CalMan 5 Release Notes and Discussion - Page 10

And it has already been stated that 0.1 steps are going to be made an available option. Continuing to belabor the point is not going to make it happen any faster.

You "forgot" a little thing: it will be ONLY available for the MANUAL DDC controllers, NOT for the autocal-routine regarding the Radiance

Hi,

is there anything specific I need to consider when creating a spectro profile for a colorimeter in Calman 5 (e.g. i1D3 profiled to an i1Pro) ?

Which stimulus level is recommended (if that does matter) ?

Does it matter if the TV is calibrated (I assume not) and therefore the signal is closer to the actual target pattern ?

I created 2 correction profiles already (both i1D3 to i1Pro) - both on the same TV - one with the TV in a calibrated state and one with the TV at factory settings... Looking at both matrices, the values do differ, up to 0.056, not sure if that can be disregarded or what a threshold would be that is of concern...

Thanks.

- M

Had to use my Calman DPG 1200 to calibrate my old LG 55LH90 at my daughter's this weekend. Now I know why I put iScan Duos on all my main systems!
Calman 5 does control the DPG 1200 better than the Calman 4, but still retains alot of pain-in-the-butts! Must have had to recycle back to Patterns 01 numerous times, and that didn't aways work, and had to be recycled again. Click on 100IRE, it would come up 95%. Numerous times, the marker is left on 100IRE, but you then had to click on 90IRE, then re-move back up to 100 afterwards. After you're finished, now had to wait until the 1200 recycled back to Black 0%, before starting the next leg of the journey. Added about an additional 1 to 2 hours calibrating time versus the iScan Duo. (Please Note: 1200 was set to RGB High, and not Auto)
Couldn't figure out why when progressing to the Pattern requested, the 1200 would one time stop at Black 0%, then proceed. Other times on 10% or 20% Gray before proceeding - despite where it started from 0, 10, or 20, most times it did arrive at the pattern requested.
Downloaded the 5RC3 today to try later.

I've had a play with Calman 5 and my Radiance with the auto cal and colour cube. A couple of questions: Can the DDC control panel for the 3D LUT on Radiance show it's native values as opposed to percent? eg 940 is white instead of 91.4% (or what ever it is). With the autocal button, it does the whole series, can you specify a certain stimulus point and autocal only that one? I'm thinking of after you've done a run, you might want to work on a couple of points. I've also noticed that after I hit the + or - on the grey scale DDC control, the + and - disappeared until I entered a value into the field.

940 is 10bit value = 235 @ 8 bit = 100% White.

Yes, I understand that. I think it can be easier to know if you're changing the actual value by much. Eg there are 876 steps in 10bit video levels (64 - 940). But as a percentage with 1 decimal place there are 1000 steps, so not every .1% will change the value in the Radiance. I like being able to change a value be the actual minimum amount when it comes to fine tuning.

And it's easy to know if you're at maximum or minimum with a value of 64 or 940. With the percentage it's ambiguous.

Not at this time with the cube AutoCal. Much of the logic is based around earlier readings of the gamut.

Yes you can use the DDC controls to adjust the cube after you're done with the AutoCal. The issue with the disappearing incrementer/decrementer is a bug that slipped through, we should have a fix posted early (PSD) tomorrow.

We use percent to abstract the hardware from the UI so that you can become an expert at CalMAN without having to learn the intricacies of every piece of hardware out there.

Whenever we update the hardware we read back the value, so we will never show a change that didn't actually make a change in the hardware as well, we also show two decimal places so you'll see if it was 91.01 to 91.98, since the actual native step size is 0.0978.

Just a thought here guys......and this DOES NOT APPLY to you pro calibrators out there but for beginners such as myself.
I would love to watch a simple descriptive video, that would shows me, what I believe, are the more challenging parts of Calman 5 (such as Gray Scale Calibration, & Colour Gamut with & without the use of a (CMS)).

The AVS HD 709 disk has a video section from "HD Nation" on how to set up the basics (Brightness, Contrast, Sharpness & Colour (using Blue filter). I mean this is pretty basic and most of us enthusiasts can quickly pick up on how to perform these adjustments. Woudn't a similar type of video showing people how to adjust a Gray Scale & Gamut, using Calman5 be useful?

I attended the SpectraCal webinar just a few days ago, and honestly, I learned nothing new that I already did not know from reading the instructions manual. In fact, for anyone who would be just starting out (ie no AVS Forum knowledge or previous research) they would be absolutely lost, for that webinar ASSUMES that you already have a good understanding of display and calibration basics.

Now, I know the Gray Scale, or Colour Gamut are there in the manual as well, but I am finding it hard to understand how to adjust the Colour Gamut.
Could someone not put up a simple video on Youtube, or if there is one already could someone provide a link.

Thanks in advance.

Paul

For those all gung ho to take a short cut to AutoCal calibrating their sets, doesn't setting, Backlight, Contrast, Brightness, Colour, Tint, White Balance to Spec, and 2 Point have to come first? I'm sure these are pre-required before getting to the Auto Part!
In fact, I thought it was also recommended to do a Greyscale first as well, in case the Duo (in my case) has a failure, thusly leaving a calibrated set?

Just to make sure we are clear.

Yes you absolutely should be doing all of that first before you AutoCal.
The basics of calibration need to be done in the display, otherwise you'll end up with a sub-optimal picture.

We hear you loud and clear.
We hope to be able to post such videos for CalMAN 5 in the near future.

Thanks Joel, I keep forgetting about the BOLD letters means "shouting".....that is really good news I hope to see these videos soon.

Thanks

Paul

Outstanding look forward to it, I've upgraded to Calman 5 but havent tried it out yet, appreciate the time both you and Derek spend here lending your expertise

Now I´m a little bit confused.

In this thread http://www.spectracal.com/forum/viewtopic.php?f=94&t=4142 you confirmed setting Contrast and Brightness to their default values before AutoCal Greyscale. With my VT50E this is 60 for Contrast and 0 for brightness. This is especially for contrast far away from the right setting.

Something I noticed when it was doing the grey scale autocal, was that it got the RGB triplet perfect but then kept going, I guess to get the luminance on target, but the end result was worse RGB. Could you maybe have an option as to choose either colour or luminance as priority when "perfect" can't be achieved? It seemed to favour luminance over colour when making compromise. Or is this to do with which dE formula it's using to assess?

When it was doing the gamut autocal, it seemed to only do 1 iteration per point. I know the default target dE is 1.5 for gamut, as opposed to 0 for grey scale. Is this why? Or is only 1 for speed's sake?

I can't really go to much into the internals of how the AutoCal works, but it tries to balance everything as close as it can to the targets, both luminance and chromaticity. It doesn't favor one over the other, it's just trying to minimize error.

The dE target thresholds are simply thresholds, when we get under the threshold for that point we stop, otherwise the algorithm will continue to try and optimize the point until it thinks it's found the spot where making a one-click change will make the result worse. We use the threshold of 1.5 on Cube AutoCal for speed. We see times of about 15-25 minutes for a radiance cube with that threshold at 1.5. If you drop it to 0.5, you will see around a 1 dE improvement in the color checker chart, but the processes takes closer to 35-45 minutes.

If you have the time, and want better results, by all means lower the cube threshold.

My comment was in response to a comment about calibrating external processors with TVs. This is a bit different but similar to the way the Cube works in the radiance, the brightness and contrast controls are BEFORE the grayscale controls in the panasonics, just like the Cube being before the grayscale in the radiance.

So you have to set the brightness and contrast to that so the controls line up with the patterns. After you've calibrated the grayscale, then you can adjust brightness and contrast to their correct settings. If Panasonic moved their brightness and contrast controls to after the grayscale LUT, then you could set them first and the do grayscale.


When Calibrating things that are sensitive to being at the correct level you HAVE to work backwards through the controls. So you do grayscale first, then brightness and contrast, then CMS, on the panasonic. Several other displays are the same way.

Sorry, I didn´t see, that p5browne talks about Cube AutoCal. Thanks for your clarification.

Cube? Don't know a darn thing about Cubes!
Do know that everyone wants to do finer calibrations that are closer together - how about half clicks instead of whole clicks, or will the AutoCal and Duo do this?
Pisses me no end when 1 click one way is too much, and 1 click back the other way, the opposite. Where a half click would work out just right!

Actually I think the RC3a that just went up earlier today halved the size of the clicks on the duo DDC controls.

It's a bit less convenient, but you can always type in the parameter value instead of using the +/- controls to get things closer.

Thanks Joel, us Duo owners do tend to compare the DDC for Calman with the 'Duo Control Panel' capability.

I know you don't want to give away any details about AutoCal, and I understand we have the option to tighten the error limit for AutoCal, but do we have any other contol over AutoCal? For example, is there any way for us to change what AutoCal uses to compute errors? Experienced calibrators know that varions dE calculations produce different results for the same measured set of coordinates. Because of that, some of us look at 2 or 3 different dE calculations as we do manual calibrations to insure that ALL of them are as low as possible for each measurement. Is there any user control over what error calculations (thinking more of graycale and 6-point CMS here, this doesn't necessarily apply to the Cube/matrix calibration) are used by AutoCal. For example, If I choose dEuv for error calculations shown to me in CalMAN, does that cause AutoCal to use dEuv (instead of the default dE2000)? If there is no user control over the error calculations AutoCal does, is it doing anything internally to insure it doesn't get into any situations where, say dE2000 produces an error of 15 while dE 1994 produces an error of 3, while dE 1931 produces an error of 6.

And a follow-up AutoCal question, specific to Radiance AutoCal...

When you use various AutoCal modes... grayscale/gamma & CMS again (I don't think this applies to matrix calibrations at all). Is AutoCal using the smallest increments the video processor has available once it gets close to a low-error result? In other words, if you start out with 10% too much green and good red and blue, will AutoCal make a big adjustment first, then a smaller adjustment and smaller adjustment, down to the point is is doing "single click" adjustments as you would be doing if you were calibrating a Radiance manually with the internal Radiance controls (luminance, red, green, & blue for grayscale with values from 1-1000+)? Or is AutoCal limited to jumping 2 or 4 "clicks" at a time for each adjustment parameter (the way the DDC controls were working prior to today)?

We are looking more at the actual light measurements and trying to get as close as possible to the target.

The threshold we quit at for early termination is entirely dependent on the dE formula you use.
When we can't hit the dE, or the dE is set to 0, we basically run till every control is one click over or under the target.

The AutoCal does typically use a 0.3 step size. That's because it is an 8bit change at that point and not all displays respond to smaller changes. Also with the step size at 0.3 we get dE's that are sub 0.5, so we do get to the point where the measured value is less than 1 8bit step away from the ideal target value. I know we are always looking to do better, but I think when we get well below human perception, it's good enough.

Of course you can always go back in and tweak the DDC controls if you want to use the native step size and a dE of 0.4 isn't good enough for you.
Edited by sotti - 10/10/12 at 7:44pm

not sure if this a bug or if I'm doing anything wrong, but when I change the CMS for the Lumagen in the Display Control tab in Calman 5, it does not load the according DDC values for that CMS... example: when I have started with CMS 3, made DDC adjustments, then select CMS 4 in the select menu via the Display Control tab, the DDC values in the fly-out DDC window are still the ones for CMS 3, which could be very annoying if you commit these values for a different CMS...

Also, it would be very convenient if the DDC COMMIT button is on every page of the fly-out DDC window, not just the last page... I made changes in the past on the first DDC page and forgot to go to last page and hit commit...

The commit is only on the cube page because none of the other DDC controls need a commit.
There is a bug in RC3 where the incrementer decrementers adjust the value, but don't actually cause the display to be updated.

yeah that's what I initially thought and the auto commit did NOT work 3 days ago - I only used directly punching in numbers and working my way forward with TAB and SHIFt + TAB as the + - controls are funky at best... the rollover state is badly coded and does not respond correctly 3 out of 4 times... anyways, so did a DDC touchup via the DDC fly-out window, punching in numbers directly and thought everything got saved (that's what the red border effect around the value field is supposed to indicate, correct ?), but values were not stored...

loaded that CMS yday and values were not stored... glad I took a screenshot of the DDC fly-out window before I closed it, so I punched in the values again and then saved via the COMMIT button, they saved this time... any idea why I had these issues or is this a known bug ?

And also (regarding my other post), what about the DDC window values once I change the CMS ? Is that a bug or what do I need to do to get the corresponding values ?

Thanks !

- M