View Full Version : Cumulative Gammas


mark haflich
05-04-08, 02:32 PM
Suppose one has a processor that has single digit gamma curves. For example a Lumagen Radiance where one can insert a system gamma of say ..85 to 1.25. This is a gamma systen number divided by 2.2, the universal camera gamma. Then say you have numeric projector gammas say from 1.8 to 2.6. How does one fiqure the combined syatem gamma of the two? thanks in advance.

sfogg
05-04-08, 02:46 PM
"For example a Lumagen Radiance where one can insert a system gamma of say ..85 to 1.25. This is a gamma systen number divided by 2.2, the universal camera gamma. "

That isn't how the Radiances control works. Per Lumagen's site:

An example on using the new gamma factor setting: if you've measured your displays' gamma to be 2.2 and you prefer 2.4 then set the gamma factor to 1.1 (2.4/2.2 = 1.09)

In other words:

gamma desired / gamma measured = gamma factor in Lumagen

Shawn

mark haflich
05-04-08, 04:23 PM
Suppose you set the projector gamma at say 2.4.the choices from the lumagen are .85 to 1.25.how do they interact?
sSppose you want an end 1.2? how do you get there?.

Mark Petersen
05-04-08, 09:09 PM
Suppose one has a processor that has single digit gamma curves. For example a Lumagen Radiance where one can insert a system gamma of say ..85 to 1.25. This is a gamma systen number divided by 2.2, the universal camera gamma. Then say you have numeric projector gammas say from 1.8 to 2.6. How does one fiqure the combined syatem gamma of the two? thanks in advance.

The VP50 Pro also has a single digit gamma control, but it apparently works the opposite of the Lumagen. Lowering the VP50 gamma raises the final display gamma. Since the VP and display gamma's interact it's not possible for the VP to know what the final "end" gamma will be so I assume that's why the control works more like a relative adjustment from 1.0. If you really want to know the final gamma, the best thing to do is measure it. Often times the display gamma curve doesn't follow a single number anyway. Shadow detail/contrast at the bottom of the luminance curve is so heavily influenced by gamma that oftentimes a projector manufacturer will offer several gamma settings that primarily affect only the bottom of the luminance curve which is the case with the JVC RS-1.

sfogg
05-04-08, 09:15 PM
"Suppose you set the projector gamma at say 2.4.the choices from the lumagen are .85 to 1.25.how do they interact?
sSppose you want an end 1.2? how do you get there?."

Gamma desired / gamma measured = gamma factor

If you desire 1.2 and your projector is measuring at 2.4

1.2 / 2.4 = 0.5

Shawn

mark haflich
05-04-08, 09:34 PM
The 2.4 set at the projector as I understand it must be divided by the camera gamma of 2.2 to give the system gamma of about 1.14. How do the lumagen values of .85 to 1.25 inteact with that numerically to give some othe gamma? sory for my being so dense and inarticulate hee. thanks to all responding. I appeciate all the posts.

sfogg
05-04-08, 09:43 PM
Camera gamma has nothing to do with this. All we are talking about is the displays gamma and the factor in the Lumagen to alter it.

gamma desired / gamma measured = gamma factor

gamma desired = gamma measured * gamma factor

If your display is 2.4..

2.4 * 0.85 = 2.04
2.4 * 1.25 = 3

Shawn

Alan Gouger
05-04-08, 10:17 PM
The VP50 Pro also has a single digit gamma control,

Mark the gamma gain and offset controls on the pro are broken and do not work properly. Everyone is waiting for a fix. if you go more then one click everything posturizes.

Mark Petersen
05-04-08, 11:18 PM
Mark the gamma gain and offset controls on the pro are broken and do not work properly. Everyone is waiting for a fix. if you go more then one click everything posturizes.

Good to know, thanks Alan. For what's it worth one click is all I've ever used for critical viewing. Since we're on the topic though I have run the Gamma settings all the way up on the VP50 Pro (not for critical viewing but it was part of an "experiment" that I was doing). I saw severe dithering noise and banding on anything with a fairly smooth and continuous greyscale ramp pattern (not a greyscale step pattern but a smooth ramp). It showed up first on a Tivo "now playing" menu and then I verified it with the internal VP50 Pro greyscale ramp test pattern. Initially I assumed that it was the VP50 Pro but it looks identical (although much more severe) than what I've seen on the RS-1 before (including with the internal service menu test patterns). It'll be interesting to see how much of it is VP50 generated and how much of it is RS-1 related.

Erik Garci
05-05-08, 01:32 AM
The 2.4 set at the projector as I understand it must be divided by the camera gamma of 2.2 to give the system gamma of about 1.14.
Camera gamma is approximately 0.50, if you use a pure power function to approximate the Rec709 transfer function.

Look at this graph: gamma.gif (http://archive2.avsforum.com/avs-vb/attachment.php?attachmentid=25025)

mark haflich
05-05-08, 09:07 AM
Thanks guys. So setting it 1.0 in the Lumagen bypasses the function.

Cameron
05-05-08, 03:59 PM
That is what I was assuming too. I have my RS2 at 2.4 and the Radiance at 1.

mark haflich
05-05-08, 05:59 PM
I checked with Lumagen and we are correct. Setting it to 1.0 multiplies it by one, no change. Above or below its a multiplier. So setting it in the Lumagen to 1.1, would raise a for example 2.4 gamma to 2.64. The default in the Lumagen is 1.0.

Once again, thankls for the help guys. You are the best.