Looking for Feedback on my Professional Calibration/Concerns - Page 2

I don't know what you mean because this statement is wrong. Grayscale comes before color. Chroma errors have to be pretty significant for them to be more obvious than incorrect colors due to a poor grayscale.


Let's keep to objectivity as your subject tastes have no basis in the science


Try telling Sharp Elite owners this

I've heard this concept stated elsewhere, and to be honest, I find it disturbing. I'm under the impression that the most accurate looking (and most accurate to standards) calibration has parameters which are completely quantifiable. ALL of these parameters can be measured and can also be charted and graphed. How then can perfect charts be representative of a "bad calibration"? I feel as though I'm missing something here.

Is it that when these types of statements are made, they are referring simply to only a good Gamut chart? Or maybe only good gray scale results? Maybe that's what they mean... Because I would like to believe that if you provide enough data, covering all possible calibration data sets (25, 50, and 75 % saturations, gamma curve, lumma, etc. etc.) you can look at the charts and prove definitively that the display is calibrated properly. Am I wrong?

I just finished reviewing the JVC X30 projector and the gamma controls allow you to fine tune the presets at levels (10,15,20,30,40,50,60,70,80,90,95 I think, maybe 5 as well) for both white and then RGB. Using this I can get a rail flat grayscale and perfect gamma. The charts are amazing.

However the calculations the JVC uses to do these are either wrong or lack proper bits for accurate calculations as the image is totally posterized and looks like bad Internet downloaded video. The charts only measure so much but your eye tells you far more. Use the charts, trust your eye. Almost all controls have limitations and the key is knowing those and working around them.

Hmm, as far as by eye, yah I am pretty sure as I used to do A/B testing of projectors very OCD for many years. With projectors (especially older ones), we had to make crazy huge concessions in calibrating sometimes, and that's why I did the A/B so many times trying to live with my "magic concessions". I am sure that a tiny blue error in the gray scale is no huge deal, there are more important issues. I am not saying it is invisible in all cases, but it surely isn't the worst thing. Actually on some projectors I have owned that have BUGGY CMS's (which is practically the majority), there are NON-symbiotic relationships with gray scale and the gamut that affect each other in ways they should not.

To say there is one exact order doesn't make sense to me since it depends on how buggy the devices' calibration controls are in what order you have to do stuff. Heck, try calibrating a Viewsonic Pro8200 projector, when you figure out the best order, let me know (the controls will die at Sat 127 and all freeze up btw, that is until you get them unstuck again by a certain secret order )

So, what did you do to correct it? Or, were you able to correct it? (I looked up 'posterized')

That was basically my point, and I completely agree, because there are programming bugs in the way some video devices do certain things as well, or even processing bugs.

Take the old Sony vw60/vw70 series, some serious limitations before using a Lumagen to correct with, basically leaving the CMS completely alone was much better than attempting to fix the issues (unless you were self-calibrating and have hundreds of hours to play around with it). The few TV's I have calibrated (very few) were so much simpler than calibrating projectors, that I don't think anyone could even learn how to calibrate "with concessions" using many of these newer TV's. I learned how to do it by taking a projector with a very near perfect calibration and A/B'n it to a projector that doesn't calibrate right, then you make decisions over hours and hours of A/B'n. Why did I do it, mostly for fun and a bit of OCD I suppose.

It is also easier to notice these things when looking at multiple devices A/B'd on screens over 100"...

I set it right back to the 2.3 default, without the fine tweaks, and it's perfect. The JVC offers so many gamma options, and they're all pretty close to what they say they are, that I don't see it as a major issue. It's much better than many $3,500 projectors, but since the control can lead to a broken image easily, I'd prefer it wasn't there myself. I'm attaching a couple using a scene from The Social Network. Ignore the horrible picture quality, but you can look at the forehead and see the posterization that occurs.

If you measured gamma at 100 points, you might catch this on a chart, but I'm not certain. The fully calibrated one was even worse than this, but I didn't get pictures of it. This gives you an idea of what it does. I'm not trying to single out JVC here (it's a really nice projector), but just using it as an example of something that can be really good on charts, and really bad in real life.

You can be precise but not accurate. A good scientist wants both accuracy AND precision. That pretty much answers the question of the fact you can have perfect charts but not accurate to standards. You want both.

Mine was pretty consistent over the time that I had it, though it's a review unit and actually goes back tomorrow to JVC. I found the contrast ratio, with no dynamic iris, to be really amazingly good. Night scenes in Drive (which uses the Arri Alexis camera that also has amazing contrast) had so much pop.

I guess depends on the lamp, it is known to be an issue though and confirmed by many people in the JVC threads. If I remember correctly, there are some internal code in projectors that internally recalibrates gamma automatically to try to maintain a consistent one over lamp wear (maybe I am wrong), and people were thinking this part in the JVC that does this has a bug. 100-200 hours is where my drift got bad.

Very few reviewers will have it for that long. I put on close to 150 hours with it, and that's a lot for someone doing reviews of them, given all the other time commitments people have and loan periods.

Good to know, at least some don't have the issue as much then... Maybe I'll call JVC soon and try to get lamp replaced.

So, let me ask you this: If someone who didn't know better asked you to calibrate their JVC X30, would you tell them "No, calibration makes it look worse." How many more displays like this are out there? Is there any way to know before you waste 3-5 hours calibrating it? I mean, could you even still bill a person after spending all that time just to set it back to defaults?

Edit: And more importantly, what exactly caused this posterization in the first place? Is this some internal defect that can be isolated and identified, so that it make be tested on other displays before beginning a calibration? How many more types of (potentially) non-quantifiable visual errors have you experienced post calibration.

I'm sorry for all the questions, but please believe me when I say that none of them are rhetoric.

So fellas, how does gamma tools and lamp life of a JVC projector help out the OP with his professionally calibrated LG panel that apparently was not done up to snuff?

Or was it? .. This comment should shake the tree and maybe mover things back closer to topic.
So on a gray scale.. is there a real visual difference between a DelteE of .5 and 3 worth the pros time to fight for on a $600 panel for a typical calibration customer (not someone with a $1000 worth of tools and constant access to this forum)?
Is the calibrator Wrong if he stops adjusting when he gets a flat gray scale and hits the target gamma and his deltas are just under 3 (visual perspectives.. or say they say)?

He is getting paid for a service, but he never said that he would adjust WP to withing a DeltaE or x - x.

So just because he did not take that extra 10 minutes to get the panel as perfect as his non profiled expensive colorimeter would allow, was he wrong? If so, why was he wrong? What body of standard or certification says you must go beyond what you can see to be correct?

I am no trolling, this is a hot topic and i can see both sides .. except for the what I can't see..

The only reason someone would calibrate color (lets use an appropriate term like CMS) before adjusting the grayscale would be if CMS adjustments had zero effect on the grayscale and grayscale adjustment had zero effect on CMS.


I'm taking it up with you since you chose to use someone else's words, without fully understanding what you read, and applied them to this thread.


If you think the grayscale on this particular LG display, post calibration, that has dEuv errors above 3.5 from 50-100% stimuli is small, you either have laxed calibration standards, do not know the capabilities of the display and/or don't know the threasholds for when errors are visible.


I've already corrected you. I would also advise you not to post what you do not fully understand (and some how have classified as "opinion").


You might want to add a few additional words to your vocabulary. Stimuli level, luminance and hue would need to be included when discussing the typical 1931 CIE chart.... which of course is not the ideal chart to use. Especially when one starts talking about "perception".



Your exculsive experience with projectors does not hold any bering on this LG CCFL LCD.


There actually is a recommended order for calibration. Only in rare and unique instances should one have to deviate from the recommended order. You can see this in the layouts for ChromaPure and CalMAN.

Yes there is a visual difference between dEuv .5 and 3. Yes any calibrator worth spit should attempt to get the errors as low as possible per the actual display's capabilities. What doe the price of this display have to do with whether or not a calibrator should take the time to maximize the performance of a display?


If said display could be calibrated have errors below 1, then yes.


He is being paid to get the display as close as possible to the standards. If he chooses to practice the "good enough" ideology, IMO he is a sloppy calibrator.


IMO, yes. I guess it is the reason why there different "tiers" of calibrators.


I cannot answer this question as you seem to believe that the human eye cannot detect errors with moving video below a dE of 3. Not only is that wrong, but you might want to ask yourself what dE forumla are you using

No, it looks far better after calibration than before. The only thing you can't do is the really fine tweaking of the gamma curve, but on every display or projector that's allowed that, it's always been broken in my experience. If you want that control, but a Lumagen Radiance. My final gamma is still 2.26, just not 2.30, with near flat RGB balance and color temperature.


Mitsubishi projectors also have gamma controls that work this way I found. Toshiba TVs were having a weird issue where CMS and grayscale settings on one mode caused a rise in black levels on other modes, and the list goes on. You find issues, you work around them, and sometimes you just start over once you figure them out and do it all again. If you can't improve it during calibration (rare) then I'd not bill them and move on, learning from the experience.


The fine gamma controls don't work correctly, or at least not well. The calculations might be using 8-bits of data when they need 12 or 16 to be accurate and smooth, or using integers instead of floating points, which could also cause this. There are so many things that could cause it, that it really doesn't matter how they are generating the error, it's knowing what causes it and avoiding that in the future.

You can easily catch these by looking at real world material, or even a grayscale gradient might have caught it. I've calibrated a grayscale and gamma where it looks perfect on the charts as you go through, but then you bring up a gradient and you see big swings in color, as when the CMS did the calculations between 20 and 30 points, it did a swing to red in the middle, and then to green between 30 and 40. You have to use the charts to get things right, but then look afterwards at different patterns and material to see what is and isn't working.

@Dnice
I own Calman and I know the things you were trying to "correct" me on. You are also again telling me I said to do color before grayscale or in some specific order, but I never said that and that is not even applicable to what I was talking about. Actually with projectors there are different optimal orders often in calibrating, not necessarily to do with grayscale vs. gamut calibration, but speaking about overall issues in the entire calibration.

My initial post was simply a question...

I never said the calibration was good, as I'd have to see it to know, but just that the blue error is not the end of the world, of course he should have gotten it flatter but it doesn't mean much without a context. Always better to get it as close as possible, but you can't only look at one chart and know how the calibration looked.

I am not defending the calibrator, or saying low end equipment can be dialed in very close. I actually was working on a follow up to D-Nice but got interrupted with work so will abandon that one for this..

What I was trying to do was focus attention back on your problem and off the other tangent that got started.

You and I know that the LG can be dialed in to under .5 but apparently our guy didn't do his research. That fact aside, what is the industry teaching for when to stop making gray scale adjustments? If DeltaE or 3 is to high, what is the cut off?

What do they teach in the THX/ISF classes as to when to stop adjusting Gray scale?
Obviously experience will tell you when, but what about for the time before you get that experience, or when you have not done your home work?

D-Nice says there are Tiers of Calibrators, that's true but where is the rating list other than the one thread here on AVS. If we choose someone outside of this tiny circle how do we know what tier they are in until we in your current positions? Seems the only way is to ask lots of question and if the answers are what we expect then buy the service, other wise look elsewhere?

PlasmaPZ80U, I don't recall if you said, did you tell the calibrator you owned your own equipment and had already done the calibration many times but were unhappy with your own results prior to his accepting the job?

You should have gotten the same quality of work either way, but if you did not disclose you skills and expectation, he probably was more defensive when he found out and maybe felt setup?

It sucks that we both had the same similar experiences with this person but it was this very experience that pushed me to buy the equipment and learn how to DIY ..

Your quote:




My response:



Now, show me where in my response I was refering to calibration workflow. When I stated "Grayscale comes before color", I was refering to how video is presented. Color is on top of the grayscale. "Grayscale comes before color." Get it now? My response in post #47 regaring your "I didn't say in a specific order" in post #35 had nothing to do with the above. I was correcting you on another fallacy in your posts. Interesting you claim to want to learn yet translate corrections of your errors as offensive insults.





So, you are saying in many of your 30+ projector calibrations there has been reason to calibrate the CMS before doing the grayscale or calibrate the grayscale before the setup of black and reference white?


If I knew everything, there would be no reason for my existence as a Human. If you took my responses as insults, so sorry as that was not my intention.

I was just assuming his calibrator checked one saturation point and not all of them. I don't know anything about this TV.

There are many different calibration steps besides just gamut and gray-scale. So yah, the complete inability to use ALL or part of a CMS, or sometimes even having to use an external VP on a wider gamut preset does change the order of the calibration a tad bit :P

You try to get it as close as you can, which often might mean multiple passes, but if at the end all you can do is get it below 3, you would be OK with that. If you can get it below 1 you probably aren't going to spend another 45 minutes trying to get it below 0.5, even if it is possible, as it is likely not noticeable. If you can get it below 0.5 easily, that is fantastic, but at some point you stop picking nits and accept what you can get. If he got it below 3 in 5 minutes and then stopped with it being "good enough", I'd say that's not enough time spent. If it took him 90 minutes to get it below 3 after many, many passes, then that is probably as good as he can do. The number without context isn't always helpful.


I'll admit to falling into this space. I took the ISF Level II class and passed, but 99% of my reason for taking the class was to make certain that the reviews I do will carry more weight with readers, and so manufacturers know that when they send me a video component, I can get the most out of it.

That aside, I am listed at the ISF site for calibrators, and would happily do them for people if the opportunity came up. Outside of reviewing I have no experience with doing customer calibrations, so for me I'd have to use my reviews as a reference, and make certain to do as good a job as possible so that I can use those new customers as references for later on. Since I lack those references, I'd fully expect customers to ask me a lot of questions before and during the calibration, and that I'd have to bend over backwards to make sure they are happy with their service.

If you ask them questions and you don't get answers that you want, or you feel uncertain, then look for a new calibrator that you are comfortable with or find one on AVS that does tours. Your local calibrator might be able to do as good of a job as a touring one, or they might not, but you should be comfortable with them before you hire them.


I have to say I'd be happy to get customers that know what they are talking about and want to participate. I hear too many horror stories from people that get customers that have no idea what you are going to do to their TV and really don't want a calibration, so I'd rather have someone ask a ton of questions of me and enjoy the process, knowing they'll be happy with the outcome if I do a good job.

http://www.avsforum.com/avs-vb/showt...4#post21664644


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


Ditto...

Most of your posts in the last two pages you were attempting to create separate arguments and had absolutely nothing to do with my statement, and you were adding words and meanings to my statements try to qualify things in ways that I never said, or telling me I don't understand anything without you even getting the point which I stated very clearly from the beginning.

It did on mine. I ended up leaving the CMS at 0 since there is no luminous controls. I was told that if you move things around to much with the CMS you have to go back an tweak gray scale, then go back and tweak the cms.. repeat as needed. I was also told Changing CMS should not effect gray scale if the CMS works correctly. For CMS to really work correctly you need 3 adjustments for each color not just 2. Sometimes it is just best to leave the color alone if the controls don't work well.. My set does not have the IPS panel and is way under saturated in red. I can't fix it, it does not have all the controls and the controls it does have messes with gray scale.. That all tells me to just leave it alone.

Yah going back and forth can be very helpful.

Most CMS's don't work correctly, that's the thing. The times they often do work correctly often ends up being the devices that ship closest to accurate OOTB and might be because you don't have to use the CMS as severely. The more severe you have to change a CMS, the more often there are side effects.

Did your calibrator end up tending to your Night mode concerns?