LightSpace 3D LUT Home Cinema Calibration Software - Page 65 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 1103Likes
Reply
 
Thread Tools
post #1921 of 2391 Old 02-12-2019, 01:50 PM
AVS Forum Addicted Member
 
D-Nice's Avatar
 
Join Date: Dec 2004
Location: Columbia, SC
Posts: 16,835
Mentioned: 123 Post(s)
Tagged: 1 Thread(s)
Quoted: 1872 Post(s)
Liked: 3685
Quote:
Originally Posted by Manni01 View Post
You are entirely missing the point. At least in my case, the issue is 100% fixed by Lightspace with the LUT concatenation procedure (according to Steve, I only got the resulting LUT and a link to a webpage when I asked how it had been produced). So Lightspace is apparently 100% able to deliver excellent results with the JVCs (at least my unit, which is significantly undersaturated). I have zero complaint about the LUT that Steve sent me, except that I can't reproduce that result easily.

The problem is that this procedure is so convoluted that I haven't even attempted it, and others are not getting as good results as when Steve is doing it himself. So we're only asking for an option that would make the LUT concatenation procedure less convoluted and accessible to experienced calibrators who nevertheless don't want to spend days generating a single LUT through some kind of dark arts guessing work. Something that would automatically calculate optimal targets for the custom gamut and automatically concatenates the two LUTs would be a very good start.

And no, the map space option isn't that. At least it doesn't work in my situation. I did try that (as well as all the other LUT options).

We're not asking for the impossible. We're asking for usable solutions instead of convoluted workarounds. One workaround might be fine. Three in the same workflow isn't acceptable (to me). And that's what I'm asked to do as a MadVR/JVC user.

By the way, despite what Light Illusion would like to suggest, there is nothing wrong with the JVCs. With a good LUT, for example with the one created by Steve himself using the LUT concatenation, the results are stellar. No one would grade content with it, but that's not the purpose of the unit. And even without a LUT, the results are excellent, although clearly not as good. So it's just an excuse to *not* improve the software.

To answer your question (again, this has already been stated twice or more), an "all settings left at default" isn't possible because JVC doesn't allow a mode that has no processing and uses the P3 filter. So if you disable processing, you also disable the P3 filter, you get even less color volume, and the results are even worse. Using the autocal doesn't make any difference given the way it works, and if it does it would be a positive one (if used properly, like @bobof and @Dominic Chan are perfectly able to), not a negative one.

I'm not pointing fingers. I've spent weeks last summer trying to get good results from LS V9. The only good results (again, excellent, absolutely zero complaint with that LUT, I'm still using it today) were provided by Steve, apparently using a procedure so convoluted that I am not willing to use it given the way its documented, despite the fact that I'm an experienced calibrator and fairly technical. Even @bobof , who is an engineer and has tried, found that his head was hurting and didn't get fully satisfying results.
I don’t think I quote your post(s) with my post so how and/or why did you conclude I was missing your point I’m attempting to troubleshoot with bobof, not you.
D-Nice is offline  
Sponsored Links
Advertisement
 
post #1922 of 2391 Old 02-12-2019, 01:54 PM
AVS Forum Addicted Member
 
D-Nice's Avatar
 
Join Date: Dec 2004
Location: Columbia, SC
Posts: 16,835
Mentioned: 123 Post(s)
Tagged: 1 Thread(s)
Quoted: 1872 Post(s)
Liked: 3685
Quote:
Originally Posted by bobof View Post
Thanks, and I take all your points on board; as I say I'm an engineer and they are all thoughts which cross my mind. As you saw in my message I fully intend on doing that "factory condition" profile. That a few folk are having the same issues makes me think it won't be fruitful (unless we've all done the gamma correction, and it is just that that is faulty). But I already know that even without the gamma calibrated all these modes have poor RGB separation measured. I really don't think backing out the gamma cal is going to do anything. But it is worth testing, which is why I mention it.

For simplicity sake the gamma cal is the only one I did, this is why if you look at the profile I have a fairly substantially non-D65 white point - I didn't even want to adjust that "just in case"...

In any case the most interesting point here really for me is that there are alternate LUT softwares that seem to work out this profile fine and generate a profile whose actual shape is the same; the only real difference is the lack of the extraneous strange points (+1 ArgyllCMS!) and the better internal quality of the LUT excluding these points (+1 Lightspace!). As @Manni01 says, it looks like there is a process on the website that (almost?) resolves this; so then we're looking at different software, one which outputs broken (not poor quality, actually broken) LUTs based on certain inputs, and one that generates quite useable ones. It's hard to make an argument that given the purpose of the software that this isn't something that could be improved upon.

Anyway, I'm approaching this completely open minded and prepared to be "wrong" at every step. I am not a calibration "expert", but am more than capable of doing it. Honestly, the easy solution for me at the moment would be to say "can't be bothered, from now on I won't use it, and I won't recommend LS to other JVC owning friends". As it is I've recommended it to several friends in recent times for exactly this use and plugged it a bit where folks were looking for LUT solutions - and so now I'm actually feeling mildly embarrassed that it doesn't seem to work well in this regard which is important to me personally and I know some of them wanted to use it for this exact use case. I'm going to try and profile a friend's unit to see if his behaves differently (it's a fairly different model to mine, same manufacturer...).

For these reasons I'm dead keen to find solution. I'm not trying to apportion blame but I can see the facts of the results I have to date, at the end of the day I do need to get to a working solution on this for me or it has outlived its usefulness in this regard.
Let me know how a “stock” LS LUT goes as I’m very curious to see the output.
bobof likes this.
D-Nice is offline  
post #1923 of 2391 Old 02-12-2019, 02:01 PM
AVS Forum Special Member
 
Manni01's Avatar
 
Join Date: Sep 2008
Posts: 9,108
Mentioned: 328 Post(s)
Tagged: 0 Thread(s)
Quoted: 5438 Post(s)
Liked: 5636
Quote:
Originally Posted by D-Nice View Post
I don’t think I quote your post(s) with my post so how and/or why did you conclude I was missing your point I’m attempting to troubleshoot with bobof, not you.
Only because you talked about pointing fingers and sounded like you hadn't read half the posts. Bobof's issue is very much similar to mine, he just hasn't been through all the steps yet. Apologies if that's not the case, and looking forward to the outcome.

Batch Utility V4.02 May 16 2019 to automate measurements files for madVR with support for BD Folders
Manni01 is online now  
Sponsored Links
Advertisement
 
post #1924 of 2391 Old 02-12-2019, 02:07 PM
AVS Forum Addicted Member
 
D-Nice's Avatar
 
Join Date: Dec 2004
Location: Columbia, SC
Posts: 16,835
Mentioned: 123 Post(s)
Tagged: 1 Thread(s)
Quoted: 1872 Post(s)
Liked: 3685
Quote:
Originally Posted by Manni01 View Post
Only because you talked about pointing fingers and sounded like you hadn't read half the posts. Bobof's issue is very much similar to mine, he just hasn't been through all the steps yet. Apologies if that's not the case, and looking forward to the outcome.
I’ve read every simgle post before I did my post. Whether or not your issue is similar to bobof is irrelevant to me. If I had any interests in your specific issues, I would have quoted your post(s).
D-Nice is offline  
post #1925 of 2391 Old 02-12-2019, 02:24 PM
AVS Forum Special Member
 
Manni01's Avatar
 
Join Date: Sep 2008
Posts: 9,108
Mentioned: 328 Post(s)
Tagged: 0 Thread(s)
Quoted: 5438 Post(s)
Liked: 5636
Quote:
Originally Posted by D-Nice View Post
I’ve read every simgle post before I did my post. Whether or not your issue is similar to bobof is irrelevant to me. If I had any interests in your specific issues, I would have quoted your post(s).
FYI, quoting someone doesn't mean that only the person quoted can contribute to the discussion, or even that your post is only meant for that person.

It only means that you are referring to a specific post in an open discussion in a public forum. Every single person reading your post is free to contribute if they wish to.

If it's a one on one discussion you are after, please remember that you can use PMs.

Out of respect for @bobof , I will stay out of your exclusive discussion with him though.

And when I said that I was interested in the outcome of that discussion in the post in which I apologized to you, I meant it. I wasn't being sarcastic.
omarank likes this.

Batch Utility V4.02 May 16 2019 to automate measurements files for madVR with support for BD Folders
Manni01 is online now  
post #1926 of 2391 Old 02-12-2019, 05:33 PM
Member
 
Htnut2000's Avatar
 
Join Date: May 2004
Location: Australia
Posts: 133
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 44 Post(s)
Liked: 85
I would like to add my voice to that of @mani 01 and @bobo v

We have the following in common:

1. We all have a JVC projector, mine is an X9000
2. We all are having issues with getting a good LUT with the new engine other than by using fit space
3. We have all been variously told by Steve that it is our fault, the display's fault, a faulty meter or go read the manual

I had extensive email exchange with Steve in 2016 about the fact that the new light engine produced worse results than previous versions and his response was there will be a new fit space mode.

I then gave up with Lightspace and used Calman with lightening LUT which produced great results quickly and without any issues.

About 6 months ago I converted to 4K using a Lumagen pro and was forced back to Lightspace as Calman would not work with HDR

Again Steve recommend using Peak Chroma but the results were unusable so I am back to Fit Space.

I have been calibrating for over 15 years with software from 4 different vendors. Years ago I would be happy to keep tweaking to get a good result.

I am at a stage of life where I just want a system that would be quick, easy to use and produce good results so that I can go back to the important part, namely watching movies. This should be easy since we have 3D LUT's.

I have invested in a Lumagen Pro and a CR100 meter in order to make the process as simple and quick as possible, but clearly from what I have gleaned from this thread that this is not currently possible with Lightspace if one has a JVC projector.

Steve keeps blaming the display instead of looking into why his software can't cope with a non perfect display.

This attitude is ridiculous! If our displays were perfect we would not have to calibrate them! The aim of calibration has always been to extract the best performance from real world displays.

End rant!
Htnut2000 is offline  
post #1927 of 2391 Old 02-12-2019, 05:50 PM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,833
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 779 Post(s)
Liked: 1144
Quote:
Originally Posted by Dominic Chan View Post
I thought LighSpace targets more than just the post-production industry, especially with those “Home Cinema Options”.
I said 'we' (Light Illusion), not LightSpace...
We work in the professional film and TV industry.
We have never done 'home calibrations'.

What LightSpace does makes no difference in whatever industry.

But I do echo your comments.
The JVC is appalling, and I would be very, very upset with them for selling a display that is not fit for purpose.

(I will have a look at the Sony profile when I get a chance.)

Steve

Steve Shaw
LIGHT ILLUSION


Last edited by Light Illusion; 02-12-2019 at 06:12 PM.
Light Illusion is offline  
post #1928 of 2391 Old 02-12-2019, 06:10 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,945
Mentioned: 130 Post(s)
Tagged: 0 Thread(s)
Quoted: 4883 Post(s)
Liked: 1847
Quote:
Originally Posted by Htnut2000 View Post
Steve keeps blaming the display instead of looking into why his software can't cope with a non perfect display.

This attitude is ridiculous! If our displays were perfect we would not have to calibrate them! The aim of calibration has always been to extract the best performance from real world displays.

End rant!
Too soon to end. See the post immediately following yours
Manni01 likes this.
Dominic Chan is online now  
post #1929 of 2391 Old 02-12-2019, 11:17 PM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,267
Mentioned: 53 Post(s)
Tagged: 0 Thread(s)
Quoted: 2640 Post(s)
Liked: 1738
Quote:
Originally Posted by Manni01 View Post
Bobof's issue is very much similar to mine, he just hasn't been through all the steps yet.
Indeed, if you don't mind, did you see my mention of you above? There are a couple of questions I wanted to ask you / hoped you might have your profile etc to share so I can see where the differences in our behaviour with seemingly the same process come from. Hoping that with more data I can chip away at understanding how it is all working or not; that our results are similar but different is interesting as we're clearly somewhere on a continuum perhaps from very broke, to a bit broke, to eventually not broke at all... . Cheers

Copied here:

Quote:
Originally Posted by bobof View Post
@Manni01 ; on a technical point / clarification; I am a little intrigued about the fact that my result is a bit worse than yours - by that I mean that the custom gamut that Steve made worked for you (assuming he did the exact web process), but my gamut needs further reducing to work. I'm trying to pick apart all the differences because if Steve followed the exact process as described on the website we ended up with slightly different results. It could be variation difference between my unit and your unit, or difference between the settings / start condition we created.

1) Does the LUT you have have any stray / funny points visible? Or does it look really nice everywhere? Any chance you could share the LUT and the originating Lightspace profile? I can actually go through the web process and see that I get to the same place as you.

2) Is the LUT you have based on the DCI-P3F colourspace that you created? Or did you use one of the other colourspaces? I based mine on the Reference colourspace as that was the one with the least RGB separation of the internals, but I wonder if in using the custom DCI-P3F colourspace you've avoided some of the issue affecting my profile because it has just nipped the gamut in a bit tighter, or perhaps Reference has it's own issues inplaces in the gamut. I will try DCI-P3F and BT2020 profiles also here when I have some time..

3) Out of interest; did you ever try to reverse-engineer the JVC CMS's idea of what its own native gamut is that it is using for it custom gamut remapping? By this I mean there are co-ordinates you can put into the custom gamut creator that make a smaller gamut than native, and co-ordinates that result in a larger gamut being mapped down (like REC2020). Is it possible to make (almost) the equivalent of "profile off" with a filter by setting the gamut extents to being the same as the JVCs native gamut (as it understands it?). I wonder if there is a common set of co-ordinates that would work for all units, or whether they have some unique values loaded in the factory. Even then it might not be linear.
Manni01 likes this.
bobof is online now  
post #1930 of 2391 Old 02-13-2019, 12:50 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,267
Mentioned: 53 Post(s)
Tagged: 0 Thread(s)
Quoted: 2640 Post(s)
Liked: 1738
Ok folks, I'm going to take a little break from posting on this issue and just beaver away at it in the background; the back and forth is eating into the day when I should be doing work I can actually charge someone for! . That isn't a complaint by the way, it is just my own situation.
I've got quite a lot of things I want to try and look at to understand what is going on here, and a few theories; blow by blow on it isn't going to be useful. (hopefully @Manni01 can either reply here or via PM with the info discussed above), as at the moment each post generates a bit of a flurry.

I guess we can all see at least as common ground that:
1) there are clearly quite a few of us with these PJs trying to do this (for right or wrong) and
2) we don't (all) have a process that works for us with this, allowing us to use the greater accuracy of LS within the colour volume

Thanks for your note @Htnut2000 , I count a least 4 of us with these kind of PJs who are struggling to get "sensible" results that work for us.
If anyone has "bad" profiles, please link me to them so I can see if the web process actually would fix it for you or not.

It sounds like @Manni01 does at least have a process that might work for him (if it wasn't for the fact that it is a bit of a pain to do, and the history which I'm sympathetic to). I agree that the description within the Advanced section on the website is a bit hard to follow. As far as possible I tried to follow it and it didn't work for me, and in the scenario I'm in (with the worst profile Steve has ever seen! ) I did manage to make it work but it was very labour intensive and I'm sure most folk wouldn't twig that progressively reducing the intermediate gamut until the errant points disappear in the LUT is what you need to do.

I don't doubt there is something suboptimal about the measured gamut so I do treat this as a kind of interoperability issue and I just think it needs approaching in good faith, and I'm sure everyone is still in this case. I'm sure all of us ultimately want our products to be as good as possible within the constraints of commercial and actual realities.

Anyway, as I say I'll get on with my testing offline and feedback once I have something hopefully useful, at least in the way of a workable step by step guide that can get folk up and running with wider colour gamut non-matrix LUTs on these projectors.

For reference; I'm probably going to look in roughly this order:
1) experiment with a custom targets gamut reduction in the projector to see if it can be made to behave better. I have a feeling Manni's DCIP3F is smaller measured than Reference or more regular towards the edges, and the reason his appears to work better - I think setting the custom gamut to be closer to the projector native gamut measurements might result in a more linear response towards the edged of the input values and hence better still. Perhaps.
2) perhaps look at the different built-in gamuts (at least one test without gamma autocal to eliminate that) and see how they vary in their response to the web process as published
3) Depending on the results of the above, try and work out how LUT concatenation process doesn't work as described and improve it.

For what it is worth I reverted my LS installation to the last pre-V3 version, and yes, the LUTs are very different, but I certainly wouldn't say from looking at them that they were what I was expecting (and Steve suggested they would be suboptimal in other ways) so I'm not sure how much mileage there is in that line of enquiry. Lets' move forwards, not backwards, given the new LUT engine is significantly better for most displays.
Manni01 likes this.
bobof is online now  
post #1931 of 2391 Old 02-13-2019, 01:47 AM
AVS Forum Special Member
 
Manni01's Avatar
 
Join Date: Sep 2008
Posts: 9,108
Mentioned: 328 Post(s)
Tagged: 0 Thread(s)
Quoted: 5438 Post(s)
Liked: 5636
Quote:
Originally Posted by bobof View Post
Indeed, if you don't mind, did you see my mention of you above? There are a couple of questions I wanted to ask you / hoped you might have your profile etc to share so I can see where the differences in our behaviour with seemingly the same process come from. Hoping that with more data I can chip away at understanding how it is all working or not; that our results are similar but different is interesting as we're clearly somewhere on a continuum perhaps from very broke, to a bit broke, to eventually not broke at all... . Cheers

Copied here:
Hi @bobof ,

I saw that post and replied to it, but it looks like it got eaten in the ether.

I spent too much time last summer with this, I simply don't have the time to revisit this between my day job and my new rs2000 that has arrived and on which I'd rather concentrate.

I can tell you though that I was indeed using DCI-P3F with the RS500.

I'll post my feedback and measurements on this new unit in the JVC 2019 calibration thread, and when I get to create 3DLUTs with LS (which I will attempt *before* even running the JVC Autocal once, to rule out once and for all the idea that it could contribute to our issue), I'll report my results here, good or bad. I hope they will be better as the gamut cover is closer to target. In my experience LS produces excellent results when the display can reach the target color volume. My rec-709 baseline with the P3 filter reaches 100% of rec-709 color volume, and my LS 3D LUTs for that calibration (peak chroma) are flawless, without any need for concatenation.

The further away from the target color volume, the worse the results in my experience. In the case of the rs500 with 3000 hours on the clock, I had no choice but in the case of this brand new rs2000 I should have more choices. As I said, I'll post measurements and tips for JVC owners of the new models shortly.

One last point, which I think is important to make: I don't think this is a JVC issue at all. We do happen to be more technical users, and especially you, @Dominic Chan and I are happy to do everything we can to eek the last ounce of performance out of them. We're not newbies, we're not technophobes, and we know our units, their limits and their capabilities inside out.

Sony projectors don't have the tools that JVCs projectors offer, and have even less gamut volume as they don't have a P3 filter (except the VW1x00ES line). So as far as the best consumer projectors are concerned, they are what they are. Even yours or mine are not unfit for purpose and they produce, especially once well-calibrated, a stellar picture, which can be achieved with LS in a very convoluted way, as well as with other calibration tools in a more straightforward way (especially if the meter used is natively supported).

Light Illusion, of their own admittance, simply have no experience of home calibration, and they see consumer displays unable to reach 100% of the target color volume as unfit for purpose. I wish this was stated clearly on their website on all the product pages, it would have saved me a lot of time, frustration, and possibly money. My understanding of using a 3D LUT is to achieve the best possible results within the limitations of the display, when not used for grading. That what I've always done, with great success, with all my JVCs.

Anyone who understands the basics of calibration knows that excellent results (and by this I don't mean only visually pleasing, I mean accurate) can be achieved from displays that only reach 90% of the target, as long as saturation and luminance track at all levels up to that point within the cube without causing visual artifacts. This is the whole idea behind using a 3D LUT. Given that there is little to no actual content towards the edge of the cube, this is not an issue with more than 90% of the actual content. Well, unless the 3D LUT produces visible artifact at the limit of the native gamut, such as posterization. This is the reason why UHD Premium only requires 90% of P3 cover. They know that most consumer displays would not qualify otherwise, and filmmakers don't mind because they know there is little to no content beyond 90% anyway. As long as 90% and below tracks properly, this limitation in the display is moot with 99% of the actual content, again with a good calibration. Heck, the actual target of the container (and of more and more of the actual content) is BT2020, and if we reach 70% of that we're lucky! I know this, you know this, @Dominic Chan knows this, many others here know this and even Light Illusion knows this.

Bashing the manufacturer because they don't want to face their own responsibility is just one more attempt to pass the bucket.

I'm not sure why they entered the home cinema calibration market with this attitude, but it explains a lot.

I think we have voiced our position and made our point. I'll contribute later if I can when I have new information, I hope that Light Illusion will have a more positive attitude towards users who are only trying to do in a less convoluted way something that's entirely legitimate. Until then, I'm out too.
omarank likes this.

Batch Utility V4.02 May 16 2019 to automate measurements files for madVR with support for BD Folders

Last edited by Manni01; 02-13-2019 at 08:15 AM.
Manni01 is online now  
post #1932 of 2391 Old 02-13-2019, 02:08 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,267
Mentioned: 53 Post(s)
Tagged: 0 Thread(s)
Quoted: 2640 Post(s)
Liked: 1738
Quote:
Originally Posted by Manni01 View Post
Hi @bobof ,

I saw that post and replied to it, but it looks like it got eaten in the ether.

I spent too much time last summer with this, I simply don't have the time to revesit this between my day job and my new rs2000 that has arrived and on which I'd rather concentrate.

I can tell you though that I was indeed using DCI-P3F with the RS500.
Thanks @Manni01 , and I completely get where you are coming from, your points are very clear. I'm not asking you to spend any time on this (unless finding the originating profile to your LUT is going to be a mamoth task, might be if your profile filing system is as elegant as mine... ). If you don't have easy access to it or prefer not to share it that is fine, it would just help me out as it gives me a little steer without doing that particular test now knowing that my results are reference profile, yours are DCI-P3F, and yours are (perhaps) a little less broken than mine.

Anyway, enjoy the new toy. Look forward to reading about your opinions on it.
Manni01 likes this.
bobof is online now  
post #1933 of 2391 Old 02-13-2019, 06:10 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,833
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 779 Post(s)
Liked: 1144
Quote:
Originally Posted by Light Illusion View Post
(I will have a look at the Sony profile when I get a chance.)
Had a look.

The display looks to have had its Saturation cranked up to 11 (or more).

The internal colour volume is pushing outwards, causing multiple internal input volumetric colours to all point to the same output colour on the gamut edge.

Sound familiar?

I suspect this may be the key to the issues with the JVC too, although only on one axis.

LightSpace has never been designed for such a display configuration, as it is just wrong - in a very, very bad way.

No display saturation control should ever be set to expand internal colours outwards, beyond the display's max gamut edge.

If in doubt, crank the saturation back until the overall gamut 'just' starts to shrink, and keep it there.

Steve
bobof and Anger.miki like this.

Steve Shaw
LIGHT ILLUSION


Last edited by Light Illusion; 02-13-2019 at 06:20 AM.
Light Illusion is offline  
post #1934 of 2391 Old 02-13-2019, 07:57 AM
Member
 
Join Date: Mar 2016
Posts: 56
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 36 Post(s)
Liked: 30
Quote:
Originally Posted by Light Illusion View Post
Had a look.

The display looks to have had its Saturation cranked up to 11 (or more).

The internal colour volume is pushing outwards, causing multiple internal input volumetric colours to all point to the same output colour on the gamut edge.

.....

If in doubt, crank the saturation back until the overall gamut 'just' starts to shrink, and keep it there.
FWIW, all the processing in the TV was turned Off and except for backlight, picture, brightness, color temperature, gamma and white balance controls, all other controls were set to 0 or neutral setting. Color or saturation was not cranked up and Live Color was Off too. Can't say if the TV was internally doing that despite the neutral setting.

During our email conversations, it was told to me that this display has a rather small gamut. Ted suggested to try Live Color setting as Low and also the Standard mode as he found in some review website that it gave the largest gamut for my TV. I tried different Live Color settings and different modes including the Standard mode, and ultimately it was suggested to me to not use the Live Color as it was boosting colors. Non-linearity could be seen with Live Color in Saturation Sweeps. But with Live Color Off, the saturation sweeps were clean and there was no pushing of colors. So, for the profile which you analyzed, the saturation sweeps did not show any kind of outward color expansion. Can you please show some graph/ visual representation through which you observed this color expansion?
Manni01 likes this.

Last edited by omarank; 02-13-2019 at 08:17 AM.
omarank is online now  
post #1935 of 2391 Old 02-13-2019, 07:17 PM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,833
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 779 Post(s)
Liked: 1144
I'd you run a saturation sweep, with a high number of patchs in both primary and secondary (I think we have example .CSV sequences in the Download options?) You will see the issue.

As you move further away from the grey scale you will see the measure points becoming more saturated, and near the gamut border you will see saturation clipping.

It's basically the same as clipping in the blacks/whites, but with saturation.

The default saturation sweeps don't have enough granularity.

With the new ColourSpace graphs (which I used) it's very easy to see the issues, as every point has tangent error lines.

Steve

Steve Shaw
LIGHT ILLUSION


Last edited by Light Illusion; 02-13-2019 at 08:41 PM.
Light Illusion is offline  
post #1936 of 2391 Old 02-14-2019, 01:42 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,376
Mentioned: 207 Post(s)
Tagged: 1 Thread(s)
Quoted: 3249 Post(s)
Liked: 4043
LightSpace CMS 10.0.0.2892 (12 February 2018) Beta is available.

Release Notes

Large code re-write to maintain as much future ColourSpace compatibility as possible.

Added Remote Control and Secondary Execution capabilities.

Updated madVR dll.

Download

As is usual, for that beta download link, please contact me here only (no PM's): http://www.displaycalibrations.com/contact_us.html

Release Info

The Remote Control and Secondary Execution capabilities are very powerful tools - and those with coding skills will really understand.

The Remote Control allows external programs to control LightSpace, and for LightSpace to send data to these programs.

It means that can written calibration tools based on using LightSpace for measuring and patch generation, etc. ...so users can develop their own programs, that use LightSpace to do profiling, etc...

For example for performing DDC AutoCAL.

The Secondary Execution enables LightSpace data to be used in an external program.

About users which are interested in the Remote Control & Secondary Execution capabilities, they will receive PDF's as well an example .bat file and the info on that.

Known Issue

There is a scalling issue for users which are using scalling in Windows (150% for example) due to serious changes made to keep the ColorSpace compatibility. This will be fixed ASAP.
Manni01 and sat4all like this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #1937 of 2391 Old 02-14-2019, 02:10 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,267
Mentioned: 53 Post(s)
Tagged: 0 Thread(s)
Quoted: 2640 Post(s)
Liked: 1738
Quote:
Originally Posted by Light Illusion View Post
I'd you run a saturation sweep, with a high number of patchs in both primary and secondary (I think we have example .CSV sequences in the Download options?) You will see the issue.

As you move further away from the grey scale you will see the measure points becoming more saturated, and near the gamut border you will see saturation clipping.

It's basically the same as clipping in the blacks/whites, but with saturation.

The default saturation sweeps don't have enough granularity.

With the new ColourSpace graphs (which I used) it's very easy to see the issues, as every point has tangent error lines.

Steve
Be interesting to know if it helps @omarank 's issue.

For what it is worth Steve I did have a quick go at this on the JVC (have a lot more results to share but have a few more tests to do in the next couple of evenings). In the case of the JVC controls; while cranking down the colour massively can get you eventually to the point where the LUT builds without the region with "crazy points" becoming a problem, it then also results in a gross undersaturation compared to the actual maximum gamut of the display with the colour control at 0. So it isn't a useful option.

I'll feed back to you when I have all my data collected. I have been documenting a couple of things in one of the JVC threads but it won't be that useful to you until I add a few pictures etc.
omarank likes this.
bobof is online now  
post #1938 of 2391 Old 02-14-2019, 05:55 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,833
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 779 Post(s)
Liked: 1144
That mirrors Jamie's findings, where he made a target colour space with lower green gamut.

His approach is better as you can individually alter the separate RGB gamut targets.

Reducing Saturation has the same effect everywhere, so will make unnecessary gamut reductions.

When making a new target colour space you need to track the RGB primaries from white through the peak RGB values, and select a reduced target on the same axis.

But, all this follows our findings too.

The JVC projector basically has a fake gamut, attempting to show a wider colour range than it can truthfully make.

Steve
bobof likes this.

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #1939 of 2391 Old 02-14-2019, 06:16 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,267
Mentioned: 53 Post(s)
Tagged: 0 Thread(s)
Quoted: 2640 Post(s)
Liked: 1738
Quote:
Originally Posted by Light Illusion View Post
That mirrors Jamie's findings, where he made a target colour space with lower green gamut.

His approach is better as you can individually alter the separate RGB gamut targets.

Reducing Saturation has the same effect everywhere, so will make unnecessary gamut reductions.

When making a new target colour space you need to track the RGB primaries from white through the peak RGB values, and select a reduced target on the same axis.

But, all this follows our findings too.

The JVC projector basically has a fake gamut, attempting to show a wider colour range than it can truthfully make.
Steve, it is me Jamie (well, James, but we'll let it pass )

I think the messages have got a little confused. It was me just put the message above to say I'd tried reducing the colour also as it seemed like an interesting suggestion, and it didn't work really work as the resulting gamut is massively reduced, whereas the custom (a bit reduced) intermediate gamut in the concatenation process does seem to "work" with a bit reduced gamut. But anyway, as I say I'll feed back some more results of my testing later.

For what it is worth, I worked out that you can it seems show the issue easily enough with just a very quick and dirty 5^3 profile out of the i1d3 (about 4 minutes profiling) which I think might interest you as there isn't much of it at all... Anyway, will explain later when I get some time.
bobof is online now  
post #1940 of 2391 Old 02-14-2019, 06:12 PM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,833
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 779 Post(s)
Liked: 1144
Ooops!
Sorry James

Steve
bobof likes this.

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #1941 of 2391 Old 02-15-2019, 01:37 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,376
Mentioned: 207 Post(s)
Tagged: 1 Thread(s)
Quoted: 3249 Post(s)
Liked: 4043
Quote:
Originally Posted by ConnecTEDDD View Post
LightSpace CMS 10.0.0.2892 (12 February 2018) Beta is available.

Known Issue

There is a scalling issue for users which are using scalling in Windows (150% for example) due to serious changes made to keep the ColorSpace compatibility. This will be fixed ASAP.
LightSpace CMS 10.0.0.2897 (14 February 2018) Beta is available, where scaling issues have been resolved.

Download

As is usual, for that beta download link, please contact me here only (no PM's): http://www.displaycalibrations.com/contact_us.html
markrubin, Manni01 and Anger.miki like this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #1942 of 2391 Old 02-16-2019, 01:25 AM
Member
 
Join Date: Mar 2016
Posts: 56
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 36 Post(s)
Liked: 30
Quote:
Originally Posted by Light Illusion View Post
I'd you run a saturation sweep, with a high number of patchs in both primary and secondary (I think we have example .CSV sequences in the Download options?) You will see the issue.

As you move further away from the grey scale you will see the measure points becoming more saturated, and near the gamut border you will see saturation clipping.

It's basically the same as clipping in the blacks/whites, but with saturation.

The default saturation sweeps don't have enough granularity.

With the new ColourSpace graphs (which I used) it's very easy to see the issues, as every point has tangent error lines.

Steve
Ok, I think for such bad displays (like the Sony TV and JVC Projectors), it would be great if some processing could be applied on Map Space LUTs (or concatenated LUTs) which clears out the stray points or smooths out the LUT around the edges and corners so that the resulting LUT does not product visible artifacts. It's not surprising for consumer displays to do nasty things. I believe LS should be able to create usable LUTs for such bad cases.
Manni01 and KarlKlammer like this.
omarank is online now  
post #1943 of 2391 Old 02-16-2019, 03:56 AM
AVS Forum Special Member
 
Manni01's Avatar
 
Join Date: Sep 2008
Posts: 9,108
Mentioned: 328 Post(s)
Tagged: 0 Thread(s)
Quoted: 5438 Post(s)
Liked: 5636
Quote:
Originally Posted by omarank View Post
Ok, I think for such bad displays (like the Sony TV and JVC Projectors), it would be great if some processing could be applied on Map Space LUTs (or concatenated LUTs) which clears out the stray points or smooths out the LUT around the edges and corners so that the resulting LUT does not product visible artifacts. It's not surprising for consumer displays to do nasty things. I believe LS should be able to create usable LUTs for such bad cases.
Please don't make blanket statements like this.

I don't know about the Sony TV, but the JVC projectors can be very well calibrated for home cinema users, including using Lightspace, although it's far less straightforward than with other solutions. They are not "bad displays" for their consumer audience

It's only extreme cases such as Bobof's and my 3 year old rs500 (after 3,000 hours on the clock, I had no issues when new to cover both 100% of rec-709 without the P3 filter and 100% of P3 with it) that Lightspace becomes unable to handle them without making the process even more convoluted, especially for MadVR users.

Sure, these "bad" units couldn't be used to grade content professionally, but they still throw a fantastic picture (and by this I mean accurate for more than 90% of the content they playback) when well-calibrated!

My new JVC RS2000 is very linear out of the box and covers 99% of P3 in high lamp. I haven't attempted to run a large 3D LUT yet with Lightspace, but I expect it should return excellent result, which I'll post here if I can.

Blaming the manufacturer is just FUD, based on Light Illusion's lack of experience and knowledge of both the capabilities and limitations of these consumer displays (which they have admitted already).

I agree with you though that if Light Illusion cater for home cinema enthusiasts, which is my understanding, they should make it less convoluted for users of the best consumer projectors (JVC and SONY, but all other brands would suffer similar or worse limitation with at least some units) as well as owners of consumer TVs to get good results without having to jump through unnecessary hoops.
omarank likes this.

Batch Utility V4.02 May 16 2019 to automate measurements files for madVR with support for BD Folders

Last edited by Manni01; 02-16-2019 at 04:07 AM.
Manni01 is online now  
post #1944 of 2391 Old 02-16-2019, 04:04 AM
AVS Forum Special Member
 
Manni01's Avatar
 
Join Date: Sep 2008
Posts: 9,108
Mentioned: 328 Post(s)
Tagged: 0 Thread(s)
Quoted: 5438 Post(s)
Liked: 5636
Quote:
Originally Posted by ConnecTEDDD View Post
LightSpace CMS 10.0.0.2897 (14 February 2018) Beta is available, where scaling issues have been resolved.

Download

As is usual, for that beta download link, please contact me here only (no PM's): http://www.displaycalibrations.com/contact_us.html
Thanks, I noticed these scaling issues recently, but as they are cosmetic only (if we're talking about the shadow background behind the windows only in the interface), I wasn't too bothered. Good to know they are fixed.

Batch Utility V4.02 May 16 2019 to automate measurements files for madVR with support for BD Folders
Manni01 is online now  
post #1945 of 2391 Old 02-16-2019, 04:16 AM
Member
 
Join Date: Mar 2016
Posts: 56
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 36 Post(s)
Liked: 30
Quote:
Originally Posted by Manni01 View Post
I don't know about the Sony TV, but the JVC projectors can be very well calibrated for home cinema users, including using Lightspace, although it's far less straightforward than with other solutions. They are not "bad displays" for their consumer audience
....
My new JVC RS2000 is very linear out of the box and covers 99% of P3 in high lamp. I haven't attempted to run a large 3D LUT yet with Lightspace, but I expect it should return excellent result, which I'll post here if I can.
I meant "bad" in Light Illusion's perspective . I am too able to get good results using Argyll/DisplayCAL LUT on my Sony TV. The point still remains the same though, and that is there is a scope to improve LightSpace's handling of these displays. And that's what I am requesting for.

I look forward to your calibration results of JVC RS2000.
Manni01 likes this.
omarank is online now  
post #1946 of 2391 Old 02-16-2019, 05:59 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,267
Mentioned: 53 Post(s)
Tagged: 0 Thread(s)
Quoted: 2640 Post(s)
Liked: 1738
Quote:
Originally Posted by omarank View Post
I meant "bad" in Light Illusion's perspective . I am too able to get good results using Argyll/DisplayCAL LUT on my Sony TV. The point still remains the same though, and that is there is a scope to improve LightSpace's handling of these displays. And that's what I am requesting for.

I look forward to your calibration results of JVC RS2000.
I just spent a moment looking at your profile. While it is clear other SW can do a better job of making an "actually usable" (I won't say higher quality) LUT in this mode, if you look at it with the "custom filter" option in manage colourspace, you can see very clearly where this issue is. It's not clear if you've looked to try and improve this or not.

I've spent a fair bit of time over the last couple of evenings looking at the way LS behaves. And it is quite simple - if you have multiple points (typically as you head to the gamut edge) mapped from different input points to notionally the same measured chroma+luminance the LUTs generated are bogus in the areas that these duplicate points exist. I believe, like you, that LS could perhaps do better with this. But in your case, I would be surprised if you can't get into a better mode to start with as a pre-condition for creating your LUT.

If you open your profile in Manage Colourspace you can look for the problems, and then try and fix them if possible in your display setup. As it happens, it seems us JVC owners have no way to fix these issues, which puts us in an almost irreconcilable position with LS as it stands. You might be able to do better with what you have by spending a little time understanding your display behaviour.

I had a look at your profile with a custom filter set to "R==1 && G==0" (see screen shot). You can see from that that from RGB=255,0,0 to 255,0,63 there is no change whatsoever in the measured values. I'd be surprised if you can't find a mode that doesn't have that characteristic. One reason it is worth trying to fix these issues at the display is that even if you succeed in generating a LUT with another solution it is still quite sub-optimal. Your LUT is having to map all the possible output colours into the range B 63-255 on the HDMI signal in that direction, whereas in a linear mode it could be using the whole 0-255 range. Only if you can't find a better mode (or you have some other reason which means you have to use this mode) would you then try and LUT this mode.

You can look at all the gamut clip directions using different filters, by looking at the values near the primaries. They are (at 100% luma, change the 1's to whatever you need):
Magenta -> Red "R==1 && G==0"
Yellow -> Red "R==1 && B ==0"
Yellow -> Green "G==1 && B == 0"
Cyan -> Green "G==1 && R == 0"
Cyan -> Blue "B==1 && R == 0"
Magenta -> Blue "B==1 && G == 0"

You could also look at the Cyan, Magenta & Yellow saturation sweeps instead / as well; the custom filters are (for 100% luma):
Cyan " R == B && R==1 && B==1"
Yellow " R == G && R==1 && G==1"
Magenta " B == G && B==1 && G==1"

Look at the values close to the primaries and secondaries and make sure they are changing all the way. If they clip to the same value you will have this issue.

Wherever possible you need to start with the best possible configuration. In the case of the JVCs we literally have no option - you cannot access the widest gamut of the display in a mode where output chroma changes in a linear fashion with respect to input values.
Attached Thumbnails
Click image for larger version

Name:	Screenshot 2019-02-16 at 13.22.27.png
Views:	25
Size:	809.4 KB
ID:	2526566  
omarank likes this.
bobof is online now  
post #1947 of 2391 Old 02-16-2019, 06:38 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,833
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 779 Post(s)
Liked: 1144
After all our assessments of the JVC projector profile data there are indeed serious issues with the display that really make it unfit for the application it has been sold for.

The issue is there is never any reason for internal colour gamut values to act the way they are with the JVC. Especially when set to an uncalibrated state.

No multiple input colour points should ever map to the same output colour - ever.

Even post calibration, out of gamut colours will map to the same points on the gamut edge, but never internal gamut colours, as with the JVC.

The only time that can happen is when an artificial colours boost is applied, such as cranking up the saturation.

Other than Omar's badly configured Sony CRT we do not remember ever having seen any other display act like this.

Even when calibrating to a wider gamut than the display is capable of, no input colour values within the displays native gamut will ever map to the same output colour.

Basically, whatever is the smaller gamut will always show unique output colours for any given input colour.

To not do so show the display is entirely unfit for purpose.

I am very happy to discuss this with JVC, and will be sending our contact there this information (as we have before, but they have ignored).

Regardless, could we do something different when profiling such appaling displays?
Probably.
Would it work reliably, and to the level we aim to attain?
Unlikely.
It is worth our time?
Probably not.
Should JVC fix their projectors native response?
Definitely.
Will we have a look at other options?
Yes, we already are.

But, we really are not into turd polishing.
And unfortunately, that is what this comes down to.

The only reliable solution is to define a colour space target that is smaller than the point where the internal colour gamut mapping errors occur.

That LightSpace can already do, with the concatenate LUTs approach.

That is why such functionality exists within LightSpace.

The new volumetric graphs within ColourSpace will help a lot in understanding such appalling display behaviour, and assist in setting a viable target colour space.

Such graphs cannot be retro-fitted into LightSpace, but any LightSpace user can still define the issue with high granularity Saturation Sweeps, at different luma levels.

All this info will be added to the User Guides within the website.

Probably the Error! page, as this is a great example of what a really, really bad display looks like.

Steve

Steve Shaw
LIGHT ILLUSION


Last edited by Light Illusion; 02-16-2019 at 07:17 AM.
Light Illusion is offline  
post #1948 of 2391 Old 02-16-2019, 07:47 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,267
Mentioned: 53 Post(s)
Tagged: 0 Thread(s)
Quoted: 2640 Post(s)
Liked: 1738
Quote:
Originally Posted by Light Illusion View Post
After all our assessments of the JVC projector profile data there are indeed serious issues with the display that really make it unfit for the application it has been sold for.
Spoiler!
Sorry Steve, I've been busy making some tools to parse, spot problem regions and potentially edit profiles, I did see your message earlier but haven't managed to reply.

I take issue with this; the displays are being sold as home theatre displays, not as pro displays to be used with LUT holders. For this use they are excellent. When calibrated via other means (autocal / manual calibration) and fed with the only consumer formats that exist for WCG they track the standards very well within the limits of what you expect from a consumer display, with wider, more accurate gamut coverage than any Sony, any consumer DLP, etc. I cannot see how any of that makes them unfit for purpose as you are so quick to brand them.

I (and others) are trying to make them even better using a 3DLUT, as we're aware of the advantages of the approach. These displays don't have to expose their native response in a linear fashion, there is no "JVC native response" content out there in the world to view. Though it would be good if they did, as it would make LUT generation easier with LS. If they're unsuitable for anything at all, they're unsuitable for 3DLUT generation with software that relies on the display output having a linear mapping from input.

LS HTP/L, on the other hand, is sold as a tool to calibrate home displays. I'm not going to go as far as to say LS HTP/L is "unfit for purpose" like you are branding these displays, but at the very least, these displays in their wide gamut modes represent a hole in the models you can support. You haven't been able to offer me any solution so far. I guess you tried using the concatenate process with this LUT and still saw the issues in the gamut. So far as I can see, anyone buying your software to calibrate these particular displays (or others exhibiting the same characteristics, I'm not aware of any) in wide gamut mode will not be satisfied at present. They cannot be calibrated volumetrically by the software as it stands.

Mine can't be calibrated even using the process you have on your website (and I don't think there is anything out of the ordinary for my particular unit), so the only options I have at the moment are to either iterate manually over the exact custom gamut extent positions, generating endless LUTs until I get rid of all the error points that appear in the LUT, or to use the "fit" non-volumetric matrix+1D LUT that everyone is so quick to criticise from other software solutions...!

These displays that are already in homes are not going to change in their behaviour, so we are stuck with them. JVC are pathological in not releasing new functionality to old product (like most manufacturers of hardware, anything is an excuse for a new model number), the models with the wider gamut been like this for many model years. If new projector customers are lucky, in some future incarnations maybe if they hear enough noise on the matter they >might< do something, who knows. But the newer models actually have a wider gamut as it is (almost 100% of P3, though no linear mode for wide gamut operation). Will be entertaining to see what the profiles look like and whether LS can do anything with them.

As I said at the top of this message, I'm spending my own time at the moment looking at the issues in these display profiles (which are actually obvious now I know how the LUT engine behaves) and seeing if there is either a way to fix the profiles up so the LUT generation makes sensible LUTs (subtly difficult outside of the LUT engine itself due to things like having to waste time to create an algorithm to reconcile the drift data, etc), or at least to automatically generate the largest possible gamut targets that remove the error points from the LUT.

I've not looked at the new "remote control/secondary execution" features yet. Perhaps using it it might be possible to automate control of LS to make the web "concatenate LUT" process a bit less of a ball-ache to do. Coupled with a script to automatically detect the custom gamut extents necessary to clip the error points out you'd actually have something that would work more or less and probably get pretty good results. (I think this will result in a gamut that is slightly smaller than it could have been otherwise, however).

We can argue about where it should be fixed; I'm looking at options here to fix my own (and other's) profiles if possible. You could as you say have a go at fixing it in your engine. I don't really care how it gets fixed, but until it does I don't think any of the above is unfair.

As it stands I cannot see any way to recommend Lightspace to folk at the moment if a large part of their requirement is to generate a volumetric 3DLUT of the wider gamut modes of JVC projectors at least up to the 2017 year model series, and it sounds like you can't either.. For me wide gamut JVC represents precisely 50% of the LUTs I want to generate at the moment. Of course, for enthusiasts calibrating other displays that are not affected, or for calibrators calibrating lots of displays this is probably moot; LS will get excellent results with most displays. Pro calibrators will just do the best they can with the JVCs as they are, which will probably mean the fit LUT or no LUT or another solution's LUT.

I hope this changes and that someone works out a viable process for doing this.
Manni01 and omarank like this.

Last edited by bobof; 02-16-2019 at 08:13 AM.
bobof is online now  
post #1949 of 2391 Old 02-16-2019, 08:14 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,833
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 779 Post(s)
Liked: 1144
Professional, or home TV - makes no difference.

If a display clips/crushes luma, in either black or white, no-one would accept it.

And if there were no user controls to fix the clipping/crushing before the display was profiled for calibration, it would be rejected out of hand.

That is what the ContrastCal/BrightnessCal images are for. To setup the display before profiling.

We have never made saturation/gamut versions as it was never expected any display would be so bad as to need them.

But, this is exactly the situation with the JVC, with saturation/gamut clipping.

It is clipping in exactly the same way as the above description for luma.

If the problem can't be fixed within the display prior to profiling, it is totally unacceptable.

The display is not fit for purpose.

It is that simple.

Go shout at JVC, and demamd they fix their appaling display.

As for a LightSpace solution, I have given it to you.
Target a colour space that is below the threshold of the issue, and then concatenate with the desired colour space.

That is why LightSpace has such tools.

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #1950 of 2391 Old 02-16-2019, 08:27 AM
Member
 
Join Date: Mar 2016
Posts: 56
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 36 Post(s)
Liked: 30
Quote:
Originally Posted by bobof View Post
But in your case, I would be surprised if you can't get into a better mode to start with as a pre-condition for creating your LUT.

If you open your profile in Manage Colourspace you can look for the problems, and then try and fix them if possible in your display setup. As it happens, it seems us JVC owners have no way to fix these issues, which puts us in an almost irreconcilable position with LS as it stands. You might be able to do better with what you have by spending a little time understanding your display behaviour.

I had a look at your profile with a custom filter set to "R==1 && G==0" (see screen shot). You can see from that that from RGB=255,0,0 to 255,0,63 there is no change whatsoever in the measured values. I'd be surprised if you can't find a mode that doesn't have that characteristic. One reason it is worth trying to fix these issues at the display is that even if you succeed in generating a LUT with another solution it is still quite sub-optimal. Your LUT is having to map all the possible output colours into the range B 63-255 on the HDMI signal in that direction, whereas in a linear mode it could be using the whole 0-255 range. Only if you can't find a better mode (or you have some other reason which means you have to use this mode) would you then try and LUT this mode.
Thanks for such a detailed analysis of my display profile. It is quite helpful and interesting. I will have a deeper look later.

But alas, it is like that in all the different modes that I tried. I can't do a better pre-conditioning using the TV's controls, apparently. But yes, maybe I can set the output range as 63-255 in madVR and then do the profiling. I will have to set TV brightness in such a way then that levels upto 63 get clipped (not sure if the brightness control is that flexible). Perhaps, I will get much better results then. What do you think?
Manni01 likes this.

Last edited by omarank; 02-16-2019 at 08:38 AM.
omarank is online now  
Sponsored Links
Advertisement
 
Reply Display Calibration

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off