Our native mobile app has a new name: Fora Communities. Learn more.

What can I achieve with just a luxmeter

1532 Views 6 Replies 5 Participants Last post by Djorre, Feb 1, 2008

Djorre

Discussion Starter · Jan 22, 2008

Hi all,

I've been calibrating, or at least having a go at it, my PJ.

But I only have a Luxmeter.

I read an article once (can't remeber the link) that you can do a lot with just a Luxmeter.

What the author stated was that if you get all three colours on the right gamma track you get max grayscale resolution for each colour. Then if you find the right balance between the three colours to match a reference display at D65 you get pretty close.

He proposed a method using just a 'white' grayscale and measuring the serparate colours using filters in front of the meter.

As I don't have filters I used the HCFR program and measured the grayscale of each separate colour.

These were, as expected with an older 'business' projector, a long way off in some regions of the gamma curve.

With the ATI colour controll - program I used the separate controls for brightness, gamma and contrast for each separate colour, and was able to get all colours 'as good as it gets' on the gamma curve. The Gamma curve for the 'white' grayscale is also allmost dead-on.

As a result I did get a lot better overall contrast, in the dark as well as in the bright region. Before the changes I couldn't get both right at the same time.

The only thing left that I can change is the mix between the colours.

But can somebody with more experience give some tips how I could achieve this with just a lux meter??

I used my standard crt monitor display set at D65 as a reference (I know it isn't calibrated as wel, but it does have a very good overall picture quality and colour, I would be very exited if I could get close to that)

Then I 'played around' adding or removing some R, G or B until I was able to get near the image on my Crt monitor. (mostly by using the brightness control to adjust the clour bias towards R, G or B)

Now for my question, as I understand, D65 consists of some % R, some % G and some % B.

When I measure the relation between the amount of Rlux, Glux and Blux on my monitor, it's about the same as I measure on my PJ.

So, is there 'a magic combination' in amount of R, G and B lux to achieve reference D65 white ???

And if any of you would be so kind, could you measure the amount of lux at 255R, 255G, 255B and full White with the HCFR program on your calibrated PJ, please??

In that way I can check if I did an allright job or not, and maybe this could be a really cheap way to achieve a more scientifically based calibration than just changing brightness and contrast using a THX menu on a DVD.

Thanks in advance,

Djorre.

See less See more

Save

1 - 7 of 7 Posts

larry2007

· #2 · Jan 24, 2008

I am not sure why this thread has not gone anywhere. I have wondered about the combined use of a good lux meter and a DVD (with or without the filters) for calibration. It seems like this should be a good combination, resulting in significantly improved accuracy and speed of calibration, prior to obtaining a colorimeter of some sort.

What is a good procedure to follow, for a lux meter supplemented calibration disk (I have several, and can buy or download more)? Where to start (max light in a 100% window or 75% window, compared to max light in a smaller window or the same size window, but less light)?

As the lux reading is inversely proportional to the distance squared to the light source, what is the proper distance to measure things? I imagine the meter should stay stationary on a tripod, making pludge patterns useless?

How do these distances or procedures differ between the 6 or so display types (or at least the three that I care about; plasma, LCD, front projector)?

-Larry

See less See more

Save

derekjsmith

· #3 · Jan 24, 2008

You will get more out of a cheap Tristim meter than a good lux meter plus a good lux meter will cost more than a cheap Tristim, Spyder2 or Display2 LT.

Save

D-6500

· #4 · Jan 24, 2008

Quote:
Originally Posted by Djorre /forum/post/12888980

Hi all,

I've been calibrating, or at least having a go at it, my PJ.

But I only have a Luxmeter.

I read an article once (can't remeber the link) that you can do a lot with just a Luxmeter.

What the author stated was that if you get all three colours on the right gamma track you get max grayscale resolution for each colour. Then if you find the right balance between the three colours to match a reference display at D65 you get pretty close.

Thanks in advance,

Djorre.

If that is a light meter, it will allow you to set contrast/picture level properly, and better than any pattern.

Save

Joelc

· #5 · Jan 25, 2008

Quote:
Originally Posted by D-6500 /forum/post/12919354

If that is a light meter, it will allow you to set contrast/picture level properly, and better than any pattern.

And gamma...

Save

Djorre

Discussion Starter · #6 · Jan 28, 2008

Hi guys,

I did get a lot better picture along the grayscale.

When I compare the lux measurements along the grayscale of my original settings to those after the calibrations they are quite nicely along the gamma curve for each colour.

When I then wompare the 2 pictures, before and after, I do get a much better detail in both black and whites. I used to have crushed whites and that were jumping off the gamma curve in the high regions. Now, when I taken a screenshot of a 'snowy' scene, there's a lot more detail in it.

I could not get this problem out with just contrast and brightness.

I put my meter on I tripod at a fixed distance, this was about 60cm from the lens of the PJ.

So with my lightmeter I can get the colours on tha gamma nicely, the only problem is that I'm in search of the 'right amount' of R G and B at 100%.

When I experiment I can get whites anywhere from green to blue to red to anywhere inbetween. From blue'ish overbright white to yellowish white. What I'm looking for is what I need to get it to be in the area of D65.

Or am I completely missing the point?

Djorre

See less See more

Save

Djorre

Discussion Starter · #7 · Feb 1, 2008

Here's a screenshot from LOTR (sorry for the small size)

in the background you can see the increase in detail on the snow-tops of the mountains in the left picture (after gamma calibration) instead of the more blooming blue-white mountain snow on the right (before calibration). It actually looks like the mountain-tops are covered in clouds in the picture on the right.

In this next picture you can see the difference in white's between the left (after gamma calibration) and the right (before). The left picture looks a lot more natural, and you can see the "white's" on the right are really blue-ish and look unnatural. (there also is some added detail and the picture doesn't look so washed out in the bright area's)

I find it's rather easy to vary between yellow-ish white up to blue-ish white while still having the right gamma curves.

after all, the 3 things I can change via my my htpc are Brightness, Contrast and Gamma for each colour seperately.

So for different amounts of R, G and B in 'pure white' (at 255), I can get the gamma-curves to allign for each colour seperately along the grayscale.

What I'm still looking for is how much R, G and B I need to have at 255, and allign the rest with the gamma curve.

Could any of you guys be so kind take a little 'lux' measurement using the hcfr program at 255W, 255R, 255G and 255B.

I think it could be interesting to see if there is a sort of 'constant division' between the % R, G and B in D65 white, or if this is really meter-dependant.

Greets,

Djorre.

See less See more

Save

1 - 7 of 7 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.

Top Contributors this Month