Originally Posted by DunMunro
We don't look at on-off contrast with our PJs (except for testing): We look at intrascene contrast and clearly intrascene contrast falls dramatically in a non-optimized theatre, and here high ASNI contrast is important for the majority of the viewed content:
TW9200/W2000 Lumens = 638 / 899
White living room with blackout drapes.
50% = 63/98 20% = 162/228 10% = 317/400 5% = 549/603 2% = 1129/942 1% = 1694/1138 on-off = 4474/1634 ANSI = 251/646
~10% = average movie.
Given that we perceive an 82% drop in lumens as a 50% difference (in other words, 18% looks 50% as bright), we need a larger difference than 400 to 317 (around 20% difference) for it to be particularly noticeable.
Originally Posted by DunMunro
This despite the W2000 throwing 50% more lumens onto the screen:
So we have the ProjectionDreams people emphasizing the importance and advantages of high ANSI contrast and their own testing shows this very clearly. What their tests don't show is the importance of brightness.
I'm not sure that's what they're doing - they're showing that with a bad room the ANSI is reduced and that happens with all projectors and will equalise that performance between high and low ANSI projectors, making it something that makes little difference. Brightness may be needed to combat ambient but it's not that important for image quality. In fact it can be detrimental as it will highlight image noise/source artefacts, raise black levels and introduce a grey veil while reducing colour saturation near black. Higher contrast will improve the image far more than higher brightness.
Originally Posted by DunMunro
Yes, if there are no bright highlights in a scene and you pause the PJ and let your eyes adapt then very high on-off contrast can become visible, but this rarely happens in typical theatrical releases. Without time to adapt 800-1 is about the limit and this will decline as we get older (alas I know all about that!) The time taken for the human eye to adapt to darkness is well studied and it is something that I'm quite familiar with because I've been a very active amateur astronomer for the last 30 odd years.
You don't need to pause an image or let your eyes adapt to see the contrast differences when comparing projectors, though you usually have to to see resolution differences, like between native 4K and fauK images, even at close seating distances.
I can only suggest you do the testing yourself as I suggested, or look at the pictures from projection dream which prove that the on/off contrast difference is visible, even in a bad room - the 7300 vs the 9300 clearly show that the 9300 has better black levels due to it's higher on/off contrast - their ANSI is much the same. They're showing the effects of a bad room by measuring the ANSI, but they are also proving that the higher on/off will still give better black levels and contrast performance regardless of what happens to the ANSI, and the pictures prove that.
A bad room pretty much equalises the ANSI of a projector - if you compare a good DLP with a good LCoS, the DLP has about twice the ANSI of the LCoS, but in a bad room, that difference is greatly reduced as to almost equalise them - the DLP will still have the higher ANSI but only by a few points - say 55:1 vs 50:1 from the LCoS - easily proven with real measurements. That difference will be hard to see if not impossible in real terms, but the higher on/off of 10,000:1 (or more) vs 1000:1 will be more noticeable.
The difference between 230:1 ANSI and 400:1 ANSI (eg LCoS vs DLP in a good room) isn't that visible, and there are not many scenes that will be close to the 50% ANSI pattern to show off that difference in a good room, but the difference between 1000:1 on/off vs 10,000:1 or more on/of is always visible in a good or a bad room. When comparing projectors in split screen, you will easily tell which one has the higher on/off, but it is much harder to tell which one has the higher ANSI unless you use scenes that are closer to 50% test patterns. Projection dreams images show that in a typical scene of 10 to 15% ADL the on/off is still visually perceptible but the ANSI (DLP vs LCoS) is being equalised and less visible.
It's quite obvious that on/off CR plays a bigger part in image quality than ANSI - both are important, but on/off more so. Just compare an old CRT projector with maybe 130: ANSI and thousands to one on/off vs an old DLP with maybe 400:1 ANSI and maybe 400:1 on/off - everybody will prefer the CRT, all else being equal. In a bad room the CRT may have black parts in the image that raise and drop or even change colour with content, but that's mostly due to the lenses IIRC, and digitals don't suffer in the same way - halo's and contamination is greatly reduced. There is a reason why there has been a constant need to improve on/off contrast capability in digital projectors, and that's because it's an important image quality. It's not the only quality of course but it's usually high on the list.
As for 800:1, that too can be easily disproved by ruining two projectors side by side in a split screen comparison, and the higher on/off pj will be visible due to the better blacks or brighter whites (assuming one end of the range has been equalised) if the difference is large enough to be perceptible - we can see such a large range of contrast, and given that we don't see linearly (see my earlier comments), we need a large contrast difference for it to be noticeable which makes the 800:1 number seem ridiculous. With numbers like that we probably wouldn't be able to drive at night.
Our eyes can also be pretty quick to adapt - a split screen image as described above will also prove that - we don't need static images and long periods of time for our eyes to adapt when watching movies or comparing projectors. Like I said before, if we can only see 800:1 there would be no need for projectors that are capable of more and we could not see the difference during a movie - except we can, so that should also be a simple way to prove we are not limited to just 800:1. I don't know where people get these numbers from.
With respect to astronomy, you tend to use the peripheral vision more due to there being more rods than cones there IIRC, and we may rely on chemical adaption which is a slower process than relying on normal adaption including the iris under normal conditions. We have 'scotopic' vision that helps us with dark conditions, and 'photopic' with brighter conditions (and if you want to measure the contrast of your pj, try and get a photopic light meter). Mesopic for intermediate.
I was looking for an image I'd seen that shows the importance of high contrast vs resolution and although I couldn't find the original article, this image should show that high on/off contrast is important, and more important than resolution.
I have a room that is lined with black velvet, and I can assure you that a high ANSI low on/off capable DLP does not compare well against lower ANSI projectors that have considerably higher on/off, and I used to be a DLP owner - my first being back in 2000 (and two more after that), so I can assure you that high ANSI and high brightness are not as important for image quality as high contrast - with high lumen projectors, dimming the image brightness down with filters is a common practice (ND filters or CC filters which can also give more on/off contrast after calibration), and people wouldn't be doing that if brightness was a better quality than contrast or black levels. Prior to this room, I had a room that wasn't so good, and despite having DLP, I was always striving for more on/off. The ANSI was round 50:1 until I treated the room and achieved 200:1, but the on/off was still under 3000:1.
I wouldn't let a bad room stop me from buying a high on/off capable pj as the benefit will always be visible. The ANSI will suffer of course, but that's less important in the scheme of things and is usually less visible with content.