AVS Forum banner

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

·
Registered
Joined
·
233 Posts
Discussion Starter #1
To all the constructive keepers of the Knowledge, please help!


I have a G15 on 2 week evaluation from Medical Video Systems that was supposed to be a showroom demonstrator and is supposed to be calibrated.


In spite of very high motivation to see the positive points, I am very disappointed. Either the unit is not calibrated in the way that all of the satisfied JVC owners on this forum reference or this unit's optical path is messed up. I have it ceiling mounted and yes, I unscrewed the screw you are supposed to unscrew for inverted mounting. (shipped to me that way)


I understand that you need to tweak these machines but since I have never seen a DILA before, I don't know whether what I am seeing is correctable or is something that JVC DILA owners live with or what.


My expectations were, since this unit was supposed to be calbrated and was used in the showroom, I would at least like the picture for 5 minutes before my critical eye engaged! Not so, alas.


[Source: HDTV off DISH network at 720P with Computer2 input set to receive HDTV(720)]


Problems I am seeing:


->>the contrast is less than a runco vx101c (which is supposedly 500:1) I have used VE but since it was fed through an inferior scaler (see below), I could not get the below black bar to show and could not see mich improvement adjusting the brightness and contrast from the 0 (zero) settings it came with.


->> and the killer: on a white field screen, the left half of the image is reddish-orangish tinged and the right half is bluish tinged, both tinges increasing toward the left and right edges of the screen, respectively.


->> colors, which I expected to be a strong point of the Xenon lamp, look fuzzy, sort of like a CRT that is out of convergence, counteracting the higher resolution matrix by desaturating the color and decreasing the image sharpness.


I have the QS Elite also on evaluation feeding DVD source to the projector but my first impressions (5 hours last night fiddling with it) of the Elite are far from positive; the de-interlacer in the Elite is terrible and the color rendition on known DVD's played on the same DVD player (no doubt exacerbated by the color convergence problem) is much inferior to the runco vhd processor that comes with the vx101c.


The Elite is absolutely no help in evaluating a machine of the G15's caliber so I am only using HDTV to evaluate the image.


I had always planned on an HTPC/Dillard/DScaler upgrade soon anyway but needed a scaler in the interim.


No flames please! I want to believe, believe me! I have waited a year to upgrade to the G15 and I am puzzled by what I am seeing.


I see the rainbow effect and the motion dithering that seems to be intrinsic to single chip DLP's so I sought the superior color and resolution of the G15. I am not prepared to lose contrast, white field uniiformity or color "convergence" to gain finer grain resolution.


Is this something that a specialized DILA calibrator can adjust?

The quarter wave plate adjustment? Or is this one of the cases that rbmartin is referring to in his following informative posts:



D-ILA's faults:


1. White field uniformity is often poor. The design is flawed in several ways. The LCD-based D-ILA chips have a certain nonuniformity which is unavoidable. The projector has a set of service adjustments which are designed to electronically compensate for this. The trouble is that the service adjustments aren't that well thought out and don't allow you to fix many common uniformity problems, like a pinkish or greenish corner. To compound the problem, the optical design is flawed in that the light path is reversed between the red and green dichroic mirrors. This means that any sort of lamp or optical misalignment affects the red and green chips in an opposite manner and creates a red/green gradient in the picture. The G10 wasn't too bad in this regard, but in the G11/G15 they changed the first condenser lens past the lamp to get more light output, and made the projector even more sensitive to alignment problems.


2. The convergence on a typical D-ILA is usually off by at least a half pixel somewhere on the screen. You can't fix it because the D-ILA chips are soldered to the optical block at the factory and can't be adjusted.



There are no dead pixels and I was able to focus the lens to my satisfaction but this terrible white field uniformity and fuzzy colors, leading to an image that looks less crisp (in general) than an 800X600 DLP, make me wonder what is going on with this unit.


If there is hope for this unit, since there are no dead pixels at least, then I won't just return it immediately. Please, please give me some useful feedback on what is going on.


All G15 experts, please advise!

Thanks for your attention.
 

·
Premium Member
Joined
·
4,176 Posts
Hi,


Reddish on one side of screen, bluish on other side.... It reminds me of a CRT projector on a high-gain screen. I wonder if this is something that DILARD can adjust....?


Sharpness problems... That's strange for HDTV. Everytime I saw HDTV on a DILA, it was pretty sharp. Are you using high-quality cables between your HDTV box and the DILA? Walk right up to the screen: Are the individual pixels clearly visible at all corners and the center of the screen, as they should be when looking only 12 inches away from the screen? If so, then it's a focus problem -- you may not be aiming your DILA projector squarely to the screen (ie, pointing half degree off-axis, may even cause noticeable focus uniformity problems).


In any case, the color nonuniformity would indicate that DILA is actually uncalibrated or something is wrong with that DILA projector. If it is feasible for you to obtain DILARD (could be good insurance) you might want see how calibratable some of these defects are.
 

·
Registered
Joined
·
842 Posts
The color tinges are a result of bad shading. It is certainly true that the factory tool is miserable at adjusting this, but if you play with it enough you will find that you can in fact adjust just a corner. Gamma calibrations or extreme black level adjustments will certainly affect the shading. The shading arrays are actually quite flexible (like the gamma tables) but unless you plan on writing a program to do this yourself you will either have to use the factory tool or wait till the shading wizard ever gets released as part of dilard.
 

·
Registered
Joined
·
233 Posts
Discussion Starter #4
Thanks for the reply Mark and Kevin:


The center of the projector lens is centered on the screen to a 16th of an inch, perpendicular to the screen, and the pixels are clearly resolvable from 2 feet or so and are nicely framed by their tiny fill ratio frames at that distance and look focused in all 4 corners.


I don't claim to be an expert on DILA's, other than what I have read on the forums for the last year!, but I have had significant experience in setting up projection screens and projectors.


The cables are nice and fat, cost around 140 for 35 feet and have the small, no-nonsense professional connectors on them. Good for 450mhz bandwidth, supposedly.


I would love for this to be an "operator error" of some kind!

What is frustrating my visually is that the underlying resolution and focus is great, but either the panel convergence problem causes this or could the projector's internal scaler on Computer input 2 is dithering the 720P signal poorly, causing the image to seem like a slight, shooting-through-cheesecloth, soft focus effect.


Dillard is a tool that will probably take longer than the 12 days I have to decide the fate of this G15 unit to learn to use effectively especially since I have no DILA experience.


The search continues...

Thanks!
 

·
Registered
Joined
·
780 Posts
Robert: May I also suggest that you try the 1080i output from the Dish 6000? With my G11 DILA, when I first set up the Dish 6000, I also used the 720p output of the Dish 6000 with the G11, but found that the 1080i provide better PQ. YMMV ,as my Dish 6000 is one of the first shipped at the product rollout way back in the Dark Ages.
 

·
Registered
Joined
·
233 Posts
Discussion Starter #6
Thanks jcthomas:


Yes the 1080i feed from the Echostar6000 is much cleaner, probably because the sat. receiver is receiving 1080i and does not have to convert to 720P.


Using the 1080i SHO or HBO feed, the color convergence is much better.


Points of concern in my evaluation:


Overall contrast and black level still appear to be less/higher than a runco vx101c so I hope that this can be improved through calibration.


The non-uniform white field issue may be correctable with quarterwave plate adjustments.


I am emailing the recommended calibrators with these two points.


I have noticed that when I compare a Direct view Sony monitor (VE adjusted) to the G15's picture, though the direct CRT wins on overall contrast/ black level and consequently on apparent color saturation, the accuracy of the color or the actual hues are nearly identical!


Much better than any digital projector I have seen so far. Usually the comparison between little CRT to big digital proj. would visibly differ in color hue as well as the expected differences.


This is vindication (for me) in insisting on the xenon bulb as a light source; to get the color accuracy to match the mature CRT colorimetry seems like an engineering coup!
 

·
Registered
Joined
·
914 Posts
If you happen to have a PC handy, it would be instructive to see a DVD played through it to a 1360x1024 image. It's impressive.

-=-

Mark
 

·
Registered
Joined
·
233 Posts
Discussion Starter #8
Greetings all:


The gamble paid off!


I kept the G15 I was evaluating (it was supposed to have been already calibrated) after some hopeful email communication with a couple of the word-class calibrators who monitor this forum.


Luckily, Mr William Phelps (Meier-Phelps Consulting) was available to perform his magic on the unit. It turns out that the previous "calibration" was no more than a 16 point gamma adjustment and a (now documented) way off gray scale tracking adjustment.


My initial confusion over Picture Quality (PQ) stemmed from what looked like an almost deliberately de-calibrated state of the unit when I received it.


Gambling that my unit was not one of the recalcitrant machines that cannot be tweaked beyond (or up to) stock calibration, I sent the unit to William with the following concerns:


->>the contrast is less than a runco vx101c (which is supposedly 500:1) I could not get the below black bar to show and could not see much improvement adjusting the brightness and contrast from the 0 (zero) settings it came with.


->>black level is too high


->> and the killer: on a white field screen, the left half of the image is reddish-orangish tinged and the right half is bluish tinged, both tinges increasing toward the left and right edges of the screen, respectively.


->> colors, which I expected to be a strong point of the Xenon lamp, look fuzzy, sort of like a CRT that is out of convergence, counteracting the higher resolution matrix by desaturating the color and decreasing the image sharpness.



Post calibration results

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



Drum roll... anticipation builds... 8=>


Tech data:


Initial CR(contrast ratio): 226!! (no wonder about zero punch in image)

Final CR: 453
 

·
Registered
Joined
·
81 Posts
Hi:


Glad to read that your G15 is performing up to snuff.


I bought a G11 last Spring from MVS with the same claim of callibration. It had many of the same faults that you described. MVS claims that their callibrator was trained by Richard Martin (known guru) and that they had purchased the equipment that Richard recommended they use. At that time no one was quite up to speed on callibrations so I returned it to meet a time limit. A even worse experience with Dukane clone soured me on DILA. After a demo from a forum member of his G11, I wished I had kept the G11 I originally had purchased.


MVS people treat you really nice. If they were as good at setting up their products as they are at refunding your money; they would be one top rated company in my book.


I hope you got a good deal as AVS is selling G15s at a very reasonable rate.


Again, glad you solved your problems and enjoy your G15.


Happy New Year To All
 

·
Registered
Joined
·
4,720 Posts
Quote:
Originally posted by robobob


My hat is off to William and to all the believers in what a calibrated DILA can do!
robobob,


I second the above sentiment - William is particularily adept at turning ugly ducklings into

graceful swans.


After William calibrated my G15, he told me the original contrast ratio was 185 - worse than yours.


When William was done - the G15 had a contrast ratio of 550, and color temperature tracking graph

that was flat as a pancake.


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