JVC DLA-NP5 Projector Review

Updates

19 Jun ’22:

  • I managed to align the panels better. See under Panel Alignment. This doesn’t change my observations about sharpness overall, but at least there’s a way to align the panels better. It does help detail retrieval ever so slightly with very high-frequency detail. Blooming hurts perception of sharpness still.
  • Note about zoom mode under both Brightness and Panel Resolution.
  • Added more detail under Contrast for the ANSI measurements

20 Jun ‘22:

  • Gaming Performance added
  • Concluding thoughts added to sharpness
  • Clear Black notes under Pixel-Level Contrast (Sharpness section)
  • Amended settings used for review with MPC, CMD, etc settings.
  • Added other calibration intro comments
  • Added a note under Calibration Functions for the “reflect to all” option.
  • Added further HDR impressions

21 Jun ’22:

  • Remeasured ANSI, on/off and dynamic contrast for the Epson in preparation for the LS12000 calibration. My previous on/off was incorrect. ANSI was correct, but more detail included re uniformity. New Dynamic measures for Epson.
  • Updated panel uniformity section with more detail.
  • More detail under sharpness – lens on focusing the unit.

22 Jun ‘22

  • Note added under brightness section re over driving of pixels for smaller window sizes.

27 Jun ’22

  • Final Summary and Review Scores added
  • Compared to Sony section under conclusions

7 Jul ’22

  • Find Final NP5 Settings (and CCSS files) here
  • Find Final NP5 Thoughts and Measurements here (link live within 12 hours) including ADL measurements.

Introduction

As you know if you’ve been following this blog, I am no stranger to projectors – and definitely not JVC projectors. I have owned many different projector models for the last 20 years: DLPs, LCDs and LCOS projectors from different brands. I have also seen plenty more in action. This does not necessarily make me an expert, but I feel I have a perspective to add.

This sample was not sent to me by JVC, but delivered direct from a store for actual money. This is an important point as this sample is likely indicative of the product you can buy at retail and was not a golden sample picked by JVC. Most of the reviews you will find online are done on golden samples sent by manufacturers for the sole purpose of the review.

While preparing this review, I had a JVC X7000 (RS500), an Epson 6050 / TW9400 and a JVC NP5 calibrated for viewing comparisons. I also have an X9900 here but that would be somewhat unfair on the NP5.

Out of the Box Performance

SDR Performance

I sat down to watch The Expanse Season 3 on both streaming and Blu Ray as well as a couple of movies over streaming in 4K and 1080p. I felt that the contrast performance didn’t look great and the picture at times felt flat. Colours looked OK, but something was certainly amiss.

HDR Performance

Again, I was not super impressed with the sharpness or the contrast performance of the NP5. While this is a 5-series projector, I was expecting a bit more.

The Dynamic Tone Mapping seemed ok, but the picture felt flat even with Theatre Optimiser on and the Frame by Frame tone-mapping on Auto. Let’s see if we can improve this performance by calibrating the unit then!

Calibration

I go into detail about the calibration at the end of the article for those that are interested.

I was not at all prepared for the lacklustre out of the box performance of the NP5. Later this was confirmed by probably the worst out of the box calibration I have seen on a JVC. The REC.709 gamut was undersaturated but colour luminance for some colours was way too high to the point that I ran out of on-board controls to pull them in line.

So before I could review the unit, it was time to do an auto-cal and then manually touch it up. This was after I spent 2 hours trying to manually calibrate the unit and realising it simply wasn’t going to work. Since the autocal invalidates any manual calibration, I should have started there.

After autocal, there were still manual adjustments that had to be made, especially for SDR in low lamp. Thankfully, my high-lamp calibration was pretty spot on. Still, I made minor adjustments to CMS to get things to line up better.

Thankfully gamma was pretty flat, which is important as JVC doesn’t allow you to manually edit the base gamma curve used by the unit to calculate all the other gamma curves – including the one for the tone mapping.

I think this is a bit limiting and would like to see JVC enable us to edit the internal gamma tables in both low and high lamp to pull them in tighter still. As far as I know, there’s no way of doing that currently apart from using the custom import slots, which doesn’t affect auto tone-mapping.

Once I calibrated the unit for both SDR and HDR, I sat down to watch the same content again.

Calibrated Performance

SDR – Overall

The issue with having an undersaturated gamut with high colour luminance is that it’s difficult to pin down by eye. This is because there is some observer equivalence with a higher luminance but narrower gamut and vice versa. I just knew something was off.

Thankfully, after calibration and lowering the iris to -6, the picture tightened up to a point where SDR was pretty good. Colours were both better delineated and better saturated. Closing down the manual iris to between -5 and -10 helps the NP5 with on/off contrast performance without hurting ANSI contrast much (more on this later).

There was better detail retrieval than I remembered with the JVC X5000. The extra pixels definitely helped the image gain some detail and sharpness. Mind you, the X5000 did not cost this much.

Compared to the X7000, the NP5 lacked in contrast. Compared to the Epson, the NP5 could look a bit flat in brighter scenes, while it was significantly better in very dark scenes.

Difficult to capture colours accurately in a dark room. I also should have removed my laptop. Ops. Pictures are not necessarily representative of viewing experience as cameras can jack up colour and contrast when handled by someone inexperienced in photography (me).

HDR – Overall

After calibration, HDR performance looked significantly better. However, I had to up the DTM level to 0 from the Auto setting. I also lowered the iris position to -6 to be happier with the overall contrast performance of the unit.

I didn’t feel an uptick in sharpness compared to the X7000 which happened to have perfectly aligned panels and a very sharp lens, or the Epson UB6050 when it was in 24p. Compared to the Epson in 60p, the NP5 was sharper. If you remember, the UB6050 shifts 3x in 24p with its latest firmware reducing inter-pixel gap and providing quite a sharp image.

Dynamic Tone Mapping brought out more colour and more detail in both the dark and bright areas of the image. I am not sure it is as much of a revelation when comparing it to a higher-model X series JVC that’s been calibrated with a bright enough HDR curve, but it certainly provides the most user-friendly and set and forget projector platform in the industry.

I really enjoyed the fact that the unit was switching automatically between SDR and HDR and that HDMI handshakes were really short. Once set up for the screen, it was practically set and forget. Well done JVC.

All these observations were shared by a home cinema enthusiast friend who came to visit a few days later. I wanted to have another opinion from someone who had no preconceptions about any of the units. His observations were that the unit wasn’t subjectively sharper than either the Epson or the X7000 golden sample. DTM was noticeable in some scenes but it wasn’t as revelatory when compared to the X7000 – especially when fed Dolby Vision.

I need to note here that the X7000 was calibrated to perfection (really!) and had 3 custom curves loaded with all 3 having a 1000nit hard cut and fed both Dolby Vision (LLDV) and HDR10 content. The 3 curves allowed us to configure brightness of the tone map similar to the HDR slider on the Epson (and the NP5) – but with only 3 levels. However, we ended up watching the unit on tone curve 2, mid brightness. It had 2000 hours on its lamp which meant about 20% less brightness in high lamp than the NP5. However, at this brightness, resolving 4000nits is a big no no! What’s more, LLDV requires a tone map with 1000nit hard cut to be correctly displayed hence this configuration.

20 Jun ’22

It’s funny how memory works. I have given it a few days without turning on the X7000 and just had the NP5 on for movies. I wasn’t missing the contrast of the X7000 nearly as much as when I had them on one after the other. I can start to appreciate why people love the N series. The look of the image is quite organic. I just wish I could tack a 4K Darbee on it because image processing could be better. (Darbee promised us 4K processing which never arrived!) I think those with a Lumagen or MadVR would probably get much better detail and sharpness with the N series.

I watched Prometheus today in HDR and was thoroughly impressed with the tone-mapping on this title. There are some very dark but also very bright scenes and the NP5 was keeping up. There were colours in both darker and brighter scenes that are simply not coming through on other projectors as well. I’m still finding there’s a bit of lack of sharpness considering the upgrade to 4K panels and after aligning them pretty well (at least in the centre of the image) but I put MPC on Standard instead of High-Res 1 as Standard seems to have more contrast-adaptive sharpening and put it up to 5. This provided enough sharpness for my tastes.

Gaming Performance

20 Jun ‘22

I connected my Xbox Series X last night to check out gaming performance of the NP5. I think there’s certainly more detail retrieval here than with the Epson UB6050 or the JVC X7000. But again I was riding the MPC control to get a better sense of sharpness and depth to the image. I also upped CMD to high as motion resolution wasn’t great without it.

I tried out HDR with Forza Horizon 5 and Resident Evil 2 remake (and its new next gen patch.) I didn’t think frame adapt HDR could trip up so badly but it did with Forza Horizon 5: as the rewards were flashing on the screen, the sky was tone-mapped and lowered in brightness. This happened every few seconds and was incredibly annoying.

I didn’t think of switching from frame to frame to scene to scene setting which might have remedied the situation, just switched to static HDR. Static HDR is what you want anyway for gaming and a lower latency. This cleaned things up. (As a side note, Forza Horizon games are an edge case here – I recon DTM would not trip up on other games and it didn’t on RE2.)

HDR gaming performance was generally good and detail was excellent as I said. However, I kept wanting to switch to the Epson for its generally better image depth for this brighter game and that sharper LCD-look.

RE2 faired quite a bit better. It’s a very dark game so the JVC’s contrast near black gave the image heaps of detail and excellent depth. I could see detail that would be a smudge on the Epson with its lower on/ off contrast. This was a great way to play this game.

Ultimately I switched both games to SDR. I preferred Forza Horizon 5 this way, while Resident Evil 2 was better in HDR. But this is likely due to the HDR implementation in these games. The same is true on my Samsung QLED.

I still prefer gaming on my 85” QN85 by a wide margin. The JVC is great with darker games but it can’t provide the detail and image depth of an OLED (have a 65” B8 here too) or high end LCD TV.

If I had to choose, I’d still keep an LCD projector for gaming. The UB6050 isn’t nearly as good in 60Hz compared to 24hz, but the LS12000 remedies that. I recon that would provide a more exciting gaming session considering its higher brightness, better per-pixel contrast (that is less blooming) and that LCD pin-sharp look.

Having said all that, if you get the JVC without any other point of comparison, you will find its gaming performance excellent. But admittedly, I am spoilt for choice here and the JVC would only be my first choice for very dark games. And even then I’d choose my OLED or QLED before I’d choose the JVC unless it was a more cinematic affair. This is partly because a projector cannot quite provide the colour volume of a TV and secondly because I am not a huge fan of sitting in a dark room for a whole gaming session. Your mileage may vary.

Brightness

Here are the brightness measurements after calibration in both low and high lamp for the 3 units. Measurements were done with an i1 Display Pro Plus (rated for 2000nit max). Both JVC units had an auto-cal done with manual touch-up. This maximised calibrated brightness for both – especially the X7000 which needed it.

Please note that due to throw distances and zoom positions between different reviews, it is not necessarily comparable to other reviews and I admit, these numbers are a bit on the low side. However, this is why I wanted to measure all three at exactly the same throw distance (4m) throwing a 100″ 16:9 image. Epson measurements were in Natural picture mode. Dynamic can measure about 5-10% higher after calibration dependent on lamp and lamp age.

The reason these measurements are interesting is that you can expect the NP5 to reach the JVC X7000 numbers after 2000 hours as they are using a very similar lamp platform with the exception that the NP5 lamp might be cooled a bit better and might be ever so slightly more stable over time. But this is purely conjecture on my part.

It is also interesting to note that the current lamp on my Epson at 500hrs is not really much brighter than the NP5’s brand new lamp. Thankfully the LS12000 should fix that up nicely. I should be able to repeat the same measurements for the LS12000 when I do a proper calibration in a few weeks’ time using the same method I used here so we can compare.

JVC NP5
(30 hrs on lamp)
JVC X7000
(2000 hrs on lamp)
Epson UB6050 / TW9400
(500hrs on lamp)
Low Lamp904 lumens680 lumens981 lumens
(medium lamp)
High Lamp1323 lumens970 lumens1350 lumens

19 Jun ’22: one interesting thing with the NP5 (and in fact the NZ7 / 8 / 9) is that you can zoom the picture to the full panel resolution (which is DCI 4K, not consumer 4K – see more about this under resolution). What this means is that widescreen movies can be zoomed in to fill more of the panel and gain about 15% in brightness compared to not zooming in. This is certainly an advantage. The scaling doesn’t seem to hurt resolution from what I could see – or not to a noticeable degree with moving video content. The onboard scaler is actually pretty good. A higher-end scaler like a Lumagen or MadVR might well do even better here but this looks excellent nevertheless.

22 Jun ‘22: I measured if there was brightness difference between 100% white versus window patterns. JVC seems to be driving the pixels about 5-10% higher with 1-10% window sizes. This was measured as 15% with the laser based units but I didn’t measure that high with the NP5. However, what this means is that the pixels can be overdriven for highlight content. This is especially useful for HDR.

Contrast

The NP5 is a bit of a strange machine. The move to 4K panels has lowered the contrast performance of the N line compared to the X line. I felt that when JVC moved to higher-powered lamps starting with the X5000, the light path was no longer adequate to support the extra brightness and it resulted in extra blooming, and a drop in both on/off and ANSI contrast.

Some of this has been carried over to the NP5 which shows up in some streaking and blooming at times hurting inter-pixel contrast. There’s the extra problem of the new 4K inter-pixel grid throwing light around the optical block. This is partly what the new high-contrast optical block is trying to solve in the NZ8 and NZ9. Unfortunately, I think the lack of this in the lower-end machines actually hamstrings performance as the machine could do better.

However, when compared to the X5000 for example, there are improvements here for sure and the NP5 is a better machine all around – and so it should be at double the price. With regards to contrast, I found the X5000 quite a bit worse with both streaking and blooming than the NP5. However, when compared to the X7000 / X9000 / X7500 / X9500 / etc, the NP5’s streaking and blooming is worse.

Now for the good news. There appears to be an increase of ANSI contrast for the NP5 compared to the X series. I was so surprised, that I had to re-measure 3 times. But again, since inter-pixel contrast is worse due to blooming in the immediate pixel area, it comes out looking either a wash or worse in practice and actually hurts sharpness. More on this in a bit.

Where the NP5 excels however is in dark scenes, especially compared to the Epson. There is definitely better dynamic range near black and a black floor that is about 1/3 to 1/4 of the Epson’s at the lower brightness scale when viewed by eye. (Our eyes are not linear. The JVC has about a 6-fold advantage near black).

There is a fly in the ointment with this particular point as well when compared to the x series: the NP5, just like the N5 / NX5 before it – has only a 5x dynamic multiplier, so the iris doesn’t swing as low in dark scenes as that of the X series, creating more of a gap in contrast performance between it and the e-shift units.

To get contrast looking a bit better, I recommend lowering the manual iris to improve on/off, but not all the way as ANSI decreases a bit too much after -12 iris position. Putting the unit in a black cave and employing a higher gamma can also play to the strengths of the NP5.

Below are the exact measurements for both on/off and ANSI.

On/Off:

  1. 32,850 : 1 in High Lamp, High Bright and 0 iris.
  2. 22,941 : 1 in High Lamp, D65, 0 iris
  3. 22,781 : 1 in Low Lamp, D65, 0 iris
  4. 26,969:1 in Low Lamp, D65, -10 iris
  5. 27,613 in Low Lamp, D65, -15 iris

Dynamic Multiplier: Should be 5x and measured at a bit over 110,000 : 1. There was around a 1/3 to 1/4 drop in contrast around the corners and the extreme edges of the panels. I talk about in more detail under Quality –> Panel Uniformity.

As a point of comparison, the X7000 measured for on/off at a bit over 40,000 : 1 in both low and high lamp at D65, 0 iris. Dynamic multiplier is 10-12X with Dynamic contrast measurements coming in at over 430,000 :1. This is very much visible with content as the X7000 had no major contrast drop or bright corners near the edges.

Epson measured at a bit over 4000 : 1 for on/off in high lamp Natural. (21 Jun ’22: correction. My original measure was incorrect. On / off is 3600 : 1 in the centre of the lens. 43,000 : 1 dynamic contrast, Natural medium lamp.)

Comparative ANSI Measurements:

NP5X7000Epson
Low Lamp255 : 1221 : 1327 : 1
High Lamp256 : 1197 : 1330 : 1
-5 iris232 : 1NM*NM*
-10 iris229 : 1NM*NM*
-15 iris217 : 1NM*NM*
*NM = Not Measured

The X7000 ANSI was approximately 13% lower than the NP5. The Epson ANSI was 30% higher than the NP5. All measurements were done 2m from the lenses and repeated 3x for the NP5 measurements and 2x for the other units. The Epson could have done better by measuring at the lens, but I didn’t want to do that. Instead, I also calculated ambient correction for each measurement. Without this, the measurements would all be 10% lower, but still bringing the native in-room ANSI for the NP5 to 232 : 1 when measured 2m from the lens. I was pretty excited about this.

19 Jun ‘22: I re-measured ANSI from as close to the lens as my instruments allowed (about 40cms). The ANSI numbers above are accurate. There are places on the panel that reach as high as 292 : 1 and the very centre of the lens as 178 : 1. The rest are somewhere in between. But it averages out to around 250 : 1 to 255 : 1. So this is certainly an improvement over the previous N line and about the same as the X line. JVCs lose ANSI as the lamp and panels age (in spite of auto-cal) so this NP5 should reach the same place as the X7000 after 2 bulb changes and 2000hrs on the 3rd bulb. That’s not too bad – better than an average sample NX7.

21 Jun ’22: I re-mesured the Epson ANSI closer to the lens as well. It averages out to the above numbers still so measuring from 2m was accurate. It goes as high as 413:1 near the edge of the panels to 270:1 in the centre of the panels. It’s actually the same in terms of panel uniformity for ANSI as the JVC. Except, bright corners cannot be seen towards the edge of the panel as on/off is limited to between 3400: 1 to 3600 : 1 (my previous measure was incorrect for on/off).

I also measured dynamic contrast for the Epson which came in at 43,000 : 1 in Natural medium lamp, 46,351 : 1 with iris at -7 and not measurable with my instruments in dynamic mode (medium or high lamp, it doesn’t matter – the iris shuts down almost all the way). Epson’s quote of 1.2 million to 1 is likely accurate here. There’s one more thing to try to measure dynamic contrast in dynamic mode, but that’s all I had the patience for today.

Sharpness

Pixel Resolution: The resolution of the NP5 comes in at full DCI 4K (4096 x 2160 pixels). Therefore, with consumer 4K (3840 × 2160) pixel-mapped to the DCI 4K panel, there are grey projected bars on the left and right side of the image. This can be visible if you don’t have proper masking on your screen. It is a slight annoyance on a wide screen (such as cinemascope) but it is not too bothersome most of the time.

19 Jun ’22: As mentioned under the brightness section, the image can be zoomed out all the way to the panel’s full resolution. This is a great convenience and works really well for widescreen movies or any content with black bars (e.g. some Netflix content. )

Panel Alignment: The panels were out by a fraction, but using fine pixel alignment reduces chroma resolution on a projector so I preferred the unit with it off. I did try it to see if it improved sharpness, but all it improved was the patterns and hurt detail retrieval in content. As always, unless a projector has motorised panel alignment, do not use the fine alignment controls. They are there to make patterns look lovely and stop warranty requests, but not there to objectively improve image quality. Using full-pixel alignment is fine but it wasn’t needed here.

19 Jun ’22: it turns out full pixel alignment is broken on the NP5. It seems to shift in some directions two pixels instead of one. So I had to use the fine panel alignment and move blue both to the right and down by 8 fractions (which equals to exactly one pixel in each direction). This cleaned things up on sharp text and graphics (not a huge difference with moving video), but unfortunately, blooming still remains which hurts pixel-level / per-pixel contrast.

Lens Sharpness: I feel that the lens is not really adequate to resolve a 4K pixel grid and therefore hurts sharpness. In fact, I feel the lens is better quality on the UB6050 overall which aids in sharpness in a big way and that unit has excellent focus edge to edge. Not so with the NP5 unfortunately.

21 Jun ’22 update: I decided to try a different tact and focused the lens at 1/3 picture widths. This defocuses the centre a bit more, but got more of the picture in focus. This looks a bit better. It’s unfortunate that the lens is’t super high quality. If Epson can include a sharper lens in a product 1/2 the price (and it’s sharp edge to edge), there is zero excuse for this! I do mean zero at this price! That’s my take on it.

Where’s me pixel grid? You can kind of make it out but look at the colour fringing and blooming. Likely the lens is adding to it greatly. You can faintly make out some pixel forms on that F, but not very sharply.

Panel Noise: The X series projectors were well known for some dither noise on the panels. However, this was very unit to unit dependent. The X7000 golden sample I had for the review has pretty much zero visible panel noise. The x9900 we had on hand had a higher panel noise. If panel noise is low on an X series unit, e-shift also looks pretty clean. If panel noise is high, e-shift tends to make it worse. I was expecting the NP5 to be completely absent of panel noise, but this wasn’t the case. There was some instability to the image as you were pixel-peeping. It certainly wasn’t as bad as with some of the x-series units, especially with e-shift on. It is certainly calmer than the panel noise on the X9900 I had with me. The X9900 also has higher e-shift picture noise in general so it’s a double whammy compared to the X7000 / 9000 / 7500 / 9500. However, once e-shift was turned on on the X7000 golden sample, the NP5 had more stability to the image, if ever so slightly. This was only noticeable close up to the screen, however. This is in contrast to the X9900 where e-shift is not nearly as transparent.

Detail Retrieval: This is where the NP5 was doing better than the e-shift units, or the Epson. There were particular frames of video where you could see a bit more textural detail. But I want to stress that due to lack of pixel-level contrast on the NP5, it didn’t look sharper than the X series units and or the Epson. In fact, at times it looked less sharp in spite of some better textural detail. I would say that for someone upgrading from an X5000 or X35, the picture will have better detail and will also look sharper. Anyone upgrading from a 7 or 9 series might well be disappointed dependent on whether you had a good or not so good sample and what your luck is with the NP5.

Pixel-level Contrast: I believe this is a big Achilles heel of the NP5 compared to the X-series units. Due to worse inter-pixel bleed on the 4K panels, there is less inter-pixel contrast. What makes matters worse is that JVC removed Clear Black image processing on these units, which is contrast-adaptive edge enhancement and was there to help D-ILA keep up with DLP and LCD with regards to pixel-level contrast. It would be needed here more than ever in my opinion so it is rather strange that this was removed.

20 Jun ’22: I actually now think that Clear Black was folded into MPC Enhance setting on the Standard and High-Res 1 options. On resolution patterns, it looks very similar to Clear Black when riding the MPC upwards. This is curious. It does help having MPC higher on these units than the X series as it helps counteract the inter-pixel bleed somewhat. I still have this feeling that the move to 4K panels without a major overhaul of the light path was not the right choice on JVC’s part.

Sharpness – Conclusion

I am wondering if Epson had the right idea with regards to 4x pixel shift instead of moving to native 4K chipsets. I am just not sure if the on / off contrast drop and inter-pixel bleed is worth it for slightly better detail retrieval. It doesn’t end up looking sharper overall in most cases, in spite of the better detail.

Mind you, LCD has a major advantage with pixel-shifting: a much higher inter-pixel gap, which means pixels don’t overlap as much as with LCOS chips. It also solves LCD’s biggest issue: the chicken wire effect when sitting up close.

So after having seen the NP5, I won’t be hassling Epson to move to native 4K panels. Be careful what you wish for I say, as all projectors lost contrast when moving to 4K, including DLPs. So the cost might outweigh the benefits. It’s only slowly that these projectors are remediating the backwards steps in other picture quality areas – slowly being the operative word here!

For someone who is just getting a projector for the first time, this will not be an issue as the NP5 will absolutely delight. They simply don’t have a point of comparison or even memory of what yesteryear’s projectors were capable of – for the same price as the NP5. But I see this as a 2 steps forward, 1 step back proposition or even 1 step forward, 2 steps back? You decide!

Colour

REC.709: the unit calibrated beautifully to REC.709 and colour reproduction is a strong-suit of JVC. No issues here. However, this only happened after an auto-cal. Pre-auto-cal the colours were off quite badly.

BT2020 / P3: Even though it didn’t reach full saturation points for green and cyan, the gamut mapping for P3 looked pretty good after auto-cal. There is no filter here for the P3 gamut.

Image Processing

HDR Tone-Mapping: The tone mapping for HDR is good and static tone maps have a 10-stage slider to configure clipping point and brightness – taking a page out of Epson’s book. This is even better than the previous N gen and certainly better than the X series in its default state without custom curves.

DTM: Dynamic Tone Mapping improves contrast in dark scenes by giving better delineation between dark parts of the picture. It also brings more of the colour out in very dark areas. Additionally, it insures that highlight detail isn’t clipped in very bright scenes. Overall, this is a set and forget solution and for someone who doesn’t want to mess with calibration and custom curves, this provides excellent results out of the box – provided you know how to use auto-cal to get it looking nice and tight.

I was impressed that the NP5 handled even 10,000 nit material with ease without breaking a sweat. This is top notch tone-mapping.

MPC: The enhance setting works better here than the X series units in that it can be put up higher without causing over sharpening and artefacts, likely due to the smaller pixels. It is needed badly too, as the pixel grid is not fully resolved – or not very sharply – and there is an absence of Clear Black pixel-level contrast enhancement. I put it up to 5 but it could probably go further up and still be relatively transparent.

CMD: CMD works great on the low setting even for movies. High setting is great for sports or other fast-moving content. I couldn’t see a huge improvement but CMD has been great since the X7500 / X9500 (with the firmware update). It still reduces greyscale / gamma resolution by one bit approximately resulting in some ever so slight banding. But this is only visible with patterns, thankfully. The pictures below demonstrate what happens with CMD still (on a difficult pattern).

Motion Enhance: Motion enhance works well on the low setting, especially if you don’t like CMD. It helps reduce blur due to motion. This is noticeable to some and not to others. Your milage may vary. I personally like it with movie content when doing critical viewing – and don’t want CMD messing with my greyscale / gamma resolution.

Usability

On / Off times: Generally fast, about 30-45 seconds for on and the same for off.

Sync times: Sync times are very fast, around 3-4 seconds. HDMI handshakes are pretty strong as well. Had no issues. This is a major usability improvement over the X series for example, where you could make a cup of tea while the thing negotiated an HDMI handshake. This is now as good as the Epson.

User / Menu Interface: Overall pretty good. I had no trouble adjusting to the new interface from the X series. The only gripe I had was with installation modes. I kept changing the lens settings thinking I could just save a new memory slot and then realising the new system asks you to change installation mode first then make your adjustments without having to save. Slightly confusing after having gotten used to the old system on both JVC X series and the Epsons. Not a huge issue, just an observation.

Calibration Functions: Auto-cal is great, but its manual is still really poor. The forums provide some help because of some excellent testing done by forum members on different versions – not to mention the ability to use alternative meters with some software hacks. But this is still not easy to use for the average person. I would like to see an official guide that goes through all the do’s and don’ts.

Calibration controls are good and extensive otherwise, as long as you are willing to dig into auto-cal. However, I would like to see the parametric gamma controls back on board the unit instead of in the auto-cal software. They were removed when HDR was introduced on the X series for some reason. It’s a bit finicky to have to wait around 7-10 seconds for every click for the software to transfer the 3 channels of gamma.

Also, as I mentioned earlier, I would like to be able to direct-edit internal tables for the base gamma, please. As it stands, we are at the mercy of auto-cal getting gamma perfect for DTM. Even a .1 change in low-end gamma can impact perception of contrast for higher-contrast but brightness-limited displays – which is what JVCs are – and auto-cal regularly has such deviations. I would like to be able to tweak auto-cal results (using 256 points or more) without resorting to imported gamma slots.

20 Jun ’22: Auto-cal also requires you to calibrate each iris range (which is a hassle), but there’s a new “reflect to all” function which will copy gamma and colour settings across to high / low lamp as appropriate. This is helpful if you are in a rush – at the expense of accuracy.

Installation Flexibility: Installation flexibility is excellent. However, be mindful of the weight if you are mounting the unit and make sure to get a mount that is rated higher than the weight of this unit, just to be sure.

Heat and Noise: Generally pretty low in both low and high lamp even as the unit was sitting 1.5m from my head on a stand behind the seating position. There is no e-shift noise which is a plus.

General Quality

Panel Uniformity: As I mentioned, there was almost 1/3 to 1/4 contrast drop at the edge and corners of the panels arriving at a 16,386 : 1 on/off contrast at the very edge. This isn’t exactly exceptional. Epson does a lot better with panel uniformity but this is only true most likely due to Epson’s much lower on/off. This is because panel uniformity when measuring ANSI contrast is about the same on both units.

Also, there were bright corners that are visible with some content, mainly on black frames or fade to blacks. It doesn’t help that the iris doesn’t close down as aggressively as on the X series. This can be remedied partly by closing down the manual iris. I settled on -6 eventually, for both SDR and HDR.

Comparison to Other Projectors

Overall Thoughts

The JVC NP5 is in a very strange position. It costs as much as the N7 from the previous generation and even a bit more than JVC’s last few X generation flagships: the X9500 / X9900.

So while this is a 5-series JVC, we cannot simply ignore the fact that some aspects of the technology here might be going backwards as far as the cost versus performance goes. It does feel like two steps forward, one step back.

Upgrading from JVC X Series – 7 or 9 Model Lines

I would not advise an “upgrade” from any of the X series 7 or 9 models such as the X7000/9000/7500/9500/7900/9900 (EU/AU designations). This is especially true if you have a great sample that is sharp, without much panel noise or bright corners. The on/off contrast drop is significant if you get an average sample, and I am not sure that the resolution advantage is that visible. In fact, if you have a good sample HDR-capable 7 or 9 model, you are best to stick to it and instead look at an external DTM solution if HDR is an issue for you. This can be as cheap as an HDFury with Dolby Vision LLDV support or as expensive as a Radiance Pro or MadVR Extreme. I am personally using LLDV and I am quite happy with it. What’s more, I don’t feel like I need to “suffer through” a movie if it’s in HDR10 and a few highlights are clipped. I think some people obsess about this a bit too much.

The 7 and 9 model lines have enough contrast in the low-end to be able to support HDR10 without DTM – as long as you use bright-enough curves. This cannot be said for the NP5, which would look worse without DTM.

But all of this assumes you know how to calibrate your unit, upload custom curves and modify them for your environment. Simply downloading curves from the forums isn’t quite going to cut it.

The issue with the X series JVCs is that they crush blacks with their factory calibrations. They need to be auto-caled and then the low-end gamma corrected based on re-profiling gamma, even for those custom HDR curves. I personally use a spreadsheet to calculate the new curves after re-profiling gamma as even small gamma deviations can hurt HDR on the X series in my experience. I find this method much faster than having to do PQ calibration for 3 different curves while also getting the benefit of Arve’s excellent curve building tool.

Let’s just say, you cannot auto-cal a JVC and be done with it. In this sense, the NP5 is very hands off and you have to live with the auto-cal results because you cannot edit them for DTM.

Upgrading from JVC X Series – 3 or 5 Model Lines

I think this is a great projector to upgrade from an older non-HDR capable X series 3 or 5 model line JVC. I would even consider it from an X5000/5500/5900 as the issues are shared between those units and the NP5, but what you get is DTM and a sharper projector. Now does it worth almost double the price of what you paid for those machines? Well, that is for you to decide. But you may well be delighted with the upgrade the NP5 provides.

Switching from an Epson TW9400 / UB6050 / UB5050

The NP5 is better with movies in general, especially sci fi, and has better HDR handling. Its colour performance and dynamic range near black is also much better.

However, the Epson is better with mixed content, such as normal TV, sports, gaming, etc. The JVC can look a bit flat with brighter content or with any light in the room. The Epson can handle a small light on or even some daytime viewing.

I think you need to audition them together to decide. What some have said was that they simply LOOK different. I totally agree with this statement.

I continue to enjoy using the Epson for TV shows, YouTube, Apple Fitness, daytime browsing, some gaming, even drama movies. Even though it doesn’t reach as low in dark scenes, it does provide less pixel-bleed than the NP5 and can look sharper at times.

The X series is used for Dolby Vision and HDR movies from disc as well as normal SDR Blu Rays. Any movie that has a lot of dark content automatically goes to the JVC X series now. The Epson MAY get upgraded to the LS12000, however. But this is not a pressing decision for me.

Compared to Epson LS12000

Now this is where the rubber hits the road as they pretty much have the same street price, with the Epson being ever so slightly cheaper in some markets.

The LS12000 (see my impressions here) calibrates quite a bit brighter, it has excellent laser dimming (in my opinion) and may appear sharper than the NP5 due to the better lens, higher ANSI, better pixel-level contrast, less issues with panel alignment (since the panels are pixel-shifted) and panel uniformity. It has a simpler light path that can be produced with less unit to unit variance.

However, the LS12000 – like the UB6050 before it – is not nearly as strong on colour or HDR processing as the NP5. There is no full-scale DTM on board, even though dynamic gamma and the better tone-mapping puts it closer to the NP5 than the UB6050. If configured in Dynamic mode, the laser dimming of the LS12000 is a quantum leap forward compared to the UB6050 – IF and this is a big IF – you are not sensitive to dynamic gamma and light manipulation. Some people are – so you must audition it.

The LS12000 does not reach as low as the JVC NP5 with regards to on/off contrast however, so dark movie scenes will still exhibit some greyness. If used in a pitch black dark room, I would say that masking of black bars is a good idea with both the UB6050 and LS12000 for maximum sense of contrast and to ensure you are not taken out of the movie during really dark scenes because you get distracted by the black bars.

Having said that, the NP5 is not yesteryear’s X9900 by any stretch of the imagination – even though it is the same price!!! – and you will also see projected black bars, albeit fainter than with the LS12000 or UB6050. Our eyes’ sensitivity is NOT linear however, so it won’t look like it is 6x worse due to the on/off contrast differences, but it will be there. So if you pick the LS12000, and you watch a lot of movies, think of a masking solution that will work for you!

Compared to Sony’s New Lineup

I have only seen the bulb version of its competitor, the 290 and 590ES. The Sony’s tone mapping and colour reproduction is on another level even with JVC’s DTM. The below video might shed some light on the differences with the new laser units. The below video might also shed some light on the softness and at times flatness I see on the NP5. When viewed on its own, not a huge deal but JVC might well get destroyed by Sony this time around – DTM or no DTM.

https://youtu.be/QgdBEFJ8C-g

Switching from a DLP

This is a difficult one. You are giving up a lot of ANSI contrast and image depth in brighter scenes just for better dynamic range near black. I think this very much depends which one you favour. If you watch a lot of dark movies, the NP5 might be appropriate. If you watch a lot of mixed content, especially Sports, you might want to stick to your DLP – or better yet, upgrade to the LS12000 for better blacks.

Review Summary & Scores

I wanted to assign some review scores from my perspective after spending 6 weeks with the NP5. 

These are subjective but I rated all three units in the main aspects of performance to establish some relative scores.  

The totals only reflect what we could assign if all aspects were rated equally – so there isn’t any weighting. You need to decide which aspects are the most critical to you and adjust the weightings accordingly. 


NP5X7000UB6050
SDR playback796
HDR playback876
Brightness778
Contrast8106
Sharpness 1080p789
Sharpness 4K877
Colour987
Image processing877
Usability1067
Quality788
TOTAL79 / 10077 / 10071 / 100

Overall, if rated equally across all aspects, the NP5 wins, due to its HDR processing, great usability and generally balanced performance. Its colour processing is exceptional, even surpassing the excellent X series. 

If you are after usability, or you don’t have the knowledge or ability to calibrate the X series line to get the very best performance out of it, the NP5 might be a good choice. It is incredibly easy to use, with short sync times, automatic picture mode selections and automatic HDR processing, making this the most hands-off projector I have ever used. 

One benefit of the NP5 over the previous N models is that it is in its second year, with updated internals and a new HDMI board which should be less prone to failure (HDMI board failures were an issue with the previous line.) There’s also no dynamic iris yellowing issues I could see, also an improvement from the previous N series projectors. (As a disclaimer, I am not super sensitive to gamma manipulation artifacts so might still be some but I didn’t personally come across any and I did look.) 

However, if you are a contrast freak, and you are still on an HDR capable X series with the ability to fully calibrate it and get the very best out of it, you might want to seriously consider if the upgrade is worth it. Resolution might not be as huge an upgrade by itself so I certainly recommend an audition before deciding. 

Clearly, the UB6050 is half the price. It cannot compete on colour or HDR processing, but it is an exceptionally good value, especially for mixed use. It has impressive super resolution processing which is very effective when used judiciously. It is weaker on on/off contrast but can look very well defined due to its excellent per pixel and good ANSI contrast. It has more of a digital look compared to the JVCs’ more analogue look. It really depends on what you prefer. I personally need both in my setup due to extensive mixed use – and not wanting to burn the lamps in the other projectors for anything other than movies. 

The above is clearly only my opinion and others might well disagree. However, I guess there couldn’t be much higher praise than that I have decided to keep the NP5 in my collection for now (along with the X7000 and the Epson TW9400 / UB6050) and enjoy it until laser projectors mature a bit more (or JVC move their production to Japan).

If nothing else, others in my family can operate the home theatre while I am not at home without fiddling with the projector controls. Just switch on the NP5 and enjoy. It really is that simple.

Settings and Calibration in Detail

My tools are HCFR, i1 Display Pro Plus (2000nit max) and i1 Pro 3 (5000nit max). I would encourage the reader to learn display calibration. Especially if you have or are considering a JVC, I feel that this is a must. I can be a bit clumsy with calibration, but I always get there in the end. I will continue to share my mistakes in this regard so you can learn from them.

If you would like to learn to calibrate your projector, you will find some resources here.

SDR Settings After Calibration

I used low lamp, manual iris at -6 and auto iris on Auto 2, which does less gamma manipulation. MPC on standard mode and MPC Enhance on 5, everything else at 0. CMD on low, motion enhance on low.

Since calibration was done with the help of auto-cal, sharing of calibration settings makes very little sense.

HDR Settings After Calibration

I used High Lamp with iris at -6, frame by frame tone-mapping with 0 tone mapping level (auto is default). Auto-iris on Auto 2. MPC on high-res mode 1 and MPC Enhance on 5-8, everything else at 0. CMD on low, motion enhance on low.

Since calibration was done with the help of auto-cal, sharing of calibration settings makes very little sense.

Calibration Baseline

Just to be absolutely clear, all dynamic processing or light manipulation must be disabled before calibration so dynamic iris was disabled before auto-cal.

SDR Calibration

SDR – Out of the Box

Before Calibration, greyscale had a nasty green push and gamma was like a roller-coaster.

All primaries except for yellow had really high dEs. Look at the delta L for luminance. Pretty gross errors there. I wonder if someone slipped on the controls in JVC’s factory.

Now we see why. The gamut is nowhere near REC.709, especially Red. Green and Blue have also gone for an adventure. This is not correctable with only on-board controls. This is the worst I have seen out of the box for a JVC, including any of the 3 series.

SDR – After SDR Auto-cal

Ok, so auto-cal messed up everything below 10%. Aaaarg! No biggie, this is correctable if we are only doing SDR in low-lamp – not HDR. Otherwise this needs an auto-cal re-run. This is why I said you cannot just run auto-cal and be done!

But let’s have a look at our major issue earlier, the gamut and primaries. It’s looking MUCH better and now the saturation points line up very well. We can do a bit of work on this but this is now looking great.

I did re-run auto-cal after this section was written but didn’t take a new reading – gamut still looked excellent with only minor touch-ups needed. Greyscale lined up better to go into a manual touch-up.

SDR – After Manual Touch-Ups

It is not perfect but everything is under 1dE except for 100% white, which is under 2dE. Good enough. Any more tweaking won’t be visible with content.

Gamma needs a bit more work…

After gamma manual controls in auto-cal. Yes, we could get it tighter still, but I’m getting older by the minute.

I did a final 20-step greyscale / gamma run to confirm results but forgot to save it. This is complete tardiness on my part. Apologies.

And finally the gamut after manual tweaking. It is lining up pretty well considering where we started. All dEs under 3. Major improvement for cyan and blue. Cyan is my favourite colour so it has to be in order…

HDR Calibration

HDR – Out of the Box

Again, my tardiness hits. After having seen how bad the out of the box measurements were in low lamp, I simply went ahead and did autocal for high lamp as well. I didn’t have any more time to waste. But I should have done a profile so we could compare. Apologies.

HDR – After Auto-Cal

Thankfully auto-cal went much better in high lamp than my initial low lamp auto-cal, which tells me the sensor wasn’t close enough for the low lamp measurements initially and I forgot to check for 0.000 readings during the first gamma cal in low lamp which was the issue most likely.

In any case, excellent results for greyscale in high lamp.

All dEs for greyscale under 1. Good job!

Gamma 2.2 is lining up well. Just as well because it’s this baseline that both DTM and the on-board static tone-mapping is using and it is not correctable for DTM. JVC, if you are reading, I want manual control over internal tables, as we could get this better still.

The BT2020 / P3 (P3 within the BT2020 container) lined up alright, but there’s certainly room for improvement here.

Just two examples on the kind of errors. They aren’t super bad, but they are there.

HDR – After Manual Touch-Up

The only thing I ultimately want to touch up here is the gamut which is a bit wishy washy. Here are the results after a touch-up: not too bad.

All colours under 3dE at 100% saturation, except for green and cyan as understandably the NP5 does not have a P3 filter that improves those primaries. Content should not suffer as linearity is otherwise excellent.

And finally the saturations for proof that everything is under 3dE. The below is a gallery so you will have to click the individual images. We could improve the look of the chart further by targeting more precise x,y coordinates, but at the expense of luminance errors increasing. Moreover, tacking a 3D LUT onto an NP5 is NOT the use case of 99% of users so not happening here. It is unlikely to be visible anyway. One option could be to do a colour-only autocal with the i1 Pro 3 but I’m pretty happy with this without suffering through another auto-cal then manual touch-up run. We have a couple of more grey hairs on our beard and less actual hair on our heads. Time to watch some content…

4 thoughts on “JVC DLA-NP5 Projector Review

Add yours

  1. After receiving the NP5 yesterday, I kinda see where you are coming from. I expected it to be sharper too.

    While it is better that the Epson 4050 before it, blown away I am not.

    Like

    1. Hehe. It will grow on you like it did on me. Align the panels using fine pixel align in 8 step increments. Also, try my recommended settings in my other article. See how you go! 🙂

      Like

  2. Google is doing you a disservice sir. One of the last sites to come up, but absolutely the most thorough and helpful review of all. Your pragmatic comparison of competing options, even adding a nod to the lowly DLP, is illuminating. Thank you.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create a website or blog at WordPress.com

Up ↑

%d bloggers like this: