The official Ruby Calibration Q/A Thread - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 615 Old 01-18-2006, 02:49 PM - Thread Starter
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,082
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 205 Post(s)
Liked: 73
Looks like we needed a good, centralized thread to share helpful tips, tricks, and settings related to calibrating the Ruby. I'll start by posting some of my findings and some questions.

[Mods - please leave this thread in this forum rather than moving it to the calibration forum - as most Ruby owners will not find it there (a previous thread on calibrating Ruby dynamic iris got no action there and quickly sank despite many attempts to keep it going...)]
=====

I've calibrated two different Ruby's at this point (using ColorFacts 6.0 and the EyeOne sensor) - one was a demo unit and the other was mine. The pjs were connected via DVI to a Bravo D1 outputting 1080i using patterns from both VE and DVE.

I thought you may find it interesting to know that out of the box both were quite warm on the color temp from 20-100 IRE, particuarly below 50 IRE. For example both units ranged from 5800 - 6250, generally starting around 5800 and moving up toward 6250 as the IRE increased. The demo unit had about 125 hours; mine had just a couple at the time.

I did all my calibration in the auto iris mode. I found it worked best to calibrate using window patterns. If you calibrated for the windows and then later tried with the full fields, it was pretty close at the full field. However the same was not true in reverse (calibrated to full fields resulted in widely off #s when measuing against windows).

One thing I did not try that I'm interested in would have been calibrating at the window, then going to full fields and tweaking at the full fields, then checking again on the windows. Basically the full fields were close enough, and I didn't want to spend hours on this as I know I'll need to recalibrate it soon as I only had a few hours on the bulb.

Nonetheless, if anyone knows of the best approach to find the right balance between APLs when it comes to calibrating auto iris please post it here, as I am still a bit unclear on the process despite pretty good results from my attempts.

If anyone is interseted in my calibration #s from the Ruby with just a couple hours on it, let me know and I'll post them here. Assuming some similarity between units/lamps, using these numbers may be a good starting point for those without calibration equipment.

On another note, I was quite surprised that I had to set my Color control to 57 and Tint to 48 to get the Split Color Bars pattern to look correct. Anyone else notice this? I find that the grayscale I set via the Bravo works great on my Comcast STB as well. However I wonder if the same color/tint setting is applicable on there as sometimes colors may look over saturated.

I set my Contrast at 69 based solely on how 10 step IRE bars looked. At the default of 80 the 100 IRE bar looked kinda "mushed" into the 90 IRE bar. As I stepped it down toward 69 the seperation between the 100 and 90 IRE bars became much more defined. I stopped at 69 which was the point at which there was no more definition between the 100 IRE and 90 IRE bar compared to higher settings. Was this the result of brightness compression?

Normally you have to back the contrast down on these projectors because you run out of the limiting color at the default contrast setting. With the Ruby however I had plenty of blue (limiting color in its Xenon lamp) to spare at 100 IRE. So it was a shame to back the Contrast down to 69 but seeing the 10 step pattern made it pretty clear that this was a necessity. Perhaps when the bulb ages and light output drops, I may opt to bring the contrast back up to get the increased brightness at the expense of the upper end. So in a way it is good this flexibility is there.

One tough spot for me as usual is that unfortunately the EyeOne will not generate a valid CIE graph. I had this same issue with the Sharp 10K. It shows Red right on but blue and green are way short of their rec 709 coords so I know this is wrong. Therefore I cannot use the RCP to gauge how to dial in proper rec 709 coords.

I have my Ruby in Nomal color space. Can someone please recommend any tweaks/settings to the RCP to dial in the colors a bit more accurately as Greg mentioned doing in his review?

Please chime in with your questions, tips and tricks related to calibrating your Ruby.
lovingdvd is offline  
Sponsored Links
Advertisement
 
post #2 of 615 Old 01-18-2006, 04:05 PM
Super Moderator
 
Bob Sorel's Avatar
 
Join Date: Jan 2000
Posts: 8,465
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked: 117
This is great, Ric! Just what the doctor ordered. I got off the phone with Greg Rogers a short while ago (I ordered his Accupel generator ), so I will be calibrating my new Ruby (arrived this afternoon ) soon. I would love to see what others are using for initial settings, as well as what info they find out about out of box calibration. I figured that since I am getting more serious about my projectors, I should also get more serious about the tools I use to calibrate them.

One thing that Greg cautioned about is the use of 3 color filter type (trichromat) sensors due to the use of a Xenon lamp. I'll let Greg give you the particulars, as I don't want the info to lose anything in the translation. I will be using the Gretag MacBeth EyeOne Pro Beamer, a unit which Greg gave the nod to as being suitable for Ruby calibration.
Bob Sorel is offline  
post #3 of 615 Old 01-18-2006, 04:38 PM - Thread Starter
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,082
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 205 Post(s)
Liked: 73
Quote:
Originally Posted by Bob Sorel View Post

This is great, Ric! Just what the doctor ordered. I got off the phone with Greg Rogers a short while ago (I ordered his Accupel generator ), so I will be calibrating my new Ruby (arrived this afternoon ) soon. I would love to see what others are using for initial settings, as well as what info they find out about out of box calibration. I figured that since I am getting more serious about my projectors, I should also get more serious about the tools I use to calibrate them.

One thing that Greg cautioned about is the use of 3 color filter type (trichromat) sensors due to the use of a Xenon lamp. I'll let Greg give you the particulars, as I don't want the info to lose anything in the translation. I will be using the Gretag MacBeth EyeOne Pro Beamer, a unit which Greg gave the nod to as being suitable for Ruby calibration.

I use the EyeOne Beamer. I'm not sure if that's different than the EyeOne Pro or they're just the same thing. One thing's for sure is that I've always struggled to get too readings from the primaries with this sensor. Its frustrating because I'd love to dial in right to rec 709, but due to the inability to get good readings in this regard its not possible via calibration - I have to do it more like guesswork.

I'll post my calibration settings a little later tonight when I have the numbers in front of me.
lovingdvd is offline  
post #4 of 615 Old 01-18-2006, 05:08 PM
dna
AVS Special Member
 
dna's Avatar
 
Join Date: Feb 2000
Location: Fremont, CA U.S.A.
Posts: 1,031
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I calibrated my Ruby using the AccuPel 3000, CF, and the trichromat sensor, all of which JimmyR was very kind to loan me. I haven't posted my experience with this because this was more of a learning experience for me and I'm sure there are things I should have done differently. In any case, I'll be very interested in reading the posts to this thread. I would be interested in hearing more about the problems with the trichromat with ruby. Are all my results garbage?

I have a couple really basic questions but I don't want to derail this thread so feel free to just ignore me. I decided to calibrate with Iris Off even though I normally view with the Auto mode on. The results I was getting in auto mode were bizarre and I visually see why as the iris changed with changing gray fields. I'll be interested in hearing how people are dealing with that.

I only had a DVD cable for the AccuPel so that's the only input I calibrated. The RGB gain/cut values I ended up with were fairly extreme. With green locked and set to 0 on Ruby, I ended up with the R and B being reduced near the extreme end of the range for gain but being boosted more than 50% for cut. Is this to be expected? The results looked fine, though, both visually and on the graphs (at least to the extent that I understand them). I certainly looks neutral gray to me from 0 to 100 IRE, both at 10 IRE increments and the finer increment (pushing the two Accupel buttons in). DVI inputs from my PC also look very good.

Unfortunately, these were extremely poor settings for my other inputs which is what I use most and are, surprise, way too green at brighter levels and way too lacking in green in dark areas. Is it to be expected that some inputs (or source devices) would vary so much?

- David
dna is offline  
post #5 of 615 Old 01-18-2006, 05:10 PM
Advanced Member
 
maxcat@wave.co.nz's Avatar
 
Join Date: Oct 2000
Location: New Zealand
Posts: 805
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Wouldn't it be wise to calibrate after it's gone through its initial wear-in pattern?

Given that Alan reported visible light loss at 50 hours wouldn't it be prudent to wait until this point (or later?)

What difference did you measure at the 125 hour mark versus your newer lamp?
maxcat@wave.co.nz is offline  
post #6 of 615 Old 01-18-2006, 05:53 PM - Thread Starter
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,082
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 205 Post(s)
Liked: 73
Quote:
Originally Posted by maxcat@wave.co.nz View Post

Wouldn't it be wise to calibrate after it's gone through its initial wear-in pattern?

Given that Alan reported visible light loss at 50 hours wouldn't it be prudent to wait until this point (or later?)

Yes, if I was paying for calibration I'd certainly wait until I had about 150-200 hours on the bulb. However since I do this myself and I can redo it at any time I might as well calibrate it now. For instance instead of 5800-6200 from 20-100 IRE I now am at D65 +/- 150 degrees or so. I do not know how to calculate delta E. Any one know the formula for this?

Quote:
What difference did you measure at the 125 hour mark versus your newer lamp?

Well, let's see. Here are my numbers for my Ruby with 24 hours on the bulb at the time:

Gain:
R: 113 (112)
G: 123 (121)
B: 140 (128)

Offset:
R: 127 (129)
G: 125 (127)
B: 125 (128)

The number listed first is my calibrated setting. The number in parens is the default value in the projector for the middle color temperature. As you can see, they are almost identical except that the blue gain was way short. This explains why I found the default setting to be too warm (5800-6200).

So with the exception of blue the pj was pretty much calibrated on target. I arrived at all these numbers on my own and it wasn't until the end that I noticed how close there were to the default. This gave me some extra confidence that I had calibrated things correctly without messing things up.

Here are the numbers for the calibration on the demo Ruby, which had about 125 hours at the time:

Gain:
R: 101
G: 120
B: 129

Offset:
R: 134
G: 122
B: 128

All these settings were taken with Contrast at 69 and Brightness at 52.

As you can see, the additional 100 hours on the demo Ruby changed the calibration considerably. Then again, it could just be slightly different characteristics of the lamp in that unit so who knows.

As an interesting side note, post calibration, I measured 22.8 cd/m2 on a 100 IRE full field pattern on my Ruby with 24 hours, and 19.2 cd/m2 on the Ruby with 125 hours. Now, these numbers are obviously wrong. And I don't know why they are reading so low. For instance I should be getting upwards of 65-75 cd/m2 (20 cd/m2 on my FH is only 5.7 fL - I think NOT). However all that being said, in an apples to apples comparison, that would be about a 15% drop from 24 to 125 hours. Then again, this may mean nothing, since some lamps may be naturally brighter than others. The real test is in measuring the same bulb x hours later. But at any rate I wanted to pass this along.

If anyone knows why my cd/m2 readings are incorrect please let me know. I am reading this via EyeOne from CF 6.0 directly off the screen with the meter about 18 inches from the dead center of the screen.
lovingdvd is offline  
post #7 of 615 Old 01-18-2006, 06:37 PM
CKL
Advanced Member
 
CKL's Avatar
 
Join Date: Mar 2001
Location: Hong Kong
Posts: 822
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
EyeOne has various versions. Which one has the most accurated results? EyeOne Pro Beamer is the same as EyeOne Pro?

How much is Accupel Generator? Any suggestion of good and cheap generator?

Lovingdvd,

What testing patterns did you use for REC709 Primary measurement? I found Eyeone Beamer can do proper NTSC Primary measurement.
CKL is offline  
post #8 of 615 Old 01-18-2006, 06:39 PM - Thread Starter
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,082
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 205 Post(s)
Liked: 73
Quote:
Originally Posted by CKL View Post

Lovingdvd,

What testing patterns did you use for REC709 Primary measurement? I found Eyeone Beamer can do proper NTSC Primary measurement.

That's encouraging to hear. I've never had any luck with it. As a source in this case I used the 75% RGBCMY full fields from DVE. Could this be the cause? I think I've also tried it with the 100% full field versions.
lovingdvd is offline  
post #9 of 615 Old 01-18-2006, 07:08 PM
AVS Special Member
 
gregr's Avatar
 
Join Date: Nov 2001
Posts: 2,129
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by lovingdvd View Post

I have my Ruby in Nomal color space. Can someone please recommend any tweaks/settings to the RCP to dial in the colors a bit more accurately as Greg mentioned doing in his review?

There isn't much you can do with the RCP function to improve the color. I wrote quite a bit about that in my review. Unlike the much better color management systems in the Yamaha DPX-1300 and Sharp XV-Z12000, the Sony RCP color function has very little effect on the color saturation of the primary or complementary colors, or the hue of the primary colors. (A color's saturation is proportional to its distance on the CIE diagram from the D65 reference white. A color's hue is approximately the angle at which the color lies in a circle around the D65 reference.) It does adjust the hue of the complementary colors and that was moderately useful to correct the hue of the yellow complementary color, which was shifted just slightly toward green only in the Normal Color Space mode. Other than that, the hue accuracy of the complementary colors is excellent and doesn't require any significant correction. The green and blue primaries could use some hue correction, but the RCP doesn't help there.

The more important color errors are in the saturation of the primary colors, which are each over-saturated relative to the Rec. 709 high-definition primaries, and even more over-saturated relative to the SMPTE-C (Rec. 601) standard-definition primaries. Unfortunately, the RCP function does virtually nothing to help that problem, because the RCP Color function affects the intensity (brightness) of the individual primary and complementary colors and not their saturation. If you want more accurate color you can reduce the Color (saturation) control but that doesn't give you individual control over the saturation of each primary.

Another odd effect of the RCP function is that it significantly shifted the grayscale when it was turned on, so that I had to calibrate a separate grayscale for its use.

The Ruby's Wide vs Normal color spaces appear to also use the same functions as the RCP control because they do not actually change the saturation (i.e. color gamut) of the primaries either. Instead, the Normal mode just reduces the intensity (brightness) of the primary colors (red the most, blue some, and green very little). The Wide mode actually has the correct primary brightness, although the Normal mode may look better on some sources when just the reds are oversaturated, but reducing the Color (saturation) control is probably better in most cases.

The Qualia 004 Wide/Normal Color Space worked much better and actually changed the saturation (i.e. color gamut) of the primaries to more closely match the Rec. 709 (HD) primaries. But it also had no way to match the more important Rec. 601 (SMPTE-C) primaries, leaving the user with only the more general Color (saturation) control to provide some correction.

Greg Rogers
Video Engineer/Product Designer

gregr is offline  
post #10 of 615 Old 01-18-2006, 07:08 PM
CKL
Advanced Member
 
CKL's Avatar
 
Join Date: Mar 2001
Location: Hong Kong
Posts: 822
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I think for REC709 primary measurement, you should use Signal Generator or DVHS tape as the 709 source. I doubt if it is accurated to use DVD disc as the 709 source. Although the DVD player will map 601 to 709 color space when you set it to ouput HDTV 1080i/720P, errors may occur during such mapping processing.
CKL is offline  
post #11 of 615 Old 01-18-2006, 07:39 PM
AVS Special Member
 
gregr's Avatar
 
Join Date: Nov 2001
Posts: 2,129
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by lovingdvd View Post

I did all my calibration in the auto iris mode. I found it worked best to calibrate using window patterns. If you calibrated for the windows and then later tried with the full fields, it was pretty close at the full field. However the same was not true in reverse (calibrated to full fields resulted in widely off #s when measuing against windows).

One thing I did not try that I'm interested in would have been calibrating at the window, then going to full fields and tweaking at the full fields, then checking again on the windows.

I would start by calibrating the Iris Off grayscale and the Iris On grayscale using normal window patterns. But then you must calibrate the Auto Iris grayscale using both full fields and window patterns. (You should calibrate using Contrast 80, Brightness 50 for all modes using standard 16-235 video signals.) You can start by first copying the calibrated Iris Off grayscale calibration settings to the Auto Iris mode as a starting point. The iris aperture will be open on 100 IRE and 75 IRE full fields, so you must ensure that the grayscale is corrected for those patterns. I would shoot for near 0 dE error at 75 IRE full field and 3 dE or less at 100 IRE full field. But then you must check the full range of 10-100 IRE window patterns and make sure the low end of the grayscale is calibrated on a 10 IRE window pattern. I would also check below 10 IRE window patterns and a 1-10 IRE 10-step pattern if you have them to make sure the grayscale doesn't become significantly green or red below 10 IRE. Then check a 10-step 10-100 IRE grayscale pattern (which will partially close down the iris aperture) and make sure there is no visible grayscale variation on those steps. You should also make sure that 0% APL, 25% APL, and 50% APL black-level PLUGE patterns have the correct black level, and a 50% APL white-level PLUGE pattern (such as a 98 IRE stripe against a 100 IRE half-frame background) is just visible (which ensures there is no white level clipping).

All of that is an iterative process, so repeat until it is all simultaneously correct.

It is quite a bit of work to get the Auto Iris grayscale correct, but you should end up with good results.

Greg Rogers
Video Engineer/Product Designer

gregr is offline  
post #12 of 615 Old 01-18-2006, 07:55 PM
AVS Special Member
 
gregr's Avatar
 
Join Date: Nov 2001
Posts: 2,129
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by CKL View Post

I think for REC709 primary measurement, you should use Signal Generator or DVHS tape as the 709 source. I doubt if it is accurated to use DVD disc as the 709 source. Although the DVD player will map 601 to 709 color space when you set it to ouput HDTV 1080i/720P, errors may occur during such mapping processing.

Of course I agree you should use a video signal generator.

Seriously, some upconverting DVD players do not transcode the native Rec. 601 YCbCr encoding to Rec. 709 YCbCr encoding when outputting 1080i or 720p signals. That is why it is important for a projector to have manually selectable Rec. 601/Rec. 709 YCbCr matrix decoding.

Greg Rogers
Video Engineer/Product Designer

gregr is offline  
post #13 of 615 Old 01-18-2006, 08:32 PM
CKL
Advanced Member
 
CKL's Avatar
 
Join Date: Mar 2001
Location: Hong Kong
Posts: 822
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Greg,

You mean we need to use 601 color space for some dvdplayers even it outputs 1080i/720P. If the projector has no selectable 601/709, it means it will auto detect. If it detects wrong, what effect it will happen. Abnormal color shift?

I would like to know if Accupel 3000 has the testing pattern to verify 1:1 mapping issue.
CKL is offline  
post #14 of 615 Old 01-18-2006, 08:34 PM
Newbie
 
XBRn00b's Avatar
 
Join Date: Jan 2006
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
what's a ruby?
XBRn00b is offline  
post #15 of 615 Old 01-18-2006, 09:05 PM
AVS Special Member
 
gregr's Avatar
 
Join Date: Nov 2001
Posts: 2,129
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by CKL View Post

Greg,

You mean we need to use 601 color space for some dvdplayers even it outputs 1080i/720P. If the projector has no selectable 601/709, it means it will auto detect. If it detects wrong, what effect it will happen. Abnormal color shift?

I would like to know if Accupel 3000 has the testing pattern to verify 1:1 mapping issue.

One of the DVD players that I reviewed did not change the HDMI YCbCr encoding from the Rec 601 standard to the Rec 709 standard when it upconverted to 720p or 1080i. Therefore, if a projector automatically selects Rec 709 YCbCr decoding (as most do) it would decode the signals with the wrong matrix and the colors would be quite visibly wrong (shifted in hue, saturation, and brightness). I believe some other players with the same problem have been discussed on the forum. If a player has that problem, then it can be fixed by manually selecting the Rec 601 decoding matrix even though the signals are 720p or 1080i. However, only a few projectors have that ability to manually select the YCbCr decoding matrix.

There have also been players that upconvert to 1080i or 720p, internally decode the YCbCr signals with the wrong (Rec. 709) matrix, and then output RGB DVI signals. There are no projectors that I am aware of that can compensate for that error (which requires a special RGB to RGB matrix transform in the projector).

BTW, keep in mind that this is a completely different issue than the color space and calibration of the projector's primary and complementary colors, which I discussed earlier. This is about the signal encoding/decoding matrix to convert YCbCr signals into RGB signals for the projector.

On your other question, yes, I designed several of the AccuPel HDG-3000 patterns to determine if there is a 1:1 "pixel perfect" mapping of source pixels to display pixels, and to determine if there is any edge enhancement added in the projector.

Greg Rogers
Video Engineer/Product Designer

gregr is offline  
post #16 of 615 Old 01-18-2006, 10:10 PM - Thread Starter
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,082
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 205 Post(s)
Liked: 73
Quote:
Originally Posted by gregr View Post

...It does adjust the hue of the complementary colors and that was moderately useful to correct the hue of the yellow complementary color, which was shifted just slightly toward green only in the Normal Color Space mode. Other than that, the hue accuracy of the complementary colors is excellent and doesn't require any significant correction. The green and blue primaries could use some hue correction, but the RCP doesn't help there...

Seems odd that Sony would create such and advanced pj yet leave out such important control over the primaries.

Do you recall what RCP changes you made to improve on the yellow?

Regarding overall Color control to reduce saturation, do you recall approximately what you lowered the Color control to from the default of 50? Would this same setting apply generally or would it also be a function of the source (say a STB)?

Thanks for all the great feedback Greg.
lovingdvd is offline  
post #17 of 615 Old 01-18-2006, 10:18 PM - Thread Starter
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,082
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 205 Post(s)
Liked: 73
Quote:
Originally Posted by gregr View Post

I would start by calibrating the Iris Off grayscale and the Iris On grayscale using normal window patterns.

Just curious - does it matter if you skip calibrating Iris On mode if you never plan to use that mode?

Quote:


But then you must calibrate the Auto Iris grayscale using both full fields and window patterns.

Which pattern should get precedence, the full field or window? For instance if 80 IRE can only be dialed in correctly on either the full field OR window pattern (but not both) which should be used?

Quote:


(You should calibrate using Contrast 80, Brightness 50 for all modes using standard 16-235 video signals.)

I think in your review you mentioned Contrast in the 60s. Others are reporting they are using settings in the 60s as well. Based on your recommendation of calibrating on 80, and my current setting of 69, I am wondering if I am selling myself short on brightness...?

Or perhaps you are saying to calibrate with it at 80, but then post calibration lower it? When I run at 80 contrast the 100 and 90 IRE bars tend to mush together which is why I backed it off from 80 to 69. Am I making a mistake here?

Quote:


You can start by first copying the calibrated Iris Off grayscale calibration settings to the Auto Iris mode as a starting point. The iris aperture will be open on 100 IRE and 75 IRE full fields, so you must ensure that the grayscale is corrected for those patterns.

So top priority is making sure 75 and 100 IRE full fields are dialed in very good, even if it means that this shifts when you put up a 75 or 100 IRE window (which it definately does shift away from D65 in doing this if its calibrated to the full field...)?

Quote:


I would shoot for near 0 dE error at 75 IRE full field and 3 dE or less at 100 IRE full field. But then you must check the full range of 10-100 IRE window patterns and make sure the low end of the grayscale is calibrated on a 10 IRE window pattern.

That is where I get confused. It can't both ways, can it? Its either calibrated for window or full fields, so which gets priority?

Quote:


I would also check below 10 IRE window patterns and a 1-10 IRE 10-step pattern if you have them to make sure the grayscale doesn't become significantly green or red below 10 IRE. Then check a 10-step 10-100 IRE grayscale pattern (which will partially close down the iris aperture) and make sure there is no visible grayscale variation on those steps. You should also make sure that 0% APL, 25% APL, and 50% APL black-level PLUGE patterns have the correct black level, and a 50% APL white-level PLUGE pattern (such as a 98 IRE stripe against a 100 IRE half-frame background) is just visible (which ensures there is no white level clipping).

All of that is an iterative process, so repeat until it is all simultaneously correct.

It is quite a bit of work to get the Auto Iris grayscale correct, but you should end up with good results.

Thanks so much Greg. This is very useful info indeed.
lovingdvd is offline  
post #18 of 615 Old 01-18-2006, 11:36 PM
AVS Special Member
 
gregr's Avatar
 
Join Date: Nov 2001
Posts: 2,129
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by lovingdvd View Post

Just curious - does it matter if you skip calibrating Iris On mode if you never plan to use that mode?

No, the settings in the service menu are saved separately depending on what iris mode you are in.

Quote:


Which pattern should get precedence, the full field or window? For instance if 80 IRE can only be dialed in correctly on either the full field OR window pattern (but not both) which should be used?

Well, ultimately it's up to what you think looks best. But I would get a 70 or 75 IRE full field as close to D65 as possible while still achieving a dE of 3 or less on a 100 IRE field, 70 or 75 IRE window, and everywhere else down to about a 10 IRE window. It's an iterative process.

Quote:


I think in your review you mentioned Contrast in the 60s. Others are reporting they are using settings in the 60s as well. Based on your recommendation of calibrating on 80, and my current setting of 69, I am wondering if I am selling myself short on brightness...?

Or perhaps you are saying to calibrate with it at 80, but then post calibration lower it? When I run at 80 contrast the 100 and 90 IRE bars tend to mush together which is why I backed it off from 80 to 69. Am I making a mistake here?

I don't recall mentioning 60 in the review, but I may have mentioned it in answer to a question here on the forum. What I said in the review was that you can "modestly reduce brightness compression by lowering the Contrast control setting, which of course also reduces picture brightness and image contrast." I would do your best grayscale calibration at 80 and then experiment with the Contrast to see if you want to tradeoff a lot of brightness and contrast for a little improvement in brightness compression. If so, you can go back and re-optimize the grayscale after you are sure you have a contrast setting that you really want to use. Because after that you can't increase the contrast again without creating actual clipping.


Quote:


So top priority is making sure 75 and 100 IRE full fields are dialed in very good, even if it means that this shifts when you put up a 75 or 100 IRE window (which it definately does shift away from D65 in doing this if its calibrated to the full field...)?

It's going to shift some because you are then operating at different point in the gamma range (new gamma curve) at a different iris aperture. But if you do a careful calibration you can keep the dE error at about 3 or less for most conditions. For most observers, a dE of 3 or less is hardly noticeable on real world video.

Greg Rogers
Video Engineer/Product Designer

gregr is offline  
post #19 of 615 Old 01-19-2006, 12:48 AM
CKL
Advanced Member
 
CKL's Avatar
 
Join Date: Mar 2001
Location: Hong Kong
Posts: 822
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Greg,

Thanks for your reply. Any plan to include 1080P and 576i@50Hz in the upcoming firmware for AccuPel?
CKL is offline  
post #20 of 615 Old 01-19-2006, 03:34 AM
Member
 
thomasclaus's Avatar
 
Join Date: Sep 2002
Location: Germany
Posts: 91
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
@lovingdvd,

Quote:


I have my Ruby in Nomal color space. Can someone please recommend any tweaks/settings to the RCP to dial in the colors a bit more accurately as Greg mentioned doing in his review?

I posted my experiences regarding the correction of primaries and secondaries via RCP and Color-controls in another thread here: http://www.avsforum.com/avs-vb/showt...&page=16&pp=30 in posting #462 - #464.

Thomas

The saw is family
thomasclaus is offline  
post #21 of 615 Old 01-19-2006, 06:55 AM
Super Moderator
 
Bob Sorel's Avatar
 
Join Date: Jan 2000
Posts: 8,465
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked: 117
Greg, would you please repeat the information you told me yesterday regarding the use of filter based trichromat sensors? A lot of people here use those types of instruments.
Bob Sorel is offline  
post #22 of 615 Old 01-19-2006, 07:23 AM
AVS Club Gold
 
darryl b's Avatar
 
Join Date: Feb 2003
Location: wi
Posts: 709
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
i know this is a thread about calibration, but.
what about not formally calibrating?
can one have a great experience by working with user settings and things like avia and thx optimizer?
i have had two pj calibrated to isf standards. very early on, when i did not know there may be someone locally i had a well known person, calibrator, come cross country to do an isf calibration. really looked very nice, but it also looked nice prior to that. i was never sure if my eye was not the better. similar outcome with the second pj.
so are there any tips for user calibration?
(edit) how can we quickly get up and running with a good picture while we wait for 50 - 100 hrs?
darryl b is offline  
post #23 of 615 Old 01-19-2006, 07:29 AM - Thread Starter
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,082
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 205 Post(s)
Liked: 73
Quote:


Well, ultimately it's up to what you think looks best. But I would get a 70 or 75 IRE full field as close to D65 as possible while still achieving a dE of 3 or less on a 100 IRE field, 70 or 75 IRE window, and everywhere else down to about a 10 IRE window. It's an iterative process.

What is the formula for calculating dE? What inputs do we need? I assume they are numbers that I'd get from the "Raw Data" window of ColorFacts which includes little/big x and y? For example, if with CF I measure 6300 with 101% R, 98% G, and 101% B, what is my dE? Knowing this formula will enable me to target your recommendations. I did a google search on calculating dE but couldn't come up with anything.

Quote:


I don't recall mentioning 60 in the review, but I may have mentioned it in answer to a question here on the forum. What I said in the review was that you can "modestly reduce brightness compression by lowering the Contrast control setting, which of course also reduces picture brightness and image contrast." I would do your best grayscale calibration at 80 and then experiment with the Contrast to see if you want to tradeoff a lot of brightness and contrast for a little improvement in brightness compression. If so, you can go back and re-optimize the grayscale after you are sure you have a contrast setting that you really want to use. Because after that you can't increase the contrast again without creating actual clipping.

I'm happy to get additional brightness and contrast in exchange for a bit more brightness compression (which I never notice anyway). So after reading this I want to go back and recalibrate at 80 Contrast. Or at least I'll do so in another user setting so I can go back and forth between a grayscale based on 69 contrast and one based on 80.

I'd like your opinion on the basis I used to chose 69 in the first place. The reason I did this was because, post calibration with Contrast of 80, when I put up a IRE steps pattern. It had 4 quandants full of a bunch of small 10 vertical step bars that ran from 0-100 (I think). Anyway the brightest bar was mushing/bleeding into the second brightest bar on this pattern, post calibration. As I lowered Contrast from 80 toward 69 I could see the definition in each of these bars come back (the mush/bleed was gone completely at about 69 and lowering any further made no difference).

Was this effect caused by BC? Was this adequate justification for using 69, or just adequate justification if eliminating/minimiing BC is a priority over brightness/contrast (which it is not for me)?

Also in case folks are wondering, on the pj with 125 hours, post calibration, I measured a full field 100 IRE pattern at 20 cd/m2 with Contrast at 69, versus 24 cd/m2 with Contrast at 80. That's 20% brighter with Contrast of 80, which obviously is quite significant.
lovingdvd is offline  
post #24 of 615 Old 01-19-2006, 08:07 AM - Thread Starter
AVS Special Member
 
lovingdvd's Avatar
 
Join Date: Jul 2002
Posts: 9,082
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 205 Post(s)
Liked: 73
Quote:
Originally Posted by thomasclaus View Post

@lovingdvd,



I posted my experiences regarding the correction of primaries and secondaries via RCP and Color-controls in another thread here: http://www.avsforum.com/avs-vb/showt...&page=16&pp=30 in posting #462 - #464.

Thomas

Thanks Thomas. Can you post your Color and Tint control settings, and the numbers you customized in the RCP? I have my Color control at 57 based on results from the split color bars pattern and I'm realizing that this is oversatured (based on Greg's and your reports).

Also Thomas - what are you using for the source of the color fields that you are measuring primaries? Is this over DVI/HDMI?

It is encouraging to hear you are getting similar CIE graphs on the factory defaults as GregR using the EyeOne. All these years I thought it was my EyeOne. However I am starting to become convinced that the problem is in my SOURCE and not the EyeOne.

For instance I am using the Bravo D1 outputting 1080i over DVI. Posts earlier suggest it could be an issue with the color decoding/matrix in the Bravo.

As a test I am going to try measuring the CIE using a) the Bravo set to 480p over DVI, and b) another DVD player using component output. I think this could be very enlighting. If either of those tests result in readings greater than the default gamma for the ntsc color space, then I think I can assume this issue is in the Bravo. What do you guys think about this testing approach?
lovingdvd is offline  
post #25 of 615 Old 01-19-2006, 02:00 PM
Super Moderator
 
Bob Sorel's Avatar
 
Join Date: Jan 2000
Posts: 8,465
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked: 117
Quote:


EyeOne Pro Beamer is the same as EyeOne Pro?

The EyeOne Pro is the sensor itself - "Beamer" is a software/accessory package that is bundled with the EyeOne Pro sensor. I also bought the "Beamer" package, as it includes a tripod mount and more importantly, a diffuser for taking readings directly from the light source. You can not buy the diffuser separately.
Bob Sorel is offline  
post #26 of 615 Old 01-19-2006, 02:16 PM
AVS Special Member
 
gregr's Avatar
 
Join Date: Nov 2001
Posts: 2,129
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by Bob Sorel View Post

Greg, would you please repeat the information you told me yesterday regarding the use of filter based trichromat sensors? A lot of people here use those types of instruments.

Bob,

As you recall I didn't mention any particular color analysis products by name. My point was simply that a Xenon lamp produces quite a bit different spectral distribution for the red, green, and blue primaries than a UHP lamp. Therefore, a tri-stimulus (fixed filter) based color analyzer may have filter response deviations that correspond to spectral content not usually present with the UHP lamp, which may result in larger errors with a Xenon lamp than UHP lamps. The spectroradiometer based instruments tend to have more uniformly distributed spectral response errors making them somewhat less sensitive to the spectral variations of sources. Any specific tri-stimulus color analyzer would need to be tested to see if this is a significant factor with the Ruby.

Greg Rogers
Video Engineer/Product Designer

gregr is offline  
post #27 of 615 Old 01-19-2006, 02:17 PM
Super Moderator
 
Bob Sorel's Avatar
 
Join Date: Jan 2000
Posts: 8,465
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked: 117
I'm the real newbie of the bunch here as I just got my Ruby yesterday and haven't had more than about 15 minutes to play with it so far. At this point, I am not sure how to handle even basic calibrations like brightness and contrast in the auto iris mode. Normally I use a set of reversing ramps that display BTB and WTW (below digital 16 and above 235) to get a quick evaluation of the overall grayscale and also to set contrast and brightness. Should I use this pattern with auto iris mode or will the presence of the extreme ends of the grayscale make the adjustments invalid? If so, what patterns should I use instead? Just looking at the reversing ramps with the default values, both the blacks and the whites looked crushed, and if I were to adjust contrast and brightness based on that one pattern, I would need to raise brightness and lower contrast - both drastically, in order to see a smooth set of ramps from 0% to 100%. It just seems unnatural to have to reduce the dynamic range so drastically, so I assume that I need to make my adjustments using more appropriate patterns.
Bob Sorel is offline  
post #28 of 615 Old 01-19-2006, 02:20 PM
Super Moderator
 
Bob Sorel's Avatar
 
Join Date: Jan 2000
Posts: 8,465
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked: 117
Thanks, Greg, that is how I understood it, but it probably wouldn't have come out right if I had said it...
Bob Sorel is offline  
post #29 of 615 Old 01-19-2006, 09:47 PM
Super Moderator
 
Bob Sorel's Avatar
 
Join Date: Jan 2000
Posts: 8,465
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked: 117
Wow, this thing is really going to take some work. After about 2 hours of playing with the Ruby tonight, I am very unimpressed. The picture looks flat and lifeless, no 3D quality at all - colors are not accurate (reds are orangy, fleshtones pinkish - even in normal mode) - both DVD and HD look noisy. Other than the better black level, so far I prefer the picture my InFocus 7210 and Optoma H-79 put out, as those two have more solid looking images with better instaneous contrast and much more accurate and vibrant colors.

I hope there is some magic bullet to make this Ruby shine, because so far it is a huge disappointment.
Bob Sorel is offline  
post #30 of 615 Old 01-19-2006, 09:59 PM
AVS Special Member
 
AndyN's Avatar
 
Join Date: Apr 2001
Location: Bay Area, California
Posts: 1,023
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 19
Bob,

Interesting comment about the reds. I don't think I've ever seen redder reds. I came from orange reds from a Barco CRT and I'm a little surprised you think the reds are organgy.
AndyN is online now  
Reply Digital Hi-End Projectors - $3,000+ USD MSRP

Tags
Sony Vpl Vw100 Lcd Projector
Gear in this thread - Vw100 by PriceGrabber.com

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off