Owners ONLY thread >>> 60"/70" XBR2 <<< Settings/Tweaks - Page 36

Thanks for confirming that flat black paint is probably a better choice than cloth. I was leaning that way, except perhaps for the area where the blue light leak is. But from your description, perhaps a fabbed shield from aluminum with a 'light trap' maze pattern (an opening that goes around at least two 90 degree corners... lets air circulate, but the two 90 degree turns with a divider between them stops light from being able to negotiate the trap. Thin aluminum cut to shape with snips, painted black, and held in place with hot glue should do the trick. But it will take more time to deal with than a simple cloth covering. I'm surprised there's INTENTIONALLY any air circulating behind the screen. You'd think you would want still air in there.

With more experience behind me since my original comments on the inner working of the SXRD, I must say I don’t think it will be possible to make a significant improvement to the blue blacks with modifications, as I believe there is more then just a blue light leakage causing it.
I tried masking off the optical block temporarily and it had a negligible effect on the blue tinge to blacks.
My replacement OB just does not have the blue black problem to start with, so a new OB is the real solution.

I have not found iris settings to have any significant effect on grey scale tracking or color calibration, possibly because my OB does not suffer from black discoloration.

If you really want to make a significant improvement to your SXRD, fit a neutral density filter and tweak the iris settings in the service menu as I have done.
The 60-70% improvement in black level is very impressive and takes the SXRD to a whole new level. The somewhat flat two dimensional image that the SXRD tends to display is a thing of the past.

Have you tried to decrease low end blue bias in the SM? I don't know if anyone else has found this but in the SM White Balance (Panel WB 2) numbers 23 to 28 there are 6 CDWB_XX parameters. The RB, GB BB ones are for fine adjustments of very low blacks and the RG, GG and BG are for fine adjustments of very high whites...


Panel WB 2

23 CDWB_RG Fine tune Red Gain (Very White)
24 CDWB_GG Fine tune Green Gain
25 CDWB_BG Fine tune Blue Bias
26 CDWB_RB Fine tune Red Bias (Very Black)
27 CDWB_RB Fine tune Green Bias
28 CDWB_BB Fine tune Blue Bias

I didn't know what those abbreviations meant! How did you find out what the abbreviations meant and what the adjustments did?

Thanks for sharing... maybe I CAN get the 20% and lower fields to look less blue this way. I'll definitely be having a look at these tonight. THANKS!

Disto, this is interesting.... what was some of the figures You got & what did You use to see what You had ? How much did You need to fine tune the six fine tune settings ? This will help to get it closer,but need to use something to get it close other than Your eye's can see...

Why would an ND filter do anything an iris size reduction wouldn't do? Or are you talking about using a small iris AND an ND filter?

I'm happy with the black level in darker movies using Auto 2 mode. I mean the gross contrast ratio is over 13,000:1 which seems almost ridiculous. Do we NEED a contrast ratio higher than that? Tongue in cheek... I'm still hoping it isn't too long before laser displays replace EVERY display. Can you imagine? Absolutely zero output for black and the 3 beams will be able to be aimed at exactly the same point like 3-panel solid state displays. And the system will be able to scale individual pixel sizes instead of trying to upscale and such. If the laser display knows your screen size and distance to the screen it will be able to make every resolution fit the screen by scaling the pixels. No projection optics needed... just aim the beams directly at the screen. Front or rear projection will never look as good.

I am use both iris programming alterations and an ND filter.

Iris adjustment can only give you approx 30% reduction in black levels. The ND filter adds another 50% reduction on top of that, for a total of around 65%.

The SXRD’s are very bright, way brighter then is needed in a dim or dark viewing environment.
A 13,000:1 contrast ratio allows a full black screen to be 13,000 times darker then a full white screen, but since the peak white level is so high, blacks are not very dark at all.
To maintain the same high level of brightness and provide good blacks, a CR of more like 130,000:1 or more would be required.

I had my old Hitachi CRT RPTV running next to my SXRD for a while and the blacks on the Hitachi where dramatically better then the SXRD, so much darker it was ridiculous.
Any video content with dark scenes looked better on the CRT RPTV, even with the SXRD’s lamp in power saving mode and iris full closed.
I found I could not enjoy such content on the SXRD.

With the ND filter installed and the auto iris reprogrammed to maximize dynamic range, not just closed, I now have blacks over 60% darker then standard while maintaining good overall brightness and punch.
The SXRD is transformed, and is now a pleasure to watch in a dim environment and competes very well with my CRT RPTV for blacks.

The SXRD was always crippled by the need to be very bright so that it could compete in the showroom war, where brighter is better.
With its output toned down very significantly to suit a dim viewing environment, the SXRD’s true potential as an astonishingly good home theatre display is unleashed.
Overall contrast ratio is more like 17,000:1 now, and ANSI contrast is also improved due to blacking out the cabinet interior.

Owen, do your mods do anything to mitigate SSE?

Yes, but only because the very significant drop in light output makes SSE less noticeable.
On my 70” I can only see a small patch of SSE in the centre of the screen if I look for it.
I therefore don’t consider SSE to be a significant problem for me, I rarely notice it.

Hours of sitting in front of the screen in the SM and changing every parameter one at time while trying to see what the changes did on the screen.


Normally I use GetGray for gray scale. I have the Spyder2TV and use HCFR but the Spyder2 does not seem to work well for a RPTV. The only screen the Spyder2 works well on is a direct CRT. So I got a GretagMacbeth D65 bulb and a Kodak Gray Card and do an optical comparison. This bulb is very accurate. I also like a couple of the test patterns on the Spyder2TV DVD. One is a very dark image full of varying intensity squares and the other is a very bright image full of varying intensity squares. I adjusted the fine bias and gain using these images against the D65 bulb. When I get a chance, I will take a picture of my setup and post it here.

I still say SSE is self-induced. Yes, there is a texture on the screen and that helps make the image brighter than a screen with less texture, but you are fixating on the texture instead of looking through it to the image. It's almost like a depth of field thing where something in the foreground has your eyes focused on it (a cell phone) instead of being focused where the eyes should be focused (the road while you are driving). If you can look past the "cell phone" (screen texture), you'll enjoy images without SSE. It took me months to get over obsessing about SSE... if I can do it, anybody can do it.

I messed with the CDWB (I can't remember if that's the right abbreviation) gains and bias controls last night.

I don't think I'd call them "fine tuning" exactly... a single digit change makes a pretty sizable % change in the bias settings... 4% to 5% so you can't get it perfect. But I did get the 20% field WAY closer to 6500 than I've been able to get before without messing up 30% and higher. In fact, I'm now varying between 6440 to 6575 from 20% to 100% with Auto 2 Iris mode.

The Spyder2Platinum that came with ColorFacts Pro won't measure the 0% (black field) so I question the 11,500 reading I get at 10% - it doesn't look that blue to my eye, especially when you display a gray scale with all the steps on the screen together... if there are color shifts from step to step, the shifts are really obvious within that gray scale. Yes, 10% has a little blue in it, but it seems more like 7500 than 11,500 to my eye (which is pretty good from 34 years of working on image quality problems in a variety of imaging products for pro, consumer, cinema, printing & publishing, still, motion, analog and digital).

It took a while to get the "regular" bias settings to work right with the RB, GB, and BB settings... maybe 1 hour or a little more. I won't be able to do anything better than what I have now until I get a Minolta spectroradiometer next month. That will measure everything and eliminate the limitations of the Spyder2Platinum.

Someone else said they thought the Spyder2 wasn't good for RPTV... I'm not having any identifiable problems using it on a tripod (not stuck to the screen) and ALMOST touching the screen... except that I don't trust the 10% white measurements and it definitely won't measure 0% with the iris in Auto2 mode.

So, thanks for the tip on those controls, it got me 10% closer to 'perfect' calibration.

.
So what was your strategy with the ND filter? Did you use Auto2 mode and use the Service Menu to open the iris more on the top (bright) end? Or did you close the bottom end even more with the ND filter installed? I have to admit that the screen is freakin' brignt in power save mode with a 100% white field, but it sure makes images look great. Not sure I'd want to give up any peak white - or not much anyway.

Are the Auto2 iris opening settings for min and max obvious in the service menu or is the abbreviation difficult to decode?

I thought of it as "fine tuning" the ends, tips or extremes of the total gray scale. Anyways, I am happy you were able to tweak things closer. If ever you visit Montreal, feel free to bring your new Minolta spectroradiometer with you. In the meantime, I will try the Spyder2 on a tripod.

I program the iris to close as small as possible on dark scenes for best possible blacks and open further then normal for bright scenes to help compensate for the lower peak output available with the ND2 filter installed. Overall dynamic range (contrast ratio) is significantly increased, and the SXRD is plenty bright for a dim viewing environment with iris in Auto 2 and lamp power saving on.
For a brighter environment, Auto 1 and normal lamp power can be used if required.

To get the full story and keep up with the latest developments read the SXRD thread on DTV Forum, starting with these links.

http://www.avsforum.com/avs-vb/showp...postcount=1023


I am very surprised you can use a Spyder sensor on an SXRD. I have the Spyder 2 and no matter what software I use it is way off. The Blue-Green balance is very close, but Red is way too high. I believe this is because the HP lamp in the SXRD and other digital RPTV's gives off UV light that affects the sensor accuracy dramatically.
I use my Spyder to calibrate my Sony G500 CRT PC monitor, a task it was designed to do and does well, and then use the CRT as a comparator to set the white point on the SXRD and as a visual reference to evaluate calibration. I can then use the Spyder to calibrate the full range grey scale on the SXRD based on the readings taken at white reference and then do a visual check against the Sony CRT.
While this method is a bit tedious, I have been able to make the SXRD look virtually identical to the Sony CRT reference and achieve an excellent result.

It may be possible to program in the required error correction into HCFR, but I have not attempted it at this stage

In a totally dark room, I use a Kodak gray card lit by a partly masked GretagMacbeth D65 bulb.

The Spyder2Platinum I am using is brand new and is specified by Datacolor for use on front & rear projectors, CRT, LCD, DLP, and plasma. I don't know if this is a change from what they said in documentation for earlier Spyder2 colorimeters or not as I never saw an earlier one. The one I have comes with a perforated black plastic grid covering the sensor area... it can be removed for some applications. The sensor area is covered with a cyan filter that looks milky, almost opaque. Perhaps Datacolor has modified the Spyder2 to work correctly with newer projection displays. Their ability to sell the Spyder2 would be pretty compromised these days if it had response problems with most of the 1080p displays. I can ask one of the Datacolor gurus if the Spyder2 has been changed to make it useful for calibrating projector lamp-driven displays (front or rear).

As far as I know, the Platinum is basically the same sensor as a Spyder 2
The Spyder 2 sensor and data color software are set up to measure only LCD and CRT displays. Spyder 2 also comes with the perforated black plastic filter intended for use with LCD's.
CRT displays are supposed to be analysed without the add on filter installed.
The Spyder 2 sensor can be used with lamp driven displays or RPTV's, but the software must be told that it is measuring such a display so that it can compensate.
I have yet to see software with that compensation built in.

HCFR software has the option to use the Spyder 2 with or without the filter installed, but it is recommended to us the filter and set HCFR accordingly.
There is no setting for lamp driven displays or RPTV's, but it is possible to apply a compensation factor if you know what that factor is.

I found the Spyder 2 useless with both my CRT RPTV and my SXRD, both where much too high in red after calibration.

After calibrating the SXRD by eye, using the calibrated Sony CRY monitor as reference, I have worked out the correction required for X and Y coordinates in HCFR, so that the software shows the correct readings when the SXRD looks visually the same as the Sony CRT. This is obviously not going to be perfect, but it's close enough for me.
It would be helpful if someone with a high end instrument and a Spyder 2 could do some comparisons on an SXRD to establish the correction factors for HCFR.

Man, I have NOTHING like that... in fact the gray scale is incredibly neutral, even when using a reference gray target card for comparison. I can select HDTV color space in ColorFacts Pro, but there's no selection for the type of monitor being calibrated.

Is the filter on the Spyder2 you have a milky cyan color? Cyan would be absorbing some red.

I calibrated an LCD flat panel and it came out great, also calibrated the XBR2 and it looks perfect also.

Perhaps ColorFacts can detect the lightspectrum and automatically remove colors/frequencies outside the normal illumination profile. I don't have any other software to use with the Spyder2 that would reveal if ColorFacts is correcting a response issue or whether this new Spyder2 responds differently that earlier models.

Owen: Have you changed the setting on DIC_SEL? If I set it to 1 coming from the set being in MIN iris setting in the user menu the black level goes way down.(Lower than I could achieve by changing the MINIRIS setting in the service menu) Conversely if I come from the Auto 2 setting in the user menu and set DIC-SEL to 0 the auto 2 black level goes way up. Would DIC stand for Dynamic Iris Control? The Min setting at DIC_SEL 1 is intriguing as the blacks are much better and my max brightness level does not fluctuate. I had my set calibrated at Min iris setting and feel it would be best to keep it there for optimum color accuracy. By the way at this new setting the changes by lowering MINIRIS in the service menu stop at 350 for me.

I want to retract my statement that turning sharpness way down is not the right thing to do any more... sort of. I said "modern" displays may be best off set to 50-55 rather than 10 or less (on a 0-100 scale) because sharpness doesn't work like it used to. Well, it DOES sort of work like it used to on the XBR2s... at least with a Toshiba HD-DVD player as a source... 15 turned out to be the right setting for no artifacts and crisp edges (on a 100 scale).

That said, I do have an LCD flat panel that wants to be right close to 50... I was confusing that setting with the XBR2 setting. Sorry for any confusion.

Bottom line... you can no longer make any assumptions about where to set sharpness, you really need a test image (black vertical lines on a medium gray background works best) and sit close to the set to see the false outlining appear.

I use Auto 2 for best performance with my IRISMIN, IRISMAX and I_GAIN settings as well as a neutral density filter.
DIC-SEL seems to reflect the user iris setting and does not lower black levels any lower in my case.
You are not the first to say that blacks don't improve with IRISMIN settings below about 320, but that's definitely not the case for me.
Blacks improve with IRISMIN settings all the way down below 50 with an Iris at Auto 2.

Fixed iris settings just don't have to dynamic range required for both good blacks and acceptable brightness.

The interesting thing for me is that with an all black screen the AUTO2 iris setting produced the blackest screen. With the MIN iris setting changed to DIC_SEL 1 now the MIN iris produces the blackest screen.

In regards to fixed iris settings: The MIN iris setting is plenty bright for me in the evening. AUTO2 has me squinting at times .Movie theater settings have the light output level much lower than a properly setup MIN iris setting on the Sony as I have my picture control at 87. In the daytime I put the iris in HIGH because I have a very bright room with 3 windows and even with it on HIGH the blacks look very black.

Congrats on your project though. It is good to see someone push the envelope in pursuit of blacker blacks. I saw one of the Pioneer Kuro Plasmas yesterday and it has the best non CRT blacks I have seen other than the SXRD.

If you reduce light output and black level by 50% with a neutral density filter, an iris setting of “Min” looks very flat and dull, especially after a CRT RPTV, and if you correct the Gamma curve to 2.2, the picture looks even duller.
A fixed iris setting is unacceptable in my setup, even in a dim viewing environment.

What happened to this thread ? No new calibration Ideas ?

the set is a year old now. the early adopters have found the tweeks they like and posted them. most others that used to post here have done the same or had an ISF calibration done. either way the newness has worn off and those that are just now purchasing generally arent tweekers. ocasionally someone pops in with a cool idea. such as the previous few posts. most have moved on to their newest toy or anticipating new purchases.

I bet Your right. Thought some that are geting 1080 P would give some settings ??? Must be close to 1080 I & 720 P ...

i've never seen anyone post settings for input from BD or hddvd players. dont know why. i was using aikenghotti's settings but sence have had UMR calibrate everything.

Is there anyway to set an XBR2 back to the factory default settings?

I purchased the last XBR-R70XBR2 from BB. It was a demo unit on the showroom floor and the remote was left out so customers could tweak whatever these wanted. All said, I know the factory default settings are better than the current configuration. Getting back to the factory defaults would sure make it easier to follow these 36 page long threads on fine tuning and tweaking.

Thanks,
Steve

why not just switch it to custom, warm. and then follow either KTTV or aikenghotti's settings.there's not much else anyone would have done to it. most passers by (BB employees either) wont know how to get into the service menu to make changes.