82 replies to this topic

[alanh]

All,
Digital Video Essentials.

All of the following tests and setups come with a demonstration and descriptions of how to use them.

This disc tells you how to set up a viewing enviroment

This is the only disc I know of which enables the viewer to adjust the brightness and contrast for the lighting conditions you are actually using.

Test the geometry and size of the image

Image sharpness test

Tests the sound system for frequency response, balance for stereo and 5.1 systems, identifing if all channels are equal and are identified as well as a rattle test. Subwooffer phase test.

With the use of a supplied optical colour filter, thest the matrix which separates the signals from the Y, P_r & P_b signals back to the Red, Green & blue signals for display. This is to ensure that the colour of strong colours are correct.

A test of how how well the processing in the display, scaler and Blu-ray player work.

The relative timing of sound and picture which is adjustable in some Home Theater Amplifiers.


Demonstration of the highest quality images from NASA and in a restaurant so you can see how good a display is on real pictures!

The advantage of these disks is that you know that they have gone to extreme efforts to make the highest quality pictures and sound. These discs are better than you generating test signals as well as it tests the Blu-ray player


There is Blu-Ray(HD)  DVD-PAL (SD) versions available

Google "Digital Video Essentials" and restrict the search to Australia to find out where you can get it and how much it will cost.

When you buy a display the cost of this disc is around 2 % of the cost of the display. It is worth it to get the best pictures and sound.

Please note: You only need to set your display and sound system. All sources for example all players, Set Top Boxes, Pay TV receivers and Personal Video Recorders do not alter any video or sound data provided you use the HDMI connections.

The only sources which require adjustment prior to recording are cameras and analog Videocassette replays.

If you are using a graphics card, adjust the display. This will make it correct for all other sources, then adjust the graphics card match (Set the graphics card to default positions before you adjust the graphics card.)


AlanH

[kwarrior]

Great Stuff Alan..

Also thought you might want to comment on the X-rite Eye One Display 2 calibration tool ?

[alanh]

kwarrior,
This is a light meter connected to the computer producing the image. It is principally designed for the graphics artist and would also have use in mulitple monitor situations such as TV studio control rooms.
It would not be able to be used for a projected image because the sensor is held on a self illuminating screen and not a reflective screen.

The DVE disc and filters is much more suitable for projectors because;
1. You have a single large image. This means you are not comparing the image against others and the eye/brain automatically readjusts the perception so white looks white again.
2. The geometric and focus tests can be used to accurately remove picture geometry and poor focus.
3. the X-rite Eye One Display 2 only tests the colorimetry and does not test the sound.

AlanH

[kwarrior]

alanh, on Jun 15 2009, 12:56 PM, said:

kwarrior,
This is a light meter connected to the computer producing the image. It is principally designed for the graphics artist and would also have use in mulitple monitor situations such as TV studio control rooms.
It would not be able to be used for a projected image because the sensor is held on a self illuminating screen and not a reflective screen.

The DVE disc and filters is much more suitable for projectors because;
1. You have a single large image. This means you are not comparing the image against others and the eye/brain automatically readjusts the perception so white looks white again.
2. The geometric and focus tests can be used to accurately remove picture geometry and poor focus.
3. the X-rite Eye One Display 2 only tests the colorimetry and does not test the sound.

AlanH

While looking to buy the Display 2 recently, I found (cant recollect where)...that somebody had a tripod set up with the Eye One focused towards the screen...
Obviously, you dont believe that that would be an ideal way to calibrate the PJ, do you?

[DrP]

Having used x-rite gear to calibrate other colour devices, I can safely say that if one had the money (and could tinker with the display settings to a high degree) that the x-rite would yield far superiour results....... but don't let that get in your way, alanh.  

[Mr.Bitey]

alanh, on Jun 15 2009, 12:26 PM, said:

kwarrior,
This is a light meter connected to the computer producing the image. It is principally designed for the graphics artist and would also have use in mulitple monitor situations such as TV studio control rooms.

You can also use them to calibrate projectors, and televisions (lcd, plasma, crt, rear projection)....

Quote

It would not be able to be used for a projected image because the sensor is held on a self illuminating screen and not a reflective screen.

They can be used for projectors, via a tripod (or other device to hold em still)

Quote

The DVE disc and filters is much more suitable for projectors because;
1. You have a single large image. This means you are not comparing the image against others and the eye/brain automatically readjusts the perception so white looks white again.

DVE doesnt set the colour of white/grey - just saturation.. which is completely subjective. A color meter, and access to the service menu- or advanced user menu, lets you set the colour of grey properly, across the IRE range - not just setting contrast, brightness and rudementary colour saturation as with DVE..

Quote

2. The geometric and focus tests can be used to accurately remove picture geometry and poor focus.

Its only saving grace..... however similar patterns are also available on THX dvds...

Quote

3. the X-rite Eye One Display 2 only tests the colorimetry and does not test the sound.

Thats right, but without buying a SPL meter, your pretty much flying blind anyway
You can do a pretty good job with the AVRs test-tone...

Im not saying DVE is crap..... only its not a magic bullet, and you can get similar results without spending much $ at all..
Alternatively, spend more, and get a colormeter and do it properly (video); and a SPL meter and set that properly too

Or by the time youve bought both, your about 1/4-1/3 of the way of having someone come and do it professionally for you...

Cheers,
Bitey

Edited by Mr.Bitey, 15 June 2009 - 02:05 PM.

[alanh]

All,
When Colour TV first started the way to calibrate multiple studio monitors was to use a calibrated Illuminant D6500K light source which is reflected into your eye by a 45 degree mirror, which contains a hole so that you can see the screen. You have an adjustable iris to make the brightness the same. This meant when the hole "disappears" the colour is identical. Since there is a variable iris you can match whites and the shades of grey. The image is PLUGE (Picture Line Up Generator) which is shown on the DVE disc.

The eye is much more critical compared to meters.

Yes you can put the xrite sensor on a tripod, but you must ensure that the rest of the room is in complete darkness. It would appear that this device has no lens so you have to ensure that the sensor is only "seeing" a particular part of the image.

If you wish to measure the luminance of the screen you need to use a luminance type light meter which measures in cd/m². The value that the instructions for Xrite are for self illuminating screen images which is 100 cd/m² projectors may be higher depending on the light source, the screen and the distance between the projector and the screen along with the amount of zoom.

As far as testing audio goes not only do you need to measure the sound pressure level you also need to be able to measure the reverberation time at various frequency bands.

DVE can be used for setting white and black level because it has the PLUGE signal along with a gamma corrected step grey scale. In digital systems generally do not have saturation controls because the digital system will not have gain variations between the luminance and the P_r & P_b signals. The main thing now which affects saturation is the actual colours used in the display for the Red, Green and Blue primary colours.

The Xrite sensor is about 10 times the price of the DVE disc.

Mr. Bitey
I do not wish to set the IRE levels because the levels are set digitally a the studios. The display when fed with HDMI is fed with these digital signals from the player. The display does the conversion back to analog and the only controls are those on the display.
As the instructions in the Xrite say set the graphics board in computers to their default position.  This should also apply to things like dynamic contrast in displays. You cannot measure digital signals in IRE units thats only in analog.

I did not say that you cannot use the Xrite on LCD displays you can but it is less critical if you are using a single display for watching TV. It is a different story using a display to decide colours in graphic design using a computer.

AlanH

[DrP]

alanh, on Jun 15 2009, 02:18 PM, said:

blah blah blah blah

Clearly all spoken by a person that has never performed professional colour calibration, rather reads from a text.  BTW, a human eye can distinguish between roughly 10 bits resolution (that's 10 bits per R, G, B channel)* ie, put two colour panels side by side and vary the LSB and that is the level at which a perfect human eye can distinguish.  Commercial colour calibration equipment is far beyond that.


_{*I know some people will dispute this and say that humans can distinguish about 17 million colours, but this is not what is being illustrated}

Edited by DrP, 15 June 2009 - 02:27 PM.

[Mr.Bitey]

You absolutely need to adjust the colour of grey... its wrong on most displays. And no manner of fiddling with saturation, contrast or brightness will fix it. Whether you feed it HDMI, Component, RGB, s-video or composite it will still be wrong. They might be right on the disc, but theyre likely gunna be wrong on the player and definatly wrong on the display.

You only 0 the video card, if your calibrating a computer display attached to the computer, and its a dirty way to calibrate it - using the video card to digitally fix the flaws in the display :shame shame:. Its much better to fix the display properly (its why displays have adjustments like RGB GAIN and BIAS - yes, even digital displays) then use the video card to make fine adjustments if necessary - its the same way you calibrate a display with a scaler.... you make the fine adjustments in the scaler last after calibrating the display, using the displays controls..

The pluge pattern on DVE is only useful for setting brightness. Which in simple terms has nothing to do with the colour of grey*

*I know it does really, but you cant fix it by changing the brightness/contrast.

Cheers,
Bitey

[Chopsus]

alanh, on Jun 15 2009, 12:56 PM, said:

3. the X-rite Eye One Display 2 only tests the colorimetry and does not test the sound.

The THX Setup on many DVD's (Star Wars, Reign of Fire) will do your sound (and plunge etc) for Free, as well as many other resources available on the web.

Ditto what MrBitey said.

Chops

[flukeyluke]

Sounds like I need it done. Who wants to come and do mine?

[TechWizNot]

Mr.Bitey, on Jun 15 2009, 01:41 PM, said:

Or by the time youve bought both, your about 1/4-1/3 of the way of having someone come and do it professionally for you...

Cheers,
Bitey

I have read a number of times on other forums that it is necessary to have a professional calibration done evey 100 to 200 hours or so as the lamp ages. Is this correct.
If so, would it not be worth buying one's own equipment and learning how to do a reasonable effort on one's projector. I could not afford to have it calibrated that many number of times professionally over the life of the lamp. I am pretty sure the her indoors would kill me.

Cheers.

[Chopsus]

flukeyluke, on Jun 15 2009, 02:54 PM, said:

Sounds like I need it done. Who wants to come and do mine?

Your Yamaha's YPAO *should* do a superior job to an aural THX job .... just use it to confirm that what the YPAO did is right (sometimes YPAO can be flakey .... I'm happy with mine though, after some tinkering on the LFE).

[DrP]

TechWizNot, on Jun 15 2009, 03:15 PM, said:

I have read a number of times on other forums that it is necessary to have a professional calibration done evey 100 to 200 hours or so as the lamp ages. Is this correct.

Technically correct, but TBH I doubt anyone here is that comitted to a viewing experience.  If it looks OK during normal viewing, then that's good enough (IMO).

[Chopsus]

TechWizNot, on Jun 15 2009, 03:15 PM, said:

I have read a number of times on other forums that it is necessary to have a professional calibration done evey 100 to 200 hours or so as the lamp ages. Is this correct.
If so, would it not be worth buying one's own equipment and learning how to do a reasonable effort on one's projector. I could not afford to have it calibrated that many number of times professionally over the life of the lamp. I am pretty sure the her indoors would kill me.

Cheers.

I'd say 300-500 hours and owning a I1D2 makes it a lot easier to confirm what you *think* you're seeing .... I've tinkered with the greyscale on my PJ twice with good effect and am yet to do a full blown calibration (to many other toys to play with atm).

it probably does change over the course of 100+, but I think upping the brightness/contrast via a THX/DVE pattern is just as effective in the between periods. This takes all of 2 mins, compared to 2 + hours to drag all the calibration gear out.

edit:760 hours on globe btw.

Edited by Chopsus, 15 June 2009 - 03:23 PM.

[Mr.Bitey]

TechWizNot
I dont know if thats still true with the newer bulbs.... you might be able to store a few settings tho... - 1 for new bulb, and a couple more as it ages and switch them manualy - of course, youd need to have the sensor and measure it for the initial settings - and with the lifespan of bulbs... might not be worth doing that way..

But theyre not expensive compared to the initial outlay of the projector/display... theyre ~$100 second hand and the s/w is free - you just need to supply your time, and brain in learning it (its not that hard).. the big one tho, is do you have access to the relevant settings (usually in the service menu) to change them.. and know which ones to change (sometimes theyre not called a sensible name)...

But if you can access the relevant controls, and willing to invest the initial time learning, its certainly a good way to go..

Cheers,
Bitey

[EasterBunny]

Bulbs tend to settle down after 100 hours or so, so best to wait till then to calibrate and that calibration should be pretty good for the rest of the bulbs useful life.

[alanh]

DrP,
The usual uninformed post. I have aligned many monitors.

You quote 17 million colours which is the rounded version of the number of shades specified by a colour signal consisting of 3 x 8 bit signal.
Not all of the bits in HD or SD signals go fully white or black as a safety margin so the real number of colours sent to the display is 11.2 million colours

For 10 bit colours there are 976 million specified colours.

Colour measurement is not only the accuracy of sensors and their electronics but the accuracy of colour filters in the sensor head. These are made with dyes and are not particularly stable or accurate compared to the electronics.

All others,
Yes lamps always change their colour in the first 100 hours of use. After that they generally remain the same colour and brightness until they get near the end of their life when they take longer to achieve full brightness.

Mr Bitey,
The digital definition of
black is Y signal 10 (Level 16) P_r = 80 (Level 128), P_b = 80 (Level 128) When converted to analog R = G = B = 0 mV
white is Y signal EO (Level 224 P_r = 80 (Level 128), P_b = 80 (Level 128) When converted to analog R = G = B = +700 mV
Grey is any value between 10 - E0 but P_r = 80 (Level 128), P_b = 80 (Level 128) When converted to analog R = G = B
In all of the above cases the saturation is zero.

When you have these signals you can then adjust the display for the colour of Illuminant D (6500 K) for all levels of grey and for white.

If you use the HDMI output from a BL or DVD player no conversions are made to analog so the levels will remain unchanged until the display takes the signals apart and you can adjust brightness contrast and gamma.
The problem with HTPCs is the graphics card. It would be much better if they sent the data through the card unchanged instead of being able to modify it. You still need the graphics capability to be able to operate the computer.

If you are going to use an HTPC, then you should adjust the display first so that all inputs are correctly adjusted. Then if the computer adjustments are wrong then adjust the graphics card.

AlanH

[DrP]

alanh, on Jun 15 2009, 05:31 PM, said:

The usual uninformed post. I have aligned many monitors.

Could have fooled me, and infact you did.... and not just me!  Based on your prior comments I'm sure glad that I am not one of the poor buggers having to use one of the monitors you 'aligned'.

Quote

You quote 17 million colours which is the rounded version of the number of shades specified by a colour signal consisting of 3 x 8 bit signal.
Not all of the bits in HD or SD signals go fully white or black as a safety margin so the real number of colours sent to the display is 11.2 million coloursp

Congratulations at being stuck in the analogue age.  Now go back and read how things work in an entirely digital system.  The level clamping you refer to only applies when transferring between analogue and digital systems.  As for the rest of it, WTF are you on?  Who cares if SD video is clamped at 16, 235.  It has NOTHING to do with human vision and everything to do about analogue sync levels.  Humans can perceive around 17 million colours.  Like it or lump it.  Its a medical (not Paramedical Steroids) fact!  However when comparing two shades side by side 10 bits of accuracy is required to describe perfect human vision.  This is also a medical fact.  You made a claim that using human vision to perform colour calibration is superior to that which can be done with a colour meter.  The claim is totally, undeniably and laughably false.

All the rest of your post is the usual drivel and does not need to be addressed as I'm sure others will tear it to shreds in time.

Oh, while I have your attention, would you be so kind as to explain precisely how an end user configures their DVD (or HD) player to use the incorrect colour matrix?  You have studiously avoided answering this question in the thread in which you made this latest looney claim.

Edited by DrP, 15 June 2009 - 05:56 PM.

[alanh]

DrP,
The current HD standards still have this allowance to stop digital clipping which looks awful.

You can have you one opinion but you were not there. Have you had to do it to have more than your TV?

When you use the HDMI output of a DVD or Bl player the only thing it does to the signal is to convert the MPEG 4 or MPEG2 b ck to uncompressed digital video. No matrix is used unless you use the PAL composite output. I can say this because the outputs are either digital or analog versions of Y, P_r & P_b. The matrix is used to convert these signals back to R, G & B.

Since you have not sat down and adjusted 3 primary colours from a light box to be able to detect the 17 million colours, funny this number matches 2⁸ 17 million. I have also seen this number in many advertisments for DTV receivers.

As I have asked in the past for you to justify your source I will believe it when i see it!

AlanH

[DrP]

alanh, on Jun 15 2009, 06:37 PM, said:

You can have you one opinion but you were not there. Have you had to do it to have more than your TV?

I guess you are still having your long held comprehension troubles.  Perhaps you need to go back and re-read the posts that came before these and then once you have fully grasped them, post here again.

Quote

When you use the HDMI output of a DVD or Bl player the only thing it does to the signal is to convert the MPEG 4 or MPEG2 b ck to uncompressed digital video. No matrix is used unless you use the PAL composite output. I can say this because the outputs are either digital or analog versions of Y, P_r & P_b. The matrix is used to convert these signals back to R, G & B.

Sigh.  I guess that's why HDMI can (and does) signal the colour matrix used in the video.

Quote

Since you have not sat down and adjusted 3 primary colours from a light box to be able to detect the 17 million colours, funny this number matches 2⁸ 17 million. I have also seen this number in many advertisments for DTV receivers.

I faill to see what this has to do with anything.  Are you off on one of your looney tangents again?

Quote

As I have asked in the past for you to justify your source I will believe it when i see it!

Pot, kettle, black.  You are yet to provide any supporting evidence for your looney claims that DAB+ will be better than internet radio (something that has been continually and repeatedly demonstrated to be false).

[Chopsus]

Hmmm, so all the discussion on the DVE HD disc about how TV's are NOT calibrated correctly (on purpose for marketing reasons) is incorrect?

Even if your HDMI signal is sending a perfect 6500k grey that does no help at all if the display device is not displaying it correctly - isn't that the whole point of calibration in the first place?

[alanh]

DrP,
If you want to use the matrix for the composite PAL output go for it. Since the BL & HD Broadcast transmit or play Y Pr & Pb and the HDMI is also in Y, Pr & Pb why do you need a matrix? The only function for the matrix is to convert from Y, Pr & Pb to RGB. Its been a long time since RGB has been on the outputs of a consumer device.

Following is technical detail which DrP, cannot understand but are there for everyone else.

Cameras still have clamps and clipping in the pre- amplifier stages prior to ADC. The clamping is to remove any leakage currents flowing measured in the black part of the image outside the active picture area and applied to the active areas of the picture.

The DVE disks have the correct levels for grey scale as outlined above. If you wish to measure the light output from the display go for it. Those signals have been digitally generated and recorded straight on the disc in digital form. The digital system transmits numbers not analog levels so it will not change unless you use a graphics card to multiply/divide or add/subtract the values. Players and receivers do not do this.

When a camera white balance is performed by the camera, the camera is pointed at a white surface. The camera adjusts its gain so that R=G=B = 100 Units. These signals are converted to digital and the processor produces  a Y level of E0 by adding 21.26 % Red +71.52 % Green and 7.22 % Blue. The processor also produces

Pb = (50.00 % blue -(11.46 % red + 38.54 % Green)= 0 to which 50 % is added to make an 80hex.  This addition is to allow the negative values for yellow to be less than 80hex.

Pr = (50 % red - (45.42 % Green + 4.58 % Blue) = 0 + 80Hex. Negative values are for cyan colour.

The processor in the display performs the reverse functions to create R = G = B= 100 units. When you set the LCD/Plasma/CRT contrast control and measure the white of 100 cd/m2 and if colour tinting is adjusted to produce a white which looks like 6500 K. These controls operate on the R, G and B signals.

So all that Joe Kane productions had to do is to produce 1110 0000 in the Y digital channel and 1000 0000 in the Pr and Pb channels of a digital signal for white. No camera required or desired.

AlanH

[Mr.Bitey]

alanh, on Jun 15 2009, 05:01 PM, said:

Mr Bitey,
The digital definition of
black is Y signal 10 (Level 16) P_r = 80 (Level 128), P_b = 80 (Level 128) When converted to analog R = G = B = 0 mV
white is Y signal EO (Level 224 P_r = 80 (Level 128), P_b = 80 (Level 128) When converted to analog R = G = B = +700 mV
Grey is any value between 10 - E0 but P_r = 80 (Level 128), P_b = 80 (Level 128) When converted to analog R = G = B
In all of the above cases the saturation is zero.

I'll take you word for it - but that sounds right..

Quote

When you have these signals you can then adjust the display for the colour of Illuminant D (6500 K) for all levels of grey and for white.

And how do you achieve this with DVE ? with only brightness, contrast and saturation controls; and no reference as to what d6500 actually should be...

Quote

If you use the HDMI output from a BL or DVD player no conversions are made to analog so the levels will remain unchanged until the display takes the signals apart and you can adjust brightness contrast and gamma.

The trouble is, displays dont track 6500 evenly across the IRE range... even if you manage to set it right somehow with DVE for a single pattern (ie IRE 100% - white, or IRE 0% Black), it'll be out for the rest.

Quote

The problem with HTPCs is the graphics card. It would be much better if they sent the data through the card unchanged instead of being able to modify it. You still need the graphics capability to be able to operate the computer.
If you are going to use an HTPC, then you should adjust the display first so that all inputs are correctly adjusted. Then if the computer adjustments are wrong then adjust the graphics card.

HTPC use is irrelevant.... fact is you cant use DVE to calibrate the colour of grey (without a sensor)... if you listen to the DVE narative they even tell you this all you can do is set brightness, contrast and saturation... while important, and will improve the picture - your still missing out viewing a properly calibrated image across the IRE range..

And were not even getting started on colour management / colour correction for making up for displays inabilities to render RED as red, GREEN as green, and BLUE as blue which you absolutely cannot do with DVE...

Cheers,
Bitey

[DrP]

alanh, on Jun 15 2009, 09:16 PM, said:

If you want to use the matrix for the composite PAL output go for it. Since the BL & HD Broadcast transmit or play Y Pr & Pb and the HDMI is also in Y, Pr & Pb why do you need a matrix? The only function for the matrix is to convert from Y, Pr & Pb to RGB. Its been a long time since RGB has been on the outputs of a consumer device.

Let me know when you've departed fantasy dream land and made it back to reality and display devices and corrected the errors in the above quote (how can you fit so many errors in such a small space?).  When you do arrive back in reality world, would you mind answering the question raised elsewhere in response to another of your oddball statements - namely, how an end user would be able to select an incorrect colour matrix in the first place.... or is that question a tad too stressing for you?  As a last dig, why did you bring up colour matrixes in the first place?  Have you done a bit more reading since then and decided to try to back pedal a bit?  Do you need to 'clarify' your position as you did with your endless MPEG4 AVC progressive posts?  

Quote

Cameras still have clamps and clipping in the pre- amplifier stages prior to ADC. The clamping is to remove any leakage currents flowing measured in the black part of the image outside the active picture area and applied to the active areas of the picture.

As I've said before alanh is stuck in analogue days.  Not only in analogue days, but also in the days of vidicon tubes etc.  I hate to shatter your dream alanh, but I think you'll find CCDs are used these days.  You might want to look that up with google too.     Do some real research before you try to BS your way out of this one too.  You are now wandering into my field and I know all about the actual nature of 'leakage' when it comes to CCDs and the light I'm looking at you with makes your statement quite 'entertaining'.

Edited by DrP, 16 June 2009 - 01:11 PM.