or Connect
AVS › AVS Forum › Display Devices › Display Calibration › Lumagen Announces 9-Point Cube Gamut (729 Color Points) for the Radiance 20XX
New Posts  All Forums:Forum Nav:

Lumagen Announces 9-Point Cube Gamut (729 Color Points) for the Radiance 20XX - Page 2

post #31 of 136
Quote:
Originally Posted by ConnecTEDDD View Post

Also they are all using CalMAN for different reasons, like 1D LUT Calibrations or by preparing the display for LightSpace 3D Cube Profilings.

I'm also using CalMAN's autocal for the 1D gamma cal of my JVC PJ via DDC, and LS for the big LUT cal of video processors.

CalMAN has lot's of indicators to check any kind of measured data. Maybe too much.
But because of it's "target approximation" autocal system / each cal point, I have only limited use for CM.

An autocal of 3D LUTs with thousands of points who needs to run over many, many hours isn't working for me.
Even if I start calibrating 2 hrs after powering ON, my PJ's lamp isn't stable enough in CCT and brightness over the next 3 hours.
But if I go calibrating my PJ, then the result should be as best as possible.

So for making big calibrated LUT's I'm using LS.
LS is fast because of it's profiling concept - each point need's to get only 1-times measured.
Using the final profiling data, I can play with parameters like different gamma coefficients or different targets like BT601/709 for making a new LUT, as many times I want, w/o having the need to run the entire measurement / autocal process again.
If desired, a final quick profiling can be done to check the result.


Every SW solution has it's pro and cons and we just should respect everybody's own decision, what's best for him in his case.
To be honest, I've learnd a lot about PJ calibration using CalMAN since V3. Had lot of fun with ... not only with it's bugs ... wink.gif

So I'm using CM+LS in combination, playing with both is part of my crazy hobby ... sitting next to a ****ing expensive PJ, watching dump-making test patterns over hours and hours ... weird ...
biggrin.gif
Edited by *Harry* - 9/18/13 at 4:27pm
post #32 of 136
Quote:
Originally Posted by Mark_H View Post

Indeed, and that strikes me as a very foolish position to adopt. We have industry standards, we have objective measurements as well as quantitative methods for assessing image quality. Why simply throw that away and rely on "gut instinct"? Herd instinct? Baaaaaa.

Ignore dE. Has anybody actually compared the xyY values generated by CalMAN/Lightspace/any other method. They should ALL be coming out with *very* similar numbers... based on the standards requirements. if not, THEN we can have a discussion about which is right and which is wrong...

dE is an attempt to indicate color accuracy... a good start, that is once you've made a decision which of the MANY dE formulas currently works for you and your content... rolleyes.gif

It is limited in telling the whole story on IQ....... --> IMAGE QUALITY....

image depth, perceived resolution, clarity, sharpness, color popping etc... are not being described by dE.
post #33 of 136
Quote:
Originally Posted by *Harry* View Post

Even if I start calibrating 2 hrs after powering ON, my PJ's lamp isn't stable enough in CCT and brightness over the next 3 hours.

I'm currently a CP user and been debating LS - in researching LS, I've seen comments like this several times and the comments confuse me...

If the projector is not stable after 2 hours, then most viewing will be with this level of instability and we have to accept variations in our measurements are part of the unit. The best we could do would be to take multiple full sets of readings and average to deal with long-term drifts...or use longer integration times for short-term drifts.

Is it ideal? No but it's a limitation of the device. No matter how good our profile is, it will be "wrong" at most points in the drift...so is one measurement better than another?

If the projector's drift creates an unacceptable variation, no amount of better calibration technique will fix it and the only real option is to replace it...(which would likely be a little extreme for most HT use)...

On a related side note, the LightSpace support forums currently have a discussion on-going about LS settings for the i1d3 with a JVC projector.
post #34 of 136
Using LightSpace (latest software) to create a 17^3 (4913 point) profile, that is then covered to a 9^3 LUT (729 points) cube for upload to the Lumagen 2041 works very well.

The whole process from profiling color/spectro meters, doing a 17^3 profile using LS, to uploading the 729 point LUT using LS took me about 2 hours. I already had my manual 10 point grayscale done and set in my VT60 (plasma) the night before.

The PQ/IQ is very good. Is it any better than using a eecolor box and a full 17^3 LUT, to early to tell.

ss
post #35 of 136
Quote:
Originally Posted by ConnecTEDDD View Post


BTW If you check the best 3D LUT Solutions of the pro market, none of them feature any detailed (Luminance-Saturation-Color Checker) type of dE Chart, there many other 3D LUT Solutions except TrueLight/THX CineCube/LightSpace CMS that are coming from Professional industry, there they don;t care about Detailed dE Reports/Charts becasue they trust the software.

So If the PRO Industry doesn't care about dE Detailed Reports, here at cunsumer industry, why we are so addicTED with them?

It is my experience that "pros" are not necessarily the most critical, most knowledgeable, or even the most up-to-date users. Some (rare) pros might be, but many are so invested in the way they have been doing things for years or decades that newer things go right over their heads. I had a pro painter here who was going to make a line between 2 rooms along a textured wall. He started laying out the line with normal blue painter's tape. I asked him why he wasn't using green "frog tape" and he'd never heard of it (it's about 7 to 10 years old by now). So I got what I had left, had him use that for the line and when he was done he agreed that there was no way the blue tape line would have been anywhere near as "clean" as the frog tape line. When I was working on professional imaging systems, I can't tell you how often I had to explain to imaging "pros" that the mottle they were seeing in high magnification of the scans they were making was not digital noise from the scanner, but was actually the grain of the film. I showed a pro mechanic how to heat a bolt with a propane torch to get it to come right off when liquid wrench had no effect even after waiting for 24 hours. Sure there are pro mechanics who know about heating a bolt with a torch, and that they are seeing film grain and not digital noise, and that frog tape makes sharper lines than painter's tape... but as it turns out, pros that really know their field up one side and down the other are pretty rare.
post #36 of 136
Quote:
Originally Posted by Doug Blackburn View Post

It is my experience that "pros" are not necessarily the most critical, most knowledgeable, or even the most up-to-date users. Some (rare) pros might be, but many are so invested in the way they have been doing things for years or decades that newer things go right over their heads. I had a pro painter here who was going to make a line between 2 rooms along a textured wall. He started laying out the line with normal blue painter's tape. I asked him why he wasn't using green "frog tape" and he'd never heard of it (it's about 7 to 10 years old by now). So I got what I had left, had him use that for the line and when he was done he agreed that there was no way the blue tape line would have been anywhere near as "clean" as the frog tape line. When I was working on professional imaging systems, I can't tell you how often I had to explain to imaging "pros" that the mottle they were seeing in high magnification of the scans they were making was not digital noise from the scanner, but was actually the grain of the film. I showed a pro mechanic how to heat a bolt with a propane torch to get it to come right off when liquid wrench had no effect even after waiting for 24 hours. Sure there are pro mechanics who know about heating a bolt with a torch, and that they are seeing film grain and not digital noise, and that frog tape makes sharper lines than painter's tape... but as it turns out, pros that really know their field up one side and down the other are pretty rare.

And to that end, go look at any work being done on color science in academia. You will find DeltaE 2000 is the predominate way they communicate about the accuracy of color.
post #37 of 136
Thread Starter 
ChromaPure 2.4.4.36472 Released

Download Link

Release Notes

New Features

Added support for the Lumagen 729-point gamut calibration. This feature is available for free to existing Advanced Auto-calibrate users. The 729-point calibration is only available on the Lumagen Radiance 20xx series. Earlier versions of the Lumagen hardware do not support this.

Enhancements

Improved the scaling on the Excel reports and also added support for reporting of BT.1886 gamma.

Note: Because the number of color points is so much larger, the amount of time required to complete the auto-calibrate session is much longer. Lumagen 20xx users should count on about 3 hrs. for the entire process to finish.
post #38 of 136
Quote:
Originally Posted by ConnecTEDDD View Post

ChromaPure 2.4.4.36472 Released

Download Link

Release Notes

New Features

Added support for the Lumagen 729-point gamut calibration. This feature is available for free to existing Advanced Auto-calibrate users. The 729-point calibration is only available on the Lumagen Radiance 20xx series. Earlier versions of the Lumagen hardware do not support this.

Enhancements

Improved the scaling on the Excel reports and also added support for reporting of BT.1886 gamma.

Note: Because the number of color points is so much larger, the amount of time required to complete the auto-calibrate session is much longer. Lumagen 20xx users should count on about 3 hrs. for the entire process to finish.


That zip file still has 2.4.3 executable in it.

update:
A splash screen comes up with 2.4.3, but in the about help it comes up as 2.4.4
Edited by avsform1 - 9/18/13 at 4:10pm
post #39 of 136
Thread Starter 
Quote:
Originally Posted by avsform1 View Post

Quote:
Originally Posted by ConnecTEDDD View Post

ChromaPure 2.4.4.36472 Released

Download Link

Release Notes

New Features

Added support for the Lumagen 729-point gamut calibration. This feature is available for free to existing Advanced Auto-calibrate users. The 729-point calibration is only available on the Lumagen Radiance 20xx series. Earlier versions of the Lumagen hardware do not support this.

Enhancements

Improved the scaling on the Excel reports and also added support for reporting of BT.1886 gamma.

Note: Because the number of color points is so much larger, the amount of time required to complete the auto-calibrate session is much longer. Lumagen 20xx users should count on about 3 hrs. for the entire process to finish.


That zip file still has 2.4.3 executable in it.

update:
A splash screen comes up with 2.4.3, but in the about help it comes up as 2.4.4

Don't worry about this, Tom hasn't update the setup release notes (no big deal), the installed software version is the 2.4.4
Edited by ConnecTEDDD - 9/18/13 at 4:46pm
post #40 of 136
Quote:
Originally Posted by ConnecTEDDD View Post

Don't worry about this, Tom hasn't update the setup release notes (no big deal), the installed software version is the 2.4.4
That's correct. The software is the correct version. I just neglected to update the Release Notes in the installation program.
post #41 of 136
Thread Starter 
Lumagen has update their Owner's Manual for Radiance 2021/2041/2022/2042 Video Processors.

Download Link
post #42 of 136
Thread Starter 
Lumagen Radiance XE/XE+/XE3D/XE/XE+/XE3D/Mini-3D/XD/XD3D + Radiance 20XX Series (2021, 2022, 2041, 2042) New 091013 Firmware Update

Release Notes

Fixes 480p and 576p reinterlacing and may clean up other video mode detection issues ( broken from 0618-0903. )

Download Link
post #43 of 136
Thread Starter 
Lumagen Radiance 20XX Series (2021, 2022, 2041, 2042) New 092013 Firmware Update

Release Notes

Fixes an audio issue which in some setups could result in missing channels for multi-channel PCM. Fix for audio only on output1 when using 4K video on output2. This could result in a couple of problems--no audio from output 1 depending on the AVR and occasional stutter on the output 2 used for video if the AVR was having an issue.

Download Link
post #44 of 136
Thread Starter 
Lumagen has update their Tech Tips for the Radiance Video Processors.

- Tip 4 - Automatic switching of output resolutions
- Tip 7 - Setup with a fixed anamorphic lens
- Tip 19 - Installation Notes


Download Link
post #45 of 136
Thread Starter 
Lumagen has update their Tech Tips for the Radiance Video Processors.

- Tip 4 - Automatic switching of output resolutions
- Tip 7 - Setup with a fixed anamorphic lens
- Tip 19 - Installation Notes


Download Link
post #46 of 136
Maybe a silly question but could anybody already tell about any visible improvements due to 9^3 in comparison to 5^3? I do have a JVC X30 and wondering if it is worth going in that direction (lumagen X vs LUMAGEN 20XX)
post #47 of 136
Quote:
Originally Posted by westmd View Post

Maybe a silly question but could anybody already tell about any visible improvements due to 9^3 in comparison to 5^3? I do have a JVC X30 and wondering if it is worth going in that direction (lumagen X vs LUMAGEN 20XX)

Hello.

I confirm that there is an improvement. I've checked it through the calman colorchecker (but calibration was done with Light Space CMS).

But I think the improvement depends on the display.
post #48 of 136
Quote:
Originally Posted by alex_t View Post

But I think the improvement depends on the display.

Very true. I think all X30/35 units can get an perceivable improvement by using 9^3 instead of 5^3 LUT cal points only.
But speaking for me (X75 owner) I have no real need to upgrade from my XS to a 2041 unit; my X75's gamut isn't that bad out-of-the-box.

If I would swap over to a 2041 then just for fun, because I like to play with cal SW.
But I'm not really excited about the "old" HDMI 1.4 inputs on the "new" 20xx units ...
post #49 of 136
Quote:
Originally Posted by *Harry* View Post

Very true. I think all X30/35 units can get an perceivable improvement by using 9^3 instead of 5^3 LUT cal points only.
But speaking for me (X75 owner) I have no real need to upgrade from my XS to a 2041 unit; my X75's gamut isn't that bad out-of-the-box.

If I would swap over to a 2041 then just for fun, because I like to play with cal SW.
But I'm not really excited about the "old" HDMI 1.4 inputs on the "new" 20xx units ...

About HDMI: new JVC's projector are based on HDMI 1.4 chipset.

20XX family offers also the Darbee processing.

I think that you would see a real improvement by upgrading from XS to a 2022 or 2042 (if you need 4K).
post #50 of 136
In my case I don't have a Lumagen at all yet but had my X30 calibrated directly by the retailer. So now i could go for a cheap mini or spend some more money for on in tehe 20XX series.
post #51 of 136
Quote:
Originally Posted by alex_t View Post

About HDMI: new JVC's projector are based on HDMI 1.4 chipset.
Maybe next year they will have real 4k wiregrids and HDMI2.x. ... we'll see.

Quote:
Originally Posted by alex_t View Post

20XX family offers also the Darbee processing.
Yes, but I like the JVC's eShift2+MPC also. Should be quite the same in picture improvement.

Quote:
Originally Posted by alex_t View Post

I think that you would see a real improvement by upgrading from XS to a 2022 or 2042 (if you need 4K).
Well, maybe if the price would be ok, I could swap over to a 2041 (need no analog input)

smile.gif
post #52 of 136
Quote:
Originally Posted by *Harry* View Post

Maybe next year they will have real 4k wiregrids and HDMI2.x. ... we'll see.
Yes, but I like the JVC's eShift2+MPC also. Should be quite the same in picture improvement.
Well, maybe if the price would be ok, I could swap over to a 2041 (need no analog input)

smile.gif

Each to their own. I find MPC is poor and just makes the image look noisy unless disabled. Say what you will about darbee processing, it doesn't make the image look noisy....

The other thing to remember about the 20x units is that they have more memory and much larger fpga than the mini/xs/xd/xe range. Remember how many improvements and features got added to those units in the 6? years that base has been out.. (And they still have more to add)....i'd expect that there is still some scope for even more performance enhancements on the 20xx series which have only been available for around 6 months.

either way though, a mini or a 20xx is going to make any JVC DILA look a lot nicer than if you didn't have one....
post #53 of 136
Thread Starter 
Lumagen Radiance XE/XE+/XE3D/XE/XE+/XE3D/Mini-3D/XD/XD3D + Radiance 20XX Series (2021, 2022, 2041, 2042) New 091013 Firmware Update

Release Notes

Fixes 480p and 576p reinterlacing and may clean up other video mode detection issues ( broken from 0618-0903. )

Download Link
post #54 of 136
Quote:
Originally Posted by Gordon Fraser View Post

Each to their own. I find MPC is poor and just makes the image look noisy unless disabled. Say what you will about darbee processing, it doesn't make the image look noisy....

The other thing to remember about the 20x units is that they have more memory and much larger fpga than the mini/xs/xd/xe range. Remember how many improvements and features got added to those units in the 6? years that base has been out.. (And they still have more to add)....i'd expect that there is still some scope for even more performance enhancements on the 20xx series which have only been available for around 6 months.

either way though, a mini or a 20xx is going to make any JVC DILA look a lot nicer than if you didn't have one....

Agree
post #55 of 136
Thread Starter 
Quote:
Originally Posted by westmd View Post

Maybe a silly question but could anybody already tell about any visible improvements due to 9^3 in comparison to 5^3? I do have a JVC X30 and wondering if it is worth going in that direction (lumagen X vs LUMAGEN 20XX)

There is a 3D Cube Presentation Comparison here, to visualize the calibrated color point differencies of every available cube size.
post #56 of 136
I am finding the 21XX does a very nice job for 9^3 LUT cube imo better than the 5^3 cube, using LS to do a 17^3 profile.
Also I would suggest when you do the conversion in LS you use peak Chroma not Luma.

ss
post #57 of 136
Thread Starter 
Quote:
Originally Posted by sillysally View Post

I am finding the 21XX does a very nice job for 9^3 LUT cube imo better than the 5^3 cube, using LS to do a 17^3 profile.
Also I would suggest when you do the conversion in LS you use peak Chroma not Luma.

ss

SS, it was to do with the display/pre-calibraton settings.

With LightSpace you have a lot of options after you have performed a 17-Point Cube Profiling (4.913 Color Points) with 4.913 Meter Reads of your display.

LightSpace separates profiling from calibration, after the 17-Point Cube Readings, it builds a model for your display capabilities and you can generate 10-20 different correction LUT's in 1 minute, with any target, any colorspace etc...... No need to re-measure again your display.
Edited by ConnecTEDDD - 10/23/13 at 3:17pm
post #58 of 136
Quote:
Originally Posted by Franin View Post

Hi all I upgrade to Calman 5 enthusiast license and downloaded the latest beta software Calman 5.1.2 build 1257. I have the Lumagen 2041 and now I cannot seem to get Calman 5 to talk to the lumagen it won’t recognize it. I checked all cables they’re fine( how do I know they’re fine, you ask?) I downloaded the Lumagen Configuration file to my PC with no issues at all using Rs232 to USB with a USB ext lead. I had no issues a few months back( April May ), Echo is ON im just confused to as why its not communicating. Any ideas? I have tried on both Win 7 and Win 8 no luck at all.

Thank you

Frank

Quote:
Originally Posted by Franin View Post

I gave it a try with the latest calman 5 update and updated my USB FTDI. Still no luck but still able to download lumagen configuration to PC confused.gif

Any ideas??
Thank you
post #59 of 136
Quote:
Originally Posted by Franin View Post


Any ideas??
Thank you

IIRC, CalMAN does its connection to Radiance models at 9600 baud. Perhaps your adapter is not negotiating to that rate as it should. Try setting the port speed manually to 9600 baud in Device Manager > Ports before starting CalMAN. If that fails, try changing flow control to None, with data bits to 8, parity to None, and stop bits to 1. Even if they already are set that way, it's possible one of them is corrupted in the Registry. Won't cost anything to try!

Gawd, I remember having to do all this kind of stuff on the old RS-232 external modems. I'm so glad those days are over!tongue.gif
post #60 of 136
So... Radiance 2041. CalMAN 5.2 current release version.

3D LUT Workflow selected

Connects to Radiance with no problem

But there's no setting I see anywhere that lets you tell CalMAN if it is an "old" Radiance that will only do 5x5x5 matrix or a "new" Radiance that will do 9x9x9... I did not actually try running a matrix cal yet, but want to do it. Will it automatically do 9x9x9 matrix for new Radiace processors? What if I just want a 5x5x5 matrix with a new Radiance (not 9x9x9) .

Am I missing something? I see a drop down window in CalMAN but there's nothing in that window...blank.

Have done 5x5x5 matrix with an XE 3D several times with no problem... I was expecting some kind of 5x5x5 / 9x9x9 selector somewhere... or perhaps a way of choosing Radiance 20XX series versus XE, XS, Mini but don't see either of those unless I'm missing them somewhere.
Edited by Doug Blackburn - 10/24/13 at 3:43pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › Lumagen Announces 9-Point Cube Gamut (729 Color Points) for the Radiance 20XX