As a preface to this review and summation of our recent “JVC vs. Sony” projector shoot-out, I’d like to share a little perspective gleaned from my own experience working with several of the major Hollywood studios while UHD / 4K standards were being finalized. About four to five years ago, I was involved with a project that attempted to get the major studios on board with higher resolution, anamorphic Blu-rays. Shawn Kelly of Panamorph had developed a technology where 33% more resolution could be “hidden” behind the black letterbox bars on Scope / 2.35:1 movies, and then reintegrated into the image when decoded by properly equipped projectors or Blu-ray players. At one point we had three of the major Hollywood studios seriously interested in this technology. As a side note, the timing of this corresponded with the rise of 4K flat panels in the consumer electronics industry.
As a result of my involvement with this project, I attended two events that were eye-opening. The first was one of the first tests of high resolution 4K native material at one of the major studios. It was a comparison of 4K scanned IMAX footage shown side by side on a Panasonic 1080P plasma and a Samsung 4K display. From anything approaching a normal viewing distance, differences were very hard to make out. In this test, the Panasonic "won out" thanks to its higher contrast and better screen uniformity. We were actually in the room with the XXXXXXX studio techs who were going back and forth trying to see differences in detail. They were there, but you had to get close to the screen to see them.
My take-away – the differences between 1080P and 4K material is in the ability to make out fine details and textures, not in overall sharpness. As we got deeper into this project and I started playing with native 4K footage myself, it became obvious to me that we needed picture content with ultra-fine textures or patterns to reveal any improvement at all with 4K displays and content. Examples would be fine textures on fabrics, leaves on trees, fine patterns containing lots of geometric shapes, etc. Sometimes differences could be seen in fine details on objects far away from the camera (such as being able to read the license plate numbers of cars during some of the helicopter shots in the film SICARIO).
It was also clear to me that you needed to be close enough to the display in order to actually see these differences. Typically, this means having to be less than 3 times the picture height from the screen. To put that into perspective, if you own a 65” 4K flat panel, you would need to sit less than 8 ft. from the display to see the improvement that 4K resolution brings. And you had to go looking for it, when it was present in the types of objects described in my previous paragraph.
The second event was a meeting of the Hollywood Post-Production Alliance in Palm Springs, CA. We witnessed Dolby making their argument that "we don't need more pixels, we need BETTER pixels." This is where the move toward HDR and wider color gamut was really kicked into high gear. Their argument was – and is - that differences in brightness, color and contrast can be seen at ANY viewing distance, while differences in resolution can only be seen at relative viewing distances. To be clear, this is not to say that differences in resolution are UN-important, just perhaps less important. That was Dolby’s point of view, and it has grown to be my own as well.
I’ll give another perspective, this time from a filmmaker's point of view. Having high resolution 4 to 6K cameras like the RED, Arri and others means that filmmakers can be that much looser framing the image during production because there is so much excess resolution captured by the camera sensor. Filmmakers and editors know they can go in during post-production and zoom and crop the image to their heart's content without losing sharpness (I did this myself when I edited the proof of concept trailer for my upcoming feature film, which was shot on a RED).
It’s also true that most movies – even if they are filmed and edited at 4K resolution or greater – are finalized on what is known as a 2K Digital Intermediate (2K DI). This means that the actual final product that makes it to Digital Cinemas and even UltraHD Blu-ray is actually only 2K in resolution. The reason? Right now, it’s simply because even the most modest films these days have some kind of digital, CGI based FX, and it’s much cheaper to render these FX at 2K resolutions than 4K resolutions. Of course, as with almost everything technology related, the cost and difficulty of doing 4K renders becomes less and less, so more and more true 4K movies and content will be coming down the pipeline.
(For anyone interested in finding out if their favorite UHD movie is actually sourced from a 4K master, here is an excellent resource: https://realorfake4k.com/).
All of that said, one of the great advantages of projection systems is the ability to deliver extremely large images. With a projection system, it’s not unusual for a viewer to be sitting within 3 picture heights of the screen, so differences in fine detail can be easier to make out (assuming native 4K material, of course). One of the major goals of our shootout was to determine how important ALL the different factors that go into the creation of UHD content (High Dynamic Range, Wider Color Gamut, and 4K resolution) are when viewing such content on the four excellent projectors we brought in for our shootout.
To begin, here is a list of the projectors we were able to test, with list prices and a basic feature breakdown:
Sony VPL-VW385ES: $7999 MSRP, native 4096 x 2160 4K panels, 1500 lumen light output, 13.5 Gbps HDMI inputs, dynamic iris, lens memory, 200,000:1 dynamic contrast, native contrast unpublished (from independent sources, likely around 15,000:1), approximately 80% reproduction of the DCI / P3 color gamut (unpublished, figure from independent sources)
JVC DLA-RS640: $7999 MSRP, 4K / 3840 x 2160 resolution capable via eShift5 “pixel shifting” processing of native 1920 x 1080 panels, 2000 lumens light output, 18 Gbps HDMI inputs, lens memory, 160,000:1 native contrast, 1.6 million to one dynamic contrast, 100% reproduction of the DCI / P3 color gamut (72% of REC2020)
Sony VPL-VW885ES: $24999 MSRP, native 4096 x 2160 4K panels, 2000 lumen light output, 18 Gbps HDMI inputs, lens memory, Infinity:1 dynamic contrast, approximately 100% reproduction of the DCI / P3 color gamut, 20,000 hour laser light engine
JVC DLA-RS4500: $34,999 MSRP, native 4096 x 2160 4K panels, 3000 lumen light output, 18 Gbps HDMI inputs, lens memory, Infinity:1 dynamic contrast, greater than 100% reproduction of the DCI / P3 color gamut (80% of REC2020), 20,000 hour laser light engine
Here is a picture of the projectors the day prior to the weekend long shootout. On the ceiling - the JVC DLA-RS4500. On the top shelf left, the Sony VPL-VW385ES, top shelf right, the Sony VPL-VW885ES, bottom shelf the JVC DLA-RS640.
Before we get to the results, it's worth noting that the Sony VPL-VW385ES at $7999 is essentially the same projector as the Sony VPL-VW285ES (the famed "4K for $5K" projector stirring up so much buzz), only with the benefit of an automatic iris to bump up dynamic picture contrast as well as zoom memory for use with Scope / 2.35:1 screens. Native contrast and resolution on the VW285 and VW385 are the same, so we thought the VW385 was an ideal test candidate since it is essentially two projectors in one.
Similarly, the JVC DLA-RS640 ($7999) used during our test is practically the same unit as the JVC DLA-RS540 ($5999), only with hand-selected optics and DILA chips for the ultimate "tweaked" performance. However, the RS640 and RS540 have almost identical specs and feature sets. Much like the Sony VW385 served as a stand in for the VW285, the RS640 serves as a stand in for the RS540.
NOTE: JVC also sells the RS540 and RS640 projectors under the “Procision” brand with different model numbers – the X790 and X990, respectively. The projectors are identical except for the color of the lens ring – gold for the RS series, black for the X series.
The other two projectors compared – the JVC DLA-RS4500 ($34,999) and Sony VPL-VW885ES ($24,999)– represent the state of the art of home theater 4K projection, or close to it. Both have native 4K panels, laser light engines, and can reproduce the entire P3 color gamut of the Digital Cinema Industry (DCI) spec). The JVC RS4500 is spec’d out at 3000 lumens, the Sony VW885ES at 2000 lumens. Most of the cost difference between the two has to do with the optics in the JVC RS4500 – a precision all glass lens system vs. the acrylic / glass hybrids in the others.
All projectors were compared one after the other on the 144” diagonal Scope / 2.35:1 Stewart StudioTek130 reference screen in our showroom.
We were able to compare the pictures sequentially and immediately by simply blocking light from one while examining the picture from the other.
To level the playing field, THX Worldwide Video Calibration Instructor Gregg Loewen
spent the night before the shootout calibrating all the projectors to make sure they were operating at their very best. In our opinion, Gregg’s presence was critical – he “kept us honest” and made sure we had a level playing field. We shot out all four projectors side by side for two days straight with a variety of 4K / HDR content.
On the lower end (sub $10,000 MSRP), it became a battle of Sony's true 4K resolution vs. JVC's "eShift" 4K, plus high dynamic contrast (Sony) vs. high native AND dynamic contrast (JVC). On the high-end laser based projectors, the differences mainly came down to brightness and the quality of the optics. What follows are all my personal thoughts to date, mixed in with some comments from attendees. While I spent 5 hours each day running my mouth and juggling remotes, I was also standing at the front of the room, so most of the time was about 2 ft. from the screen.
Here, broken down by model number, are my own personal thoughts:
JVC DLA-RS640: the deepest blacks and best native contrast of all projectors tested, very good picture detail, best overall HDR “pop” thanks to the high native contrast of the projector panels. Fine details in native 4K material were slightly obscured at close seating distances compared to the “native” 4K projectors (certainly at my vantage point of about 2 – 3 ft from the screen), some video noise in high detail areas noticeable right at the screen (literally inches away), slight red cast to the picture even after calibration (Gregg commented that color biases could be completely dialed out with a more sophisticated calibration; our calibrations were deliberately done with the calibration software built into each projector), slight black crush and loss of shadow detail in very dark areas of the HDR picture (though this could be mitigated by raising the “dark level” control in the JVC’s HDR Gamma setting menu), definite sense of “three dimensionality” and depth thanks to the high native contrast
Sony VPL-VW385ES: very good blacks, excellent picture detail (slightly better detail than the JVC DLA-RS640 in fine textures and patterns), transparent auto-iris operation, definition in items such as lettering on menus slightly crisper than the JVC RS640, during dark and moody scenes looked somewhat “flat” compared to the JVC DLA-RS640, very good HDR implementation when the HDR contrast setting was set to a maximum of 60, slight greenish cast to the picture (again, Gregg commented that color biases could be completely dialed out with a more sophisticated calibration, such as what is attainable from a Lumagen video processor), very good shadow detail, inability to get very fine focus at the screen resulting in some softness at the pixel level (could be a limitation on the lens)
Before we get to the laser projectors, I’d like to comment a bit on the two projectors discussed above to put things into context. Make no mistake – both projectors threw an excellent image. With bright, native 4K material (such as PLANET EARTH II), at typical seating distances most people were hard pressed to tell a difference between them at all. There were slight differences in color reproduction, contrast, and brightness, but when we “blinded” the comparison people were just as likely to think the JVC was the Sony and vice versa.
Here are some comments from others at the event. From attendee Todd Smith:
":…what amazed me the most is just how similar the 640/385 are when calibrated. I purposely sat at about 1.5 screen widths (when leaning forward from the back row) since I sit 1.45 screen widths at home so all my comments should be read with that in mind. In motion especially, it was really splitting hairs for me trying to pick ANY differences in the material we watched even doing an A/B. Both projectors looked fantastic and I could be very happy with either. Some random thoughts...
-Sony was a HAIR sharper, but I couldn’t see it in everything and really needed something like letters, numbers, or some type of texture (in Sully for example when he's running through Times Square) to see it at all. There were even times when I didn't know which projector was playing and I would say to myself "that must be the Sony" and it was the JVC and vice versa. Splitting hairs most of the time.
-Contrast there was a slight real-world difference in favor of the JVC, but nowhere near as big as I expected and I really couldnt see any difference most the time. This surprised me. - I couldn’t see any difference in native motion going back and forth."
From attendee Chris Lind:
"My wife and I were lucky enough to be the last ones there tonight, but our 8 hour round trip of driving meant we needed to get the most out of the event. Like (Todd) said early this was an awesome event and I had a load of fun, the only thing I regret is not being able to leave the demo with a JVC RS4500. I feel the like results of the 640 and 385 were the same as when I did a small shootout in my own room. The 385 is sharper when it comes to text and fine detail, but you have to pause the image and look for it. There were times that myself along with (Todd) and my wife who attended were forcing ourselves to be blind to which unit was playing and we would get it wrong at times on which was the sharper unit. One thing for me, that was easier to spot was the contrast advantage of the JVC, I felt like I could pick out the JVC when focusing on contrast and depth of the picture. At the end of the day, we are truly splitting hairs or pixels when it comes to picking a winner."
My own thoughts: in bright scenes, I would agree almost completely with the comments above. With fine details like clothing texture and street signs in the distance, etc, the Sony was sharper up at the screen. However, sometimes the "fine detail" differences disappeared at a typical seating distance. In dark scenes, though, the difference in image contrast was obvious to me in most shots, especially in the below scenes from SULLY. This is where the JVC excelled. As a result, the JVC image had more “pop” and a greater sense of three dimensionality where the Sonys looked a little flat (IMO). The JVC E-shift 5 seems to be an upgrade over previous iterations, and the new “HDR” preset presents a wider color gamut than the Sony while retaining brightness and contrast.
To illustrate some of the differences graphically, I have pulled a couple of screen shots from the SULLY Blu-ray to better help me illustrate the differences. NOTE: these are NOT 4K screengrabs, nor are they HDR. They are simply used for illustration purposes. Even if I could grab 4K UHD images, the time and effort would be wasted unless you were reading this on a HDR, 4K capable monitor. However, I can still use these standard Blu-ray screen grabs to make some points:
Looking at the above screen shot, you can see that this Times Square scene from SULLY really poses a challenge for HDR. We have the black of the night sky combined with the intense highlights of the neon and LED signage. With the JVC DLA-RS640, the black of night was rendered a bit darker than the Sony VW385, while at the same time the white highlights of the electronic billboards and American flag were brighter. This would be attributable to the high native contrast of the JVC combined with its higher lumen output. This is what I mean by HDR “pop.”
If you look just to the left of Tom Hanks, you can see a food truck with the name “Tapas Papas” emblazoned on it far in the background. This logo was easier to read on the Sony VW385ES (and 885ES, for that matter) from my vantage point right up at the screen:
Of course, since this is a standard 1080P Blu-ray screen grab, the lettering is a bit blurry and hard to make out. However, it was quite distinct on the two Sonys with the 4K Blu-ray as its source. Up at the screen, on the JVC the lettering was readable, but not as crisp (but still more legible than the Blu-ray screen shot you see here). Quick note – on the native 4K laser JVC RS4500 this was even crisper than the two Sonys, but more on that in a bit.
Here's another “torture test” screen grab from SULLY:
Note the “pools of light” on the side of the aircraft carrier, and the bright highlights of the actual lights themselves (on the UHD Blu-ray, these pools of light and highlights are even more pronounced). With this scene, the JVC once again showed brighter highlights while retaining the deep blacks of the letterbox bars and night sky. In comparison, the Sony looked somewhat “flat” and lacked a certain sense of dimensionality. However, this scene also revealed slightly crisper detail on the Sony, noticeable in the superstructure area of the aircraft carrier. The JVC also crushed the blacks a bit in the dockyard area on the lower left of the screenshot. Elevating the “dark level” control on the JVC helped bring back out the shadow detail, but it was at the expense of some three dimensionality and “pop.”
NOTE: as mentioned, we chose the Sony VW385ES for the shootout as disabling the automatic iris essentially turns it into a VW285ES, the $4999 Sony entry level projector. It was in scenes like what you see above that the difference became most apparent. For example, in the shot of the aircraft carrier, disabling the automatic iris resulted in a brighter overall picture with brighter highlights. However, at that point black levels suffered and appeared more like a dark grey. Turning on the iris distinctly improved the black level, but at the cost of some brightness. As the blacks got deeper, the white lights on the ship got a bit dimmer as well.
One other title we used to show the differences between these two $7999 projectors was another torture test: BILLY LYNN'S LONG HALFTIME WALK. This title is unique in that it outputs a full 60 frame per second image in a 4:4:4 HDR P3 color space. As the 13.2 Gbps HDMI inputs of the Sony VPL-VW385ES won’t handle the full bandwidth of the UHD signal coming off the disc, it will reduce the 10 bit color to 8 bit, which can result in visible “banding” in the image (banding is where fields of color don’t show smooth gradations between various shades, instead breaking up into what appears to be “bands” of color adjoining one another). Actually, we did not see any real evidence of banding on the Sony VW385ES, however, what was most noticeable to me was the limitations in the overall color gamut. As the JVC DLA-RS640 is capable of reproducing approximately 100% of the P3 color space (vs. approximately 80% on the Sony VW385), the colors appeared richer and more vivid on the JVC than the Sony (most noticeable in the reds and greens). This was obvious in scenes like what you see in the screengrab below (although, of course, the reds and greens you see below are limited by the capabilities of your computer monitor or mobile device screen, plus the fact that this screen grab is from the standard REC709 Blu-ray). Once again, though, the Sony revealed a bit more fine detail in the bright LED lights making up the backdrop, and in the texture of Billy Lynn’s uniform when you were up close to the screen: