Originally Posted by coderguy
The average gain and factoring in the difference in seating position for any non-uniform or retro-gain screen should also be done to the post-converted fL measurement AFTER you measure the regular ANSI LUMENS, not from measuring fL points.
Let's see how this holds up. Remember, you were the one who wanted to make sure people took non-uniformity into account, which is what taking 9 (or 5) measurements is about.
Here is a hypothetical case.
For some setup there is a screen with 50% falloff from the center to each of the 8 points outside the center that correspond to the points used for ANSI lumens.
Projector A and Projector B have the same ANSI lumens, but Projector A is completely uniform for the 9 points while Projector B is twice as bright for the center point and half as bright for 2 of the edge points (thus the same average and ANSI lumens as Projector A).
If those projectors are shot onto that screen with that setup and we call the ft-lamberts for the center point from Projector A 1 unit, then Projector A and B have the following breakdown of points with different units of light:
1 unit: 1 point
.5 units: 8 points
2 units: 1 point
.5 units: 6 points
.25 units: 2 points
The average coming off the screen for Projector A is thus .56 units. The average coming off the screen for Projector B is thus .61 units.
Using coderguy's logic that you first need to average the 9 points from the projector for ANSI lumens before calculating the ft-lamberts creates mathematical error. It results in the same ft-lamberts for both cases when the average ft-lamberts are different (and so is the center ft-lamberts). The whole point behind doing ANSI lumens is to get an average for the thing you care about, not to do the math in an incorrect way when figuring the ft-lamberts that introduces error which was not necessary.
That was just an example and the amount of error can vary. Of course there are other errors in the system, but why add one that isn't necessary and is just doing the math wrong?
To be clear, I'm fine with people doing things like just using the center or brightest spot from the screen if that is what they want to do as long as they say what they did. It was coderguy who said he wanted to make sure people took non-unformity into account and all I did was apply that to the ft-lamberts question and his claim that you first get the ANSI lumens before taking screen gain and size into account.
Originally Posted by Seegs108
Not everyone is looking for perfect results. FtL reading especially. I don't think people are taking half gain numbers of the screen into consideration when they're calculating the screen brightness.
As I said, I am fine with people not taking non-uniformity into account if they want to. Just say (or at least imply) what you did.
I wasn't the one who brought up taking non-uniformity into account. I don't have a problem with that, but if a person should be taking the non-uniformity of the projector into account when figuring ft-lamberts why should they ignore non-uniformity of the screen? If a projector has 50% falloff toward the edges with a uniform screen should you treat that different than a uniform projector with a screen with 50% falloff as far as calculating ft-lamberts? That is, if two setups have the same amount of light coming off the screen for all points should we say they have different ft-lamberts just because the non-uniformity came from a different factor?
Seems to me that if you really want to make sure you take non-uniformity of the images into account when determining ft-lamberts you should take it into account whether it is from the projector or the screen. And if you are going to take the non-uniformity of the projector and the screen into account why not do the math correctly instead of averaging one before applying the other (which can throw your result off from the real average and isn't necessary)?
--DarinEdited by darinp2 - 10/24/13 at 10:52pm