or Connect
AVS › AVS Forum › Display Devices › Display Calibration › The Official ChromaPure thread
New Posts  All Forums:Forum Nav:

The Official ChromaPure thread - Page 3

post #61 of 5355
I've made three attempts to calibrate my 1st generation Sony SXRD RPTV. The 1st time thru I used an earlier version of Color HCFR along with my i1 Pro meter. The results were less than stunning, although some improvement was noted (the main issue at that time was not being able to correct gamma response which varied from 1.8 to 2.6). Shortly thereafter my set started degrading with the infamous "Green Blob" issue and for various reasons it took quite awhile before I finally got around to asking Sony to fix the problem. After it was fixed it became watchable again but it clearly needed to be calibrated (by then I was watching most everything on another display though). When I finally got around to calibration attempt number two I used the latest version of Color HCFR along with my i1 Pro meter. This time around though I was armed with the knowledge of how to correct the gamma response (many thanks to Chad B for posting this info). After hours of tweaking the picture was showing an immense improvement but it was still not where I wanted it, especially at IRE levels below 30%. I reasoned that the low light weakness of i1 Pro was the main obstacle to achieving a better calibration. Cutting to the chase I ended up buying a Chromapure/Chroma 5 package from Tom. While waiting for the new stuff to arrive I dialed in the calibration as best I could with HCFR and the i1 Pro. By doing this I figured I could find out how well the Chroma 5 and i1 Pro correlated and I also wanted to see how calibrating with Chrompure would compare to calibrating with Color HCFR, which by now I was getting fairly adept at using.

Along came my ChromaPure/Chroma 5 package via insured Priority Mail. It was nicely packed and arrived safely and quickly. Asides from watching demo's on the ChromaPure site I dove into the calibration not knowing what to really expect (other than having to learn new software which can be anywhere from easy to complex). Going into this I had a preconceived notion that the Chroma 5 probe was going to be the just the ticket for getting the calibration fine tuned (especially below 30% IRE) and that using ChromaPure instead of HCFR would be perhaps a bit more straightforward but would yield similar end results as HCFR.

It turns out that my i1 Pro and Chroma 5 readings were quite similar above about 30% IRE but the Chroma 5 is clearly more adept at lower IRE's (this is just as Tom had stated in various posts here and in an email exchange between us). As to ChromaPure vs HCFR I am no expert but I do think one could get similar results using both software packages. That said it's what ya gotta do to get to those end results. IMO Chromapure gets you there quicker and in a far more straightforward (and intuitive) fashion. Don't get me wrong, HCFR is decent freeware but Chromapure is professional grade calibration SW that is easier for the novice (me) to use than HCFR. Its lay out is terrific. Being able to view gamma response in real time is brilliant. I jumped around a bit to tweak gamma before grayscale (grayscale adjustments on this Sony don't affect the gamma response too much but gamma response settings do affect grayscale because there are individual R,G,& B gamma [cut and gain] adjustments). The end result is terrific - grayscale Delta-E errors are quite acceptable and the gamma response is fairly flat above 30% IRE (it's starts drooping below that but it never gets horrible). Unfortunately I cannot generate a report because that requires Excel and being a Mac guy, I do not own a PC version of Excel.

One minor annoyance I noted is this - I used the AVS Calibration DVD and Chromapure does not follow either the HCFR or Calman order for colors (ChromaPure wants them is this order (IIRC): W, R, G, B, Y, C, M. It would be terrific if this DVD is updated to support Chromapure in addition to Calman and HCFR. I may take one more pass at calibrating using Tom's DVD patterns which should be in the proper order (but the AVS DVD is fantastic and it would be a shame if it isn't updated for Chromapure).

The bottom line is I'm very happy with the purchase. The ChromaPure/Chroma 5 package cost $600, which is not cheap but I can calibrate my displays whenever I choose. Alternatively I could have had my displays professionally calibrated just twice for that price so in the long run I'll be money ahead. The Chroma 5 meter works better than my i1 Pro at low light levels and I'm quite confident I could not have gotten the calibration as good as I did below 30% IRE using the i1 Pro. Going into this I was less sure of the Chromapure benefits over HCFR (other than not being able to use Chroma 5 colorimeter with HCFR) but in this instance one gets what one pays for. HCFR is decent but Chromapure is several notches better: It's much more refined and far easier to use (IMO). For me it was well worth the $200 price of admission. Perhaps I should not have been surprised - Chromapure is software written BY an ISF Calibration expert who has a keen understanding of the calibration process. But I expected the software to be competent and fitting for the job at hand. What I wasn't sure of is how well crafted and user friendly it was going to be to use but it turns out that it's quite excellent in that regard and is quite intuitive to use. I never pushed the help button once during calibration: Only when I was done did I look at the help section and see how really excellent those files are. They are clear and concise and contain all the info needed to help you understand what you need to do. All that is left to really figure out is where to find the proper adjustments in the service menus.

All I can say is WELL DONE TOM!
post #62 of 5355
It sounds like you've come to a similar conclusion to me: HFCR is free and gets you there in the end, but using Chromapure is so much easier and quicker If only I had a more accurate sensor than my now 1 year old i1.

I did another run and focused on getting the red lower as suggested by Tom and using the HDQ it was quite easy to keep the gamma spot on and reduce the red point by point (well nearly every point as a couple were 'bouncing'up and down slightly while I was reading them). I've attached the report which shows very low delta E and mostly the red below the green and blue on the greyscale. Gamma nicely around 2.2 even slightly above in places. I had another go at the colour gamut, but it is a series of compromises so I went for the lowest delta E average rather than any particular colour...it just happened that magenta ended up lowest, but yellow and green were the two hardest to get right (and mutually exclusive with my limited controls).

However, this near 'perfect' result on paper has a major issue with greyscale ramp or bars patterns I checked with: The grey bars look like stripes of pink or green candy and the ramp shows subtle changes mostly in the middle IREs. As I've used fairly aggressive adjustments in the HDQ I wonder if this is the cause? I plan to go back and just use the TV's RGB controls to get greyscale as close as is reasonable and then just use the HDQ to get the gamma correct. I'm not sure where to ask the question as it's as much a HDQ issue, but I wonder if there is any further testing in Chromapure that I could do to help adjust this out. All I know is turning off the HDQ's greyscale control cures the 'candy stripe' issue, while making the greyscale measure worse....but it actually looks better.

As I'm mostly using Chromapure for projector I will only be using it's own greyscale and gamma controls I think given this dissapointing result. This is no slight on Chromapure as I'm sure I had this issue before using HFCR as well, but I'm wondering if there is anything else I can check?

 

Second Attempt.pdf 240.16796875k . file
post #63 of 5355
Quote:
Originally Posted by Kelvin1965S View Post

However, this near 'perfect' result on paper has a major issue with greyscale ramp or bars patterns I checked with: The grey bars look like stripes of pink or green candy and the ramp shows subtle changes mostly in the middle IREs. As I've used fairly aggressive adjustments in the HDQ I wonder if this is the cause? I plan to go back and just use the TV's RGB controls to get greyscale as close as is reasonable and then just use the HDQ to get the gamma correct.

I don't think calibration software will help you here. This sure sounds like a limitation in the HDQ's processing. Make sure that you are not using the HDQ's Color Decoding controls (Color Red, Hue Red, etc.). They are notorious for causing these types of banding problems. Otherwise, relying on the TV for grayscale and the HDQ for gamma sounds like a good approach.
post #64 of 5355
Quote:
Originally Posted by Geof View Post

Unfortunately I cannot generate a report because that requires Excel and being a Mac guy, I do not own a PC version of Excel.

This limitation will be addressed in version 1.2, which will be out very soon.
post #65 of 5355
As mentioned in my post above, I could not generate a report because I do not have a PC version of Excel (during calibration I was running Chromapure from my Bootcamp partition (meaning I booted up my Mac in Windows XP). I have since rebooted in Mac OS 10.6, and used Parallels Desktop to run Windows XP. I then tried generating a report in Chromapure and (to my surprise) the data is exported to Excel 2008 (the latest Mac version of the program) but it does not look right. I'll keep playing with this though....

Tom,
It would be nice to have more lead length on the Chroma 5 - have you tried using it with a USB extension cable?
post #66 of 5355
Quote:
Originally Posted by TomHuffman View Post

This limitation will be addressed in version 1.2, which will be out very soon.

Fantastic!
I guess there will be no need to keep experimenting with exporting it to Excel on the Mac side.
post #67 of 5355
Quote:
Originally Posted by TomHuffman View Post

I don't think calibration software will help you here. This sure sounds like a limitation in the HDQ's processing. Make sure that you are not using the HDQ's Color Decoding controls (Color Red, Hue Red, etc.). They are notorious for causing these types of banding problems. Otherwise, relying on the TV for grayscale and the HDQ for gamma sounds like a good approach.

Thanks Tom, I've just had another try and managed to get the greyscale flatter without the HDQ and only used it for gamma correction. The delta Es are a little worse, but the stripes issue is gone now.

The only issue I seem to have at the moment (and it really is minor as I know what the result is from my HFCR days) is that the 'Contrast' reading doesn't seem to work anymore. It did the first time I ran Chromapure and not since.
post #68 of 5355
Quote:
Originally Posted by Geof View Post

It would be nice to have more lead length on the Chroma 5 - have you tried using it with a USB extension cable?

Yes. It works fine.
post #69 of 5355
Tom,
How do we go about downloading the new (version 1.2) update?
post #70 of 5355
Quote:
Originally Posted by Geof View Post

Tom,
How do we go about downloading the new (version 1.2) update?

In my case Tom sent me an email with a download link to 1.2
post #71 of 5355
Quote:
Originally Posted by catmother View Post

In my case Tom sent me an email with a download link to 1.2

Thanks. I did not get an email so I'll send one to him.
post #72 of 5355
I found that putting a box over the meter on the screen helped. The box has been spray painted as close as D65 gray as possible: made the below 30IRE easier. Problems arise due to the LCDs menu calibration adjustment menu radiating into the meter. I also set the meter more towards the top of the calibration window as well. These two items help to separate the two entities (Adjustment menu and calibration Window) Also put the box edge just below the meter. Several people have mentioned using Gray Blankets but I don't see how that will isolate the meter from the adjustment screens. Has anyone else tried this or have or thoughts on isolating the meter from the adjustment menus?
post #73 of 5355
Version 1.2 is indeed ready. The main improvements are

- automation via video card or Accupel signal generator
- integrated reporting (Excel no longer required)

The complete release notes are available on the News section of our web site.

It is our intention to send e-mails to all customers with a download link, but we are currently away from the office and won't be back until next week.

If you don't want to wait, send me an e-mail and I can send you the download link.
post #74 of 5355
Chad B uses x=.3130 and Gamma at 2.2. Mr Huffman is using x=.3127 and Gamma at 2.2 to 2.35 - is there a reason for these differences that I need to be concerned about?
post #75 of 5355
x.0.3127, y0.329 is the correct specification for D65. Chad may just be rounding.

I am firm believer that there should be some latitude for gamma to accommodate different viewing tastes and environments. No display is a perfect 2.22 across the grayscale in any case.
post #76 of 5355
I'm glad you said that about gamma Tom, I did a calibration on my HD350 last night (using the 1.2 version of Chromapure) with the main aim of improving the gamma. I managed to get it around 2.2, though it does go a bit high around 50-60 IRE (I need to adjust the blue gamma channel a little more) and tails off a little at 90 IRE. Even so the picture does look better than before and Chromapure certainly helped me get there a bit quicker, though I'm sure I can improve it further with a bit more time as the HD350 controls are a bit fiddly as they keep 'smoothing out' the other gamma points after each change.
LL
LL
post #77 of 5355
Kevin,
how does your gray scale tracking look? Your gray scale should not be compromised to improve gamma.
post #78 of 5355
Quote:
Originally Posted by Gregg Loewen View Post

Kevin,
how does your gray scale tracking look? Your gray scale should not be compromised to improve gamma.

A bit of a bump in the middle 50-60IRE range with the blue. I'm thinking of resetting the JVC gamma and using external adjustment as it seems so much quicker and easier, though for now the image does look better than it was.

BTW It's Kelvin like the temperature measurement.
LL
LL
post #79 of 5355
I see the option is there for an i1 Pro/display 2 combo. How does one go about using both sensors with this software? I own the Pro and am very interested in the Display 2/Chromapure package to have a more accurate reading below 30 IRE.
post #80 of 5355
Quote:
Originally Posted by slosvt View Post

I see the option is there for an i1 Pro/display 2 combo. How does one go about using both sensors with this software? I own the Pro and am very interested in the Display 2/Chromapure package to have a more accurate reading below 30 IRE.

http://www.chromapure.com/demos-offset.asp
post #81 of 5355
Quote:
Originally Posted by TomHuffman View Post

http://www.chromapure.com/demos-offset.asp

I saw the demo, so I guess I should have phrased my question a little different. With the i1 Pro/Display 2 combo, it would be best to use the i1 Pro as the reference meter and take all field readings with the Display 2, correct?
post #82 of 5355
Quote:
Originally Posted by slosvt View Post

I saw the demo, so I guess I should have phrased my question a little different. With the i1 Pro/Display 2 combo, it would be best to use the i1 Pro as the reference meter and take all field readings with the Display 2, correct?

Correct.
post #83 of 5355
Quote:
Originally Posted by p5browne View Post

I found that putting a box over the meter on the screen helped. The box has been spray painted as close as D65 gray as possible: made the below 30IRE easier. Problems arise due to the LCDs menu calibration adjustment menu radiating into the meter. I also set the meter more towards the top of the calibration window as well. These two items help to separate the two entities (Adjustment menu and calibration Window) Also put the box edge just below the meter. Several people have mentioned using Gray Blankets but I don't see how that will isolate the meter from the adjustment screens. Has anyone else tried this or have or thoughts on isolating the meter from the adjustment menus?

The settings towards the top edge of the calibration screen are far different than those dead center. Which leads to the question - are your calibration settings only good for the exact spot under your meter?

ie: Off center Flatlined Green: 10IRE: 6 20IRE: 9 30IRE: 11 40IRE: 17 50IRE: 22 60IRE: 28 70IRE: 32 80IRE: 38 90IRE: 40 100IRE: 8 (Without Box)
Centered Flatlined Green : 10IRE: 2 20IRE: 2 30IRE: 2 40IRE: 2 50IRE: 5 60IRE: 7 70IRE: 6 80IRE: 12 90IRE 14 100IRE: 4 (Without Box)
post #84 of 5355
This has crossed my mind before now. I even think this may be the route cause of different readings on different days. I've wondered about taking say 5 measurements: Centre, then four towards (but not in) the corners. Average out the results to deceide on the optimum setting. This method would require the use of full field test patterns, which would work fine with an LCD TV, but might result in peak white compression at higher IREs with a Plasma TV. With an LCD TV and using an i1-LT I find that if I take the sensor off the screen and replace it before repeating the run, it will be slightly different as well, so it's more a question of consistancy. Again taking a few 'averaging' readings with the sensor only moving a few mm from the centre position can help mitigate for this.

I'm more confident about calibrating the projector using the diffuser with the sensor facing the projector as the image is more condesed onto the sensor, so kind of causing a natural 'averaging' effect. The only issue there is that I use the sensor 'training' mode to create an offset to allow reading directly from the projector. How do I know that the initial 'off screen' reading is a good base? My idea was to put the projector out of focus when taking the off screen readings, but I'm not sure if this really achieves anything this way. Of course you can't put a TV out of focus anyway, hence my 5 point averaging suggestion.

At least it is quick and easy to take a greyscale using Chromapure, so it minimises the hassle. I also found that fully warming up the display and sensor first (including pluging it into my PC and leaving Chromapure taking real time white balance readings) seems to help with consistancy.

At the end of the day I'm at the mercy of the accuracy of my i1LT anyway.
post #85 of 5355
Anyone having issues with Windows 7 Ultimate?
post #86 of 5355
Quote:
Originally Posted by p5browne View Post

The settings towards the top edge of the calibration screen are far different than those dead center. Which leads to the question - are your calibration settings only good for the exact spot under your meter?

Sadly, yes. This is why it is important to purchase a display with good white field uniformity.
post #87 of 5355
OK - what can you do with these results? What do they mean?

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

Quote:
Originally Posted by catmother
With local dimming of all zones are at equal light output so the meter could be placed anywhere.
Good question, suggest you post this in Tom Huffman's thread or the HCFR thread.

BTW I have been following your posts on those threads

OK here goes: Local Dimming is turned off. Divided screen into 8 quadrants. Took measurements in the center of each quadrant - then dead center.
Results: 1 to 4 across the top. 5 to 8 across the bottom and DC - center of screen - Screen set to Full Screen 100%
Top:
Quadrant 1: x=.3041 y=.3119 Y=56.9088
Quadrant 2: x=.2978 y=.3182 Y=56.6575
Quadrant 3: x=.3050 y=.3100 Y=59.9141
Quadrant 4: x=.3051 y=.3142 Y=56.9200
Bottom:
Quadrant 5: x=.2502 y=.3406 Y=58.3131
Quadrant 6: x=.3079 y=.3150 Y=59.6779
Quadrant 7: x=.3112 y=.3157 Y=64.2058
Quadrant 8: x=.3109 y=.3159 Y=60.6877

Center: x=.3054 y=.3173 Y=62.2260 So much for overall uniformity!

Results of Test #2 - last night LG was perfectly flatlined - turned TV off:
Tonight - turned TV on - warmed up set with 100%
White - Tested the settings for 100% White versus
last night - not even close!!! What gives?
Meter was not moved, calibration program was left running, no settings changed. What changed? Meter can't be left on? LG changes settings everytime you turn it on and off? Is this why, after calibrating I have a drop dead picture and can't understand why it doesn't look so hot the next day, then the day after it looks OK again? Anyone have an answer? Does this mean the day after your set is calibrated it's not the same calibrated set anymore, and the day after, and the day after? Has anyone done followup tests on their LGs or for that matter, any other brands? Chad B, at last questioning, mentioned he'd had never rechecked a calibrated set the day after to see if values had changed. Personnally, this is getting frustrating! Like, what's the sense - the same problem was also on my July build!
post #88 of 5355
Quote:
Originally Posted by robbyrockets View Post

Anyone having issues with Windows 7 Ultimate?

Tom now has a 64 bit version of ChromaPure which installs and runs in W 7 Ultimate.
But it depends on the meter.

I have installed 64 bit drivers in 7 for the DPT94 and windows accepts those but ChromaPure does not yet work with my meter.
Tom is working on a fix for this.

I would presume the Chroma V would be OK. Ask Tom.
post #89 of 5355
Okay, question from a noob. If I want to use Chromapure to set light output to 30 ftL using a free masure, how do I know what this value is when I just have x, y, and Y displayed?
post #90 of 5355
Quote:
Originally Posted by slosvt View Post

Okay, question from a noob. If I want to use Chromapure to set light output to 30 ftL using a free masure, how do I know what this value is when I just have x, y, and Y displayed?

In the Options module set luminance to fL. Now the Y value will display the info you want.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › The Official ChromaPure thread