Home Theater Forum and Systems banner

i1Display 2 questions

Tags
i1display
6K views 47 replies 9 participants last post by  jacare 
#1 ·
Like several of you perhaps, I bought the D2/CalMAN DIY license bundle from SpectraCal a few weeks ago. I participated in SpectraCal's Video Calibration webinar last week, I've read through the DIY workflow overview posted in this forum several times, and I've gone through the Standard workflow three times now. However, each time I start the CMS calibration I get different initial readings, with greatly out of whack Delta L and Delta H values. This morning (the third time) was the worst. (Example: This morning, getting green to a Delta L < 3 resulted in a Delta C of nearly 100, and green is pegged near white on the CIE chart. Same with magenta. I wasn't seeing anything like this yesterday.) I'm beginning to think I may have a bad meter.

It's well established that the D2 is not the greatest meter, but I don't have the budget for the Pro, so I need to figure out a way to work with what I have. If SpectraCal sells the D2, they must have some confidence in the meter, right?

Before I formally conclude that my meter is bad, I want to make sure I'm using it correctly.

  1. Does this meter need to be initialized every 15 minutes? The DIY workflow overview posted here says "yes," but CalMAN has never prompted me to initialize it. When I go through the initialization steps in the program, clicking Initialize brings up a small blink-and-you'll-miss-it pop-up window. I assume it's initialized, but the same thing happens regardless of whether I have the meter against a neutral dark background or a bright-colored one.
  2. How does one measure ambient light with this meter? I assume what's what the ambient light shield is for, but when I use it my luminance readings plummet from ~45 fL (appropriate for my dimly-lit room) to around 7 fL or so. This is true regardless of whether the lights are on or off.

These are my initial questions for now. The monitor I'm calibrating is an LG 42LW450 plasma. I was hoping to calibrate ISF Expert 2 for lights on, and ISF Expert 1 for lights off.
 
See less See more
#2 ·
I can answer the 1st question as I asked Mech when I started playing with mine - that meter does not need to be reinitialzed every 15 minutes.

I needed to get a tripod, so I have not gone much further than the pre-calibration readings so unfortunately I cannot answer the second question. :huh:
 
#4 ·
Go into meter settings. Check the 'Advanced Options' box - this is tough to see whether it's checked or not but you will know when it is checked. What do you have selected under the Low Light Handler? I believe these will vary depending upon the meter and I don't have a i1D2 anymore so I can't check for you. What options does it give you to the right of the LLH? I'd change those settings to 5 samples if that's available and try it again.
 
#5 ·
Make sure that your meter and display have warmed up for about an hour. Place the meter on the screen to get it up to proper temp. Too many variable to give you an exact answer. Ensure that you are in the same mode with the same settings either on or off each time. Keep Intelligent Sensor and Dynamic Contrast off. Use window patterns and not full screen patterns.
 
#6 ·
@mechman: LLH is off, but set to 1.5 seconds by default. The value for Low Trigger is 0; is this the value you recommend setting to 5?

@donnymac: I've been good about having my monitor on for at least an hour before beginning a calibration. I always start in ISF Expert 2, and have never used Intelligent Sensor or Dynamic Contrast (yuck). My meter is in the same room as the monitor, but it's obviously not as warm as the set is after an hour of operation, so I'll keep that in mind for my next try. I was using window patterns, but switched to full screen patterns partway through my calibration yesterday because I noticed IR from the window patterns in the center of my display. I'll go back to window patterns next time.

Great suggestions and reminders guys, thanks.
 
#7 ·
@mechman: LLH is off, but set to 1.5 seconds by default. The value for Low Trigger is 0; is this the value you recommend setting to 5?
No. For other meters the seconds are replaced by samples - I was talking about setting it to 5 samples. :T

@donnymac: I've been good about having my monitor on for at least an hour before beginning a calibration. I always start in ISF Expert 2, and have never used Intelligent Sensor or Dynamic Contrast (yuck). My meter is in the same room as the monitor, but it's obviously not as warm as the set is after an hour of operation, so I'll keep that in mind for my next try. I was using window patterns, but switched to full screen patterns partway through my calibration yesterday because I noticed IR from the window patterns in the center of my display. I'll go back to window patterns next time.

Great suggestions and reminders guys, thanks.
Set the meter up on the display while it is warming up. I'd bet this is the problem in that the i1d2 is temperature sensitive from what I recall.
 
#8 ·
In your original post you talk about the color/cms readings being out of whack. How about the greyscale readings? How does your gamma look? Are you having the same instability with the greyscale?
 
#9 ·
I didn't get that far with my calibration this morning. Honestly, I was pretty pleased with the way things (i.e., CMS + grey scale) turned out yesterday, and only fired up CalMAN again this morning to see if my settings looked as good today as they did then. Of course, when the luminance and Delta H values were far from what I adjusted them to yesterday, I started tweaking them . . . and then realized I forgot to write yesterday's settings down. I stopped what I was doing after my "correct" green and magenta screens were nearly white. So, I need to redo my work tonight. Practice makes perfect, eh?

Gamma set to Medium on my set is fairly flat at 2.2, which is right where CalMAN says it should be for a dimly lit room.
 
#11 ·
Thanks again to you all for your suggestions. Last night's calibration was a much better experience after I let the D2 warm up with the TV.

Tangential question: once calibrated, can grey scale values be kept the same across all inputs? I ask because the LG has an "Apply to all inputs" option for those values. What about CMS settings? If those need to be adjusted for each input, will that have an effect on the grey scale values? (In other words, do I need to calibrate each input from scratch?)
 
#12 ·
When applied to all inputs then all inputs should be the same as far as calibration. If you need confirmation then just run a quick greyscale and gamut reading on each input after copying the settings. Congratulations on working out your calibration issues. Enjoy your new calibrated picture.
 
#13 ·
Like several of you perhaps, I bought the D2/CalMAN DIY license bundle from SpectraCal a few weeks ago. I participated in SpectraCal's Video Calibration webinar last week, I've read through the DIY workflow overview posted in this forum several times, and I've gone through the Standard workflow three times now.
Sorry to glom onto your thread, but quick question: Do you by chance know if the webinar is available for later viewing? I bought the package, too, but in advance of the new TV I intend to use it for, which I hope to bring home next week. So I wasn't paying attention and didn't note there was a webinar coming up...

Thx for being the guinea pig and getting us the necessary tip of warming up the D2 on the screen before using it, that'll come in handy once I start using mine. :)
 
#14 ·
It's not online, and even though I got to participate for free (thanks to a code I received when I ordered the package), they normally charge $99 for it.

At the time I viewed the webinar, mechman's CalMAN DIY thread wasn't yet up. To be honest, the webinar covers the same material, with the added bonus of the trainer doing a live calibration of an LG set as he spoke. I think you'll do just fine reading his posts instead of paying for the webinar.
 
#20 ·
So is it normal to see the initialization message blink and go bye bye when initializing this meter?

Whenever I have tried it, I get the same thing as the OP. I did get an initialization ONE TIME where it took a second or 2 that said 'initializing meter' and it was a slow scroll right to left which is what I thought it was supposed to do in the first place. That was one time though, and every time before and after was a 'blink and you miss it' message and sometimes I have to click it a few times just to get the box up.
 
#24 ·
Is it possible to mess up your meter when trying to calibrate it?
I tried calibrating it last night thru calman and it hung on 'calibrating meter' for a good 5 minutes. Since it wouldn't stop I tried to kill the process but that was hung too. The only way I could get it to stop was to unplug the meter.

Is there a chance I messed it up?
 
#27 ·
Hmm

Well something is def wacky, when I try to calibrate the tv now (it's a sickness lol) on the greyscale area where I was trying to calibrate it (gains) I'm getting readings that are way to high and I'm literally having to drop the RGB gains into the minus 20s and 30s just to get it anywhere close to zero. That's not right at all, I've never had that happen before and I didn't have this problem until I tried to calibrate the meter last night. It only affects the gains though, and not the offsets or 'bias' on my set.

When I do make adjustments they are VERY SMALL and not taking big lunges like they used to while the bias settings are taking the lunges like they always have, and as the gains always have.

I think I messed it up, I don't know why it would do this. Every other reading is perfectly fine during the calibration but the gains on the greyscale.

Again, completely stumped. If I leave the picture like that with the gains way down like that it looks horrible and very very bland and skin tones are really pale looking. :( :(
 
#28 ·
Are you using RGB absolute?
Are you taking a 100% reading first?

Sounds like the gamma is screwed up.

Also when you say calibrate the meter, do you mean creating a profile or initializing?

Typically re-calibrating a meter is something that is only done in a calibration lab.
 
#31 ·
I believe its RGB absolute, it is to what it is defaulted to. I didn't touch any settings.
Yes, 100% reading first.
The gamma on the TV is not anywhere near a straight line at all except for when it first starts reading. I think the gamma on this TV is not very good, I would think it should be more of a straight line.
I meant initializing.


Sounds like you're calibrating an LG. I can't give you too much help with the issues your having but ill offer you some general advice for your display.

Local dining, dynamic contrast etc...off
HDMI black level -high
Edge enhancer - off (but check to see if low looks better)
Set your color profile to 709 I think it defaults to wide/expanded on your set.

1)Measure your gamma to be sure it's getting close to 2.2 you can adjust contrast/brightness like high/low controls for gamma. Just don't go too far with it as you are directly adjusting white/black levels doing this.

2)Calibrate WB before you attack CMS as CMS used white points as reference. Also your workflow is set for WB target to be gamma 2.2 so the closer that is to start the less work you will have to do.

3)green will likely be off on your LG, even after you tweak it there is a good chance it won't be 100% (test setting color and tint with blue only mode prior to CMS, sometimes it gets you closer)

Measure your display with the menus up and with controls off. You may find a difference in your readings between the two. If so use the measurements you capture with the menu/controls off screen. (most LGs drop some green when the menus are pulled up.)

Good luck
-Rabbit
Thanks for the tips, but its a sony KDL-46V5100. I have either off, low, med, or high for gamma settings. In order to get a decent calibration I have to choose low. And all of those settings are off - anything like dynamic contrast, pure white, etc., all off except for the low gamma setting.

Are calibrating 2pt or 10?
Remember your target level is the same as gamma level 2.2. After you adjust one part take a full read to see how your adjustments are effecting the whole curve. Also remember your most important target is to get each step to .3127 .329 if you must have an error it's best to have it trending to lower values for the two coordinates (trending to blue) as it will be less noticeable as if you have green or red too high.
Its 10pt I'm doing. I have some screenshots of the gamma curve and the final readings. What's strange is that the picture looks really good like this and isn't much different than other calibrations I did last week. It's just that the numbers are looping around to give the same result, if that makes any sense. In other words, the higher I have to bump the bright/contrast etc, the lower I have to move the white point settings.

For example, here's 3 calibrations I did the other day:

1) picture 80
bright 58
color 50
hue R1
gamma low
R-gain -3
G-gain -10
B-gain max
R-bias -7
G-bias -10
B-bias -13

2) picture 82
bright 46
color 49
hue R1
gamma low
R-gain -3
G-gain -7
B-gain max
R-bias +7
G-bias +8
B-bias +3

3) last night (check out the gains)
1) picture 90
bright 53
color 50
hue R2
gamma low
R-gain -28
G-gain -33
B-gain -19
R-bias -2
G-bias -2
B-bias -9

Taking pics of the display with my camera and comparing them, all of these settings gave practically the same picture with the second one having the least amount of detail, or softest looking, while the first and last gave the best 'pop' of colors and the last having the greatest detail and has slightly warmer skin tones and most even color across the RGB balance. But the difference you would not see with your eyes, the only way I saw the difference was to compare the pics side by side and the differences while there, are very minute. Last night's calibration did give the best picture though, I'm not sure why I freaked out at the gain settings. I would not think that was normal.

But here's a question you guys might know? As a general rule, doesn't a low brightness/high contrast setting give the best quality in a picture everything else aside?

Here's the pics of gamma and chart from last night.
 

Attachments

#29 ·
Sounds like you're calibrating an LG. I can't give you too much help with the issues your having but ill offer you some general advice for your display.

Local dining, dynamic contrast etc...off
HDMI black level -high
Edge enhancer - off (but check to see if low looks better)
Set your color profile to 709 I think it defaults to wide/expanded on your set.

1)Measure your gamma to be sure it's getting close to 2.2 you can adjust contrast/brightness like high/low controls for gamma. Just don't go too far with it as you are directly adjusting white/black levels doing this.

2)Calibrate WB before you attack CMS as CMS used white points as reference. Also your workflow is set for WB target to be gamma 2.2 so the closer that is to start the less work you will have to do.

3)green will likely be off on your LG, even after you tweak it there is a good chance it won't be 100% (test setting color and tint with blue only mode prior to CMS, sometimes it gets you closer)

Measure your display with the menus up and with controls off. You may find a difference in your readings between the two. If so use the measurements you capture with the menu/controls off screen. (most LGs drop some green when the menus are pulled up.)

Good luck
-Rabbit
 
#30 ·
jacare said:
Hmm

Well something is def wacky, when I try to calibrate the tv now (it's a sickness lol) on the greyscale area where I was trying to calibrate it (gains) I'm getting readings that are way to high and I'm literally having to drop the RGB gains into the minus 20s and 30s just to get it anywhere close to zero. That's not right at all, I've never had that happen before and I didn't have this problem until I tried to calibrate the meter last night. It only affects the gains though, and not the offsets or 'bias' on my set.

When I do make adjustments they are VERY SMALL and not taking big lunges like they used to while the bias settings are taking the lunges like they always have, and as the gains always have.

I think I messed it up, I don't know why it would do this. Every other reading is perfectly fine during the calibration but the gains on the greyscale.

Again, completely stumped. If I leave the picture like that with the gains way down like that it looks horrible and very very bland and skin tones are really pale looking. :( :(
Are calibrating 2pt or 10?
Remember your target level is the same as gamma level 2.2. After you adjust one part take a full read to see how your adjustments are effecting the whole curve. Also remember your most important target is to get each step to .3127 .329 if you must have an error it's best to have it trending to lower values for the two coordinates (trending to blue) as it will be less noticeable as if you have green or red too high.
 
#35 ·
the RGB gains are just individual versions of the picture control.

rasing contrast and then lowering RGB gains is exactly the same thing.

Getting brightness set correctly and contrast set correctly are both extremely important.
The numbers in the user menu, aren't particularly relevant, it's the numbers the meter is picking up that make all the difference.
 
#36 ·
That is quite a swing in numbers on the RGB settings and the Brightness setting. Brightness from the first and second calibrations should not change by 12 with only a 2 point change in contrast. I'm sure the reason it changed so much is because of how much green you used. A general rule of thumb is not to change green in the RGB settings unless absolutely necessary. The big reasonis that once you set contrast and brightness changing green will require you to go back and readjust contrast and brightness. Changing the green so much will also affect gamma.
If you are happy with the picture then leave it alone and enjoy it. If you want to play some more then I would recommend resetting the 2pt and 10 pt rgb settings to 0. Do the 2 pt without using green. Then do the 10 pt.
Have you possibly been changing the color temp between runs? Are you allowing the meter and the display to warm up before taking readings? Are you making sure you place the meter in the exact same spot each run? This is important for an lcd which may have issues with screen uniformity. If yo do another run take a pic of the greyscale before chart to post along with the after chart. Good luck whichever way you decide to go.
 
#39 ·
A general rule of thumb is not to change green in the RGB settings unless absolutely necessary. The big reasonis that once you set contrast and brightness changing green will require you to go back and readjust contrast and brightness. Changing the green so much will also affect gamma.
The gains/highs are just individual contrast channels. The better rule of thumb for gains is to only turn them down and not up. If you have contrast set correctly, and red is low, if you turn red up it will likely start clipping.

So by turning any of the gains up you'll likely cause clipping if it was already at the edge of it's contrast setting.


Use green as the anchor to balance off of is more relevant when doing gamma adjustments on a 10 point system. On a 10 point system at the 100% point you'd still only turn the values down.
 
#37 ·
Do you have seines set to general or auto on the Sony? Also does your TV have seating detection on?

I've seen LEDs have way too high swings in upper gamma levels. That's why pulling up a stepped grey scale test pattern (bars that for from black through grays to white) will let you visually verify that the gamma looks right. That is you should see a steady increase from black to white through all shades of grey. A split gray scale is the same test pattern but it's been cut horizontally and flipped so you have black and white on both sides of your screen.
 
#38 ·
donnymac said:
That is quite a swing in numbers on the RGB settings and the Brightness setting. Brightness from the first and second calibrations should not change by 12 with only a 2 point change in contrast. I'm sure the reason it changed so much is because of how much green you used. A general rule of thumb is not to change green in the RGB settings unless absolutely necessary. The big reasonis that once you set contrast and brightness changing green will require you to go back and readjust contrast and brightness. Changing the green so much will also affect gamma.
If you are happy with the picture then leave it alone and enjoy it. If you want to play some more then I would recommend resetting the 2pt and 10 pt rgb settings to 0. Do the 2 pt without using green. Then do the 10 pt.
Have you possibly been changing the color temp between runs? Are you allowing the meter and the display to warm up before taking readings? Are you making sure you place the meter in the exact same spot each run? This is important for an lcd which may have issues with screen uniformity. If yo do another run take a pic of the greyscale before chart to post along with the after chart. Good luck whichever way you decide to go.
That's a good point.
Take a look at your CIE chart (the color triangle) you'll see that green accounts for about 70% of your vision. Since adjusting RGB is in fact adjusting luminance then green will have the most effect on luminance where as blue will have the least, but our eyes will see increases in blue as adding sharpness and brightness. Go figure.
 
#41 ·
I think the confusion was when you were asked if you were doing a 2pt or 10pt and you replied 10pt.

You are measuring with 10pts but only have a 2pt system.

A 2pt system, the cuts/lows are basically RGB brightness controls and the gains/highs are just RGB contrast/picture controls.

So if you turn contrast way down, then you can turn gains up higher and vice versa.


If you only have gamma presets, just pick the best one, and don't worry about gamma. If you've got a kink in your gamma curve, you'll never fix it without a 10 point control system.
 
#42 ·
I thought the question was about the gamma curve? I guess that's what I meant with the 10pt answer, taking a reading of the gamma from0-100% in the 10 steps. I guess I misunderstood the question. :bigsmile:

Ah yes, its a 2pt system, exactly how you described it. So is there anything I did wrong or can correct somewhere?

I have to say the picture looks good, probably as good as its going to look with this particular set. I understand it's not a top of the line model (I wish I knew that when making the purchase, I had the $$ to get one at the time :rant: ) Even with the discrepancies in the settings, it all generally points back to the same PQ, and looks basically the same as the last calibration, with only a very slight modification to the picture.

So is there a way to set contrast the right way? I'm confused as to how to set it other than using it as a luminance guide because it says on the AVS HD709 disk that to make sure anything past 235 (if I remember right) should not be visible but it's ALL visible no matter how high or low I set it and the bars never get that pink hue color that some people talk about getting if its set too high..
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top