View Full Version : Bulb ageing recognition on graphs ?


008
05-23-08, 09:19 AM
Say a bulb on a rear projector looses red over time ( quite typical ) how will this show up in rec, gamma, grayscale graphs ? This assumes you have already taken measurements when the bulb was new.


Cheers

nashou66
05-23-08, 09:25 AM
Do you mean Tube? Not a bulb in crt's. The picture will get more blue to green tint. the Graph for greyscale rgb will have a lower red readings, at least with a crt you can increase G2 and drive to get output back up, but eventually you'll over work it and need a new tube. How many hours Are on your set?

Athanasios

008
05-24-08, 05:18 PM
Do you mean Tube? Not a bulb in crt's. The picture will get more blue to green tint. the Graph for greyscale rgb will have a lower red readings, at least with a crt you can increase G2 and drive to get output back up, but eventually you'll over work it and need a new tube. How many hours Are on your set?

Athanasios

Im talking about bulbs in a digital rear projectors. Unfortunately they loose significant ammounts of spectrum.

Im using an RS1 and its lost about 20% of its red at 400 hrs.

I guess this would show up on a Gamut graph as the white point drifting towards Cyan ?

HappyFunBoater
05-30-08, 08:45 AM
Im talking about bulbs in a digital rear projectors. Unfortunately they loose significant ammounts of spectrum.

Im using an RS1 and its lost about 20% of its red at 400 hrs.

I guess this would show up on a Gamut graph as the white point drifting towards Cyan ?

Just curious, but how did you determine that the bulb is losing red? Are you using a Spyder?

I've got a Sony XBR2 with a VERY old bulb (almost 7000 hours) and recently tried to calibrate it with a Spyder. I cranked red bias and gain to its max and couldn't get enough red. In fact I also couldn't lower blue enough. Yet when I looked at the picture afterwards (using various grayscale ramps, etc.) the picture was HORRIBLY red. Apparently my old bulb is producing plenty of red, but the Spyder just had trouble seeing it.

Anyway, I'm just wondering if you're using a Spyder and are seeing the same thing I am.

Michael TLV
05-30-08, 09:23 AM
Greetings

Spyder is not a good device to pair with LCOS technology. Wrong readings.

regards

HappyFunBoater
05-30-08, 09:45 AM
Greetings

Spyder is not a good device to pair with LCOS technology. Wrong readings.

regards

I'm on a tight budget. How about the Display LT?

Michael TLV
05-30-08, 09:58 AM
Greetings

Check with Tom Huffman on that. I have never used the Display LT before. If it is a tri-stim device ... then there may be a problem.

Some programs have a built in offset for the technology type to give better results.

The bulb gives off UV light that the Spyder probe picks up and interprets as blue light. It generally thinks that there is much more blue in the image than there actually is. Hence a "corrected" image that looks way way too red.

regards

HappyFunBoater
05-30-08, 10:30 AM
From Tom's "Calibration meter shootout" thread it appears that the Display LT/2 is quite a bit better than the Spyder - especially on an LCOS. But it still seems to be not nearly as good as the iPro (which is outside my budget). I'm trying to figure out if the LT is "good enough". It may be better than the Spyder, but if it isn't good enough to trust then there's really no point in spending the money.

Thanks for your help.

sperron
05-30-08, 10:42 AM
For the money you are looking to spend, the only options you will have are the display 2 LT and DTP94. A display 2 is probably the better option as it's easier to get ahold of one. I ended up upgrading from a display 2 LT to an i1 pro because I wasn't quite satisfied with the results calibrating my SXRD.

On my SXRD I went from:

spyder 2 - way off
display 2 LT - pretty close
i1 pro - hopefully spot on (seems good)

Michael TLV
05-30-08, 10:43 AM
greetings

For those that are stuck using these devices, a good rule of thumb is that the error is always from 500 to 800 degrees to the red green side. aiming for 7200K toward the blue usually gets you closer to where it really needs to be.

Regards