X-Rite's Revolutionary New Colorimeter - Page 22 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #631 of 1402 Old 09-09-2011, 08:37 AM
 
cf181user's Avatar
 
Join Date: Sep 2010
Posts: 82
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
On the Calman Certificate Of Performance it reads that the c6 was reference with the an instrument and standards "traceable" to NIST. Is this the same thing as being NIST certified?
cf181user is offline  
Sponsored Links
Advertisement
 
post #632 of 1402 Old 09-09-2011, 12:33 PM
 
cf181user's Avatar
 
Join Date: Sep 2010
Posts: 82
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by avsform1 View Post

This is true! I guess its just frustration coming out. Calman is a great product and they have great engineer like Sotti and others, but the sales team, program managment and supply chain needs some work. I think it would have been better if they let you know that it would be sometime before we received the product.

Yeah, I guess I have to apologize to CM, I didn't read the email about the shipping of my unit. It clearly stated that the case will ship later.
cf181user is offline  
post #633 of 1402 Old 09-10-2011, 11:06 AM
AVS Special Member
 
Rolls-Royce's Avatar
 
Join Date: Apr 2003
Location: Victorville, CA
Posts: 1,991
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 61
Quote:
Originally Posted by cf181user View Post

On the Calman Certificate Of Performance it reads that the c6 was reference with the an instrument and standards "traceable" to NIST. Is this the same thing as being NIST certified?

Yep.

...Royce...

"I never drink...wine."
Bela Lugosi, DRACULA, 1931
Rolls-Royce is offline  
post #634 of 1402 Old 09-10-2011, 12:13 PM
AVS Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 2,184
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 46 Post(s)
Liked: 424
Quote:
Originally Posted by cf181user View Post

On the Calman Certificate Of Performance it reads that the c6 was reference with the an instrument and standards "traceable" to NIST. Is this the same thing as being NIST certified?

This method of correcting a colorimeter is the recognized industry standard.

It was devised at the National Institute of Standards and Technology (NIST) and has been published in several peer-reviewed scientific journals. (The classic article is: Yoshi Ohno and Steven W. Brown, Four-Color Matrix Method for Correction of Tristimulus Colorimeters, Proceedings of the IS&T Sixth Color Imaging Conference, 1998).

SpectraCal is using the same process that the manufacturer (X-rite) does at the factory in a completely blacked out climate and humidity controlled room.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS + CalMAN ColorChecker
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, CalPC, ControlCAL
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #635 of 1402 Old 09-11-2011, 10:16 AM
AVS Special Member
 
bodosom's Avatar
 
Join Date: Aug 2002
Location: Niagara Frontier
Posts: 6,367
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 5 Post(s)
Liked: 21
Quote:
Originally Posted by cf181user View Post

On the Calman Certificate Of Performance it reads that the c6 [is] "traceable" to NIST. Is this the same thing as being NIST certified?

Certified mean a product was compared to a NIST "standard" (SRM). Traceable means a product was compared to a certified product. Assuming everything was done correctly it's equivalent. NIST does not verify/audit the chain of comparison.
bodosom is offline  
post #636 of 1402 Old 09-11-2011, 03:27 PM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 594
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by ConnecTEDDD View Post

This method of correcting a colorimeter is the recognized industry standard.

Four-Color Matrix Method for Correction of Tristimulus Colorimeters,

SpectraCal is using the same process that the manufacturer (X-rite) does at the factory in a completely blacked out climate and humidity controlled room.

X-Rite tries to determine the spectral sensitivity characteristics of the sensors. This is why it's a "revolutionary new colorimeter". The correction method (to even out the imperfections of the filters) is based on spectral characteristics instead of simple matrices.

I don't know what exactly SpectraCal or ChromaPure guys do in their own lab. May be they try to refine the factory spectral calibration with an additional matrix correction or may be they try to create their own spectral sensitivity based calibrations.

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list
janos666 is offline  
post #637 of 1402 Old 09-11-2011, 03:36 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,585
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 164
Quote:
Originally Posted by janos666 View Post

X-Rite tries to determine the spectral sensitivity characteristics of the sensors. This is why it's a "revolutionary new colorimeter". The correction method (to even out the imperfections of the filters) is based on spectral characteristics instead of simple matrices.

I don't know what exactly SpectraCal or ChromaPure guys do in their own lab. May be they try to refine the factory spectral calibration with an additional matrix correction or may be they try to create their own spectral sensitivity based calibrations.

We do the same thing X-Rite does for adding additional display types, using the native tools.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #638 of 1402 Old 09-11-2011, 05:18 PM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 594
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by sotti View Post

We do the same thing X-Rite does for adding additional display types, using the native tools.

So you replace the EDR files in the firmware? That sounds nice (better than creating another corrections on top of the existing factory ones and much better than using the old matrix method...).

But I still don't like the idea to create general "PDP", "LCD", etc corrections instead of start building an online display (backlight+)panel database.

Two random PDPs are more similar than a random CRT and a random LCD but there are differences between the different PDP panels as well as between different CFFL based LCD displays (different manufacturers may use different phophors for their different models...).

You are only using a very tiny bit of the possibilities of what this method can provide.

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list
janos666 is offline  
post #639 of 1402 Old 09-11-2011, 05:38 PM
Senior Member
 
avsform1's Avatar
 
Join Date: Jan 2004
Posts: 494
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 21
Ok I used a C6 to do a greyscale calibration, everytime I select the 10% reading slider for the DVDO it throws the entire reading off. I thought with this meter I would have no problems reading at 10%. Is there something I am doing wrong?

Thanks in advance.
avsform1 is offline  
post #640 of 1402 Old 09-11-2011, 06:42 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,585
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 164
Quote:
Originally Posted by janos666 View Post

So you replace the EDR files in the firmware? That sounds nice (better than creating another corrections on top of the existing factory ones and much better than using the old matrix method...).

But I still don't like the idea to create general "PDP", "LCD", etc corrections instead of start building an online display (backlight+)panel database.

Two random PDPs are more similar than a random CRT and a random LCD but there are differences between the different PDP panels as well as between different CFFL based LCD displays (different manufacturers may use different phophors for their different models...).

You are only using a very tiny bit of the possibilities of what this method can provide.

Your right that is an extremely interesting idea. We've only had all our tools for a short period of time.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #641 of 1402 Old 09-11-2011, 07:39 PM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 594
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by sotti View Post

We've only had all our tools for a short period of time.

You don't need anything else to create this database but a trusted spectrometer and physical access for many displays. It's nothing else but measuring the spectral distribution of the display primaries.

The tricky part was to describe the spectral sensitivity characteristics of the sensors (which should be done individually on every individual instruments -> or could be batch averaged to save time over quality...) but if I get it right then you didn't replace this data in the sensors (but X-Rite did it with NIST certified tools, so nothing to worry about...)

Of course it would be the best to do it with reference grade spectros in a lab but I think the traveling ISF/THX calibrators could also help with this: any time they calibrate a display which doesn't have a database entry then they have to use a spectro. If CalMan finds an entry in the database then the user can decide if he uses it or he creates his own i1d3 correction with his own spectro. If he uses his own spectro then he can decide to upload the results to the database (and next time he will be able to recall that, no need to use the spectro again for that particular display model, not even he uses a different i1d3). As more entries get uploaded, they can be automatically inspected and sorted/averaged. And as the database grows, almost every popular display models would have a trusted entry (-> no need to use spectros on the filed except some rare cases).

I think your software could ignore the EDR files in the EEPROM of the instrument and use a custom (downloaded) one. Or is this correction step hardcoded at driver level? If so, then it's a shame (for X-Rite). Even if you don't create this database, the on-the-field colorimeter corrections should be done with this method anyway (instead of the old matrix based method which is strictly bounded to the actual colorimeter).


Of course, all of this would only work with the i1d3 (at this time - but old colorimeters could also be spectrally characterzed in a lab, you just can't store the data in the instrument's EEPROM. But here comes another online database which stores the colorimeter characteristics for the serial numbers... * Doesn't it sound like "endless possibilities"? This is the "revolutionary" in this new colorimeter, the spark of these ideas...).

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list
janos666 is offline  
post #642 of 1402 Old 09-11-2011, 07:47 PM
AVS Special Member
 
Rolls-Royce's Avatar
 
Join Date: Apr 2003
Location: Victorville, CA
Posts: 1,991
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 61
Quote:
Originally Posted by janos666 View Post

So you replace the EDR files in the firmware? That sounds nice (better than creating another corrections on top of the existing factory ones and much better than using the old matrix method...).

But I still don't like the idea to create general "PDP", "LCD", etc corrections instead of start building an online display (backlight+)panel database....

You are only using a very tiny bit of the possibilities of what this method can provide.

The kind of thing you're speaking of, using display-specific corrections, has been done before, and could well prove to be unworkable today. As well as CalMAN, I own Progressive Labs' CA-6X software. At one time, there were individual screen-specific correction files that could be read into CA-6X to establish offsets for calibrating projector-based home theater setups. These were needed for a particular meter supported by the program. As you might imagine, the number of offsets shipped with the program was quite large and as other screens came on the market, more offsets were needed. Thankfully, newer and more capable meters did away with the requirement for these offsets to be provided by Progressive Labs.

...Royce...

"I never drink...wine."
Bela Lugosi, DRACULA, 1931
Rolls-Royce is offline  
post #643 of 1402 Old 09-11-2011, 08:03 PM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 594
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Yes, it's similar, but not exactly the same.

The sensor (i1d3) is spectrally characterized by X-Rite (hopefully individually, or not... but I think it's good enough, at least for home users and semi-professionals...). The display measurements should only be done for the different display models, not for every individual displays. The correction could be created on-the-fly by comparing these two (spectral sensitivity of the sensor and spectral characteristics of the display model).

Of course, it would be hell of a job to measure every display models at a lab by Spectracal (or anyone). But note the idea about the traveling agents (it could also help their own work and the upload of the results would be voluntarily...)

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list
janos666 is offline  
post #644 of 1402 Old 09-12-2011, 08:06 AM
AVS Special Member
 
Rolls-Royce's Avatar
 
Join Date: Apr 2003
Location: Victorville, CA
Posts: 1,991
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 61
Quote:
Originally Posted by janos666 View Post

Yes, it's similar, but not exactly the same.

The sensor (i1d3) is spectrally characterized by X-Rite (hopefully individually, or not... but I think it's good enough, at least for home users and semi-professionals...). The display measurements should only be done for the different display models, not for every individual displays. The correction could be created on-the-fly by comparing these two (spectral sensitivity of the sensor and spectral characteristics of the display model).

Of course, it would be hell of a job to measure every display models at a lab by Spectracal (or anyone). But note the idea about the traveling agents (it could also help their own work and the upload of the results would be voluntarily...)

Why not just use an I1Pro, CM, or other spectro to profile the colorimeter on the display you're working on? It would certainly beat waiting for SpectraCal or someone else to come up with measurements if you are an early adopter type who just has to have the newest and greatest thing to hit the stores.

...Royce...

"I never drink...wine."
Bela Lugosi, DRACULA, 1931
Rolls-Royce is offline  
post #645 of 1402 Old 09-12-2011, 01:33 PM
Senior Member
 
Jeroen1000's Avatar
 
Join Date: Nov 2008
Posts: 314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
I managed to read the entire thread, but skimmed some sections. I do still have a few questions.

Firstly, It seems that ChromaPure cannot calibrate a PC monitor? Correct?
So, Xrite has software that can, but if I purchase the CP OEM version, the Xrite software will not work? So, is there another option?
I've also come accross ArgyIIcms which does not work yet and and there there is also Icolor display 3 software from Quato. I do not know how good these alternatives are.

2nd question is regarding Window patterns. They seem to vary in size between 10% and 15%. What size are the Iscan Duo's patterns? Or how does one calculate their size? Are they suitable for a plasma device?

Additionally, how accurately can patterns be sent from a laptop instead of the VP or a test disc. I seem to be a bit wary of introducing a computer as a video device.

Thanks for your comments on this!

My gear: Panasonic TH-42PF11EK pro plasma display. -- Iscan Duo video processor -- i1 display 3 colorimeter -- i1 pro 2 spectrometer
Jeroen1000 is offline  
post #646 of 1402 Old 09-12-2011, 03:59 PM
AVS Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 3,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 23 Post(s)
Liked: 328
Quote:
Originally Posted by Rolls-Royce View Post

Why not just use an I1Pro, CM, or other spectro to profile the colorimeter on the display you're working on? It would certainly beat waiting for SpectraCal or someone else to come up with measurements if you are an early adopter type who just has to have the newest and greatest thing to hit the stores.

A C5 or D3 profiled from a Spectro will be more accurate than the C6 no matter what the display type. No exceptions.

Buzz
THX Certified Video Calibrator

 

buzzard767 is online now  
post #647 of 1402 Old 09-12-2011, 04:04 PM
AVS Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 3,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 23 Post(s)
Liked: 328
Quote:
Originally Posted by Jeroen1000 View Post


Additionally, how accurately can patterns be sent from a laptop instead of the VP or a test disc. I seem to be a bit wary of introducing a computer as a video device.

Thanks for your comments on this!

It depends on your video card output and the only way you will know for sure is to compare the computer output to a reference output. I use ChromaPure and the output is nearly identical to my Accupel DVG-5000. If I didn't have the 5000 I'd never know. Same thing goes for Blu-Ray output of content from calibration DVDs and Blu-Rays. You have to make the comparison.

Buzz
THX Certified Video Calibrator

 

buzzard767 is online now  
post #648 of 1402 Old 09-12-2011, 04:06 PM
Senior Member
 
avsform1's Avatar
 
Join Date: Jan 2004
Posts: 494
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 21
Quote:
Originally Posted by avsform1 View Post

Ok I used a C6 to do a greyscale calibration, everytime I select the 10% reading slider for the DVDO it throws the entire reading off. I thought with this meter I would have no problems reading at 10%. Is there something I am doing wrong?

Thanks in advance.

Anyone?
avsform1 is offline  
post #649 of 1402 Old 09-12-2011, 04:07 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,052
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 187
Quote:
Originally Posted by buzzard767 View Post

A C5 or D3 profiled from a Spectro will be more accurate than the C6 no matter what the display type. No exceptions.

What if the user had the same display as the one used in the lab (same model with matching panel and possibly backlighting source)? The ColorMunki and i1 PRO spectros are not quite reference accuracy.
PlasmaPZ80U is online now  
post #650 of 1402 Old 09-12-2011, 04:12 PM
AVS Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 2,184
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 46 Post(s)
Liked: 424
Quote:
Originally Posted by PlasmaPZ80U View Post

What if the user had the same display as the one used in the lab (same model with matching panel and possibly backlighting source)? The ColorMunki and i1 PRO spectros are not quite reference accuracy.

For That Display Used In The Lab it will better, for the other ~5.000 different types & models available then profiling from i1PRO will give you better performance. i1PRO for it's price gives you near reference performance.

It's approved also by THX as a Minimum Recommended Reference Tool.

The closest other option that performs better as a spectroradiometer is the 8.500$ JETI Specbos 1201.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS + CalMAN ColorChecker
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, CalPC, ControlCAL
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #651 of 1402 Old 09-12-2011, 04:19 PM
AVS Special Member
 
buzzard767's Avatar
 
Join Date: Dec 2005
Location: Naples, FL & Wausau, WI
Posts: 3,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 23 Post(s)
Liked: 328
I'm in THX school right now and my i1Pro compares very well to a Jeti 1211. This is why they're authorized for THX and ISF.

I spent over an hour at the Spectracal booth at CEDIA and Derek verified my above statement about profiling.

Buzz
THX Certified Video Calibrator

 

buzzard767 is online now  
post #652 of 1402 Old 09-12-2011, 04:47 PM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 594
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Jeroen1000 View Post

Additionally, how accurately can patterns be sent from a laptop instead of the VP or a test disc

You can output RGB (both limited and full range) very accurately from a PC.
But proper YCC output is currently impossible http://forums.amd.com/devforum/messa...&enterthread=y
And I always prefer to output YCC (preferably 4:2:2) from a player unless a bug forces me to do otherwise. Limited range RGB and standard YCC test signals should yield similar calibration settings but it's the best to use the same format for calibration and actual playback.

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list
janos666 is offline  
post #653 of 1402 Old 09-12-2011, 05:02 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,585
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 164
Quote:
Originally Posted by janos666 View Post

You can output RGB (both limited and full range) very accurately from a PC.
But proper YCC output is currently impossible http://forums.amd.com/devforum/messa...&enterthread=y
And I always prefer to output YCC (preferably 4:2:2) from a player unless a bug forces me to do otherwise. Limited range RGB and standard YCC test signals should yield similar calibration settings but it's the best to use the same format for calibration and actual playback.

That's not actually a big Deal

YCC -> RGB and back is a lossless set of math as long as you keep it in floating point.
The real issue is they don't actually do the RGB->YCC conversion correctly which leads to some green error in the low end of the grayscale.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #654 of 1402 Old 09-13-2011, 08:43 AM
AVS Special Member
 
Rolls-Royce's Avatar
 
Join Date: Apr 2003
Location: Victorville, CA
Posts: 1,991
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 61
Quote:
Originally Posted by avsform1 View Post

Anyone?

OK, I'll take a run at it. It likely isn't the C-6's fault. If you are talking about all the point gamma values changing when reading 10%, and you are following SpectraCal's recommended interactive method of starting at 100% and working down to the low end (0 or 10%), this is quite common. I see it from time to time, and I'm using either an I1Pro or a C-5 profiled to it, not a C-6. In this case, IMHO, the change comes because that low-end reading is significantly different from the predicted reading. The lowest reading (0 or 10% depending on program settings) gets factored into the overall gamma calculation along with the 100% reading and thus affects the results. Until you read in the actual value, the program appears to be using a predicted target value calculated on the 100% reading and your desired gamma. It's one reason why you should go back and do another full run as recommended by SpectraCal after all your adjustments are made.

...Royce...

"I never drink...wine."
Bela Lugosi, DRACULA, 1931
Rolls-Royce is offline  
post #655 of 1402 Old 09-13-2011, 10:02 AM
Senior Member
 
Jeroen1000's Avatar
 
Join Date: Nov 2008
Posts: 314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
I see, I could compare Chromapure, the AVS-disc and the iscan duo. When using the same pattern (at the same stimulus) all readings _should_ be (nearly)identical.

However, the size of the window patterns in CP, AVS, and the Duo will probably not be identical. Is this going to be an issue?

My gear: Panasonic TH-42PF11EK pro plasma display. -- Iscan Duo video processor -- i1 display 3 colorimeter -- i1 pro 2 spectrometer
Jeroen1000 is offline  
post #656 of 1402 Old 09-13-2011, 11:05 AM
AVS Special Member
 
Rolls-Royce's Avatar
 
Join Date: Apr 2003
Location: Victorville, CA
Posts: 1,991
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 61
Quote:
Originally Posted by Jeroen1000 View Post

I see, I could compare Chromapure, the AVS-disc and the iscan duo. When using the same pattern (at the same stimulus) all readings _should_ be (nearly)identical.

However, the size of the window patterns in CP, AVS, and the Duo will probably not be identical. Is this going to be an issue?

It depends on your display and its circuitry. Window patterns used on plasmas need to be small enough that they don't trip the set's brightness limiters, which would throw off the results. IIRC, up to full-screen can be used with standard LCDs (not sure about the LED-lit ones). On my CRT RPTV, I've used various sizes of windows, all with good results.

If you are using a disc player, its output can also be a factor. Windows from the AVS AVCHD and Get Gray discs measure brighter from my XA-2 HD DVD player than they do from my Panasonic 210 Blu-Ray player with both players' picture settings at neutral, for example.

...Royce...

"I never drink...wine."
Bela Lugosi, DRACULA, 1931
Rolls-Royce is offline  
post #657 of 1402 Old 09-13-2011, 02:50 PM
Newbie
 
fredtoft's Avatar
 
Join Date: Mar 2002
Location: Denmark
Posts: 11
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by janos666 View Post

You can output RGB (both limited and full range) very accurately from a PC.
But proper YCC output is currently impossible http://forums.amd.com/devforum/messa...&enterthread=y
And I always prefer to output YCC (preferably 4:2:2) from a player unless a bug forces me to do otherwise. Limited range RGB and standard YCC test signals should yield similar calibration settings but it's the best to use the same format for calibration and actual playback.

Quote:
Originally Posted by sotti View Post

That's not actually a big Deal

YCC -> RGB and back is a lossless set of math as long as you keep it in floating point.
The real issue is they don't actually do the RGB->YCC conversion correctly which leads to some green error in the low end of the grayscale.

I would disagree with that.

As Janos wrote: If you calibrate with a PC source sending out RGB, you bypass the color decoder in the display device, and any errors it may introduce. Then when you connect your "real" source device you'll usually be sending out YCbCr 4:2:2, and might(!) get a different result. It may not be a problem, but there's a chance it will be.

Also, I would expect YCC -> RGB conversion to only be lossless if you do both in 4:4:4?

Or am I missing something?

My Danish blog about Calibration, Flatscreens and Home Theater: www.av-blog.dk
fredtoft is offline  
post #658 of 1402 Old 09-13-2011, 02:53 PM
Senior Member
 
Jeroen1000's Avatar
 
Join Date: Nov 2008
Posts: 314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
My color decoders in both TV and VP are accurate. But it is a good idea to make sure this is true nonetheless.

My gear: Panasonic TH-42PF11EK pro plasma display. -- Iscan Duo video processor -- i1 display 3 colorimeter -- i1 pro 2 spectrometer
Jeroen1000 is offline  
post #659 of 1402 Old 09-13-2011, 03:03 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,585
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 164
Quote:
Originally Posted by fredtoft View Post

I would disagree with that.

As Janos wrote: If you calibrate with a PC source sending out RGB, you bypass the color decoder in the display device, and any errors it may introduce. Then when you connect your "real" source device you'll usually be sending out YCbCr 4:2:2, and might(!) get a different result. It may not be a problem, but there's a chance it will be.

Also, I would expect YCC -> RGB conversion to only be lossless if you do both in 4:4:4?

Or am I missing something?

I didn't say the difference between RGB and YCC isn't significant. It most certainly is. My TV behaves very differently between the two.

I said the video card decoding mpeg content to RGB then re-encoding back to YCC while it's all in floating point in the video card buffer isn't a big deal. RGB doesn't have a 4:4:4 concept. All YCC (BD, DVD, cable, ect..) sources are 4:2:0 or 4:2:2. But if you did start with 4:4:4 converted to RGB, then pushed it back out as 4:2:2 then you would give up some color resolution. But then there would have be reason why you converted to 4:2:2, right? Whatever that reason was would likely have prevented you from using 4:4:4 content to begin with.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #660 of 1402 Old 09-13-2011, 04:15 PM
Newbie
 
fredtoft's Avatar
 
Join Date: Mar 2002
Location: Denmark
Posts: 11
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by sotti View Post

I didn't say the difference between RGB and YCC isn't significant. It most certainly is. My TV behaves very differently between the two.

I said the video card decoding mpeg content to RGB then re-encoding back to YCC while it's all in floating point in the video card buffer isn't a big deal. RGB doesn't have a 4:4:4 concept. All YCC (BD, DVD, cable, ect..) sources are 4:2:0 or 4:2:2. But if you did start with 4:4:4 converted to RGB, then pushed it back out as 4:2:2 then you would give up some color resolution. But then there would have be reason why you converted to 4:2:2, right? Whatever that reason was would likely have prevented you from using 4:4:4 content to begin with.

In that case I don't disagree anyway

I just didn't want anybody not familiar with the difference to conclude that it's not a big deal to calibrate with a PC RGB source, and then use a YCbCr 4:2:2 output from their DVD/BRD player to watch movies.

That is what I took Janos' post to be about, and I've seen it done far too many times with unpredictable (but rarely positive) results. I'm sure you're correct about there being no problem doing it all in the video card buffer.

My Danish blog about Calibration, Flatscreens and Home Theater: www.av-blog.dk
fredtoft is offline  
Reply Display Calibration

User Tag List

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