AVS Forum banner
Status
Not open for further replies.
1 - 10 of 10 Posts

·
Premium Member
Joined
·
24,981 Posts
Discussion Starter · #1 ·
Hi Thumper,


Have you ever tried to improve contrast on an LCD projector? I don't see any reason to expect that LCD's have any better control of internal scattered light than DLP's, with potentially similar possible improvements.


In particular I'm wondering how the Sanyo XP21 would respond.


Thanks


------------------

Noah


[This message has been edited by noah katz (edited 08-29-2001).]
 

·
Registered
Joined
·
1,515 Posts
I'm sure Thumper will respond to this, but I think that most of his black level improvement comes from blocking the clear section. Also filling in the gaps in the color wheel.


The thing is, the reason LCD's have poor blacks is because when the panel is supposed to be completely opaque, it's not, and light still gets through. He can't do anything about that. With a DLP, for black, the mirrors are off, and that does a much better job of stopping light than an LCD. So the DLP mods address the most significant black level issues for a DLP. But for an LCD, unless you can darken the LCD panel, darkening the chamber and stopping reflections would probably only make a small improvement.


Mike



------------------

Join the NYC-area HT club! Send me e-mail!
 

·
Premium Member
Joined
·
24,981 Posts
Discussion Starter · #3 ·
Mike makes an interesting point. Could you guesstimate how much of the total improvement on DLP's comes from the various mods you do?


Thanks


------------------

Noah
 

·
Registered
Joined
·
375 Posts
Hi Noah,


Interesting question. As a matter of fact, early on I tried some mods on a couple of LCD units; a Sony 400Q and a Mitsu LV-2000. The primary reasons (and I won't go into much detail here) for low contrast ratio is the "lack of perfection" in the polarizing elements of the engine. Namely the primary polarizer plate allows a bit of non-polarized light thru, and the same goes for the LCD panel's "twist" as it cannot completely turn the 100% of the beam.


I tried stacking a second (vertical orientation) primary panel and this helped the contrast ratio but at the expense of about 20%+- of the overall brightness. A secondary horizontal "cleanup" polarizer was added after the LCD panel with the same results: better contrast but also at the expense of 20%+- brightness. Together they cut the brightness 40% but did yield better contrast. Its understandable why manufacturers don't do this; their primary goal is still the brightness spec and thats unfortunately what sells the most projectors http://www.avsforum.com/ubb/frown.gif


As to the amount of improvement with DLPs, it depends on a lot of factors within a given make and model. 30-60% contrast improvement is the norm but it does vary greatly. For instance, a very dirty DLP engine will have poorer initial contrast ratio so the end result after mods is more improvement.


Interestingly, as the contrast ratio has improved from the manufacturers, the better the mod results yields (the little LT150 is a good example). In lower contrast units the correction of scatter light, etc. makes less of a difference (in proportion) than in high contrast units.


Thump
 

·
Registered
Joined
·
779 Posts
I pulled some of this out of a previous post of mine cause it is probably more suited here. A process for improving contrast on the 10ht, which is a 3 panel LCD projector, has been developed. I have no idea if this would work for other LCD projectors but I might be worth a try. I guess it would depend on if the other LCD projectors have the same problem with generating a level of red equal to green and blues. If I understand correctly the red generation problem it is more a function of the bulbs ability to produce red at the same level as the greens and blues, so this process might be useful to other UHP bulb LCD projectors. While the filter process does knock down brightness the increase in RGB gains offsets the light loss.


The process is relatively strait forward. Like I already said, it has been observed that the 10ht lack sufficient red. By using a red color correction filter you can boost the gain on green and blue with out blowing out either one. I am using a red 40 cc filter with my RGB gains set to the highest level. If you would like to understand it better, here is the website of the guy who came up with a process. http://home.pacbell.net/steve367/ Using a $2 photo electric sensor and spreadsheet to so you can make the adjustments while keeping the color balanced. One of the key assumptions is that apparently Sony has done a good job in calibrating the color accuracy at the factory. The process uses use this baseline to allow you to adjust gains and apply filters while keeping the ratios aligned with the original calibrated gamma curve. For a better understanding of the entire process read http://home.pacbell.net/steve367/Intro.html . If you are not overly concerned with exact color accuracy you can use filters with out the spreadsheet and just use AVIA to work out the proper contrast, brightness, and check color calibration. The CC filters are stackable, I purchased a 4X4 20CCR filter which I later cut in half and stacked to create a 2X4 40CCR filter. Before you do this you would want to make sure that the size of the throw as it exits your lens and where it would hit your filter is less than 2 inches. On my 10ht I have about a 1/4 inch buffer on each side. I have attached the filter to a cardboard cutout which is affixed to the lens casing by masking tape at the top and bottom.


I have applied the color correction filter technique to my 10ht and the visual perception of improved contrast is obvious but I can only make a purely a qualitative observation. I wish I had the proper equipment to measure the actual quantitative change in the black level/contrast. I have no idea how it now compares with lt150 or other projectors, and refuse to join in the on the my projector is better than your is cheerleading, that is so prevalent in many posts. There has been some speculation that Sanyo may have implemented a similar color correction filter fix, which they integrated into the projector, again that is pure speculation. Sanyo uses 16:9 LCD panels from Sony and they are the same panels used in the 10ht.


Thumper, you have lots of experience in working to improve contrast do you have any thought about this process?


Regards,


Brian



[This message has been edited by btmoore (edited 08-30-2001).]
 

·
Registered
Joined
·
375 Posts
Hi btmoore,


Yes, I'm familiar with the panel yield "balancing" technique you described. The process increases contrast mostly by increasing brightness yield and not by reducing black levels.


In the case of DLP's it is the opposite; modest or no decrease in brightness (discounting the color wheel' clear section's cheater boost) with increases in contrast derived from lower black levels.


I believe this has been covered here before, but the recent advances by Sanyo in contrast ratio have been from the use of more efficent polarizer plates that do a better job of blocking stray non-vertically polarized light (along with some other changes such as changes in the light integrator). I don't really think lightning struck twice with Sanyo when comparing contrast ratios between the XP18/21 to the PLV60 (I know what the spec says, but...)

I was somewhat disappointed in the Sanyo PLV-60. All I could obtain was a 430-1 ratio. A little better than the HT-10 but not leaps and bounds, particularly after a full Sony calibration.


Thumper

 

·
Premium Member
Joined
·
24,981 Posts
Discussion Starter · #7 ·
Thumper,


Thanks for the education about polarizers.


That's interesting about contrast improving more on projectors with good contrast to start with. Any plans to try your hand on the Sharp 9000?


------------------

Noah


[This message has been edited by noah katz (edited 08-31-2001).]
 

·
Registered
Joined
·
460 Posts
Thumper,


Did you try to fully calibrate the PLV-60 with the same techniques that are used on the 10HT? I am helping Steve create a version of SMART for the PLV-60. So far, I have seen a 50% improvement in contrast on my PLV-60.


If you were seeing 430:1 without the tweaks, you should be able to get around 645:1 with the tweaks which is closer to their stated 700:1 ratio. Does this sound right?


Robert
 

·
Registered
Joined
·
8,781 Posts
Other threads have discussed the improved black level of the new crop of 16:9's, like the 11HT and the Sanyo PLV60.


I thought I read in several of these discussions some months ago that one of the reasons for the improved blsck level was due to increased voltage sent to the LCD panel.


I am assuming that on most LCD projectors, "black" is created by the application of voltage to the LCD panel, rather than the absebce of voltage. Hence, driving the LCD panel with higher voltage makes the LCD panel more opaque- kinda like "overclocking" your LCD panel.


The point here is, could current or older LCD projectors be modified by bumping up the voltage to the LCD to improve the opaque level when "black" is present in the video signal.

 

·
Registered
Joined
·
343 Posts
CC XXR filters do actually improve the black level, and especially the perceived black level as they do significantly attenuate (e.g. 50%) the green and blue light levels, thus reducing the LCD panel light leakage component of the black level.


If at the same time you proportionally turn up the "gain" on green and blue you get back most all of the light intensity at higher IRE levels, essentially double the contrast, and maintain proper grayscale tracking. Obviously doing this accurately requires measurements.


One of the reasons this works so well is that the eye is most sensitive to green in perceiving luminosity, especially at low light levels (where blue is also important). So the perceived improvements in black levels using this process can seem very real indeed. With a CC 30R filters my black levels are at least as good as when I was using a 50% transmission ND filter, but with the CC filter I have twice the contrast and light intensity for the brighter scenes.


Steve



------------------

Steve Smallcombe
 
1 - 10 of 10 Posts
Status
Not open for further replies.
Top