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

[Sittler27]

On my JVC NX9 I performed a gamma+color autocal with my Spyder5Pro for DCI mode (iris=0).  Everything looked perfect.
I use madvr + Kodi as the player, with madvr set to “display calibrated = DCI-P3”.
After adjusting Brightness to -3 and Contrast to +1 & Gamma to 2.4 PT3 DG3, I then went into HCFR with my i1D3 to fine-tune off screen (0.98 actual gain).
Using Custom 1/6500K, I found that I needed to drop green gain substantially to -11 just to get red and blue to come up. 

Does that sound normal? If not, what could I be doing wrong?
Are these projectors known to have this much green push?

Other adjustments in this DCI mode included offset red to +4, offset green to +2 and offset blue to -2.

Doing the same in Natural mode (6500K; gamma 2.4 PT3 DG3), I found I still needed to drop green gain to -5, red gain to -1 and red offset to +3.

It should also be noted that after an autocal of HDR10 mode (no filter; HDR color) I then had to do some really crazy adjustments of gains/offsets (with ATV4K as source) to bring greyscale inline.  Had to drop green gain to -15 and some other large adjustments of red and blue offsets.  Again, just wondering what is normal and what I could be doing wrong overall.

[Manni]

This wouldn't be normal if you were using the i1pro, but as you're using a Spyder that's probably where the adjustments come from.

I tested a Spyder X and it's unusable for color. It actually made the calibration significantly worse than it was (my previous color autocal had been made a few hundred hours earlier with my i1pro2).

There is nothing wrong with the panels or the projector, but if you use an inaccurate meter for color, then when you finetune it you'll find it needs significant corrections.

Plus you're not using a reference meter. The error in the i1display3 could be in the same direction or the opposite direction. So you might be under or overcorrecting with the i1display3. There is no way to know unless you use a reference meter for such corrections.

As advised in the basic section, I wouldn't recommend using a spyder for color autocal. It's fine for gamma only (and better than the i1pro2 for that), but unless you've established the degree of accuracy of the spyder it's better not to use it for color at all.

[SirMaster]

On my JVC NX9 I performed a gamma+color autocal with my Spyder5Pro for DCI mode (iris=0).  Everything looked perfect.
I use madvr + Kodi as the player, with madvr set to “display calibrated = DCI-P3”.
After adjusting Brightness to -3 and Contrast to +1 & Gamma to 2.4 PT3 DG3, I then went into HCFR with my i1D3 to fine-tune off screen (0.98 actual gain).
Using Custom 1/6500K, I found that I needed to drop green gain substantially to -11 just to get red and blue to come up. 

Does that sound normal? If not, what could I be doing wrong?
Are these projectors known to have this much green push?

Other adjustments in this DCI mode included offset red to +4, offset green to +2 and offset blue to -2.

Doing the same in Natural mode (6500K; gamma 2.4 PT3 DG3), I found I still needed to drop green gain to -5, red gain to -1 and red offset to +3.

It should also be noted that after an autocal of HDR10 mode (no filter; HDR color) I then had to do some really crazy adjustments of gains/offsets (with ATV4K as source) to bring greyscale inline.  Had to drop green gain to -15 and some other large adjustments of red and blue offsets.  Again, just wondering what is normal and what I could be doing wrong overall.

Dropping green that far could be normal but remember you aren't adjusting that from stock as in "Are these projectors known to have this much green push?". You ran color autocal with a spyder which could just not be very accurate compared to the i1 Display Pro. Your spyder could have added a bunch of green during autocal. Or perhaps they are both off somewhat from each other in opposite directions meaning that after you use one, using the other to correct it could be a bigger adjustment.

Also, you should not really use positive offset adjustments as that will raise your black floor and lower your contrast ratio pretty considerably. Instead if you want to tune the calibration with a better meter after doing the automatic autocal you should use the autocal gamma adjustment tab along with HCFR and your i1D3 which is a manual process.

Open the gamma editor in autocal, and in that screen set the gamma to Custom1 and the correction value to whatever you are targeting, like 2.2 or 2.4.

In the autocal gamma screen you will see that you have 12 gamma points: 5, 10, 15, 20, 30, 40, 50, 60, 70, 80, 90, 95.

On the projector set the gamma value to Custom1 as well and set the correction value to "Import".

In HCFR make sure that in references (Advanced -> Preferences -> References tab) you set the gamma target to the same one that you selected in autocal correction value, like power law gamma 2.2 or 2.4. Also make sure you have the correct white point selected like D65. Then open the Grey Scale view and right click on the line graph part on the bottom and make sure the "w/Gamma" option is enabled. Finally go to Measures -> Parameters, and change the number of greyscale levels from 10 to 20.

The first thing you should do in HCFR is click the 100 in the % White row and click the green triangle to display white on your screen. On the JVC in your 6500K color temp adjust the gains to balance your white to 100%.

Now, before you begin adjusting in autocal, click the run greyscale button to measure all 20 grey scale levels.

When that is done, pick a point from my number list above, click on that number in the % White row in HCFR grey scale view and then click the Green triangle to display the pattern on your screen. Then in autocal gamma screen, use the first slider to move to that same number point that you chose to adjust. With the second slider you can then move white, red, green, or blue up or down. Moving white moves all 3, so adjust them as you need with the second slider to bring all 3 to 100%. This is calibrating both white balance and gamma to your target for that percent. Repeat this for all 12 points.

When you are done you can save this in autocal if you want, but there is nothing else left to do if you just want to use it, just set the JVC to the Custom1 that you used to do the adjustment.

When you are all done with this, finish by double checking your brightness and contrast values with the black and white clipping test patterns.

Hopefully I made this process clear. If you are at least somewhat familiar with HCFR, autocal, and JVC menus you should have no trouble. Otherwise ask if you need clarification on something.

[grendelrt]

On my JVC NX9 I performed a gamma+color autocal with my Spyder5Pro for DCI mode (iris=0).  Everything looked perfect.
I use madvr + Kodi as the player, with madvr set to “display calibrated = DCI-P3”.
After adjusting Brightness to -3 and Contrast to +1 & Gamma to 2.4 PT3 DG3, I then went into HCFR with my i1D3 to fine-tune off screen (0.98 actual gain).
Using Custom 1/6500K, I found that I needed to drop green gain substantially to -11 just to get red and blue to come up. 

Does that sound normal? If not, what could I be doing wrong?
Are these projectors known to have this much green push?

Other adjustments in this DCI mode included offset red to +4, offset green to +2 and offset blue to -2.

Doing the same in Natural mode (6500K; gamma 2.4 PT3 DG3), I found I still needed to drop green gain to -5, red gain to -1 and red offset to +3.

It should also be noted that after an autocal of HDR10 mode (no filter; HDR color) I then had to do some really crazy adjustments of gains/offsets (with ATV4K as source) to bring greyscale inline.  Had to drop green gain to -15 and some other large adjustments of red and blue offsets.  Again, just wondering what is normal and what I could be doing wrong overall.

I did not autocal color and I was deficient on red for grayscale which caused me to have to drop back green and blue by around -13 I think on gain and barely any on offset just for blue. I am on a studiotek 100 screen with the adjustment put in the jvc. So may just be normal. 

[Sittler27]

This wouldn't be normal if you were using the i1pro, but as you're using a Spyder that's probably where the adjustments come from.

I tested a Spyder X and it's unusable for color. It actually made the calibration significantly worse than it was (my previous color autocal had been made a few hundred hours earlier with my i1pro2).

There is nothing wrong with the panels or the projector, but if you use an inaccurate meter for color, then when you finetune it you'll find it needs significant corrections.

Plus you're not using a reference meter. The error in the i1display3 could be in the same direction or the opposite direction. So you might be under or overcorrecting with the i1display3. There is no way to know unless you use a reference meter for such corrections.

As advised in the basic section, I wouldn't recommend using a spyder for color autocal. It's fine for gamma only (and better than the i1pro2 for that), but unless you've established the degree of accuracy of the spyder it's better not to use it for color at all.

This makes a lot of sense.  Now I'm wondering if I should revert back to my init file and just do the autocal for gamma only with SpyderPro5 and then do the manual gamma adjustments SirMaster graciously provided instructions for below.

In the end, I really just need 3 modes calibrated:

1.  Natural 709 - for TV (mainly sports) - which I usually autocal and then use patterns in madvr set to calibrated to 709 (since I can't get patterns from my TV cable source)

2.  Custom 1 (DCI) - for everything madvr - which I usually autocal and then use patterns in madvr set to calibrated to DCI-P3

3.  HDR10 (BT2020) - for HDR from my ATV4K streaming content - which I usually autocal and then use patterns in Plex to calibrate

[Sittler27]

Dropping green that far could be normal but remember you aren't adjusting that from stock as in "Are these projectors known to have this much green push?". You ran color autocal with a spyder which could just not be very accurate compared to the i1 Display Pro. Your spyder could have added a bunch of green during autocal. Or perhaps they are both off somewhat from each other in opposite directions meaning that after you use one, using the other to correct it could be a bigger adjustment.

Also, you should not really use positive offset adjustments as that will raise your black floor and lower your contrast ratio pretty considerably. Instead if you want to tune the calibration with a better meter after doing the automatic autocal you should use the autocal gamma adjustment tab along with HCFR and your i1D3 which is a manual process.

Open the gamma editor in autocal, and in that screen set the gamma to Custom1 and the correction value to whatever you are targeting, like 2.2 or 2.4.

In the autocal gamma screen you will see that you have 12 gamma points: 5, 10, 15, 20, 30, 40, 50, 60, 70, 80, 90, 95.

On the projector set the gamma value to Custom1 as well and set the correction value to "Import".

In HCFR make sure that in references (Advanced -> Preferences -> References tab) you set the gamma target to the same one that you selected in autocal correction value, like power law gamma 2.2 or 2.4. Also make sure you have the correct white point selected like D65. Then open the Grey Scale view and right click on the line graph part on the bottom and make sure the "w/Gamma" option is enabled. Finally go to Measures -> Parameters, and change the number of greyscale levels from 10 to 20.

The first thing you should do in HCFR is click the 100 in the % White row and click the green triangle to display white on your screen. On the JVC in your 6500K color temp adjust the gains to balance your white to 100%.

Now, before you begin adjusting in autocal, click the run greyscale button to measure all 20 grey scale levels.

When that is done, pick a point from my number list above, click on that number in the % White row in HCFR grey scale view and then click the Green triangle to display the pattern on your screen. Then in autocal gamma screen, use the first slider to move to that same number point that you chose to adjust. With the second slider you can then move white, red, green, or blue up or down. Moving white moves all 3, so adjust them as you need with the second slider to bring all 3 to 100%. This is calibrating both white balance and gamma to your target for that percent. Repeat this for all 12 points.

When you are done you can save this in autocal if you want, but there is nothing else left to do if you just want to use it, just set the JVC to the Custom1 that you used to do the adjustment.

When you are all done with this, finish by double checking your brightness and contrast values with the black and white clipping test patterns.

Hopefully I made this process clear. If you are at least somewhat familiar with HCFR, autocal, and JVC menus you should have no trouble. Otherwise ask if you need clarification on something.

Wow, thanks so much for these instructions.  Yes, I am familiar with these tools.

I didn't know about adding gains/offsets upsetting contrast.  Good to know.

I also just learned that the color autocal affects the white balance - I had thought it was only adjusting the CMS (which I didn't touch in HCFR afterwards).  I had no idea it was also affecting the gains/offsets even though they are still all set to 0 once done.

So, should my calibration approach now be?
1.  Reset projector using autocal init file.
2.  Perform a gamma only autocal (with/without filter - all in low lamp -for now)
3.  Use HCFR and my i1D3 to perform the manual gamma calibration as per your instructions above.

I'd want to calibrate ideally all 3 of the user modes I listed in my previous post replying to Manni.
Would I be able to do that with this method by just switching it between each mode and repeating the steps you provided?

Also, finally a question about performing a DCI 3DLUT for use with madvr Custom 1 DCI mode after performing the above on that user mode:
As DisplayCal has that calibration window of it's own to lineup the bars in the middle with adjustments to gains/offsets - should I be making further adjustments in the gains/offsets to line that up before commencing the profiling and subsequent 3DLUT creation & installation?

[SirMaster]

Wow, thanks so much for these instructions.  Yes, I am familiar with these tools.

I didn't know about adding gains/offsets upsetting contrast.  Good to know.

I also just learned that the color autocal affects the white balance - I had thought it was only adjusting the CMS (which I didn't touch in HCFR afterwards).  I had no idea it was also affecting the gains/offsets even though they are still all set to 0 once done.

So, should my calibration approach now be?
1.  Reset projector using autocal init file.
2.  Perform a gamma only autocal (with/without filter - all in low lamp -for now)
3.  Use HCFR and my i1D3 to perform the manual gamma calibration as per your instructions above.

I'd want to calibrate ideally all 3 of the user modes I listed in my previous post replying to Manni.
Would I be able to do that with this method by just switching it between each mode and repeating the steps you provided?

Also, finally a question about performing a DCI 3DLUT for use with madvr Custom 1 DCI mode after performing the above on that user mode:
As DisplayCal has that calibration window of it's own to lineup the bars in the middle with adjustments to gains/offsets - should I be making further adjustments in the gains/offsets to line that up before commencing the profiling and subsequent 3DLUT creation & installation?

Negative adjustments to gains/offsets don't harm black or contrast. Positive adjustments to offsets do, but I don't think positive adjustments to gains do.



Your approach sounds fine. You can fine tune each if you need by using a different custom gamma slot for each. Just use custom2 and custom3 for the other 2 modes you are making.



If you are doing a 3DLUT in DisplayCAL then you don't even need to do any adjustment for that user mode. Just leave your gamma in the JVC at 2.2 or 2.4 and do the 3DLUT. The 3DLUT will take care of calibrating absolutely everything as the i1D3 sees it. You do not need to do anything with the RGB bar screen that pops up before the 3DLUT. The 3DLUT will calibrate everything correctly including white.





Technically there is one other thing you could do that would potentially improve your color gamut. Normally this would be taken care of with color autocal, but you would want an i1 Pro 2 for this. However, this is a way to basically do it with any meter.



It's referred to as the Chad B method.




What he is describing here is a method of fixing the color gamut after doing a color autocal with a spyder. He talks about using CalMAN, but you can easily use HCFR as well. In HCFR under Measures -> Parameters, just set the number of saturation color levels to 10, then you can measure 50, 60, 70, 80, 90, and 100 color saturations to get the x and y values to fill in his spreadhseet like he says.



It's a somewhat advanced method, but it would potentially expand your color gamut since you are performing the measures with color profile off in the JVC (i.e. native bulb gamut) and then using the measurements to create a new custom color profile which will have the correct primary coordinates (according to the better meter being used in HCFR rather than the spyder via color autocal) which you can create and load via autocal.



I think technically automatic autocal wouldn't even be beneficial here.



1. You would do The Chad B method to create a custom color gamut based on color saturation measurements taken in HCFR with i1D3.

2. You would do adjust the gains to set 100% white with HCFR with i1D3.

3. You would do the 12 point RGB gamma calibration in autocal with HCFR with i1D3.



I guess you could still put a gamma autocal with the spyder in between steps 1 and 2 because you can get a more detailed 33-point gamma cal then and the spyder is OK for that.

[Manni]

The above method works fine if the target gamut entirely fits withing the gamut covered by the projector.

If any point is undersaturated, it doesn't work as well as it cannot make the difference between a meter error and a limitation in the projector's native gamut, so it can over-correct unnecessarily. It usually works fine for rec-709 (unless green or red can't be reached), but it doesn't work as well for DCI-P3 or BT2020.

I recommend correcting only white, unless all the target primaries can be reached at 100% sat.

I have put together a modified method that produces better results with DCI-P3 and BT2020 (as well as rec-709 when the unit doesn't reach 100% sat for each primary), but the i1pro2 produces perfect results so my motivation for posting about this was pretty low.

If anyone is interested, I'll try to put it together next month, when I have a bit more time.

I haven't posted on the manual gamma calibration method either as no one asked and as it's pretty much unnecessary after a properly done gamma autocal, given how quickly these projectors drift. By the time you've tuned your last point, the first point has probably already drifted. I'm only slightly exaggerating

[SirMaster]

The above method works fine if the target gamut entirely fits withing the gamut covered by the projector.

If any point is undersaturated, it doesn't work as well as it cannot make the difference between a meter error and a limitation in the projector's native gamut, so it can over-correct unnecessarily. It usually works fine for rec-709 (unless green or red can't be reached), but it doesn't work as well for DCI-P3 or BT2020.

I recommend correcting only white, unless all the target primaries can be reached at 100% sat.

I have put together a modified method that produces better results with DCI-P3 and BT2020 (as well as rec-709 when the unit doesn't reach 100% sat for each primary), but the i1pro2 produces perfect results so my motivation for posting about this was pretty low.

If anyone is interested, I'll try to put it together next month, when I have a bit more time.

I haven't posted on the manual gamma calibration method either as no one asked and as it's pretty much unnecessary after a properly done gamma autocal, given how quickly these projectors drift. By the time you've tuned your last point, the first point has probably already drifted. I'm only slightly exaggerating

I think that would be cool.

I agree it's not needed if you have an i1 Pro 2, but for people who don't, I think there is some value.

Personally I was looking into stuff like this so that I could calibrate my cousins RS400 which does not support the i1 Pro 2, only Spyder4/5.

I find that running the color autocal with a good meter actually expands the gamut coverage at least on my NX5 since it's able to carve out a new gamut from measurements taken with profile off on the native bulb.  I was looking for a way to accomplish this for people who don't have a spectrometer or older JVC that don't support it.

Also I was wondering if you know what part of autocal is responsible for sometimes resulting in an increased peak brightness after running the autocal?  I was hoping I could try to reproduce that potential brightness gain again for people who don't own an i1 Pro 2 or for older JVCs that don't support the spectrometer.

My assumption is that it's some internal calibration tables that we just cant access, but maybe via custom color gamut or similar, we can achieve it?

As far as the manual gamma cal I again agree that it's also not really needed if you can do the color autocal with the i1 Pro 2, but if you can't (for some above reasons) I find that it's a good way to perform the white balance if adjusting the gains alone isn't sufficient.

For example, here is my NX5 measured stock: 
https://nicko88.com/misc/nx5cal_old/stock.pdf

If all I had was a Spyder5 and i1D3 so all I could do is gamma autocal, the white balance is still pretty out of whack and the 12 point gamma cal is a good way to dial in the rest of the white balance without using the offset/bias adjustments which always seem to hurt more than help.

[Manni]

I agree, I used all this with my rs500 as it only supported the spyder 4/5, but with the rs2000 and the i1pro2 support on top of the spyder 5/X, autocal is so much faster/easier

No spreadsheets, no fastidious, unnecessary manual corrections. Just gamma with Spyder, color with i1pro2, and you barely need a 3D LUT, all in less than 10 minutes.

That's why my motivation to explain manual processes that are not needed with the right tools has gone significantly down.

They are still useful with older gens or if you want to cut corners, but if your time is money, investing in a Spyder X / i1pro2 combo is a no brainer.

Given the cost of these units (especially rs2000/rs3000), I'd say that it's a very reasonable investment for "maintenance".

I'll try to put something together next month though.

To answer your question about restoring brightness, it's a combination of two things:

1) The meter used. Some spyders are really off (up to 10dEs or more), so you can lose a lot of brightness using the wrong one for color calibration, as it will overcorrect and shave brightness unnecessarily.
2) Once the internal tables are corrected, you don't need the external gains corrections (or much less) and you can restore some brightness that way too.

No there is no way (that I know of) to access these tables, but it's not really necessary. Just make sure you're using a meter that doesn't hobble brightness with color calibrations because it's inaccurate (could be the spyder with autocal or any other meter used to adjust the gains at 100% white).

I find that if you use bias/gains at the optimal IRE for best balance and use the gamma controls in the menus (dark/picture/bright) you can get very close. That's much faster than using the manual gamma controls, but I agree it's less accurate.

[Sittler27]

Does the color autocal drift like the gamma does as bulb ages?

I have an option to borrow a far away friend's i1photo pro 2 (is that the same as an i1pro2?) but obtaining it in the future for maintenance will be tougher.  Just wondering if it might just be a one time calibration and that's it?

[Manni]

The gamut doesn’t shift much (though it can shrink a bit after a couple of thousand hours) but the rgb balance keeps changing as the lamp ages. You could always do a color autocal now with a borrowed i1pro2 and keep checking. If your lamp has stabilized (it needs 300-500 hours) then you might be fine for a while.

Best would be to get an i1pro2 or ES2000, new or second hand.

[Sittler27]

The gamut doesn’t shift much (though it can shrink a bit after a couple of thousand hours) but the rgb balance keeps changing as the lamp ages. You could always do a color autocal now with a borrowed i1pro2 and keep checking. If your lamp has stabilized (it needs 300-500 hours) then you might be fine for a while.

Best would be to get an i1pro2 or ES2000, new or second hand.

So is the i1pro2 the same as an i1photo pro 2 (E02PHO)?  My friend has the latter as said on the box, but I think that's just the bundle - isn't it the same i1pro2 between the i1basic pro2, i1photo pro 2 and i1pubish pro 2?

[Manni]

So is the i1pro2 the same as an i1photo pro 2 (E02PHO)?  My friend has the latter as said on the box, but I think that's just the bundle - isn't it the same i1pro2 between the i1basic pro2, i1photo pro 2 and i1pubish pro 2?

No idea. There used to be differences with UV-cut between versions, but I can’t even remember which is the correct one for display calibration.

[CyberPaul]

I applied a variation to Chad method because I was concerned to work on the boundaries of the gamut.
So what I do is reading RGBW at 70,80,90 ire but with saturation 75%. I think it is more reliable because it is an area more used by the contents and far enough from VPR limits. I then interpolate to estimate xy values at 100% and then create a custom color profile. Same speculative approach for BT.2020, deriving the gaps with a proportional calculation.
Everything in excel. Very good results! Not applicable to DTM though