Film-Tech Forum ARCHIVE: 3D peak white test pattern

This is topic 3D peak white test pattern in forum Digital Cinema Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=16;t=001207

Posted by Antti Nayha (Member # 4980) on 08-29-2012, 04:50 AM:

According to "The EDCF Guide to 3D Cinema" (March 2011), page 25, you should always use a dedicated "3D peak white test pattern" when measuring 3D light levels through the glasses.

I've seen a lot of people getting weird results when trying to measure 3D brightness with active shutter glasses and the projector's full-white 2D test pattern. Apparently that way you are ignoring the 3D dark time, resulting in a measurement which is much too high.

Is such a 3D peak white test pattern available somewhere? Or is it as easy as creating one from two full-white still images with OpenDCP?

Posted by Carsten Kurz (Member # 5396) on 08-29-2012, 07:10 AM:

How should a 'different' white pattern solve the issue of measuring through shutter glasses?

- Carsten

Posted by Ken Lackner (Member # 1002) on 08-29-2012, 11:50 AM:

Because when playing back a projector's internal test pattern, 3D settings are not activated, so the glasses probably would not sync and the pattern will not triple flash. Additionally, the desired MCGD is not activated. Playing content off the server, with the server set for 3D playback and the correct title on the projector activated, would solve this.

Posted by Antti Nayha (Member # 4980) on 08-29-2012, 12:02 PM:

Exactly. Just for the sake of science and fun, I made two peak white DCP's - one in 2D and another in 3D - and played around with them a bit. The brightness readings were as follows:

- 33.4 fL (2D peak white)
- 26.5 fL (3D peak white, without glasses)
- 2.8 fL (3D peak white, through XpanD X101 glasses)

The last two measurements were made with the projector in 3D mode, with 6:2 triple flash and XpanD's current recommended 3D settings (a dark time of 1100 µs).

I have to say was a bit shocked about the through-the-glasses reading. XpanD's marketing claims 17–18%, but I only got around 11%. Admittedly, the pair of glasses that I used for testing was two years old and not 100% clean... but I guess that just gets my test closer to an average real-world scenario.

If anyone's interested, I'll be happy to upload both test DCP's.

Posted by Monte L Fullmer (Member # 2797) on 08-29-2012, 12:35 PM:

quote: Antti Nayha
2.8 fL (3D peak white, through XpanD X101 glasses)

Worse than Drive-in fL's .. and with no 3D. Being that low, is the picture on the screen viewable being so dim?

Posted by Antti Nayha (Member # 4980) on 08-29-2012, 12:54 PM:

I didn't test with actual content yet, but these were not the "house settings" for 3D anyway. Just playing around with the lamp set to minimum wattage.

Still, my result just can't be right. If they were, the 3D light output would actually be only around 8% compared to the 2D test pattern with the same lamp settings. I'm going to do some more testing tomorrow to see if there was something wrong with my test clip and/or procedure.

Surely the Harkness Digital Screen Checker can handle triple-flash 3D?

Posted by Carsten Kurz (Member # 5396) on 08-29-2012, 03:34 PM:

Ah, alright, didn't get it you were comparing an in-projector test pattern to a DCP test pattern. Of course it doesn't make sense to measure through the glasses without the triple-flashing and glasses-sync going on.

Now the question is what kind of instrument did you use? And why 'minimum wattage'?

Of course it's irrelevant for relative efficiency, but you could easily get to decent levels by adjusting the lamp current? I guess that Xpand's numbers are 'best case', probably with select glasses and optimized flashing parameters.

- Carsten

Posted by Marco Giustini (Member # 4544) on 08-29-2012, 03:41 PM:

Try disabling temporarily the colour correction. If you gain a lot of light it's probably wrong.
Also, as said, those default values are usually improvable.