Official JVC RS20 / HD750 Calibration and CMS thread - Page 3

Originally Posted by SOWK

Why would you prefer a 3 chip design?

That only calls for more "issues", at least in my thinking.

Originally Posted by TomHuffman

Here are the CIE94 numbers for the THX preset.





Here are the CIE94 numbers on Mann01's latest "improved" technique using the CMS.





This is no better than what was available with the RS1/2.

Not an encouraging result. Other than red, none of the colors are within acceptable limits, and some of them (green, yellow, and cyan) are grossly inaccurate.

This is the result, despite the fact that the 1931 CIE chart looks reasonably good.



This just shows the inadequacy of relying exclusively on the CIE chart (especially the 1931 version) to grade color performance.

I haven't worked on one personally yet, but from the data of those who have it is beginning to look like the functionality built into the RS20 isn't a true CMS at all, but a pseudo-CMS with limited (if any) value.

You can select the THX preset, which gives reasonably good color but a quite yellowish gray scale; or, get a good gray scale with a custom calibration and live with the poor OOTB color. The included CMS controls don't seem to be able to improve it.

There may be very good reasons to like the RS20 (better optics in particular), but unless something new comes up we haven't seen yet or JVC offers a fix, better color performance isn't one of them.

I'm sure that this statement will annoy some of you, but that's what the data shows. If we get better data, then I'll be happy to revise this judgment.

Originally Posted by TomHuffman

You can select the THX preset, which gives reasonably good color but a quite yellowish gray scale; or, get a good gray scale with a custom calibration and live with the poor OOTB color.

Originally Posted by Manni01

Tom, I beg you not to take my data as meaningful (apart from proving I'm inexperienced but motivated!)

This is my first CMS calibration, I did train on my ACER fro greyscale, but it had no 11 steps gamma adjustments, and of course no full CMS. I am learning as I go, and as I progress I started to get a clear feeling that someone with more experience would nail this pretty easily. Of course seeing that LovingDVD who is much more experienced than me is hitting more or less the same walls crushes this theory, but I still believe there is a way to get there!

Originally Posted by TomHuffman

Not just LovingDVD. Cine4Home reported much the same thing as well, though not as explicitly.

The fact that you are new to this is not particularly important. Did you use a white pattern that was the same level of stimulus as the color patterns? Unless you did not, the data is the data. It doesn't change just because the person collecting it hasn't has a lot of experience. The mere fact that both of you report that lowering the green saturation tool has almost no effect on green saturation is, by itself, pretty strong evidence that the CMS has a pretty severe design flaw.

What you guys were finally left with was just lowering the main color control and then trying to compensate for the loss in luma by raising the brightness controls on the various colors, which is at best a kluge. You might want to see if you can get the brightness of green, yellow, and cyan higher without also raising their saturation levels back to where they were.

Originally Posted by Lawguy

Stuck in New Orleans due to snow in NY. It is unclear when I will get home. RS20 awaits.

Originally Posted by Bulldogger

Well at 6:27 pm, it's 71 degrees here. If you are out in Kenner at the airport, that's boring but at least you can enjoy the warm weather.

Originally Posted by Manni01

Cine4Home posted an almost perfect corrected gamut (although without the raw data so it's difficult to know how the Y values were, but I would trust someone like Ekkehart).

Originally Posted by Manni01

One question to help me understand what's still wrong in my values: when you say that apart from red, my colors are grossly inaccurate, could you quantify/explain why? Is there something flawed in the theory I am exploring, which is to lower the color value (we still need to find the optimum parameter) to be able to use hue and saturation to bring the colors spot on on the x,y space, and then use the brightness control to compensate for the loss of luminance generated by the lowering of the color control and get Y closer?

Originally Posted by TomHuffman


......What you guys were finally left with was just lowering the main color control and then trying to compensate for the loss in luma by raising the brightness controls on the various colors, which is at best a kluge. You might want to see if you can get the brightness of green, yellow, and cyan higher without also raising their saturation levels back to where they were.

Originally Posted by WOLVERNOLE


Not that I was one bit prophetic, but I sensed that the CMS in the RS-20 was going to be thee deal-MAKER over the RS-2/HD100. It still would appear to be a better product at a great price, but does this sense of the CMS being...well, at least somewhat ineffective, make it a deal-BREAKER ? Dunno.

Originally Posted by KTTV Images

Tom,
One set of controls that no one has brought into the picture are the 6 controls ( Gain and Offset ) for adjusting the gray scale to D65. Yes, I know juggling these will destroy the gray scale over the full 0 to 100 IRE input range but it seems to me there is one possible new strategy here no one has tried.

First let me state an assumption I found to be true on 2 different Sony SXRD Rear projectors I've had: It is possible to achieve a correct gray scale of D65 with more than one set of settings of these 6 controls. For example- suppose you found good settings of the 6 controls for D65. Now if you reduce the settings of, say, the red Offset (Dark part) by 3 clicks , you will find that reducing the Offset of the other 2 colors in a similar way will bring the darker portion of the gray scale back into the D65 range. You could do the same with the 3 Gain (Bright part) controls for the brighter portion of the gray scale. For various reasons I tried this on the Sony SXRD's and it worked.

Now, what has happened if we make the changes described above is that we have changed the ABSOLUTE levels of the operating points for Red, Green and Blue. IN electronic analog terms the normal operating points for all 3 primary colors is now at a lower voltage than it was before. Now, this is the point .. if there is a constraint on the Green primary Saturation control ... perhaps by shifting the operating point downward the Green saturation control will begin to function .. at least a little bit. There is no good reason to believe it will.. the idea is to simply try it out and see what happens.. we are desperate here.

I do not predict this will in fact help .. it is just that in an exhaustive attempt to get the Green CMS Saturation control to work it is easy to try and may be worth a wild shot.

What do you think?

Originally Posted by SOWK

What would you give up, except better black levels? And yes I know this can make a big difference.

The Samsung I think would be just as good/superior in most every other area of an image quality standpoint... no?

And before anyone posts, SOWK is bashing JVC again, I'm not bashing, just honestly asking.

Originally Posted by Hughman

Kris,

Darin has commented a couple times since the RS20 comparison that you've tweaked your RS2's color gamut and it looks very good, are you using an outboard processor or have you discovered something worth sharing?

Originally Posted by Kris Deering

...I've found a very nice settling point with my projector. Saturation is far better than it was before but it took its toll on overall luminance. But looking at the image and not the computer monitor with the readings results in a picture that is more the livable, it is gorgeous. I wish I could dial it in a bit better in terms of overall accuracy but I've found a great place to settle on that works for me and I've yet to have a complaint from vistors; videophiles or not.

Originally Posted by lovingdvd

Aside from the color gamut which I haven't attempted to tackle yet, I was able to get very good results from a calibration I did last night.

In summary I was able to dial in D65 with a dE of less than 3 for the entire range of 5-100. Most of the dE readings were about 1.5, but I could have had most of these under 1.0. However doing so would have left 100% with a dE of 3.6. So instead I decided to sacrifice a tad (moving the dE from 0.5 to 1.5ish) in most of the range to get the dE at 100 under 3.0 (OTTOMH I think it was about 2.6).

At first I dialed in a flat grayscale from 20-100 (all with a dE under 3) just by using the global gains/offsets. This went pretty quickly. However this left 5-15% with high dE. For instance I think 5% was over 12 dE and 10% was 6ish.

So I set out to use the RGB controls in the custom gamma to flatten things out. This turned out to be a major undertaking which I spent many hours on. Ultimately though it was worth it to get 5-100% all under 3 dE.

One slight downside is that my gamma dropped a bit from 2.2 to 2.17. I ran out of time to work on correcting it. I understand from GlenC that with his RS2, if you touch the White point it then throws the grayscale off. So this may cause me to have to rebalance everything. Sigh. Not sure if that will be worth it, as this calibration was pretty grueling. You can read about some of the challenges I hit by scanning back in this thread.

I am very pleased to report, and I'm sure many of you will be happy to read, that I did measure on/off CR at 45,685:1! This is at mid throw (maybe a bit beyond mid throw - about 15' from a 106" diag) with the iris fully closed (-15) and with high lamp. It yielded 305 lumens post calibration in this mode. This equates to about 12fL on my 1.3 gain Firehawk which is PLENTY bright. As the bulb ages I will slowly open up the iris to maintain this same brightness level.

I believe this measured CR comes in at the higher end based on what cine4home published with his pre-production unit. In particular from max throw he reported 48,100:1 with high lamp D65 and iris -15 with 230 lumens. I say "higher end" because even though this measurement was less than 48,100:1 it was done at mid throw. I would anticipate that at max throw that his number is based on it would surpass that number. Supposedly there is a fair amount of variability in CR from unit to unit, so YMMV in either direction.

I have confidence in my measurements. I am using a Spyder II meter trained to an i1 Pro. I have used this combination over the years and found it very effective because it combines the color accuracy of the i1 with the low level light reading abilities (and ability to read directly in front of the pj) of the Spyder. As a frame of reference I measured about 14,000:1 with my RS1 using this same equipment with the same exact setup (same throw etc). So getting 3x the contrast is certainly a nice improvement!

After working with the unit I get the feeling that calibration settings may not carry over so well from pj to pj. Nonetheless I will post the values in a few days once I have time to double check everything and see if any other tweaks are needed.

I can tell you that the OOTB grayscale in the default 6500K mode has too much green and comes in at about 5.5 dE across the scale, at least on my unit. It is pretty clear when viewing that its too green. This was a bit surprising because the RS1 (and presumably the RS2) had near perfect grayscale OOTB (at least as far as factory configured grayscales go).

Originally Posted by Kris Deering

I've found a very nice settling point with my projector. Saturation is far better than it was before but it took its toll on overall luminance. But looking at the image and not the computer monitor with the readings results in a picture that is more the livable, it is gorgeous. I wish I could dial it in a bit better in terms of overall accuracy but I've found a great place to settle on that works for me and I've yet to have a complaint from vistors; videophiles or not.

Originally Posted by TomHuffman

Just as I did. Please read my post again. You can show a nearly perfect gamut on a CIE chart, and yet have grossly inaccurate color for precisely the reason you mention.

To answer that question all you have to do is look at the LSH values. Those need to be no more than +- 2%. They are not even close.

There is nothing wrong with the theory. The problem is in what's possible given the tools you have to work with. Since the green saturation control doesn't work AT ALL, the only thing you can do for green is to lower the main Color control, which lowers both saturation and brightness for all colors. In theory if you can then raise green brightness back to where it should be without simultaneously raising the saturation again, then this would work. But I assumed that you'd already tried that and found either that the green brightness control either doesn't have sufficient range of adjustment OR that it also raises saturation as well leaving you right back where you started. IF using the Color control alone to reduce saturation and then using the hue/brightness controls to individually adjust each color's brightness and hue, then it is possible. Do the controls allow you to do this?

It is a simple proposition. Can you get the green xy where it should be and the brightness close to 71% of reference white both at the same time? Your most recent set of numbers had green at just over 40% of reference white and it was still considerably oversaturated and the hue was screwed up as well.

Originally Posted by millerwill

Well, in reading this thread I'm glad I chose to get Jason's 'calibration service'; I hope he had the magic hand. My RS20 is scheduled to arrive Tuesday, so I'll find out soon.

I have recently gotten a I1LT, though, and want to start learning about how to do gray scale, cms, etc., but it seems clear that the RS20 is not the easiest unit to start with. It will be nice to have Jason's results in User1, and then I copy them into User2 and start screwing around from there. At some point down the line I presume that some of you 'old hands' will post your settings so that us novices can use them as a starting point. Merry Christmas to us all.

Originally Posted by lovingdvd

I will post the values in a few days once I have time to double check everything and see if any other tweaks are needed.

Originally Posted by Manni01

I find it quite worrying that you seem to imply that someone like Ekkehart would post a perfect gamut not realising that the Y values are completely wrong.