1. Join Now

    AVForums.com uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

How many projector owners would be interested in hiring a colour analyser?

Discussion in 'Projectors, Screens & Video Processors' started by John_N, Feb 26, 2003.

  1. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Hi Guys

    As you know, I own an HS10 and at the moment am taking a keen interest in calibrating it.

    Colour analysers - used to set the colour temperature - are not cheap bits of kit. They are expensive and tend to be upwards of 1 to 2 thousand pounds sterling to buy.

    This is only SPECULATIVE but I was considering buying one and then possibly renting it out (obviously carried by a carrier like securicor and with insurance) for a small fee in order to recoup some of the large costs of the purchase.

    So. How many people might think of going for that? If I had spent more than a grand on this bit of kit I would be tempted to rent it out at about £75 / week to be honest. Might sound like a lot but not compared to employing a pro to perform the calibration for you.

    In addition, bear in mind you need to be technical to use this sort of kit. You need to know what you're doing in other words.

    Comments? ideas?
    Cheers
    John
     
  2. Dangerous

    Dangerous
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Would this piece of kit enable you to calibrate in the same way as Gordon calibrates a projector? If so then I would be interested.
    Dave
     
  3. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Ho hum.

    I don't want to damage Gordon's business. I'm sure Gordon would do a better job since he would be performing a greyscale calibration as well as colour balance to D65 and he works in the field and therefore knows what he's doing.

    However, if hypothetically one used a colour analyser and a tool like SMART you should be able to get your projector calibrated to D65 and have correct greyscale tracking. Then again you might mess it up badly since this business is not necessarily all simple you know.. I'm just trying to give a balanced view.

    My current view is that I don't know if my HS10 is calibrated correctly. I don't particularly want to be paying £200 - £400 for calibration when I might get rid of the unit in a few years and have to do it again. I also don't particularly want to spend £1500 or whatever on a bit of kit that sits in the cupboard and never gets used because that's a worse waste of money so in all honesty I'm just throwing some ideas in the air and seeing what happens.
     
  4. Joe Fernand

    Joe Fernand
    Distinguished Member AVForums Sponsor

    Joined:
    Jan 20, 2002
    Messages:
    23,762
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    The Borders
    Ratings:
    +2,464
    John_n

    Not wanting to pour cold water over your plans but a few points to consider:

    01. Are you infringing the manufacturers licensing rights when offering out these products for hire?

    02. What's the total cost of your hire kit - Flightcase, Spares (you really should have a back up unit in case your main unit fails), In-life service (can you vouch for the accuracy of your system).

    03. Insurance and liability - what will be your terms and conditions be and what happens if someone blows up there system when using your kit; they (or there home contents insurer) may come after you saying it was your faulty kit that caused the problem.

    04. Shipping insurance - it may well be insured but you try getting your money out of a UK shipping company in a hurry - expect to take up to SIX months to get your money back on a successful claim!

    05. Damage and shortages - what will be your policy and how will you police it; a lost mains cable is the most annoying and common problem (and costly).

    06. 24 Hour tech support - your going to be inundated with calls; can you handle them.

    Sorry to seem so negative but having run an AV hire company for 10 years I can tell you its not all easy street.

    In the hires world you should be looking at a daily rate of around 1/20 of the purchase price of the kit - and a weekly hire is normally charged out at 3 x daily rate; your kit is going to be knackered and requiring full or partial replacement far sooner that you think.

    I'd also assume most of your 'Hires' will be over a weekend so your going to have to have time mid week to take the system back in and check it over and then get it ready too ship back out again!!!

    Best regards

    Joe
     
  5. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Hi

    Thanks for the information - I wasn't really planning it as a business since I work quite hard I don't really have the time for it. I was thinking more about a means for me and other users to get to use a colour analyser cheaply.

    However, you are right to point out the legal and contractual issues and IF it ever came to it of course I would have to investigate those fully. I'm not looking to make a profit out of it - more just get some use out of a colour analyser and allow others to do the same. It may prove to be too much hassle.

    If you have a colour analyser for rent let us know!

    John
     
  6. JohnS

    JohnS
    Active Member

    Joined:
    Oct 31, 2000
    Messages:
    2,580
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    71
    Location:
    Herne Bay, Kent
    Ratings:
    +33
    I'd be up for it on a £75 per week basis, do these things really wear out? How many new projectors might you buy over the next 10 years? £1K is one thing, £2K is another. £1k is a lot for a pj like the HS10 but if the market place keeps on improving at the current rate it'll only be 18-24 months and a newer pj with decent res, better contrast and less noise will be replacing it.

    Just thinking out loud really.
     
  7. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Hi John

    That's a lot of good points.

    I don't know whether a colour analyser would wear out quickly but I would imagine it would outlast an HS10! :laugh:

    I really have no idea how many projectors I plan to buy over the next 10 years... As few as possible!

    I know = next to the depreciation on a car we're not talking about much but nonetheless I don't like chucking my money down the drain.

    What do other people think? This is a wider picture really. The big question is:

    IS IT WORTH SPENDING MONEY CALIBRATING YOUR PROJECTOR OR IS IT BETTER TO PUT THAT MONEY TOWARDS A NEWER, QUIETER UNIT IN THE FUTURE?

    What do you think? At the moment, seems to me that it's gonna cost about £200-£400 to get a projector properly and fully calibrated to D65 by a pro.

    It's going to take someone to stick his neck out and buy one and hire it out for access to colour analysers to come down.

    Another question:

    IS IT WORTH IT?

    I know all the technical stuff.... But I was wondering whether you could setup using a tool like SMART and the greyscales in AVIA and maybe a reference D65 light source which I think you can buy... I wonder whether in the final analysis we are talking about big differences here?

    Cheers
    John
     
  8. Klippie

    Klippie
    Active Member

    Joined:
    Dec 1, 2002
    Messages:
    623
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    31
    Location:
    Scotland
    Ratings:
    +55
    John_n,

    I must agree with you in saying is it worth it, this is a lot of money to shell out just to see if there is any difference in the picture quality. If you have your projector setup to your lliking and are happy with it why bother.

    What happens when the bulb starts to fade which in turn sets all the colours and contrast levels way out will spending another pile of cash be cost effective in the long run just to have the perfect picture until the bulb dims a little more.

    Does anybody know what is supplied with the Smart system that makes it so special if it is not too complex I am sure we could all come up with suitable components to make our own calibration tool then we could all have one.

    Just a thought...

    Klippie.
     
  9. JohnS

    JohnS
    Active Member

    Joined:
    Oct 31, 2000
    Messages:
    2,580
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    71
    Location:
    Herne Bay, Kent
    Ratings:
    +33
    Since I bought a rear projector tv about 4 years ago and the two projectors since I've taken a great deal of pleasure from tweaking it to get the most out of it posssible, I've not paid for a proper calibration and I'm not likely to (no offence Gordon). I dont see the value in having the whole chain calibrated when I change stuff so often.

    When I 1st bought Video Essentials I struggled to justify the £25's odd. Even more so when I bought Avia @ £40 though I wouldn't be without it now. The hoya FLD for my ae100 was probably the best £20 you can spend but then you also need Avia afterwards so that was a good investment.

    When the HS10 came along it was no trouble to justify the expense of a CC40r and new dvi cables to get the best I could from it. How far off I am now from "how its suppose to be" I couldn't say, I'm certainly not unhappy with the picture I've got, Dave Boswells settings have certainly pushed me in the right direction with where to go and Avia to finish it off. Could I justify the expense of a colourmeter....I dont know.... I'd want to know the improvement I'd get and of course untill its done you'll never know:(

    At the moment I dont beleive that an extra grand spent on the projector would give me any picture quality improvement (only maybe quieter). The only panel pj's I would shortlist are all over £4K. Maybe the colour calibration would help but....maybe it wont.
     
  10. Gordon @ Convergent AV

    Gordon @ Convergent AV
    Distinguished Member AVForums Sponsor

    Joined:
    Jul 13, 2000
    Messages:
    13,976
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    Living in Surrey, covering UK!
    Ratings:
    +2,779
    John has put his case well so far as has Joe. All I can add is that there is more to getting a good image than tracking D65, just as John points out.

    You can track D65 quite well and still have your set configured in such as way as you are shortening its life expectancy.

    I am about to go on to my third tool for part of this job. The first was returned to US 3 times in space of 9 months. This at MY expense in shipping BOTH ways, including a 3 month fight with customs and excise on two occasions for charges to not get paid on top. The last time it went back was for calibration. You see the measuring device needs to be calibrated every year as well to make sure it is "seeing" correctly. They can be damaged making them less accurate. If you have no experience of whether it is working correctly then it's hard to know if you have got a good result.............................

    The last unit was sent back as I had a chance of something better and more technically accurate.

    With an understanding of what you are trying to achieve you can possibly get yourself a better picture.

    I would also say that I have never charged anyone £400 for a calibration.

    Cheers,

    Gordon
     
  11. rooster

    rooster
    Standard Member

    Joined:
    Mar 18, 2002
    Messages:
    216
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Edinburgh
    Ratings:
    +2
    So what calibrates the machine that calibrates the calibrator:D

    How much do you charge then Gordon. If i can be so forward.
     
  12. Gordon @ Convergent AV

    Gordon @ Convergent AV
    Distinguished Member AVForums Sponsor

    Joined:
    Jul 13, 2000
    Messages:
    13,976
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    Living in Surrey, covering UK!
    Ratings:
    +2,779
    I don't actually know what calibrates a calibration device....probably a very expensive device indeed:D

    My charges depend on device, sources and software watched. Some displays only have one global colour temperature setting. Others store them for PAL, NTSC, PAL progressive, NTSC progressive, RGB625i, RGB525i, 720P, 1080i etc.

    I have a basic rate for a device and then an additional £50 for each extra cal requiured. If it's a crt projector and other work is required then I negotiate a daily rate. If someone has a slew of devices and the units stores each one seperately then they may get a discount to make it more affordable.

    On top of this there is a shared cost for travel and accomodation. Usually between £35-£50 per head. For exact costs folk should email me.

    Gordon
     
  13. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    No I wasn't implying that the £400 cost was for the calibration - I really meant that over the lifetime of the unit, the cost of two or three calibrations would probably be in the order of 200-400 pounds depending on circumstances. I am reluctant to give specifics and I'm sure if anyone wanted a quote Gordon would give them one over a private email but I don't think anyone would be too upset at my saying that this was probably the ball-park since it gives a fair amount of latitude.

    Yes I've noticed that various companies then calibrate the Photospectroradiometer to ensure that it is calibrated correctly. Part of the problems with these devices you see is that the photocells that they use have a non-linear response to the spectrum - in other words - they tend to be more sensitive to green light than say blue light. This tendency changes from photocell to photocell and the calibration can drift (because this is analogue electronics) due to the aging process of the components, or even because the unit is habitually used in a warm environment rather than a cool one for example.

    The whole thing is a bit of a can of worms I think and as Klippie points out, sometimes people just set it so they are happy with the picture which is fair enough and what I've done.

    At the same time, I do get a lot of pleasure messing around with the settings as does John S and I'm interested to know whether I'm getting the 'best' out of the unit as I'm sure everyone else is.

    The bit about tracking D65 and shortening your unit's life expectancy interests me.. I wonder if this is because you could track D65 correctly at a level which compresses black detail making the whole picture too dark and therefore causing heat buildup on the LCD panels? My theory would be that since D65 is a spectral energy distribution, the relative levels can be OK but they can all be too 'low', meaning the projector is not letting all the light through that it could.

    This leads me onto another calibration issue that I've been thinking about:

    in the HS10, we have gain values (0-255) and Bias values (0-255).

    I would imagine that the final colour intensity on the screen is decided by something like:

    Colour intensity = Input Level * (Gain/A) * (Bias/B) (for the sake of simplicity I'm ignoring any other factors like Gamma).

    Now. the LCD panel will be 100% driven at a certain voltage. This is the maximum colour intensity that the projector is capable of because the panel can not let through any more light.

    So how do we tell whether the projector is driving all the panels at 100% when given a 100IRE white field ?

    My answer would be that if we consider the entire set of gain/bias values that give an acceptable colour balance (set to D65) I would imagine that some of the gain/bias values would be too low and therefore would result in the panels being underdriven and therefore reducing contrast and losing black detail. Likewise if the gain/bias is set too high we will overdrive the panels and will end up crushing white detail.

    I would imagine that this is the process that SMART helps us with since AFAIK SMART is basically:
    a) a light meter
    b) a set of filters
    c) a spreadsheet.

    Now - if we assume that our light meter is calibrated so that it is equally sensitive to all colours of the spectrum we could use this to take reading from the screen. I think SMART uses a photocell and a multimeter... cheap & simple stuff. You can pick up a photographic light meter from EBAY though quite cheaply.

    If I was doing this I would send the unit a 100IRE white field and basically increase the gain/bias on each colour with a filter in place and the others set to zero until I got to the point where the colour 'clipped' and didn't give me any increase in intensity. At that point I would know that the panel was fully driven for that particular colour. That would be the maximum gain/bias that that particular colour could handle. If we repeated that for each gain/bias, we would have a set of values that we definitely could not exceed without crushing white detail. I would imagine you could do the reverse with a 0IRE black field and see how 'low' you could go until the readings stopped dropping. I would imagine this would determine the points at which the panels are at maximum opacity (letting through the least light) and this might be different for each colour too.

    Then I suppose you could try to set the colour balance without violating any of those values. And then I guess you would juggle with the gain/bias as various points on the greyscale to try to get the linearity correct across the scale.
    This would be my take on doing it manually without a colour analyser and just a light meter and a set of filters, which to be honest I think is all the SMART is.

    So - does this all sound reasonable? I really need comments from Gordon here but I've read it all again and it doesn't sound unreasonable to me. I'm still trying to figure out what relation the contrast - which I think is the white level setting - and the brightness - which is the black level setting - have in all this - and I think you would need to set the contrast and brightness using the black bars / white bars or PLUGE pattern before you started.

    Cheers
    John
     
  14. Gordon @ Convergent AV

    Gordon @ Convergent AV
    Distinguished Member AVForums Sponsor

    Joined:
    Jul 13, 2000
    Messages:
    13,976
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    Living in Surrey, covering UK!
    Ratings:
    +2,779
    John,

    My comments about longevity were more to do with other devices that a colour analyser can be used to calibrate , ie Plasma and CRT devices where you could track d65 relatively well while having the thing in meltdown mode. Plasma in particular.....

    Perhaps a solution that may be best would be to buy a unit then re-sell it.

    Gordon
     
  15. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Hi Gordon

    Yes - I didn't realise you were talking about plasma&CRT too - then I can appreciate that the "decrease in lifetime" is applicable due to overdriving the display / power supply issues etc - but if we confine our discussion to LCD/DLP projectors for a moment then does the above text sound reasonable?

    What I'm trying to do here is come up with a reasonable methodology for calibrating a front projector without spending a billion pounds...... :)

    Cheers
    John
     
  16. Gordon @ Convergent AV

    Gordon @ Convergent AV
    Distinguished Member AVForums Sponsor

    Joined:
    Jul 13, 2000
    Messages:
    13,976
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    166
    Location:
    Living in Surrey, covering UK!
    Ratings:
    +2,779
    John:

    Not going to give away all the trade secrets on here but if you want to maximise contrast ratio, without clipping or crushing detail and still track D65 you r methodolgy seems sound to me.

    Gordon
     
  17. Mr.D

    Mr.D
    Well-known Member

    Joined:
    Jul 14, 2000
    Messages:
    11,040
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    133
    Ratings:
    +1,111
    I'm assuming the smart system just gives specific intensity aims for certain IRE levels along RGB intensity curves: in order to approximate idealised response in a known system. This should sort of work OK depending on how good the idealised system is and how close your system can approximate it but you might find a little innaccuracy here and there and I doubt the sensor techniques are all that consistent form one kit to another.

    With regard to the colour analyser hiring. I suggest a better alternative would be for a group of people to collectively buy one and pay for maintenance and calibration. ( regardless if it gets damaged in an individuals care : the whole group is responsible for any repair work: this is the risk you take ).

    Anyone refusing to meet any group maintance obligation is ejected from the group and no longer given access to the equipment. Initial buy in is void. Anyone wishing to sell on their share is allowed to do so with the consent of the rest of the group. All maintenance responsibilities and conditions fully pass on to the new member.

    Anyone wishing to buy in at a later date pays the required dividend which is then redistributed equally amongst the initial group.

    Rental of equipment to individuals outside the group is at the consent of ALL the members of the group : any monies arising being equally redistributed to the buy in members. Any damage incurred to the machine will be considered the total liability of the hirer.
    Likewise no individual of the group shall be held responsible for any damage occurring to equipment as a direct result of calibrating it with the equipment in question. Whether the damage is incurred by either a member of the group or an outside hirer.

    The philips colour analyser we use is about 300quid a time for yearly calibration and it normally takes a few weeks to turn-around. Its costs about £5000 new last time I checked. 10 people £500quid initial buy in .

    I'm not interested myself.
     
  18. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Hi Gordon

    No of course not - I wouldn't expect you to. :)

    OK. First things first.

    To everyone else: Does anyone have experience of photographic light meters? Are they calibrated to be equally sensitive to different light levels? are they calibrated to match the sensitivity of photographic film?

    What I intend to do is a bit more research, a bit more maths and I think I'll read a few more books on film, video, photography and colour theory. When I've done it I'm going to write down a short textbook on "how to calibrate your projector without spending a billion pounds" and I'm going to make it available.
     
  19. Mr.D

    Mr.D
    Well-known Member

    Joined:
    Jul 14, 2000
    Messages:
    11,040
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    133
    Ratings:
    +1,111
    Any photographic light meter is set to indicate correct exposure for 18% gray with regard to a log film curve type response ( ie 18% grey looks like mid gray to your eyeball when depicted on film).

    The critera you set on the meter can vary from unit to unit: film speed , shutter speed/angle , aperture are the usual ones.

    You can also take incidence or reflectance readings.

    Still or motion picture makes no difference.
     
  20. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Hi.

    OK. If we leave the colour balance / setting to D65 for a moment as a separate issue, we can concern ourselves with maximising contrast.

    About the photo light meters... Let me rephrase the question.

    Suppose we have 3 light sources of precisely equal intensity, one Red, one green and one blue.

    If we use a photo meter to measure the intensity of each source, will it read the same for each colour? or will it read different readings for each of the 3 colours due to a non-linear spectral response? I don't know much about photography but I would imagine that colour film is more sensitive to a certain colour light than others... Is this characteristic therefore modelled in the light meter? or is it ignored and compensated in the developing stage?

    I don't really care about what the markings on the scale are - all I need is a light measuring instrument that treats all colours of the spectrum in an even=handed (or at least predicatable) manner.

    Cheers
    J
     
  21. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    PS... About the 'buying a colour analyser collectively' part... Sounds good in theory but I'm really not sure that we would manage to get enough people willing to spend the cash...

    In any case. Surely if we have a calibrated light meter and a set of filters, we should be able to calibrate to D65 with a lot of effort since we are merely talking about a mathematical illuminant here. I know it wouldn't be easy. I know it would take a lot of measurements. I know that it would be a real pain in the neck. But it might still be possible.....
     
  22. Mr.D

    Mr.D
    Well-known Member

    Joined:
    Jul 14, 2000
    Messages:
    11,040
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    133
    Ratings:
    +1,111
    Any lightmeter reads intensity , colour doesn't make any difference: otherwise it would be a colorimeter not a light meter.

    In answer to your question a red patch and a green patch and a blue patch of equal intensity will read the same precisely because they are the same intensity. Think of a tonne of feathers and a tonne of coal ( although I'm not too sure that helps!)

    You want to use a light meter to calibrate a display device yes?

    The problem isn't measuring the various intensities of say a 50IRE red window patch vs a 50IRE greenpatch its knowing what they should actually be to give you the desired result for a given patch. Its the aims you need in addition to the measuring. If someone else has calibrated a given type of display they can run a bunch of patches through it of various IRE values and list the intensities to give you aims to go for.
    If they do this for red , green and blue patches they have gone some way to giving you the aims for correct grayscale tracking as long as you can use exactly the same measuring device and your initial black and white point is the same ( intensity wise) and the response curve of the display is roughly similar. Chances are they won't be so you'll get a margin of error thats either large or small and varies across the grayscale. (SMART is like this ?) This technique is known colloquially in the trade as "ripping curves"

    A colour analyser/colorimeter measures chromaticity as well as luminence ( intensity) it also tells you how far off you are RGBwise from a set of idealised aims (ie rec709 ,) so you can modify the RGB balance accordingly. The reason you can trust it to give you an accurate picture of how your display relates to a set of known criteria is that you spend 300 quid a year sending it off to be calibrated itself. Modifying the RGB levels is also not so cut and dried: there are good ways and bad ways to get to 6500K ( or 5400K in my case most of the time)

    There are other areas of calibrating a display device that are more involved. The grayscale and tracking is only a part of having a good display.

    To maximise your intensity range ( ie set black point and white point) your eyeball is best in conjunction with suitable test patterns.
     
  23. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Hi Keith

    The reason I asked the question is that although a light meter is indeed supposed to measure intensity and in theory independent of wavelength, it actually tends to use a light sensitive detector such as a CDS cell to actually perform the measurement.

    This piece of electronics is unfortunately non-linear in it's spectral response and for this reason, if you take a CDS cell and apply blue light of a known intensity to it, it will read a different resistance to if you apply red light of the same intensity to it. For this reason, a correction needs to be applied internally to the meter which is very unlikely because the cell does not 'know' the wavelength of the light it is measuring - it just has a resistance value and unfortunately this resistance is a function of both intensity AND wavelength. This is inevitable because of quantum theory.

    In fact - I've answered my own question. A photographic meter would be no use at all unless it was certified to have a flat spectral response in the detector. I don't think that such a device exists in the world of photodetectors. I would be very surprised if a photographic light meter was entirely flat in it's spectral response.

    So. Sadly my idea of using an uncalibrated photo meter will not work either. We still have the non-linearity to contend with.

    On the other points regarding what they 'should' be, we should be able to derive this by applying the mathematical model of the D65 illuminant to our particular case. This is what will be performed when calibrating a colour analyser, otherwise all we have is a series of approximations. At the end of the day you have to refer back to the black body spectral function that is the root definition of D65. This can be described mathematically and precisely and therefore it is possible to calculate EXACTLY what the R,G,B levels need to be if only you are able to measure them accurately enough! It is this measurement process that is the difficult and costly part.

    Kind Regards
    John
     
  24. JohnS

    JohnS
    Active Member

    Joined:
    Oct 31, 2000
    Messages:
    2,580
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    71
    Location:
    Herne Bay, Kent
    Ratings:
    +33
    You guys just sort out what we need and I'll go with the best option:p
     
  25. severnsource

    severnsource
    Active Member

    Joined:
    Jan 19, 2003
    Messages:
    322
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +9
    John

    Can I ask why you want to calibrate your projector to D6500? AFAIK no domestic displays are set to anything like this colour temperature yet the results are perfectly acceptable, because you have no visual reference.

    Yes, the greyscale needs to track, and I would be interested in an easy way of doing this, but the absolute colour temperature is of only academic interest, surely.

    Bill
     
  26. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Hi Bill

    Lots of domestic displays are calibrated with the colour balance way off and the colour temperature too high. This is so they look good and have impact on the shop floor.

    Away from the shop floor and when viewing, the displays often display unnatural colours and poor flesh tones.

    If we are trying to recreate the "theatre" experience, this in an ideal world needs the reproduction chain to be calibrated to D65. This is the standard that the engineer was using then they mixed the movie and set the colour balance, and is using a display system set to anything else, you are not seeing the movie as the director intended.

    Granted, you may find the picture perfectly acceptable, but we aren't talking about a £400 telly here. Given the total cost of my entire home theatre system approaches £8000, rather to my surprise, I'm inclined to spend some time and effort to make sure that it is calibrated correctly! :)

    In any case, an image looks more "three dimensional" and the field appears to have more "depth" if the colour balance is set correctly.

    Cheers
    John
     
  27. Klippie

    Klippie
    Active Member

    Joined:
    Dec 1, 2002
    Messages:
    623
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    31
    Location:
    Scotland
    Ratings:
    +55
    John_n

    I have read through your posts a few times and as I understand it the whole point in calibratring the projector is to get the white output set to D6500K which is not easily done without the expensive calibration tool you describe.

    This may be a hell of a long shot but if you were to construct a photo cell and rgb filter type of affair then set it up to a D6500K
    light source ( 98 CRI fluorescent tubes ) would this give a rough base figure to set the projector white output to.....said it was a long shot!

    Everybody else,

    When I posted there were over 260 hits on this thread, has anybody tweaked their projector ie fitted a filter and adjusted the gains and bias settings then had their projector professionally calibrated, I am just wondering if the picture quality was greatly improved over their own tweaked settings.

    Klippie.
     
  28. BabylonP

    BabylonP
    Active Member

    Joined:
    Dec 10, 2002
    Messages:
    98
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    11
    Ratings:
    +5
    Klippie

    Thats an interesting point and one that Gordon might be able to shed some light on.

    This thread has been highly informative and I for one have learn't a lot. Thanks to those who are far more knowledgable than I.

    Phil
     
  29. benmc

    benmc
    Active Member

    Joined:
    Jul 2, 2002
    Messages:
    175
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Location:
    Berkshire UK
    Ratings:
    +7
    What we would really need to arrange is 2 HS10 and Gordon. Calibrate one HS10 and then compare them side-by-side. We could then give some good thoughts about the value of calibration.

    I know that John_N and I only live about 15 miles apart.

    Worth some more thought?

    Ben
     
  30. John_N

    John_N
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    That's a good point - about the side-by-side comparison - and one that is well-worth considering.

    Klippie:

    Calibrating the projector needs you to know:

    a) What spectral balance you need to acheive (ie D65).
    This can in theory be calculated using a mathematical model since D65 is baselined to the "black body radiation" of a 6500K ideal black body and this can be defined mathematically. It's a classic physics/maths derivation.

    b) What your current spectral balance is.
    This is harder to measure - because your photocell and measuring equipment will have a non-linear response to the spectrum. If this isn't compensated for, all the readings will be completely wrong.

    One of way of doing this would be to measure the output of our photocell using a light source of a known spectral output and then measure our photocell output for different wavelengths. At this point, we can then calculate the deficiencies in the response of the photocell.

    However. The output of our test fluorescent tube may not be flat or accurate enough for that. It would be easy if we had a guaranteed source that is absolutely flat, but of course we do not.

    My latest idea is that there is one source whose properties are well known and the source is shared throughout the planet - The sun. I'm investigating what would happen if we use our photocell and RGB filters on a sunny day to allow us to measure the relative sensitivity of the cell to R, G and B and therefore characterise the response of the photocell.

    However, the intensity of the sunlight would need to be moderated using a neutral filter such that it placed the photocell in the same range of it's impedance/intensity curve as the real video signal would. If we failed to do this, we may have to take into effect the non-linear response of the cell to light intensity as well! Added to this the fact that the spectral response of the cell may alter depending on where in the intensity scale we sit (this function is not a constant) and therefore we end up with a complex function of two variables for the output of our cell. This therefore involves a lot of guess work, assumptions etc. At the moment I'm just trying to find out a bit about how to make intelligent assumptions and mitigate the factors.

    At the end of the day, setting a display to D65 is basically going to be a compromise between a lot of factors. The whole measurement process will have errors in it. The equipment will have been designed with assumptions in it, and the calibration performed on the equipment may well have certain assumptions made. There will have been compromises made in the calibration between acheiving maximum contrast and tracking D65 exactly. The gamma of the system is based on an assumption and therefore our system gamma may not be correct. We may even have to contend with the fact that certain material may not be correctly balanced in the first place! In addition, people's eyes respond to colour in different ways and your eyes don't have a perfectly logarithmic response. The sensitivity function of your eyes to various colours is also dependent on light intensity.

    The more I think about it, the more I think that it is actually a quite imprecise science and at the end of the day I reckon that a calibrator would use their eyes to set a pleasing image once in the correct 'ball park' given by these figures which are no more than guidelines.

    Also - for what it's worth - after thinking about it, I reckon that to most users it would be more valuable to get the colour balance approximately right and pleasing and then perform a calibration to get maximum contrast like SMART - than to be too rigorous about tracking D65 or what you perceive to be D65 because the whole chain has quite a lot of errors in it in the first place.

    Therefore I'm also thinking about:

    a) Making a SMART system without spending £180 quid in order to get maximum contrast and greyscale linearity.

    b) Can we get a D6500 reference which (by eye) would allow us to try and get our AVIA colour balance in the right ball park. Is there a way of making this process more efficient? After all - in your head you have two very good measuring instruments that are pretty sensitive and free to use. There is not a lot of point in making changes that you can't see the effects of anyway - since at the end of the day we're involved in watching movies, not measuring using graphs. For all I know, you might be able to get a special optical instrument like a telescope with a built-in D65 source that allows you by eye to compare and contrast a know reference with the area viewed in the instrument.

    whew!
    That's all for now.
    J
     

Share This Page

Loading...