Sunday, December 25, 2016

No More Room In The Case: DAWN OF THE DEAD 4K Remastered Blu-ray Review

Midnight Factory's 6-disc Blu-ray/UHD-HD collection for DAWN OF THE DEAD is an extremely frustrating package, all in all. On the one hand it's simply the best presentation of the film that's ever existed... on the other, it's a huge mess in its' own right, and if you're waiting for clarification on if it's worth the asking price, it really does depend on what you've expected out of it.



As has been stated elsewhere, the presentation of the George A. Romero approved "Theatrical Cut" and the "Extended Cut" are a 1:1 match with the audio and video files found on the substantially more expensive Japanese box set released by Happinet in 2013. The M2TS files are literally the same here - copy pasta, all the way down. Both transfers are pulled from the same 35mm elements, with the Extended Cut being the 'source' and trimmed down to match the length and audio mix of the Theatrical Cut. The only new additions to these discs are Italian forced-subtitles, and a Region "B" lock.

Mind you, the entire 6-disc Midnight Factory box set under scrutiny here costs about the same price as just the Extended Edition Japanese Blu-ray, so if you're clever enough with a BD-ROM to bypass the forced subtitles it may well be worth it to go for all the way. On the other hand, if you only want the Extended Cut and could care about the European Cut, that release doesn't have forced Italian subtitles. So invest accordingly.


Not something I usually talk much about, but for a $50~ import with a half-dozen discs, I guess I may as well...

Plenty more photos available at High-Def Ninja.
(And yes, they all have the same stupid disk art.)

Despite this being a hefty 6-disc set, the whole collection is contained in a single, somewhat opaque blue case the same height and depth, but double the width, of a standard BD keepcase. The case has two flaps - one for each side of the case - each holding 2 discs, with one disc in the front and rear of the case proper for a total of 6. To call it "complex" would be an understatement, though I assume the 4-disc edition comes missing a flap, and is likely less of a pain in the ass to deal with.

The Italian language booklet (seemingly containing a new essay, and interviews with Dario Argento and George Romero) covers the 4K UHD, and the Post Cards - kept in a small, sealed plastic bag, just the way I like it - flop somewhere around in the middle, which means you'll be attacked by them every time you open the case. It's kind of a shit-show to be honest, and none of this is helped by the fact that the case itself is chunky and only likes to close when it seems to want to close when it feels like it. Not gonna' lie, I've had worse, but this may be the most cumbersome piece of "Deluxe" packaging since that equally ridiculous fat-case for Dust Devil on DVD nearly a decade ago.

The only real upside is the fact that - much like Arrow Video's old "Special Edition" Slipcases! - the case has a total of 4 poster art pieces you can swap around as you like - the iconic American, Italian, Spanish and Pre-Release posters are all given a full panel. The art itself is decent quality, amazing historical stuff, and... kind of pointless, considering you don't have a window to show any of it off from. Goddamn, Midnight Factory, get your shit together on this thing!  Instead it's wrapped in a thin, glossy cardboard slip the International Post loves to ding up as much as humanly possible - but it's that weird Refen themed art, so, fuck it, I can't convince myself to be as upset as I was at the less-serious damage on my copy of the Anchor Bay Ultimate Edition.

If you're a packaging fetishist and you're more patient than I am, I say wait until the inevitable German port of the same content in a Mediabook, or 3D Steelbook, or a Leather Bound Zombie Skin Edition - pretty much anything would be less excruciating to deal with than this goddamn thing. I don't ask for much from a box set, just functionality, and somehow this has come up just short enough to annoy the crap out of me.

But we all know I'm ripping these to my hard drive to fuss with them anyway - and may have gotten so sick of waiting I grabbed a Nautical Freelance Copy to start this review. Moot point, but hey, full disclosures and all that...


As expected, while the bonus features with Tom Savini and Nicolas Winding Refen feature English audio and Italian subtitles, the rest of the interviews with Dario Argento, Claudio Simonetti, the LVR restoration crew, and the archival interviews with Alfredo Cuomo and Claudio Argento are in Italian with no translation. If you're surprised by this, you probably haven't imported many Italian DVDs.

Archival interviews and trailers are all SD PAL, and as such will likely not play on most North American displays. Not much to see here unless you speak Italiano, which - considering the countless hours of bonus material already available in English - is really no skin off my nose. I was hoping the Restoration featurette would show some cool before/after samples, but unless you really want to see a graded vs ungraded shot, that's as good as it's going to get.

The Open Matte transfer is arguably the HD highlight of this collection, and is - for all intents and purposes - identical to the 1.85:1 transfer aside from having all that sweet, sweet headroom. It's a little bizarre knowing that the 1.85:1 area on the center-right of the scan has been cleaned of dirt and scratches while the rest of the transfer hasn't, but as the 35mm Master Positive has been kept in relatively good shape over the last nearly-40-years, there isn't a whole lot to complain about that can't be equally leveled at the "Final" 1.85:1 master.

Arrow Video's variant of the "Divimax HD Master".



Before we talk about the new master - the good, the bad, and everything in between - it's worth noting why this was such a big deal. This is hardly the first time Dawn of the Dead has been given a new, high-resolution transfer, but it's never quite gotten the attention and care it so desperately asked for...

In 2004, Anchor Bay restored the American Theatrical cut of the film to what was - at the time, certainly - one of the most impressive presentations of a 70s horror film outside of the big studio wheelhouse. Scenes were corrected on a shot-by-shot basis to get consistent, rock-solid black levels and somewhat neutral skin tones, despite highlights often having a too warm (and sometimes blatantly brown) cast. Sadly, as this was still the era of DVD where heavy film grain led to terrible, clumpy compression, heavy DVNR that led to blatant ghosting and edge-filtering to prevent the whole film from looking "soft" as a result led to an unnatural, digitally processed look that's hard to look past. Going forward, this will be known as the "Divimax Master".

Perhaps only adding to the confusion were reports that the Arrow Video Blu-ray - which also included the two "Alternate" cuts on PAL DVD - was regularly cited as being 'higher quality' than the Anchor Bay transfer. This is certainly a matter of taste, but the biggest difference between the two was that Anchor Bay later applied an additional pass of scratch-removal tools, which did their job (somewhat) but also caused the usual high-frequency detail loss that comes with the territory. The master itself is exactly the same otherwise, and sadly, most of the damage was done during the initial creation of the master over a decade ago.

The Extended Edition of Dawn of the Dead was first released on Blu-ray by Happinet Japan in 2013. I suspect the Happinet transfer is actually the restoration that served as the foundation for Dawn of the Dead 3D, a conversion project that was first announced as in production all the way back in 2007! That would, at the very least, explain why the new transfer is so distractedly bright... hmm... either way, this will be the "Extended Master" when comparisons are unavoidable.

The Happinet release also included a "New Master" for the shorter American Theatrical cut - in reality it was the same exact scan as the Extended master re-cut to the soundtrack of the George A. Romero approved 1979 Theatrical Cut, since the former was effectively just an unpolished, longer cut of the latter anyway. As such I'll refer to the 2013 Theatrical Master and the 2013 Extended Remaster as if they were one and the same, because... well, they are.

Since we're being slightly pedantic anyway, I'll point out that while Elite erroneously released the 'Extended' version on Laserdisc in the 90s as the "Director's Cut", Romero has since clarified that he willingly trimmed the extra 7 minutes or so of footage and that he considered the 127 minute "Theatrical Cut" his personal prefered version of the film. Romero seemingly had no direct involvement in the creation of the 118 minute "ZOMBIE" cut released in most of Europe.

Presumably Alfredo Cuomo and Claudio Argento hold the master positive that makes up "ZOMBIE", the Dario Argento approved 118 minute version that serves as the leaner, meaner, less satyrical answer to Romero's more pensive and playful take on the zombie apocalypse. I like Zombie just fine - I'm not convinced it's the "best" version of the film, no, but I'm also convinced that each version has their own strengths, and Zombie is no different - though for better or worse, the elements appear to be no better or worse than the 35mm CRI that's seemingly served as the basis for every new home video transfer of the film dating back to the mid-90s.

Foot Note: Anchor Bay possibly created a new 35mm print in the late 90s, but the inclusion of "extra" scenes on their initial video releases - which were later removed for the 25th Anniversary Edition DVD - suggest they either have been cleaning that old girl up for decades, or perhaps struck a low-contrast print from the same vintage CRI. The connecting thread is the moire whorl patterns that aren't present on the ZOMBIE elements, and thus can't be on the camera negative. Take a look at Danny's Dawn of the Dead Collector's Blog for some fascinating info on this seemingly rock-solid theory. 

Enough context!
Let's do this.



First the good news: The promise of an all new, high quality 4K scan of the "Zombie" Master Positive print have been delivered exactly as promised - by which I mean it's not an upscale or anything like that. The Japanese "4K Remaster" of The Crow has proven that this isn't always the case on overpriced imports.

It's clear that efforts have been made to maintain the overall integrity of the element at a scan level, from the correctly centered 1.85:1 aspect ratio to the carefully regulated highlights and a level of consistency between shots for things like fluorescent lighting and time of day that no print of Dawn of the Dead prior has ever seen. In many ways it may indeed the best, most natural looking version of the film ever released... but that doesn't mean it's a particularly attractive presentation on its' own. Romero's principle of substance over style means the original lighting and focus on the film was never particularly great to start with, and anyone expecting Dawn of the Dead to be a crisp, glowing film for the 21st century is probably not someone particularly familiar with the film to begin with. It's always been an underexposed, inconsistent, grimy looking little movie and while there's a few problems with the presentation, it's clear the overall project did everything it could to preserve the occasionally underwhelming world George Romero and DP Michael Gornick created.

When I first looked over the 4K European Remastered 1.85 disc, my heart sank a little. Shadows were an indistinct gray, flesh tones had a dull, washed-out look compared to every prior print worth mentioning, and while the appearance of visible on-set lighting was no longer a big, hot colored blob of light, I found myself struggling to call what we have an "improvement" - from any raw, aesthetic standpoints, at least. I suspected that there was an incorrect gamut conversion at first: Outside of the optical titles none of the unexposed areas of the print are ever "black", and while the new 4K sourced BD looks perfectly serviceable during well lit exteriors, the vast majority of the while film looks very... pale, to put it kindly.

It wasn't until I looked the whole thing over with a histogram that it started to make sense - even if I find myself frustrated with the result. While the Extended HD Master released in 2013 seemingly favored midtones over highlights or shadow details - getting a consistently vibrant, natural looking color pallet at the cost of virtually every lamp and light fixture being blown out into a glowing vortex. In short, the Extended HD Master pushed the exposure to increase contrast, and with it find some sensible looking midtones at the cost of making very bright scenes way too bright, and very dark scenes a little on the thin side. There was a little clipping on things like the mall lighting in the 2004 Divimax transfer as well, but nothing outside the margins of error (or good taste).

By comparison, the new 4K transfer seems to have preserved the highlights perfectly, with the brightest scenes consistently topping out at IRE 100, just as they should, with the muddled midtones and weak black levels the natural result of leaving things as they are. The result - while technically sound and perfectly respectful to the source material... still strikes me as very underwhelming, from any aesthetic point of view at least. The cool cats at CAPS-A-HOLIC have made many of the things I'd otherwise write completely irrelevant (huzzah!), but all the same, their comparisons can be a bit limited and may not always tell the full story, as might be the case here, so I've included a bunch of purty pictures for your personal pleasure.

I've tried to get a wide variety of examples so that anyone on the fence about taking the plunge can make the right call.

But so what if Kentai thinks the movie doesn't "look" attractive - low contrast is good, right? It brings out more detail than a high contrast transfer which will only crush shadows and boost highlights! This is actually all a good thing and he's just being grumpy because it doesn't look the way he wants it to... honestly, I was worried that was the case for about a week, but something just didn't feel right, and I think I've finally put my finger on it. See, going lower-contrast will usually yield more shadow detail, that hasn't been the case here. The shadows on the 4K Remaster are oddly vague and murky, and actually have even less information hiding in the shadows than either of the competing HD masters. I was expecting a hundred shades of gray, but there's no shades to speak of - just the same damn off-gray!

To give you a quick, simple example of how completely different all three masters are - and to give an excuse to use the scene where Tom Savini calls Ken Foree "Chocolate Man" - here's a quick frame-match between all 3 major Blu-ray releases. Obviously, how close to calibrated your monitor is will have a huge impact, but you can at least see the big issue for yourselves.

European 4K (Top) - Extended HD (Middle) -  Divimax HD (Bottom)

As you can see, both the Divimax and Extended masters show what I'm assuming to be a round smoke detector and a sprinkler to the right of the grate... but the European 4K remaster has only the vaguest hint of the former, and no visible instance of the latter at all! The shadows are simply a murky haze of nothing, and seeing the "actual" black 1.85 matte bars against it only makes the weaknesses around shadow detail within the new transfer that much easier to spot.

That said, the blue cast on the European transfer is one of those "intentional" changes that works - another example is the dark blue shot of the mall exterior, which plays off previous shots of the dark night sky, playing their arrival up as the true "Dawn" in the film's title. It's not in the original, but it's a clever enough touch I'll happily take it. It's little things like this that show me that far too much time and effort was put into this new transfer for the issue to simply be that the people working with the master didn't know what they were doing: Even if it's only in a small way, the work making Dawn of the Dead consistent from shot to shot and scene to scene elevates the film into a more cohesive, structured whole than it ever has been, and it's given me a newfound respect for how much care and effort can go into work I still may not, ultimately, be all that happy with.

Only adding further fuel to my personal speculation fire is the Open-matte transfer, shows countless instances of small black debris on the print registering as "true black", while completely unexposed areas of the original film are the same milky gray color as the finished 1.85:1 transfer. I can't say anyone treating only optically printed debris as "true black" (because no light passes through them) is wrong, since that's what a 35mm print looks like on projection, too... and yet, even in that context, these seem to vary between "looks pretty normal" to "what the heck am I even looking at"?

Some of these instances - such as Peter and Roger emerging from the doorway - look... fairly okay, I guess? The screen is an almost consistent fade to optical-black (ie: "this is as dark as the prior element is getting on this print"), and I can believe that's how the scan is 'supposed' to look... but then again, the shot of Roger fighting a zombie in the truck cab is just gnarly and washed out, far as I'm concerned. Compare it to a similar (not frame accurate) example from the Divimax HD master - or even the Extended HD master! - and despair at how flat and sad it all looks:

Divimax HD (Top) - Extended HD (Bottom)

So... Is the grading just poor, or is there more at play than meets the eye?

In the end, I can only guess that the different 35mm elements themselves are the limiting factor here, and the new color grade only reveals those weaknesses. After all, the prior HD master was struck from the same 35mm elements and the results were VERY SIMILAR as far as contrast and shadow detail goes - but when you factor in how piss-poor so many other elements of the previous masters were, it was easy enough to chalk those up to the mastering process rather than the elements themselves. If that's the case - if the black levels on the low contrast 35mm IP elements simply don't exist - I can't blindly call the transfer poor. Sometimes, a cigar is just a cigar, and as disappointing as the black levels are here, I feel that far too much care has been spent elsewhere for this to have been anything but a known compromise.

I will say - if my hunch is anywhere close to correct - that I would have handled it a bit differently... but that's the beauty of being an arm-chair critic, isn't it?



If this were the only issue I'd do my best to look past it, but the compression is frustrating as well. The good news is that the artifacts aren't as blatantly ugly could be, but they're better described as an issue of consistency, rather than overall quality. The phrase "I-Frame Pulsing" has been used to describe the issue, and it's not a bad one because it does sum up the issue... but in terminology that means bugger all to anyone who's never actually encoded video before. So forgive me for giving a primer on this, but it's actually relevant to understanding where things may well have gone wrong.

Most of the video compression algorithms we use to this day - MPEG-1, MPEG-2, AVC, and now HEVC - are based around the notion that a sequence of frames have commonality that can be exploited and replicated between frames with minimal changes. These smaller segments compressed together are called Group Of Pictures - or a "GOP". While I sure didn't vote for them to be the defacto standard to Make Video Quality Great Again, the way their trickle-down compression works goes something like this:

I-Frames have the least level of compression, and are thus easiest to decode. They're also the largest frames in the GOP as a result. I-frame only videos are typically used for broadcasting and editing purposes for these very reasons, but have to have a dramatically higher bitrate to match quality with "Long GOP" compression as a result.

P-Frames reference the previous frames in the GOP, sequentially, until the next I-Frame is hit. These are smaller, but are slower to decode since they're ultimately referencing previous macroblocks. They're handy enough, and easier to read for certain applications without playback funkiness (ie: Media Servers), but if you're going down this road you may as well go all the way in.

B-Frames are even smaller at similar quality levels, but they work by pulling blocks from the frames both ahead and behind, making them an absolute beast to play back on weaker hardware, and more or less impossible to get to function outside of standard forward playback without hitches.

A typical MPEG-2 GOP may look something like "IBBBPBBBPBBB", at which point another I-Frame gets punched in and the pattern repeats. AVC is a bit more flexible in that it can have GOP sizes from 1 to 250 frames without playback taking a hit, but smaller GOP sizes lead to higher quality, and as a way to avoid decoding issues Blu-ray specifically demands a 24-frame max GOP for standard progressive content.

Still with me after all that nerd shit? Cool! So, what happens when your bitrate is adequate and your I-Frames look great, but your B-Frames and P-Frames aren't encoded to the same standard - either because you did a one-pass encode that doesn't properly allocate the bits where they need to go, or because your multipass algorithms are straight up crap? It means that some frames are nice and crisp and sharp and look like your high quality master, and then the next several frames - while perhaps not terrible - are substantially softer, more diffused, and uneven looking compared to the I-Frames. In other words, the good frames "pulse" in and out, while the bulk of the transfer is the fuzzier, blotchier B/P-Frames that aren't anywhere near to the same standard.

In properly encoded two-pass content, the difference in overall clarity between an I-Frame and a B/P-Frame should be undetectable, but that sure as shit isn't the case here. Here's a pair of frames from the same seconds which perfectly illustrates the problem:

You see the huge gap in the quality of the grain structure? This, my friends, is what poor temporal compression looks like; not single instances of compression creeping up on a specific frame, but regular intervals of the source video being preserved and then smoothed over or turning blotchy, and then the crisp, accurate texture is back again before you can even blink. It's an abstract kind of Hell, to be sure, but as someone who's job requires assessing visual quality on a wide range of source materials, this is the kind of shit I just can't unsee. It's possible the more sane among you will see the artifact, shrug, and then move on with your day... but if so, you're not one to care what I have to say about compression to begin with, are ya'?

The above screenshots are from the "Restoration" themed featurette, and are slightly worse than the main feature - but as what we're talking about is a temporal effect, and thus impossible to show properly in stills, I'm using a more-obvious-than-normal instance to give you the idea of what to look for in the main transfers. Sadly, both the 4K Remastered 1.85 OAR and the 4K Remastered 1.33 Open-Matte transfer have this issue - don't have the hardware to watch the 4K UHD disc yet, so no comment on that one.

The encoding - fascinating as the reason behind it might be - just isn't particuarly good. I've seen far worse on releases that were highly reviewed, so I imagine this'll annoy those who are already prone to snort and shake their heads at the quality of "grain structure", and that most others will be perfectly happy that they see any grain at all - much less plenty of it on brightly lit footage. The transfer may be far from ideal, but compared to the smeared, waxy texture of the 2004 Divimax HD master, it's still a pretty dramatic step forward for a film that's never been able to catch a break on Blu-ray.



Fuck it. Moving on.


Both the 1.85:1 and Open-Matte presentations feature an English 5.1 remix and the original mono mix encoded as lossless DTS-HD Master Audio. Sadly, they also feature forced subtitles... assuming you aren't using an HTPC and some AACS cracking software. Hint-hint.

The English 5.1 remix - which sounds more or less identical to the Dolby True HD 5.1 mix on the Japanese BD - is about as goofy as you'd expect; sound effects are too loud as dialog is too quiet, foley echoes and trails when it clearly shouldn't, and the concept of directionality is optional, at best. The high end cuts out suddenly, and there's some odd humming in certain scenes with a lot of ambient noises which, I can only guess, is a residual artifact of the hiss removal DNR. It manages to be simultaneously bass heavy and tinny, and the fact that I had to turn my player volume all the way up to comfortably hear the dialog is nothing short of terrifying.

In short, like every 5.1 English remix of Dawn of the Dead, it sucks... and you've either made peace with it, or you'll go with the original mix. As you'd imagine, I went with the latter.

The original mono track sounds... dated, like it was pulled from a finished 35mm optical track (and likely was!), but under the circumstances I've no real complaints with the overall fidelity or volume of the mix. Unlike the Japanese release, the mono track seems to remain consistently in sync with the image and is presented as lossless. Like the visual presentation, it's not stunning, but once you know what you're getting into there's not much to complain about either.

The saddest part? Listen to the restored Italian 5.1 track for all of 15 seconds. It's fucking amazing - clear, nuanced, pans like it should, the whole nine. Listening to the English 5.1 track leaves me wondering why anyone bothers to waste time remixing 70s low-budget movies, but listening to the Italian track - in short, morbidly curious bursts - makes me wonder what could be if Rubenstein ever goes digging in the vaults and Dawn of the Dead is given the proper remix it deserves.


This is far from the only BD release of the 4K Remastered European version of Dawn of the Dead we're going to see over the next few years, and the combination of region locking, forced subtitles, middling compression and a shoddy limited edition package all make this an incredibly rough purchase to recommend to anyone who isn't a huge fan of the European cut to begin with. The transfer is underwhelming, the audio competent, the bonus features are mostly useless if you don't speak Italian, and even the limited edition box is a swing and a miss. I feel bad saying that, knowing how much work and dedication clearly went into the preservation of one of the most unique horror films of the 70s... but the finished disc just ain't that good.

Unless you absolutely need the 4K UHD-BD and the Open-Matte transfer, I'd say hold off for now and see what the rest of the world starts doing with the same materials. It can always be worse, sure - but at least it might be more English friendly or region free in the process!

Happy Holidays, friends.

See you all in the new year!

Monday, December 05, 2016

Dawn of Ultra-High Definition: PSA on DAWN OF THE DEAD 4K, UHD-TVs, and HDR


No clue, friends. But that aside, we've got some good news!

For those who may not be aware, Italian label Nightmare Factory is releasing several editions of their all new 4K restoration of DAWN OF THE DEAD. Yes, this release is reported to be Region B locked, and yes, rumor has it that the "restored" transfers will all have forced Italian subtitles - but I'll just have to find out when my copy arrives, won't I?

What you need to know is this: The 4K restoration is the first new transfer minted for Romero's 1978 classic since the Rubenstein Company started their 3D conversion back in 2008, which has so far only been shown in a handful of theatrical screenings. For better or worse, the new Italian handled restoration is based on the 118 minute European cut - "Dario Argento Presents Zombie", if you know the film's convoluted and multinational history - but with Rubenstein having funneled a fortune into a 3D version literally nobody asked for, and reportedly holding the home video rights hostage until he's able to make his money back on a fat license fee, this is probably the only viable alternative we're going to see. If you're a Dawn of the Dead fan and you're craving a state-of-the-art 4K release... this is pretty much the only game in town for the foreseeable future.

For those who aren't exactly fluent in Italiano - and that includes myself - here's what the 4-disc edition contains in a nutshell:
  • - Restored HD version (2016 4K master) of the 118 minute European cut of the film
  • - HD version (2013 master) of the 127 minute American Theatrical cut of the film
  • - HD version (2013 master) of the 133 minute Extended Workprint
  • - Bonus BD with 2.5 hours of new content plus vintage trailers
  • - 5 postcards designed by fans of the film
  • - Booklet

I won't get too in-depth on the bonus features here since... well, it's a safe bet that 90% of them will be in Italian with no English subtitles. The 18 minute interview with Tom Savini will likely be in English, as may the 8 minute introduction by Nicolas Winding Refen (who seems to have gotten the ball rolling in this 4K Remaster project) but I'll be shocked if anything featuring Dario Argento, Claudio Simonetti and so on features any English dialog or subtitles of note. That said, if you already own any of the exhaustive, absurdly special-features packed DVD or Blu-ray releases from the last 12 years and you still need more, I... don't know what you're expecting to find... I'unno, maybe that 8 minute restoration featurette will show some cool Before/After footage?

The 6-disc edition contains all of the above, but adds two very enticing extra discs into the mix:
  • - "4K" UHD-BD of the 118 minute European cut of the film
  • - HD "unrestored" 1.37:1 open-matte version of the European cut
You should also know that there's reportedly a fairly major issue with the 4K Restored Blu-ray which I can only describe as "I-frame pulsing". Basically, every I-frame (once a second or so) is much sharper, and noisier, than the B-frames and P-frames that follow, which have something of a softer, more diffuse look. It's not even that the softness is the problem - it's the inconsistency in the middle of a scene to suddenly "pulse" a sharper frame after and to be followed by a string of more neutral, blurred frames. If you're an autistic crazy person who knows what GOP structure is, you'll instantly know (and probably hate) what you're gonna' see in this set... but, it's impossible to know how bad these things are via stills, so I'll just have to report back when I see it myself in motion.

For those so inclined to import - and can deal with the combination of region lock frustrations and possible forced subtitle shenanigans that always crop up on titles like this - the 4-Disc HD Edition can be had for about $36, and when I pulled the trigger on the 6-Disc 4K Edition it was for just over $50 - and that's with standard shipping included! I'm not getting a dime should you follow those links, I just want the world to know that this exists, and that while there's already talk of some compression annoyances, it's still going to blow the similarly themed Japanese box set out of the water - and that damn thing still sells for about $120!

There's also a 2-disc DVD for those who just straight up hate quality. I have little doubt a non-limited edition of the European remaster will be released next year for a lower price - likely without the American and Extended cuts, of course - but for the sheer volume of content you're actually getting in this particular package, it'd be a little crazy to hold out for something better.


Sadly, no. But having weighed this one out for weeks, I'd rather take the time to explain why I won't be doing it - and why nobody but those with way too much money to burn should bother, either.

This is primarily because I haven't personally made the jump to "4K"/UHD, and while I've been horrendously tempted to pick up an Xbox One S just to review the 4K disc at 1080p... that would be rather pointless on a lot of levels. As such, I'll be doing a write-up on the BDs, which - considering how many things are wrong with all the HD versions of this film - should still be plenty fascinating to dig through on its' own, anyway.

This is a good time to rant about the current state of UHD, though: To be blunt, the display hardware - while absolutely mouth watering on its' own - simply isn't ready for anyone who actually understands what these displays are (and aren't!) capable of doing. I've been tempted to spend a small fortune on an overpriced OLED and lord it over all of my friends as they huddle around their sad, pathetic, peasant-LED's for warmth... but try as I might, I just can't convince myself it's worth it. Not yet, anyway.

Consider the following a primer of sorts for anyone who's about to take the plunge on a new display. My advice is "wait" - but if you're still dead set on being that guy, I completely understand. At least know what the hell you're getting into.


Thankfully, "4K Resolution" itself is pretty straight forward* - double the pixels in width, and height, meaning 3840:2160 UHD has exactly four times the resolution of 1920:1080 HD. The standard also allows for up to 59.94fps, meaning that James Cameron's fantasy of people wanting high refresh rates outside of video games and pornography could be a thing... if, y'know, anyone wants to actually produce content that isn't garbage dramas shot interlaced. (Or Hobbits. I guess.)

* Other than the fact that "4K" is referencing horizontal resolution while "1080p" was referencing horizontal, which literally makes fuck-all sense. Also, 4K spec is 4096 wide, the same way that 2K is 2048 wide, meaning that not only is "4K UHD" not by definition 4K resolution, but that "2K" should be "2K HD" - unfortunately people assume that 2K is inherently better than HD as a result, when pointing out "2K SCAN" vs "HD TELECINE" has more to do with how the image was captured and restored, rather than the actual resolution thereof. Because fuck TV marketing.

UHD-BD has some interesting quirks as a format; not only are discs available in single-layer 50 GB, dual-layer 66 GB and triple-decker 100 GB, but each of those discs has its' own maximum bandwidth limit - 82 Mb/s, 108 Mb/s, and 128 Mb/s respectively! Audio has largely been Dolby Atmos encoded, which is basically a 5.1 mix with metadata to "shift" the individual sounds around a grid of tiny satellite speakers; It's actually pretty cool tech, but completely impractical outside of an actual movie theater with a massive array of speakers installed to cover a wide area. Video - the part that interests me the most, I admit - is now handled by 10-bit HEVC, which - having done some preliminary test encodes myself - I can confirm it holds an almost shocking level of efficiency improvements over Blu-ray's most common codec, AVC, and suspect that in most cases, 100 GB is more than enough for an excellent, reference-quality end-user transfer.

Yes, of course, word is that there are already a handful of UHD-BD's with visible compression issues, and I'm sure that bitrate starved Netflix and Hulu streams will always be disappointing when it comes to grain structure - but if you didn't expect that, you didn't pay much attention to the launch of DVD, Blu-ray, or Netflix HD, did you? There are still some minor disappointments in the spec sheets - video is still subsampled to 4:2:0, and for another there's still "Limited" and "Full" color spaces to confuse and annoy everyone who can instantly tell the difference between PC and TV levels - but honestly, chroma bumped up to a full 1920:1080 is probably enough to satisfy even my crazy self.

But by far, the most promising part of this whole process is the introduction of HDR - or High Dynamic Range. In super-simplistic terms, it means two things; expanded color space, and higher peak luminances (whites). The actual colorspace for Rec. 2020 expands red notably to include those glowing, fire-engine-light reds that Rec. 709 HDTV colorspace simply was never designed to acknowledge, to say nothing of the other-worldly greens that... ironically, nobody's ever seen before in a movie. Seriously, the technology to produce them on a digital screen didn't even until recently exist, so unless you insisted on the most amazingly fabulous bright-ass neon green pride float to ever burst info flames in NoHo live, you've probably never seen colors quite like what 2020 is capable of producing once you hit those outer reaches of the gamut.

Thanks for the visual aide, Google Image Search!

It's worth noting that for these colors to even exist as a spectrum of light the human eye can see, there has to be a lot of light being pushed out by the display - considerably more light than we've ever had on consumer or even public exhibition screens until quite recently.

So... where's the problem? Kentai's down for 10-bit color, and this Deep Color stuff is pretty rad, right? Mo' Reds, Mo' Greens, and screens so bright you'll go blind - what's not to love?! Well... there's a couple things really chaffing my HDR boner, and if you've looked into it as long as I have you'll feel thoroughly cuckolded by the mistress of Rec. 2020 yourself.


To put this another way; have you ever looked at a movie and went, man, this scene is visually stunning - but if only there were deeper, more vivid greens? Well... I mean, maybe you have.  But to Rec. 709's credit, it's basically covered the overwhelming majority of the blue spectrum humans are capable of seeing, and while red can certainly be improved, the only major gains we get are in the side of the color spectrum that we associate with... golf courses. Dramatic lighting tends to be white, blue or red, and while higher color fidelity will lead to greater visual contrast in some titles and subtle improvements on things like color banding for pretty much everything, this is very much a subtle refinement in 95% of real-world uses.

Arguably, though, Rec. 2020 is a bit less important than DCI-P3, which was a color space specifically made to capture every possible color range of 35mm film. Well, more relevant as far as movies are concerned, anyway - I'm thinking of the obnoxiously saturated colors in games like the DOOM reboot and salivating at what might be with a new HDR profile...

Many titles in the initial batch of UHD-BDs - The Martian, Mad Max: Fury Road, and so on - were all, seemingly at least, re-graded from scratch specifically to show off how vibrant and pretty and magical this new format was. Stuff released by Sony... wasn't. Pineapple Express was hardly the sort of title I expected from the first wave, but a decent one to show that even with all the expanded color in the world, a movie shot with a drab, overcast visual style is always going to be a drab, overcast movie. I imagine the world will be surprised to see that the inevitable UHD-BD release of Caddyshack doesn't glow like the primary heavy hues of an X-rated Ralph Bakshi cartoon, but these are things that people will slowly figure out on their own.

They were all HDR enhanced transfers, make no mistake, and there's a difference in just how vibrant those titles can get in terms of red and green and particularly highlights on the bright end of the grayscale, but the difference is subtle refinements and increased fidelity in things like traffic lights and grass lawns - not show stopping crazy neon explosions. I have little doubt that young and enterprising film makers will make full use of these new expanded colors over time, but for now, this is basically just additional chroma headroom for your favorite movies; nice to have, sure, but the odds of you noticing a huge difference, even during an A/B comparison, aren't that bloody likely.


Even if we assume that P3 is more important than 2020, there isn't yet a consumer level device capable of displaying that entire range of color yet anyway. Even when properly calibrated and setup, there will be a point of roll-off where the brightest green simply stops short of the full range of signal. The main reason for this is - assuming the hardware all down the chain is capable of accepting the full signal to begin with - to generate a wider color gamut, you need to produce a brighter peak white to carry the wavelength the color exists in.

TV manufacturers love to talk about wide color and High Dynamic Range and all the amazing stuff their new models can do that the competition can't - yet they're always oddly hesitant to talk about the actual cd/m2 rating -or number of "Nits" produced - because they know even the high-end OLED and LCD screens out there fall damn short of the recommendations by a wide margin. Simply put, if you're watching an SDR screen calibrated to the industry standard of 100 Nits, the brightest, deepest blue is only going to be 7 Nits and be pretty fucking dull - but if you're watching on a high quality OLED screen with a maximum output of 800 Nits, you suddenly have 56 different stops of blue to play with before maxing out. So it's not just the brightest, peak blue - it's all the subtle gradations between the light baby blue of the sky of early morning, and the deep hues of midnight that'll have a new level of depth to them.

For the time being, any OLED TV labeled HDR has to have a maximum output of 800 Nits, while any LCD with the label has to output 1,200 Nits - but even that's only on miniscule highlights, where a full screen of white is notably dimmer for both. The reason they have different scales is because OLED can actually turn individual pixels off, giving it greater perceived contrast than LCD, even with a lower light output. That said, both technology are fudging this stuff - OLED have the Automatic Brightness Limiters the same as Plasma, which means that while a small reflection can have a crazy high light output, a full fade-to-white will be substantially lower, and you can watch the whole screen dim as large, bright colored objects come into frame. LCD uses local-area based dimming, which... well, kinda' sucks, but in a completely different way.

In both cases, however, they all pale to a reference Dolby grading monitor that outputs a terrifying 4,000 Nits! We're likely not going to see that for home use until we come up with some crazy new technology to power it, sadly, but mentioning the word "Dolby" brings up the other big stinker in this new tech:


Without delving into I-could-be-compromising-NDA's-here territory, I don't mind saying that my day job consists of a lot of transcoding-server-based wizardry that'd come off as mundane and even disappointing to most readers - developing templates to convert one kind of file into another, automating broadcast friendly audio normalization to different international specs, dumb stuff like that. But the 4K content rollout happening right now has left me to be the front line in telling clients who want to get their feet wet what we can and can't give them, and holy hell, has it been equal parts enlightening and infuriating to follow.

The short version is that because each TV has different color and light capabilities, they've developed what they call "PQ" - or Perceptual Quantization - to make sure every display is capable of displaying the movie as closely as possible to what the graded master was intended to be shown at. In other words, the source media is always encoded as Rec. 2020, but a metadata setting tells the TV what levels it was graded at - so if the movie was graded on a 3,000 cd/m2 (or "3,000 Nits") monitor and your TV can only handle 1,200 Nits, it'll actually shrink the color gamut/dynamic range by 60% so that you keep as much of the color as possible without resorting to clipping and blowing things out of proportion. This includes values for absolute red, green and blue data as well, which means the chroma and luma of the signal are scaled to the appropriate points separately - which means if your TV can't handle the maximum brightness but it can handle the maximum red, it's not going to compromise one for the other.

It's actually a really, really cool idea, and having seen demos of it in action, I can say that PQ is a damn fine thing. Now, my understanding of HDR is that even if the color never falls outside of Rec. 709 (or P3), the contrast between black and white is effectively infinite, and only limited by the display itself. This, admittedly, makes the lack of HDR enhancement on the DotD 4K release something of a missed opportunity, but considering both how niche this title is nearly 40 years later, and how experimental the hardware is, I can almost forgive seasoned professionals not familiar with HDR for not bothering to dive into the new format incorrectly, and instead focusing on delivering the best quality SDR presentation they know how to... particularly when you factor in that, even with current HDR content, the average brightness between an HDR and SDR output is typically comparable - it's just got a lot more dynamic range, which lets you not clip highlights and crush shadows while still staying within the realistic 500~1,000 cd/m2 light output of the average consumer TV.


The bigger problem is that there's different kinds of HDR Metadata. The two most common right now are HDR10, and Dolby Vision - the former has fixed coordinates, meaning the whole movie has a max output for each color from start to finish, while the latter is a dynamic solution, meaning each scene can be tweaked individually to compensate for low lighting - say, making sure a camp fire is displayed properly alongside the following scene taking place on an overcast afternoon. Adjusting PQ on a shot-per-shot is great, particularly when it comes to downscaling HDR content back to SDR - something you could easily do via algorithmic automation on a Dolby Vision master since it would scale everything right down to the Rec. 709 limit, but would require plenty of hand-holding on an HDR10 source to avoid crushed shadow detail and clipped highlights.

So what's the industry standard? There... really isn't one - and certainly not a permanent one. Oh sure, the Society of Motion Picture abd Television Engineers (SMPTE) put their stamp of approval on HDR10, but Netflix is 100% behind Dolby Vision, which requires hardware on the display that knows how to handle the dynamic metadata outside of apps. Samsung knows that Dolby Vision's dynamic properties are here to stay, so they're currently working on an open-source dynamic equivalent to Dolby's equivalent,.. but whether it'll require all new hardware or be shunned due to partnerships with Dolby is anyone's guess. (There's also rumor that HDR10 Metadata is limited to 1,000 Nits, but having seen it for myself I can tell you that's certainly not true. It is limited to 10-bit, however, while Dolby Vision is up to 12-bit.)

That's before we even talk about Hybrid Log Gamma, a really clever method to make HDR content backward-compatible with SDR hardware without any Metadata at all - it just outputs a fixed (pair of) gamma curves, and trusts that the device it's being fed to is calibrated somewhere in the ballpark of "correct". The downside is that it features no PQ, which means all levels have to be fixed on a static 1,000 Nit scale. That doesn't sound terrible at first, since most displays on the market are only hitting about 800~1,100 Nits anyway, but as it's only being used for what could charitably be called experimental broadcasts in Europe for the time being it's all kind of a moot point... for now.

Odds are there would be more support for it if the only thing that used it wasn't VP9, but word is H265 will be implementing HLG profiles for 2017, so buckle up kids - this shit's just getting started.

Праблема нумар чатыры:

Even with all the exciting changes in perceptual contrast and color gamut... there's also the limitation of the resolution itself. Namely the fact that most Hollywood movies finished over the last 15 years or so are limited to 2K resolution.

Let me clarify that for a second: Despite professionals slowly but surely moving towards digital photography as a more cost-effective and (arguably) flexible atlernative to celluloid, movies are still - by and large, at least - shot on 35mm film. These days, 35mm is typically scanned in at 4K and then edited on a digital image sequence called a DI - or "Digital Intermediate". Despite confusing naming conventions surrounding it, 2K 1.78 is exactly the same resolution as 1080p HD - 1920:1080 - and since virtually all 2K sources will ultimately be seen on Blu-ray or HD streaming anyway, the difference between 2K and HD is really negligible, at best.

The only reason it's been sononymous with "Better Than HD" for some time is because of the difference between an HD Telecine - that is a real-time transfer of 35mm to HD video - and a 2K scan to uncompressed DPX image files, which are then graded and assembled on a 2K DI. In short, the "scan" part of the "2K Scan" is the important distinction - not the resolution itself.

Things get more complicated when you talk about a modern movie - something like, say, The Martian. While more or less every shot of Matt Damon was shot on 35mm film, it was then scanned into a digital file, and placed on a DI for digital effects matting, and even for grading and general clean-up on shots filmed entirely on-set. Those scans may well have been scanned at 4K, but the DI itself - which gets differing resolutions from different cameras, post houses and so on - is at 2K. Even if the 2K DI was kept (which may or may not be the case), all of the color grading and effects compositing was done at 2K resolution, meaning the ultimate digital file that represents the final output of the movie is limited to 2K.

"But wait!" I hear some of you thinking. "Isn't THE MARTAIN already out on UHD-BD?" It sure is! But the dirty little secret is that Fox didn't actually re-create the entire movie in 4K. They simply went back to that same 2K DI source and upscaled it by 4X to produce a faux-UHD master. To be fair, upscaling HD to UHD is a much less butt-ugly process than upscaling SD to HD... but it still isn't "really" 4K. And that's what about half of the supposedly "4K" UHD titles kicking around in the market right now are. Crazy, right?


Dead. Fuck'n. Serious.

But it's hard to blame them in some cases, I admit. MAD MAX: FURY ROAD - another early "4K" home video release - was shot entirely on Arri Alexa cameras at 2.8K resolution (2880:1620), and then scaled down to a 2K DI. So even if the 4K transfer was re-rendered from scratch, the resolution shot on location was never full 4K resolution to start with.

It's understandable that all-digital movies would be limited to 2K, but why the limitation on 35mm sourced films? The short answer is "3D". Unless a movie was actually shot using 3D cameras, you're basically having people manually trace and rotoscope individual objects in post to add depth, and that means you need multiple layers of the same footage. And since this is a process done effectively at the last minute, and is also the way most Hollywood blockbusters make guaranteed extra money at the box office, it's seen as a necessary evil... even by directors like Guillermo del Toro and Zack Snyder, who are about as open in their disgust for the process as they're allowed to be before the producers give them a talking-to about shitting on the profit margins.

So why not just use 4K DI's? The short answer is that it's just not feasible to do it for all projects on the sort of turn-around time expected of a feature film - and it will have a sizable impact on budget, even if it's a fraction of the overall cost. It comes down to raw resources, and if you're a Hollywood producer who's trying to maximize profits, and the tech guy in the office says that 2K work is a lot cheaper than 4K work, odds are that's the first thing that's getting axed. This is why despite Sony scanning everything at 4K to get the most out of the raw scan possible, we get full 4K restorations of Taxi Driver but a title like Fright Night is scaled down to 2K for further clean-up and grading - the higher resolution scanning is baked into the cost of having the hardware to do it, but taking the time and bandwidth to continue working in 4K adds up pretty quick. As an example, a feature length 2K master of a 90 minute film will take up about 1.5 TB of disk space alone! A 4K DI has four times the resolution, and yes, will take up four times that space, to say nothing of the added strain on whatever network is forced to decode that much raw bullshit at any given time.


For better or worse, I spend every day dealing with network latency, side-eyeing mismatched source content and queueing up transcodes for both production and delivery purposes on a server that was effectively built to handle HD material in real-time. Think of it this way: Even when all you're doing is jamming cuts of meat down a conveyor belt into a meat grinder, suddenly cramming four times as much meat into the same sized pipes is gonna' cause all sorts of panic you, and the rest of your team, simply aren't ready for. You've got three seasons of SD content? Beautiful, spread the cheeks on that thing and go to town. You've got THREE 4K source files?! We're basically fucked the rest of the day, and anything not-4K-related that you expect to get out the door by 5 is going to require a very special request.

Even if your goal is simply to take the finished master and convert it to 3D, keep in mind that conversion is effectively rotoscoping, and multiple people are going to be working on the same title at the same time, and the bandwidth required to hit the same 6TB file from multiple workstations is ridiculous. It can be done - native 4K DI's for titles ranging from The Smurfs 2 to Elysium are proof enough - but if 3D is part of the picture, it's barely worth the expense, and allows you to farm work out to multiple, less insanely equipped sub-studios. There is, technically, "4K 3D IMAX" in select theaters... which are always, without exception, upscaled from the 2K 3D masters even if a 4K master for the "Flat" version exists.

The process of rendering a full length movie is often simply too data-intensive (ie: "too expensive") to justify doing at resolutions beyond 2K. The market for higher-quality 3D is already negligible, and quite frankly, the overwhelming majority of consumers are so clueless they wouldn't know the difference between 720p and 2K - forget HD and UHD. Hollywood caters to the lowest denominator with the biggest payout... which happens to be 2K 3D. The saddest part of this reality is that while I'm not thrilled at the idea of purchasing a 2K DI sourced movie in "upscaled 4K", the added benefits to H265 compression, PQ contrast and HDR color are enough to get me to at least consider buying the more expensive version, even if I don't have a fucking display that'd know what to do with any of it yet. The fact that virtually all UHD-BDs released so far include a "Restored in 4K" Blu-ray copy only makes that jump a little easier to swallow.

In short, 2K DI was easy to convert to 3D, which means nobody wants to invest in 4K DI. Weak.


Anything finished on 35mm can be re-scanned at 4K. Otherwise I'd be warning you about Dawn of the Dead rather than throwing fitty bones at it. But realistically, the leap between 2K resolution and 4K resolution is probably going to be quite a bit more subtle in most cases. Most sane people don't sit within 6 feet of a 55" TV, and the real-world gains in resolution are going to be pretty subtle. If anything, the most obvious improvements will come from the variable block-sizes and 10-bit refinement of H265 as a codec... but neither of those things actually needed higher resolution to begin with. Much like "2K" vs "HD", the difference has more to do with the mastering process than the actual output resolution. 4K is better, make no mistake, but it's also not why most 4K masters are a dramatic improvement on dated HD scans on its' own. That's why the "Remastered in 4K" versions of Ghostbusters and Sam Raimi's Spider-Man looked so much better than the prior Blu-ray releases, even at 1080p resolution.

Don't get me wrong, you should buy 4K restorations of classic films if you have the money to taste. But the market is going to be even smaller than Blu-ray has already become, and I have a feeling the titles we see are going to be titles that the guys running the labels know will sell, and titles they happen to love personally. Dawn of the Dead is something of an outlier, and I fully expect the majority of 4K movies to be trash nobody would ever actually want to watch outside of penance.


I may talk about the PS4 Pro in another post. Short version is it isn't really 4K, just clever upscaling and some supersampling for any schmucks buying this for playback at 1080p (complete with frame-drops not featured on the base hardware!), but for $400, it's "close enough" to UHD resolution.

Holding aside the Metadata and Display Gamut nonsense, one of the few times where that crazy expanded colorspace really would be an amazing boon would be video games. Digital creations aren't limited by 35mm exposure, in-camera sensors or available local lighting, and the thought of the boiling reds and alien greens in something like, say, the 2016 DOOM reboot with enhanced grayscale and color fidelity could be absolutely brilliant if done properly - and a handful of PS4 and Xbox One games already have full HDR support, even on the older-model consoles, though of course as displaying HDR samples on SDR monitors is pretty much impossible there aren't many useful A/B comparisons floating around yet...

The issue, however, is how HDTV's handle those signals. Most displays process incoming footage in different ways - upscaling, deinterlacing, color correction and so on - and in the case of movies, it... doesn't really matter. In virtually all cases movies are shot, edited and shown at 23.98fps and play at a locked framerate from start to finish. No problems there. Anyone who plays games - well, anyone who plays shooters, rhythm games, anything where instantly reacting to game stimuli is required - will know that reflexes are important... but also rendered completely fucking moot if your display has any major input lag.

Back in the days of NTSC CRT, there was no delay to speak of - the input was virtually instantaneous to the analog connection - but HDTV's and now, UHD displays, require time to properly process whatever signal they're being fed. Most game consoles and anyone gaming on PC that isn't using a specialty designed high framerate monitor is locked to 60fps (well, 59.94Hz if you wanna get technical) which means there's exactly 16.66ms between each frame. Theoretically. Slowdown and frame-pacing are a thing, but let's ignore that for the time being and just say that a great TV will give you one frame of delay, while an average TV will give you about two frames worth. And, yes, video game mechanics themselves are slower now than they were a decade ago to account for this phenomenon, if you can believe it.


Yes, you do. And all of that is irrelevant when discussing delay, because it's based on actual milliseconds going into the display. Desktop style monitors typically have ~15ms delay these days, even on cheap garbage monitors, but you're never gonna' get that sweet, sweet HDR profile for Deus Ex: Mankind Divided on a 1440p screen that runs 144hz, now are ya?

And yes, I've recently upgraded to a GTX 1080. Feels good to play 2160p, even if it's currently being scaled down a 1080p. What is aliasing, again?

When you buy a TV and you know you're going to play games on it, you probably set it to "Game" or even "PC" input. Why? Because that disables plenty of the internal processing and lets you get data in as fast as humanly possible. ~33ms is typically as good as a large display is going to get, and is fairly playable at anything that isn't a competitive shooter or fighting game - but anything after that is going to introduce an obvious delay that could be annoying at best, and unplayable at worst. The only monitors with better response times tend to be smaller desktop monitors that forego a lot of the basic conveniences of "TVs" - nothing wrong with that, but if you're sitting 2 feet away, you probably don't want a 55" UHD monitor to start with. (Well, I do. But I also want a three foot erection that breathes fire, so grain of salt there.)

With all that in mind, let's remember that UHD with HDR is not only feeding four times the resolution and an expanded gamut over the same connection as an HD master, but it's also got to equate the HDR metadata so the image is displayed properly. In other words, HDR was designed to be used for movies more than games, and the hardware implementation reflected that... at the expense of games being actually playable. While plenty of TVs have had firmware updates to speed up response times, plenty of sets still have over 60ms of response time - or about  a 4 frame delay - while Sony's current Bravia line - the one marketed as the "Perfect Match" for the PS4 Pro, no less! - has over 100ms of lag on UHD with HDR enabled. In other words, it's fucking unplayable.

To be fair, a lot of current models have had firmware updates that dramatically cut down on input lag once proper diagnostics by third parties. For the love of Pete, always check, though I assume the first generation or two of HDR sets (which'll probably be really cheap if you can find them kicking around) are never going to be updated, largely because people who do play games don't dive in when it's crazy expensive and can burn-in like a mother. Ask anyone

Even so, the fact that this wasn't even a consideration is shocking to me. Truly, if there's one thing I want out of a $1,500~5,000 TV, it's for my natural shitty skill at vidya to be exponentially amplified by a full tenth of a second. I need to respawn in every room six times on Ultra Violence anyway, so the last thing I need is to have the game trailing behind my obvious lack of skill.

I'll stop mentioning Doom 2016 when I'm damn good and ready to.


Light output needs to improve before the colorspace is ever going to reach P3 levels (with "100% Rec. 2020" being science goddamn fiction for the foreseeable future), and unfortunately edge-lit LEDs - the cheap, lightweight garbage people love to waste money on - aren't capable of that to begin with. Despite being incredibly pretty, OLED isn't as bright as a full on baclit LED (yet). As the tech improves and the price comes down, it'll start to be feasible to have both full-P3 and at least 1,200 Nits - which, from any sane perspective, should have always been the bare minimum for "HDR". Instead it's being used as a gimmick to separate the cheap 4K sets from the expensive 4K sets, which is fucking deplorable.

There's nothing wrong with a 4K SDR monitor, and I've considered getting one myself - but the HDR monitors that are available right now are so bare minimum, it kinda' makes me want to have no part of this shitshow until the 2017 models have locked horns long enough to get at least a 100% P3/1,200 Nit standard. Maybe by then the HDR Metadata will have been figured out, too - though I wouldn't get my hopes up on Streaming and Disc having a single, unified standard, much as that would make
every consumer's life a lot easier.

Unless you get a crazy good deal on something really, really good, I say save your money for another year. The tech has to mature, and it's only going to get cheaper anyway, and unlike Blu-ray and the PS3 a decade ago, the Xbox One S just... isn't an appealing enough console to justify getting as a cheap media player with games. Not for someone with no love for Halo and Gears of War, at least.

As for the Dawn of the Dead box, we'll have to talk about that another day.