Sorry, I am talking about controlled environments. If anyone is watching with light sources in the room or through a window, you won't be able to see elevated blacks anyway.
The right amount of D65 backlight behind the TV is enough to make the blacklevel look "black" and not elevated.
For the eye only? Yes. From a neurological point of view when it comes to perception, no. More in the 800-1000:1 ballpark.
Color shift can lead to variation in brightness.
Moving closer deals with the "point light" issue from small TVs. It still doesn't give the feeling of a large screen. As long as one has clues about the size of the screen, such as visible frame or an object next to the screen, the perception of size will only change with the perception of distance. So one still knows it's a small TV sized screen. That's why moving an iPad right up to your nose doesn't work, it's still perceived as a 11" or 13" screen. In addition, if cinematic experience is considered, moving up to the screen can be problematic when it comes to audio. Nothing can replace a properly sized screen. But of course this comes at a cost which is certainly more than a $2k or $10k TV. For professional environments price doesn't matter. For home use, it's a hobby. Some buy sneakers, some are into photography with expensive lenses, others are collecting cars and some decide to have home theaters. Never question a hobby.
Hate to break it to you, but 3D is dead. The whole concept with glasses is flawed. And don't forget the hardware to do it properly is expensive. The best 3D I've seen (and actually enjoyed) is the Sim2 HDR double stack which is using triple flash @144Hz. It looks great when the source material looks great. Unfortunately there's few such material. And at that point, I'm not sure if spending $150k on two 1080p machines with 5000 lumens for 3D is worth it at all, given there's better 4k with HDR in similar price ranges. I've never been a fan of 3D on TVs, way too small. But it's a personal preference, so to each their own.
1) I can't recommend these cheap colorimeters. There was a study a few years back published to ISF members that basically said more than half of the Spyder stuff is coming out of the factory having false readings, are extremely inaccurate (particularly in the low IRE region) or have non reproducible (random) readings. I bought a bunch of <$1500 sensors back in the day and never found them to work I expected. They're also not suited for every type of display device. If one is serious about this stuff and willing to put the time and effort (one has to learn the science behind it, it's not a push of a button), then I always recommend the Klein K10 as an entry level instrument. However, it's $7k (sometimes on sale in the $5k range). Anyone who isn't willing to deep dive into this, is better off hiring a ISF certified calibrator to do it.
2) OLED is the better choice when you're in a pitch black room without any type of ambient light and you don't have a LED panel at hand that can compete with the black level of OLED. That depends on number of dimming zones, if it's a single or dual LCD design, etc.
Btw, technically it is a misconception that OLED is totally black. There's still a very small amount of light coming from a "turned off" pixel, it's just so low that most sensors can't read it (and eyes can't see it), you need to go into the NIST certified six-figure $ range to be able to measure it.
OLED is also better for integrating the display in your environment (check the wall series from LG). You can also curve certain OLED screens (we've also seen curved LCDs in the past), as they're flexible (some). You can completely hide them with roll out screens too. Power consumption can be point, depending on model.
3) Yes, properly done is not distracting at all. You've probably seen this on Philips TVs, they call it Ambilight. But it's not D65, so not color neutral which will result in changed perception of colors and it's way too bright. They have the right idea, but the implementation is more a gimmick.
4) Lumagen has different models, so as always, pick the right choice for the job. Street price for the top-model is more around $5k. I recommended it for better upscaling of SD material. But in reality it does so much more. It does not only upscale, it does frame rate and aspect ratio conversion. It can do sharpening, softening, can remove edge enhancement artifacts, noise, artifacts from older film prints, etc. It has full 3D LUTs for calibrating source devices and dynamic tone mapping for each frame which brings HDR to a whole new level of quality by having a fully calibrated chain from source to display unlike standard HDR10 and Dolby Vision, which is generic, static (at least for now) and does not consider your specific setup at all. You can also have different setups depending on situation. For example with projectors, 3D as pointed out requires more light than 2D. So you can have calibrated 2D and 3D settings, even with multiple lamps (if not laser based) turned on in the projects. I've seen installations with up to 4 lamps in a single projector, depending on required light output.
All of that is only useful, when your display device is up to it of course. It makes no sense at all for bad displays. I'm going to go one step further and say, that anyone who is really serious about the best video quality possible will need one of these. It's a must have device and not a gimmick and the difference is not small (again, depending on display quality).
Let me add a few things in general, because I think there's a lot of misconception when people talk about "nits" and brightness. The term these days is mostly used as a marketing thing. More = better. But technically a nit is the amount of light equal to one candela / meter^2. So this is specific to the area of one m^2 (about 10.8 sqft). Change the area and it doesn't work anymore, the definition isn't valid anymore. No one is arguing that an OLED isn't bright enough when putting on a full white field. It will probably be so bright, that you have to look away or close your eyes. But again, this is not what it is about. What we want is small objects to be super bright, such as stars in the black of space. Lets say your display is rated 600 nits, then it can't reproduce stars at 600 nits, because in order to do that, the star would have to be one m^2 in size on your display. However when the star is only a few pixels large, you only get a fraction of those 600 nits from the area of one m^2. So in order to show these stars at a very high brightness, you need a much higher brightness in one m^2, that is why 2000 nits and more displays make sense. Not to have a full white field and blind you with it, but to be able to have high brightness for smaller objects.
Here's a practical example. Buy a small, dim flashlight and have someone shine it into your eyes from some distance. You will be ok. Then use 100 of those flashlights and do the same. Then 1000 and so on. Many flashlights will be too bright to look at, however if you want extreme brightness from a single light source (flashlight), you have to buy a brighter one.
So in addition to the simple nits number, we have to look at what we want to do and put it into perspective. If anyone is familiar with the series "Home Before Dark" on Apple TV, the following is specific to episode 5 of season 2 I recently looked at from a technical point of view. Color primaries are BT2020, the color primaries of the mastering display were P3. The mastering display luminance is between 0.005 cd/m^2 and 1000 cd/m^2. The maximum content light level in that episode is 1597 cd/m^2. That is the brightest pixel in the entire episode. However, the maximum frame-average light level is 163 cd/m^2. That is, the frame with the average brightest luminance level in the entire episode. It seems "dim" at only 163 cd/m^2, however it could be mostly dark with the bright pixels going up to >1000 cd/m^2 (remember this is an average over all pixels). And that is why you need high brightness, to properly show those small bright parts of the image. It becomes less of a problem, the larger the objects get. Some definitions here:
https://docs.microsoft.com/en-us/windows/win32/api/dxgi1_5/ns-dxgi1_5-dxgi_hdr_metadata_hdr10
So in addition, as someone mentioned Dolby Cinema theaters above. One has to consider that home HDR content is usually mastered to at least 1000 nits. We're already seeing 4000 nits and the industry is making approaches to push to 10000 nits. HDR in Dolby Cinema theaters however is mastered to 108 nits. So you really can't compare these two or use Dolby theater content at home without further processing (leave aside the issue on specific equipment needed). So why is that? Well, here comes human perception which is not linear and in general a little strange (but hey, it works). A bunch of papers exist (also from Dolby) that say given two objects (one small, one large) with the same brightness, the larger object is perceived is brighter. So in general, large object seem brighter than smaller objects. Given the large screen size in Dolby Vision theaters, that might make sense when it comes to mastering to 108 nits (I personally think it could be more, but less than home HDR. There's a bit more involved, but lets ignore that to not blow up the thread). That also mirrors the general experience for our home theaters where screens are usually between 15' to 25' wide for the enthusiastic movie fan. I have seen some 35'+ wide home theaters, but they're not that common. Mastering nit wise, they could fall between the massive Dolby screens and the small TVs. I'm saying could because depending on equipment and quality one can use both home content and cinema content.
All of that brings me to answer your question from above:
To have that large dynamic range between the "the black of space and that super bright tiny star far, far away". So it's not really about that super massive on/off contrast ratio of millions : 1 (infinite :1 is nonsense, its marketing). It is your intra scene contrast ratio that you want for HDR.
I hope this puts things a bit more into perspective, even if I've simplified it and skipped some stuff.
