Film-Tech Forum ARCHIVE: Luminance meter

This is topic Luminance meter 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=000845

Posted by Drew King (Member # 6349) on 07-11-2011, 03:36 PM:

http://www.bhphotovideo.com/c/product/329952-REG/Spectra_Cine_18007SABL_Spot_Meter_System_.html

could this piece of equipment be an acceptable solution for measuring fL in a professional setting?

My instincts tell me no, but curious of your opinions.

What would be the cheapest industry standard piece of equipment for fL measurements?

The minolta LS-100 is out of our price range.

My readings would be of a 182.5 inch silver screen.

Posted by David Zylstra (Member # 4230) on 07-11-2011, 07:39 PM:

I use the Sekonic L-758C for most readings, currently available from B&H for about $822 - it works well for everything but 3D through the glasses.

Posted by Marcial Feliciano Ramos (Member # 1989) on 07-11-2011, 07:59 PM:

We have a Minolta CS-200 and for us it works perfectly. But the price is quite high.

Posted by Phil Ranucci (Member # 3820) on 07-11-2011, 10:54 PM:

Are you using a film or digital projectors? A few threads on this already, but the shutter in a film projector makes it hard to use a regular photo meter. The Spectra Cine Spot is the film standard, I guess and is about $2500.

I'm curious as to why the Sekonic 758Cine won't work through 3D glasses? I just got one on e-bay and haven't gotten it yet.

Posted by David Zylstra (Member # 4230) on 07-12-2011, 08:51 AM:

My thoughts are that reading through the glasses is similar to reading 35mm - there is dark time involved (i.e. at 144hz every other frame is the left eye) - I would welcome any corrections if I am mistaken on this.

A quick test I did a few years ago between Sekonic and a Minolta:
With Z-screen in place, no glasses - Sekonic 14FL, Minolta 14FL
Through the glasses - Sekonic bouncing with a peak of 2.2FL, Minolta 6.5FL

Posted by Steve Guttag (Member # 268) on 07-12-2011, 09:59 AM:

Interesting, David.

Out of curiosity...were you measuring using a white test generated from a player or using a projector's built in pattern? You normally can not use the projector's pattern since the 3D device won't be receiving the pulses to sync up. If you use a pattern generated from the player, then the 3D device will sync and the flash rate will be quite high and your meter would have less of an integration issue, in my opinion.

Next time you do your test, wear a set of 3D glasses when looking at the white pattern...if there is a notable flicker to it...then you are not in sync and the reading, regardless of whose meter is invalid.

-Steve

Posted by Scott Norwood (Member # 30) on 07-12-2011, 10:09 AM:

Semi-related question: using non-D-cinema video projectors, I have measured "white" using both the internal test pattern generator on the projector and off of a tape (usually the white patch on SMPTE color bars, which should be 100 IRE, but also scenes in which the video fades to white, which I presume is also at 100 IRE). Invariably, the "internal test pattern white" is significantly brighter than "video white." This is with an SD-SDI or HD-SDI link between the deck and the projector, so cable loss (etc.) should not be an issue.

Why is this? It seems to be the case for several different models of video projectors (including the Panasonic 7700 and Christie HD10K-M).

I use a Spectra SC-600 ("Cinespot") that was calibrated about a year ago.

Posted by Robert Minichino (Member # 3495) on 07-12-2011, 11:06 AM:

On the SDI signal (assuming 10 bits for luminance), code 64 is black and 940 is white. If the projector is mapping 0 to black and 1023 to white, and using the corresponding values in its internal test patterns, this would explain the discrepancy. I don't know what the DCinema levels are supposed to be; maybe you can find an SDI signal with blacker-than-black and whiter-than-white on it, which should both show up the same as black and white if everything's set properly (64 should look the same as 0, and 940 the same as 1023).

Posted by Steve Guttag (Member # 268) on 07-12-2011, 11:14 AM:

One easy mistake some make is to not ensure that the color correction/color space is correct when making measurements. If you use a projector's uncorrected white, then yes, it is going to be brighter than the white REC709 signal that has to be mapped properly.

This is another reason to always use the test patterns off of the player when doing 3D calibrations (even non-Dolby). In fact, on some projectors I've found that just asking for "white" will get you the uncorrected white pattern...this will have the resulting picture too dark (or darker than you think it is).

My tests have shown that, on average, color correction alone accounts for about 3% light loss. Screen age, port glass discoloration, notch filter alignment...etc will generally make that number worse.

I have not done any tests to specifically show variances between DCDM XYZ space versus P3 space versus REC709...etc using various patterns versus using internals or brightness variances between each type but it would be interesting (if I could ever find the time).

-Steve

Posted by Bajsic Bojan (Member # 4928) on 07-12-2011, 12:00 PM:

but can basic luminance meters measure color reproduction accuracy?

i always thought a spectroradiometer was necessary, like the pr655.

whats the most affordable tool for this?

Just for doing fL measurements i would use much cheaper 'photographer' units, unless you are doing installs on a regular basis, a 800$ unit that does fL readings of luminance on screen is a tad much for something you will probably do once a year or so? And wont you be wanting to get correct color set up at about the same time anyways?

Posted by Drew King (Member # 6349) on 07-12-2011, 12:33 PM:

NEC1200c throwing to 182.5in silver screen.
Dolby DSS200
realD "Post Production System"

Not doing installs. Work at one location and need to make different macros for different client specs.

Posted by Carsten Kurz (Member # 5396) on 07-12-2011, 06:01 PM:

I think, for simple checks, a digital photometer is sufficient. Of course it does not measure screen brightness, but projector lumens. But most people roughly know their screen gain anyway. A photometer should be okay for comparative measurements, e.g. between bulb changes, adjustments, etc.

Of course, if the money is there for a better instrument...

But I think todays average presentation quality would be better if every projectionist had a 30US$ photometer.

- Carsten

Posted by David Zylstra (Member # 4230) on 07-13-2011, 10:55 AM:

Steve - I was hoping you possibly had a better answer than I did on the through the glasses reading. . . . . yes the "bouncing" readings with the Sekonic are evident both on the projector (Christie) test pattern and DCP test content. I just double checked with a series 2 Christie and the built in test pattern gives the RealD XL a sync signal so there is no flicker with glasses on - 20FL from the projector test pattern yeilds a bouncing 1.5-2.2FL through the glasses, then 17FL without glasses from a 3D DCP white pattern also yeilds the same bouncing 1.5-2.2FL reading through the glasses. On one reading I did get the Sekonic to read a semi-solid 6-6.6FL but I could not repeat it.

My initial tests were to prove to our service company that their Sekonic meters did not reliably read 3D through the glasses - we were getting 15+FL without glasses but the tech was reading 1-2FL on his Sekonic with the glasses and telling me I needed a larger lamp, but other tests with better meters proved that 14FL without glasses yeilds about 6FL with.

I recall a comment somewhere about a way to configure one of the older Sekonic models to read 35mm, but since we are all digital I did not pursue doing it (I think it may have been a comment by John Pytlak years ago).

The Sekonic is definitely a good easy tool to use for setting and checking light levels.

Posted by Drew King (Member # 6349) on 07-14-2011, 12:01 PM:

Seems I should find another solution other than the photo meter since the majority of my calibration is 3D.

If anyone can get consistent 3D luminance readings (through the glasses) with a $1k or under piece of equipment please share with me.

David, I'm curious what yielded your "semi-solid" reading at 6-6.6 and how you CAN repeat it. Let me know if you learn anything new.

Thanks!

Posted by David Zylstra (Member # 4230) on 07-14-2011, 01:52 PM:

The Sekonic may still work, it is just a matter of knowing what measurement without glasses equates to your desired target with glasses. It would just take a side by side session with a meter that can read 3D properly with glasses to give yourself a basic chart of what to get without the glasses for various levels with glasses - the tech doing your install might be nice enough to help out while he has his meter out shooting colors.

When I have time I will experiment more with trying to repeat a solid reading on the Sekonic through the glasses.

Posted by Steve Guttag (Member # 268) on 07-16-2011, 08:10 AM:

Seriously...what is the value of a cheap meter whose reading is in question? I understand that the cost of proper meters is out of the reach of most individuals but that is truly besides the point. A device that outputs a number does not necessarily make it of value. Maybe it makes you feel better because the gadget in your hand says 14fL or 5fL or whatever but it doesn't mean that you are really getting that value.

And just where are you pointing that meter? Light is not a point-n-shoot situation to set levels. Even more absurd is the thought of using an incident meter to figure out the actual reflected light. It presumes that the screen's specular response remains consistent as it ages. It does not (neither in color or reflectance). The odds you hold the meter in the exact same spot each time also brings a flaw into the equation. If your cold mirror, or reflector or lamp gets changed, moved...etc your control point is moved and makes the reading completely invalid (no two lamps are made the same...hence they need to be adjusted on each installation).

What would likely work better as a "cheap" meter is to, right after a technician with a CALIBRATED light meter sets up your system...get a camera with adjustable shutter and aperture and a tripod. Clean the port where the camera will shoot through. First project a crosshair target (there should be one in all DLP projectors) and point the camera's center to that point, then switch to white light and adjust the shutter speed and f-stop for a "perfect exposure".

At least with that set up, you have a repeatable set up...your calibration will vary a bit as the camera's sensor ages (or if you damage the camera)...you will likely find that if you check from month to month, it remains pretty close. Don't bother using the 3D glasses (or 3D apparatus in front of the lens/light) because all you are looking for is a change in light level. Whatever the magic setting is for 2D or 3D for your camera...that is what you are looking for.

Note, you can obviously use your camera in the theatre, instead of the booth (especially since booths are becoming less accessible or non-existant but try to make sure you pick the same seat each time, if your lens is a zoom, make sure it is set to the exact same position each time (best to have it at an extreme to it is repeatable).

Cheap meters are not the way though...it may give one comfort to think they have measured a desired light reading but it is just comfort, not necessarily a real reading. It gives one a false-sense of accuracy.

What surprises me is this sudden concern about light levels...it is no more critical for digital than it was for film. Yet now, folks seem like they want to have the means to measure the light. I love the idea that people are concerned enough to try and put out the best quality show and don't want to discourage that but using devices that do not necessarily tell a true picture (no pun) are not the way.

Even the good meters need regular calibration if they are be of any value (not just light meter, most any meter). We spend a small fortune annually to keep our test equipment in proper calibration...both light and sound meters. The net result, I can take three different light meters (from different manufacturers even) and yield identical results...it gives credibility to the readings. I wonder how many people that set up the color on these projectors actually also VERIFY that when the correction is applied that the color is correct.

-Steve

Posted by Louis Bornwasser (Member # 3063) on 07-16-2011, 10:23 PM:

Thank you, Steve. You are 100% correct regarding the light meters.

We are, however, in a "Feel-good" universe where it is not required to be good, only to appear good. I have sent at least 20 meters back to Spectra since SONY projectors arrived. Louis

Posted by Phil Ranucci (Member # 3820) on 07-17-2011, 03:13 PM:

So what's the verdict on the Sekonic Cine for 2D digital? 3D? I know it won't work for film, but there is no way that I will be supplied with a Spectra. I like Steve's idea about the camera, but all of the installs we have are already done, so no opportunity to benchmark a camera.

Posted by David Zylstra (Member # 4230) on 07-17-2011, 05:07 PM:

In my opinion the Sekonic L758Cine is a good tool, and in my side by side tests with other meters (Minolta, Spectra, USL) has proven accurate in measuring 2D.