces a distinct pixel structure. Each pixel is extremely small of course, but you can see clearly distinct pixels in rows and columns when you examine the image close up.
On the UHD50, when you are at the same distance of 12 inches from a 5-foot wide screen, you can detect an extremely subtle hint of pixel structure, but it is nowhere near as distinct as on the UHD60. There is no sense of discrete individual pixels. This is true of the ViewSonic PX727-4K as well which uses the same 0.47" chip. The indistinct pixel structure on these projectors is a feature of the four-phase pixel shifting going on with the 0.47" chip.
Once you back up to a viewing distance of two feet from a 5-foot wide screen (that is, 0.4x the screen width and a lot closer to the screen than anyone would want to be when watching a movie), all hint of visible pixel structure disappears on both projectors. At this distance the UHD60 and the UHD50 look the same, at least in terms of detail resolution.
Is there is any visible difference in image sharpness? The answer is clearly no -- in practical terms you see exactly the same amount of image detail on both models. The fact that one projector's chip has a mirror matrix of 2716 x 1528 and the other a matrix of 1920 x 1080 (half the number of mirrors) has absolutely zero impact on the sharpness and detail in the image.
This does not mean the pictures look identical. There are other contributing factors on a projector that affect what you perceive as image sharpness. In this case digital noise is a factor - the lower the noise the cleaner and sharper the picture looks. And in this regard the UHD50 tends to edge the UHD60 with an overall lower noise factor.
In addition, one of the biggest factors influencing what you see as image detail and sharpess is contrast. In this case the UHD60 is higher in contrast than the UHD50, and in many scenes this contributes to the perception of it being the sharper of the two projectors. But at the end of the day, both the level of digital noise and the projector's contrast have more impact on one's perception of the sharpness and the ability to resolve detail on these two projectors than does the physical resolution of their chips, which is a non-issue.
This is a different comparison entirely. The ViewSonic PX727-4K has the 0.47" DLP chip with a 1920 x 1080 mirror matrix, generating a 4K picture (8.3 million pixels) via four-phase pixel shift. Meanwhile, the Epson HC 4000 uses 3LCD chips with the same 1920 x 1080 pixel matrix, but it has only a two-phase pixel shift. So it does not produce 8.3 million pixels, and Epson has never represented that it did -- they market this class of projectors as 4K-enhanced. So the HC 4000 paints two slightly offset 1920 x 1080 images per frame while the PX727-4K paints four. In theory the PX727-4K should produce a sharper picture in terms of image detail, and it does.
So, what differences do we see? When viewing the Epson HC 4000 from our close-up distance of 12 inches from a 5-foot wide screen, there is no distinct pixel definition. The two-phase shift in this implementation tends to blur any hint of a discrete pixel matrix. In this regard it is extremely similar to the PX727-4K. When viewed very close up they look not quite perfectly identical as far as the indistinct pixel array is concerned, but they are pretty darned similar.
Now, let's back up to a more practical viewing distance. This time we will stand 5 feet from our 5-foot wide screen (1.0x the screen width). Can we see any difference in image sharpness and detail? Yes, the PX727-4K is the sharper of the two. And this is true no matter if you are displaying a native 4K video signal, or if you are displaying regular HD 1080p and having the projectors upscale it. It is not surprising that the PX727-4K can deliver more 4K detail from a 4K signal, but it is surprising (to me anyway) that it can even upscale HD 1080p with a higher apparent level of precision. This four-phase pixel shift is remarkably potent as far as rendering detail is concerned.
HOWEVER. And this is a huge however - the Epson HC 4000 is much higher in contrast with much more solid black levels than the PX727-4K. And as we know, high contrast makes a picture look sharper. So when we back up from the screen to a viewing distance of 1.5x the screen width and put on some 4K HDR material, the Epson 4000 actually appears to be a bit sharper than the PX727-4K.
This is due to a combination of factors - as you move back from the screen the advantage the PX727-4K has in rendering fine detail becomes less visible so contrast becomes a more dominant factor in your overall perception of the picture's sharpness. So as your move back your perception flips and the Epson 4000 looks like the sharper of the two projectors.
The point at which this occurs depends on the contrast of the subject matter, but with the 4K HDR movie Lucy it appears to happen at a viewing distance of about 1.3x the screen width (with my 20/20 vision). In point of fact, the effect is somewhat of a hybrid, with some elements in the picture looking sharper on the PX727-4K, and other elements looking sharper (or at least as sharp and more three-dimensional) on the Epson 4000. In this no-man's land, where the latent competitive strengths of both projectors are visible simultaneously in the same picture.
In the end, this comparison illustrates that contrast is a huge mitigating factor that can outweigh the perception of image detail, sharpness, and clarity. Small differences in image detail which are obvious when examined close up become insignificant or even invisible when you sit back to enjoy a movie in a typical theater set up. Meanwhile, contrast is of utmost importance.
The Optoma UHZ65 ($4495) is a laser-based 4K projector using a single 0.66" UHD DLP chip with two-phase pixel shifting. The Sony VW285ES ($4999) uses three native 4K resolution SXRD panels. So these two models compete directly both in resolution class and price. The big question is, can you see the difference between a projector using authentic native 4096 x 2160 resolution chips, and a projector using two-phase pixel shifting on a native 2716 x 1528 chip?
Once again, we start by doing some pixel-peeping close up, from a distance of 12 inches on a 5-foot wide screen. And at this distance, the pixel structure of the two projectors is obviously different. Both have discernable discrete pixel structure, but the pixels on the UHZ65 are (by comparison) larger and more well-defined. The VW285ES's pixel structure is so small as to be almost invisible, but on very close examination you can detect the pixel pattern.
Thus, in theory, if the number and size of discrete pixels has anything to do with it, the VW285ES should be able to resolve more detail than the UHZ65, assuming the three chips on the VW285ES are perfectly aligned. To detect any visible differences, the detail must exist in the source to begin with, so we need to use a 4K source. Once we've put the 4K material on the screen, the issue then becomes whether there is a level of fine detail that can be resolved by the VW285ES that cannot by the UHZ645, and whether this incrementally resolved detail is visible to the viewer at a typical viewing distance.
Since few people would ever want to sit closer than 1.0x the screen width, we set up to view at a distance of 5 feet from our 5-foot wide images. Popping in the 4K HDR disc Lucy gives us a good test sample for video as it is loaded with fine details in hair, leather and cloth fabric, etc.
And the results? In this movie there are definitely certain elements in some scenes which are rendered with slightly more detail on the VW285ES. This occurs in the first scene, where detail in the straw hat and the subtle texture of the leather jacket are more successfully resolved by the VW285ES. However, at the same time, the hair detail in the Scarlett Johannson close-ups appears to be better defined on the UHZ65. Why is this?
Once again, the big issue is contrast - the UHZ65's 4K HDR picture is substantially brighter and noticeably higher in contrast than the VW285ES. The increased vibrancy of the UHZ65's picture makes it look subjectively sharper even though it isn't. Once you move to a viewing distance beyond 1.0x the screen width, the eye's ability to resolve the subtle differences in detail between these two projectors evaporates. At this point the physical resolution of the chips in these two models has nothing to do with perceived image sharpness. Contrast (and other factors like digital noise, lens resolution, and video processing) become the determining factors in one's perception of the picture's overall sharpness. In other words, in this case we have a projector that many will dismiss as "faux-K" actually beating a "true native 4K" projector in perceived image sharpness.
Let's now consider the display of an HD 1080p picture rather than 4K HDR. With an HD 1080p source, we no longer have the extreme detail in the source itself. Both projectors are upscaling a 1920 x 1080 signal to a 4K-like image. With these two projectors, with respect to detail definition itself, there is very little if any discernable difference between the two pictures, even when viewed close up.
However, with HD 1080p we once again encounter a big difference in contrast, except this time the VW285ES wins. In default Cinema presets, when displaying HD 1080p source material, the VW285ES is much higher in contrast and saturation than the UHZ65. So the VW285ES looks subjectively sharper and more three dimensional despite the fact that there really is no additional detail in the picture. At any normal viewing distance beyond 1.0x the screen width, the differences in perceived sharpness of upscaled 1080p material is entirely attributable to factors other than the physical resolution of the chips. Whether it is true 4K or faux-K has nothing to do with it. Contrast is king.
Since "4K" is all about resolution, it is not surprising that so many folks have been preoccupied with native 4K vs. pixel shifting 4K, and debating the merits of each. Though many folks use the term faux-K as simply synonymous with pixel-shifting technology without implying anything derisive, many use it precisely because it insinuates inferiority. When it is used in this manner it is counterproductive and serves no good purpose.
This preoccupation with the physical devices is unfortunate since it causes the unsuspecting consumer to get fixated on what is the least relevant aspect of 4K projection technology (at least as far as home theater is concerned) -- which is the number of elements on the device. All of these 4K projectors regardless of chip type are able to create images in a class that is legitimately understood as 4K resolution. Even the JVC and Epson models with their simple two-phase shift on 1080p chips can outperform competing 4K projectors on the basis of contrast and black levels and produce pictures that look equal to, and in some cases better than, a competing 4K projector at a similar price.
So the bottom line is that there are many important factors that contribute to a 4K projector's success. Certainly contrast is a critical one--both with HDR and SDR. But also the level of digital noise, the image enhancement video processing, the color balance and saturation, and the optical precision of the projector's lens, all contribute to making a 4K projector's picture great or less than great.
Unfortunately, all of these vital factors tend to get sidetracked when the consumer becomes preoccupied with the almost irrelevant issue of device level resolution. While this used to be a hugely important spec in the projector business, it is no longer an issue of consequence in the world of pixel-shifting 4K for home theater projectors.
Now, before you go dropping me a flaming email, I am NOT arguing that native 4K imaging devices are irrelevant, or that native 4K devices cannot be a critical feature in a projector. They most certainly can be. There are projection applications that WILL benefit very directly from native 4K resolution devices. If you are displaying still complex graphics in large scale, with viewers being particularly close to the image, the incremental resolution that can come from native 4K chips (like that which can be seen close up in the Sony VW285ES vs Optoma UHZ65 comparison) can be both visible and relevant to the application.
In point of fact, Digital Projection, Inc. just staged a demo of the first 8K projector in Las Vegas at the NAB tradeshow this week. So not only does native 4K imaging have a set of applications for which it is uniquely advantageous, 8K does too, and it is coming sooner than you think.
And even in home theater, if you have a much larger than typical screen and you like to sit in the front row of a movie theater, the incremental picture resolution from a native 4K projector may be visible and relevant to you. But this is simply not the typical desired set up in consumer home theater.
So what exactly is a 4K projector?
The term "4K projector" is open to any of several interpretations, and there is no chance of a consensus--it is an emotional subject for a lot of folks. Strict fundamentalists will say the term should be reserved exclusively for projectors that have native 4K imaging devices and that no pixel shifting machine should qualify. At the opposite end of the spectrum are the tech liberals who believe that if the projector takes a 4K signal and does something to make it look like an exciting, high resolution 4K picture on the screen, that is really what it's all about.
While I prefer the latter definition myself, what I care more about is that buyers come to realize that the differences in detail resolution between all of these various 4K implementations are extremely trivial compared to the major issues of contrast, brightness, color balance and saturation, and image stability and clarity that ultimately define the greatness of a home theater picture.