JVC Autocal Software V12 calibration for 2019 RS1000/RS2000/RS3000/NX5/NX7/NX9

[Manni]

Please read my posts more carefully too.

• I was simply saying that you should compare the meters between them, not the meters with the reference, as you can have an imperfectly calibrated VPR as a starting point. So if DisplayCal told you that you have dE=+2 with SpyderX and let's say -1 with i1pro2, you may have to sum the two to get the total difference (depending on the direction of the deviation). I was asking confirmation that you considered the real difference between the two.
• As I explained very clearly, the point of the correction matrix is that with HCFR forces you to select one so I was asking if DisplayCal does the same and in that case which one had you selected.
• I was NOT comparing results of different softwares. Where did you read that?

Nothing offensive or stupid I believe...

I would humbly say keep calm and be open

I am perfectly calm and open, I am simply asking you to read my posts more carefully (editing yours aftwerwards doesn't make any difference to the fact that when you posted it, as it is above, you still had not read my original post or my reply carefully. Making a mistake is fine, we are indeed all human. Not reading a post twice is not a mistake, it's carelessness).

For the last time, I am not using DisplayCAL (or HCFR), full stop, yet you keep mentioning it. Why?

Even after I tell you in my last post that I DON'T USE DISPLAYCAL. Not shouting, only making sure that you are not going to miss this information a THIRD time, as I said from the beginning that the measurements were taken with Lightspace.

I have posted measurements using different software, and you are saying that the differences are bigger in one than in the other, without backing this up with any data. I am simply saying that you can't compare directly measurements from two different software anyway.

I have posted the exact measurements, with the relative dE for each meter. If you want to calculate the absolute dE for each meter because you disagree with my estimate, you can do so and post the actual dE for each measurement.

For the last time, please either provide information if you wish to do so so that it can be discussed, or be patient and wait for the next chunk of data that I said I would provide when I have the right software and the time to do it.

If you keep being disruptive, taking the thread off topic or wasting my time, you will be banned from this thread, or I will simply close it, temporarily or definitely.

I have no interest in debating or arguing. My time is limited and I'd rather use it to provide information and advice and learn things.

So please keep the thread on topic, read all the posts carefully before making assumptions, raising questions and questioning the data. If you question the data provided, do some work and provide more data yourself instead of asking more work from me.

This is the last warning. There won't be another one.

[Jwb2]

Hi Manni, 

thank you for this great resource. This thread is so helpful to an autocal novice like me.


Ive gotten good results on color and gamma with my idpro2 and spyder x.  I just do a little touch up on 5%-20% greyscale range with chromapure/i1dpro.  

However, I still get a fairly decent brightness loss.  For HDR, I go from 95 nits to 83 nits when my white balance/color is autocaled.  Do you think that is acceptable/normal?  Does changing the screen size option in the autocal software still potentially affect brightness?  Ive tried hard to follow your recommended steps, but perhaps this is just the normal brightness loss when starting out with a white balance that is fairly off (I have excess green and short on red and blue before autocal).  

I have a JVC rs2000, 135" 16x9 neve screen at min throw.  

Would you have any advice, or perhaps this is just the norm with my particular setup?  Thank you

[Manni]

Hi Manni,

thank you for this great resource. This thread is so helpful to an autocal novice like me.


Ive gotten good results on color and gamma with my idpro2 and spyder x.  I just do a little touch up on 5%-20% greyscale range with chromapure/i1dpro. 

However, I still get a fairly decent brightness loss.  For HDR, I go from 95 nits to 83 nits when my white balance/color is autocaled.  Do you think that is acceptable/normal?  Does changing the screen size option in the autocal software still potentially affect brightness?  Ive tried hard to follow your recommended steps, but perhaps this is just the normal brightness loss when starting out with a white balance that is fairly off (I have excess green and short on red and blue before autocal). 

I have a JVC rs2000, 135" 16x9 neve screen at min throw. 

Would you have any advice, or perhaps this is just the norm with my particular setup?  Thank you

Hi there,

Losing 5-10% brightness is perfectly normal when you go from uncorrected (all gains at 0) to autocal, especially if you have an excess of green, as that's where most of the luminance (around 70%) lies. Next comes red (around 23%) and last comes blue (around 7%).

So if you have to reduce green to get to D65, you'll lose a lot more brightness than if you have to reduce blue for example.

On the other hand, once you have an autocal, if you correct the gains over time more and more to get to D65 as the lamp ages and the balance changes, you might end up getting some brightness back doing an autocal, because the autocal can adjust some internal tables and you might end up with more brightness.

One thing you might want to try is see if after the autocal another color temp preset gives you more brightness after adjustment than 6500K without, but usually this will not give you results as linear for the rest of the steps.

As far as I know, the screen size, distance etc is only used if you use the environment settings, which I personally don't recommend to use if you have a bat cave, so I wouldn't play with these.

One thing you could do if you wanted to be sure that autocal doesn't lose any brightness, is restore your INIT file and see how much brightness you get when setting 100% white to D65 manually with the gains from the 6500K (or 7500K). My guess is that it would be as much or possibly a bit less that what the Autocal delivers. If it's significantly more, post the measurements and we'll take it from there.

If you don't have to make any significant corrections after the autocal (no more than a few clicks), pat yourself in the back for a job well done, and enjoy the projector

With the new DTM in your rs2000 (assuming you're using f/w 3.10) and even less with madVR/Radiance Pro, 83 vs 98 nits doesn't make a significant difference anyway.

[Jwb2]

Thank you Manni, that was incredibly helpful and I appreciate it.

I do get about the same/slightly worse brightness loss when I manually adjust white balance.  Im extremely happy with the picture quality I have now and will sit back and enjoy!  Youre def correct on needing less brightness with good DTM.  I am on the current firmware and am very happy with the DTM.   Im planning on getting a Mad vr Envy when it arrives to take it to the next level.

Thanks again for your help!

Regards!

[Manni]

Thank you Manni, that was incredibly helpful and I appreciate it.

I do get about the same/slightly worse brightness loss when I manually adjust white balance.  Im extremely happy with the picture quality I have now and will sit back and enjoy!  Youre def correct on needing less brightness with good DTM.  I am on the current firmware and am very happy with the DTM.  Im planning on getting a Mad vr Envy when it arrives to take it to the next level.

Thanks again for your help!

Regards!

You're very welcome, sounds like you are all set

The Envy will definitely take you to the next step. The JVC DTM is a huge improvement over the old static tonemapping, but it's still significantly behind madVR/Envy, especially after the latest madVR developments (particularly the HSTM contrast enhancer).

Even from an ease of use point of view, with the Envy you won't have to make any manual adjustments, while with DTM Auto isn't always the optimal choice when the metadata isn't reliable and occasionally you'll get better results overriding to low/medium/high manually.

Still, you'll be able to enjoy a great picture in the meantime

[Manni]

I have done some quick tests to see if my RS2000, which is very slightly undersaturated on red in P3, could benefit from using profile off with the filter (which is the widest gamut the projector is capable of) instead of the DCI-P3 color profile, and the answer is no. It would be the same for BT2020 by the way, I only used DCI-P3 as it's closer to the native gamut of the projector and makes measurements easier/clearer.

After my last autocal, I get above 97% of P3 cover for a max dE under 2 (that's before my madVR 3D LUT) and 99% color volume. The calibration is perfectly linear and barely needs a 3D LUT.

With profile off with filter, I get 98% of P3 cover but that's because of extra cover of blue and magenta that compensates my deficiency in red. Red is limited by the native gamut, so profile off doesn't give me anything there.

Same with color volume, I get 104% of P3, but that's because it takes into account excess in green and blue, it doesn't make any difference for my slightly deficient red.

However, it's completely non-linear and unusable without a very large LUT (5,000 points, which is 2.5 hours here at least) while using DCI-P3 is perfectly usable after an Autocal without a 3D LUT, and a 10-minute lightning LUT in Calman (100 points) gives me near reference results (according to the i1pro of course).

So no hesitation here, at this stage, profile off is not an interesting option with this specific unit. It could be different in other situations, but here I won't bother. Too much hassle (you have to enable the filter using a specific IP or IR command as there is no option for profile off with the filter) for no visible gain.

I post some gamut and color volume measurements that show the result of my last autocal (V11, Spyder 5 + i1pro2) re gamut, no 3D LUT applied. After a few days, it's nice to see that it hasn't drifted at all.

[jj-34]

So as for now you recommend a gamma autocal with V11 and spyder 5 and not V12 and spyder X ?

As I have both it's interesting to know.

[Manni]

So as for now you recommend a gamma autocal with V11 and spyder 5 and not V12 and spyder X ?

As I have both it's interesting to know.

Until further tests are done, yes.
I'm hoping to get a beta build of Calman with Spyder X support any day now, so I might be able to provide more definitive data over the week-end.

The only thing I can say is that with my specific Spyder X and Spyder 5 units, I get better results with V11 / Spyder 5 / i1pro2 than with V12 / Spyder X / i1pro2. Could be the meters, could be a software bug or a driver issue, we should find out soon.

[jj-34]

As in my physical environment it is not very easy to position the 5 accurately, I'll wait your findings to do another gamma autocal 

[Manni]

As in my physical environment it is not very easy to position the 5 accurately, I'll wait your findings to do another gamma autocal

Yes, it’s certainly much easier to position the Spyder X facing the screen in most rooms.

[jj-34]

But wait, I can do a color autocal with my I1pro2 in the mean time, I believe it's not gamma autocal dependent, or is it ?

[Manni]

But wait, I can do a color autocal with my I1pro2 in the mean time, I believe it's not gamma autocal dependent, or is it ?

It’s not gamma dependent but it impacts the greyscale, and there might be some issues with the i1pro2 in V12 as well, it’s not just the Spyders.

I’m not going to make any further comments on the subject until more extensive testing is done to figure out exactly what’s happening and why.

[CyberPaul]

He can do a color autocal with i1pro2 using v11

[Manni]

He can do a color autocal with i1pro2 using v11

If you read my posts, you'll see that it is recommended to use V11 if you can.

So he can do anything he wants with V11: color only with the i1pro2, gamma only with the Spyder 5, whatever he fancies.

It's only with V12 that I'm saying that running a color only with the i1pro2 can produce results inferior to what can be obtained with V11. The issue with V12 might not be only with the Spyder X.

At least that's what I observe here.

[CyberPaul]

I have always had a dilemma on this: since grey scale is anyway impacted by color calibration, which one do you recommend to autocalibrate first, gamma or color?
Also, is gamma autocalibration performed separately on the three primary colors? I think so and this can have an implication on the answer to the above question.