New xrite spectracal c6 questions?? - Page 2

Your professional calibrator was wrong.

The Klien K-10 is good to .0002cd/m

http://www.kleininstruments.com/k10TechnicalSpecs.html

I trust Derek!

And for me I can say, my C5 was not bad for calibrating Grayscale & Gamut.
But my experiences are, it was better to calibrate my Mitsu HC5000 in direct mode reading (faced to projector), cause the bulb wasn't any of the brightest. This has 2 benefits: 1. I can read 10 IRE without problems, 2. readings are faster cause the meter gets more light.

And to get a more correct D65 calibration in direct mode, I also dertermined the (little) screen offset by performing first a reflective and afterwards the direct mode reading (same IRE). As delta I got the screen offset x=0.0038/y=0.0104; as you can see, very low and not really significant for D65.

Out of interest, how stable is the C6 over, say, 50 readings of the same point?

I found there to be considerable variance with the C5 where other cheaper meters returned stable, repeatable readings. Less accurate, but that doesn't matter when they're being profiled to a spectro for low-light work.

I've seen other people report this too.

If you read the specs carefully, the accuracy and repeatability are specified only for >1 nits. So, methinks the pro calibrator may be right!

The meter is better than those specifications suggest. I would trust the guy making the calibration software, with his own lab, than an end user.

So if is 3 nits.. do you really think you could tell the difference watching content? I mean after all, isn't that what we do when it is all over and done.. watch Movies and sports and god forbid.. HD TV... it's all a bit "nit" picky isn't it?

I think I would trust the manufacturer even more - those are his specs after all!

From your link:

"Accuracy(Illuminant A):
Chroma +- 0.002 x,y (displayed value > 1.00 nits)
Luminance +- 1% Y, +- 1 LSB (displayed value > 0.01 nits)

Repeatability:
Chroma +- 0.002 x,y (displayed value > 1.00 nits)
Luminance +- 1% Y, +- 1 LSB (displayed value > 0.01 nits)"

So, it may be accurate for 10% gray but not 0%.

Accuracy doesn't fall off a cliff below 1 nit.

And it does specify the measurement range as .0002-10,000 nits.

With the latest improvements in CalMAN, the C6 is very stable on the displays I've used it on (plasmas, front PJs, and LED LCDs after profiling off my Jeti). I did have some repeatability problems initially, though.

Hm, this is something that was supposedly solved on the C5 with CalMAN updates, but was not the case in my experience. (maybe it didn't like the displays I tried it with? I'm not sure I want to take my chances with this.

Frankly, I don't find much benefit to most of the very low light readings as it is anyway. I've yet to find a display or processor that has fine enough controls to make useful adjustments there that doesn't introducing colour banding. (it seems to either be that, or the controls are so coarse that you can judge it by eye) Greyscale patterns may look fine, but shadow details start looking discoloured/murky in films.

What's the difference between the C6 and the EyeOne 3 Pro? Are they identical except for the fact that one is for use with CalMAN and the other ChromaPure?

TIA

No the C6 has a different firmware.
It has an adaptive exposure mode that slows down for accuracy in low light and speeds up to be as fast as possible in bright light.

The C6 also has more calibration tables and since the C6 is field upgradeable, we are able to push out new tables with updates to CalMAN.

Thanks for the info Sotti

With CalMAN v4.6 we have added another table to the C6 for Wide Gamut White LED.

We have found that White LED LCD displays types need two tables normal gamut and wide. So we are splitting the one we have into two. They are very close but had enough discrepancy we felt it was best to have two. Looks for this with CalMAN v4.6 which is in beta now.

Also in CalMAN v4.5 we added Quad pixel LED for RGBY devices like the Sharp quad.

I ordered one this weekend. Curious to see how good it is.