So I finally had some time last night to put my Display 3 meter & ChromaPure calibration software to the test and measured how my RS56 does out of the box after ~500 hours of use. I've always thought that the results (pre-calibration) looked pretty phenominal and were probably pretty close to perfect and now I've confirmed it.Greyscale/Gamma:
The important thing to look at is the CIE94 error (DeltaE) values. This is a numerical value of how far off the greyscale is from the target from 10% to 100% light output (stimulus). While 0 is perfection, the goal is to try and keep DeltaE under 10 for each point if possible. A DeltaE of 2-3 or under is so close to perfect that most people cannot detect the subtle imperfection. My points are (IMHO) very good with the highest being 5.7 at 50%.
A little too much blue in in the middle range as you can see. Lowering blue or possibly raising the other two at the low end should fix it. A little bit of tweaking and these DeltaE numbers will get a bit lower but it's really questionable with already deltaE's so low if the difference will be noticeable.
Gamma tracks pretty well between 2.3 and 2.2 (which is where I want it). About 2.2-2.4 is ideal. Depends on your room (darker room = higher gamma ok), your preferences, and so forth). The RS56 has some nifty build in gamma adjustment tools at 12 points including 5%, 10%, 20%, etc... so it should be hopefully easy to tweak this slightly at the lower end. Again, questionable how much would be noticed on source material.
Note that my RS56 is set to the 6500K setting and gamma set to 2.3 but that doesn't mean someone else doing the same will get the same results as above (hence the whole reason why we calibrate and why there are controls to begin with).
The observant reader will note that I'm "only" getting 9.5 ftL of light output at 100% when SMPTE really says you "should" be doing around 12-14 ftL. I find this 9.5 in my darkened basement to be very bright and couldn't imagine or want more even after trying it out for a while. I find 12-14 ftL overly bright. Part of this may be how I've been using CRT projectors for years which tend to have good peak white but poor ANSI. ie: A small white window on CRT (or plasma) will be bright but a full white field will be dimmer - digitals don't do that. Case in point, my previous CRT projector was calibrated to do 13.8 ftL on 100% white window but looked a lot dimmer on average picture level (APL) content when the whole screen was filled compared to this projector doing 9.5 ftL on the same white window.
I have the lamp running on low lamp with the iris set to -6 (it goes from -15 [closed] to 0 [open] where 0 is brightest). So there's certainly room to go brighter if needed.Colour Gamut:
Again, the CIE95 (DeltaE) are pretty good with the highest being 3.6 for the green. This means that the primary/secondary colours are pretty good on my unit when using the "Film" colour profile/picture mode.
This doesn't tell us anything however about how well the colours track inside the triangle (inside the gamut). Some displays can be really off in there (example pictures here
) but still be great at the primary/secondary points.
So I still want to go in and measure inside the gamut to see how much adding a 125 calibration with somewthing like a Radiance video processor would help.
Even if it doesn't help much I do like the fact that the Radiance can also do 20 point greyscale and gamma, and do it all automatically in ChromaPure without me having to mess with the menus, or tweak the unit every ~6 months (because I know I won't), or even really have me have to understand what I'm doing. (My brain is full).
So I measured the primary/secondary numbers again at 100% saturation as well as at 75%, 50%, and 25% to see how well it tracked inside.
ChromaPure has a really handy feature to do this and the free AVS HD709 test disc has the patterns needed.(Note: Oddly enough the patterns under their ChromaPure section aren't in the order that ChromaPure asks for - they're laid out as 75%, 50%, and then 25% in the disc when you skip through but ChromaPure asks for the patterns in the 25%, 50%,75% order. So some jumping around was required. Still only took 5 mins to do however, just seems odd).
Here are the results:
So as you can see, it tracks pretty well inside the gamut (between 0 and 100% saturation). Very linear. The highest deltaE errors are mostly at the 100% saturations which can be corrected with the built in projector CMS and the others will likely follow a bit.
The highest dE is the green at 100% saturation but it's still under 4 which is (IMHO) pretty good and well below the 10 target that we want to achieve. Anything below 2-3 is considered (for the most part) perfect and not something we'll notice with the human eye. So it's even questionable whether the 100% primaries/secondaries are worth adjusting. They certainly don't look 'off' with regular source material which is what I noticed when I first set up this projector.(Note: I've got ChromaPure set up to use the CIE94 method of calculating dE since it's the default and what they recommend. ChromaPure does recommend keeping dE below 2.0 if possible for greyscale measurements and below 1.5 for primary/secondary colours, not below 2-3.)
While this doesn't show all of the billions of points possible it gives you a good indication of how well the projector does inside the gamut.
Using something like the Radiance 125 point colour cube correction through the advanced auto-calibration offered by ChromaPure to adjust inside the gamut will likely have little to no noticeable result.
Some others have noted that their RS56 was pretty darned close in greyscale/gamma/colour gamut out of the box and mine is definitely the same. Only minor tweaking will be required. Just make sure to go with "standard" gamma, "Film" colour profile/picture mode and/or 6500K.