1. Join Now

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

Projector Contrast Ratios...

Discussion in 'Projectors, Screens & Video Processors' started by cyberheater, Jun 28, 2005.

  1. cyberheater

    cyberheater
    Well-known Member

    Joined:
    Apr 3, 2003
    Messages:
    3,190
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Scotland
    Ratings:
    +175
    Are manufacturers deliberately misleading us or is there a problem accurately measuring contrast ratios.

    Here is an example.

    Take two projectors. The BenQ PB6200 DLP and the Panny AE700.
    Both have a contrast ratio of 2000:1.

    At an Australian website http://www.ausmedia.com.au/projectors_benq.html They did the ansi checkerboard pattern test and here are the results:-

    "The BenQ 6200 when tested using the "Real World" ANSI contrast ratio "checker board" was actually 28% better than the Panasonic AE700 overall. That's to say if you average the CR of all the modes tested.

    The Presentation mode was 20% better than the AE700 Dynamic mode, the Economy mode (the one we like) was 40% better than the Normal mode on the AE700. The Video Mode was about equal to the Cinema1 mode on the AE700."

    So why the difference. If the measurement is accurate. The 6200 is outperforming the AE700 by quite a bit.

    Are they any other sources of good quality non-supplier derived metric information.
     
  2. jriihi

    jriihi
    Standard Member

    Joined:
    Mar 6, 2005
    Messages:
    196
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    16
    Ratings:
    +0
    Its DLP vs LCD.
     
  3. cyberheater

    cyberheater
    Well-known Member

    Joined:
    Apr 3, 2003
    Messages:
    3,190
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Scotland
    Ratings:
    +175
    But surely it doesn't matter if it's dlp of lcd. The quoted contrast ratio should measurebly be the same.
     
  4. phillfyspoon

    phillfyspoon
    Banned

    Joined:
    Oct 21, 2003
    Messages:
    2,128
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    51
    Location:
    Hampshire
    Ratings:
    +76
    in the real world they can differ greatly depending on lots of factors so they just round it off to 2000:1 and I know the ae700 can do just over that but with colour way off.
     
  5. cyberheater

    cyberheater
    Well-known Member

    Joined:
    Apr 3, 2003
    Messages:
    3,190
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Scotland
    Ratings:
    +175
    Which explains a review I once read of the AE700 and the use of colour correction filters to get the required CR.
     
  6. jriihi

    jriihi
    Standard Member

    Joined:
    Mar 6, 2005
    Messages:
    196
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    16
    Ratings:
    +0
    In real world DLP vs LCD contrast ratios are rated very differently it seems. So thats probably why manufacturers contrast 2000:1 dlp vs lcd is so amazingly different.
     
  7. Gary Lightfoot

    Gary Lightfoot
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    12,478
    Products Owned:
    1
    Products Wanted:
    1
    Trophy Points:
    166
    Location:
    Surrey. UK.
    Ratings:
    +2,120
    You can increase the CR of a projector by ramping up the red green and blue contrast levels, as well as the common contrast setting. This produces a very bright image with no colour balance and little white detail. Brightness is reduced as well, and if it has white peaking (DLP), that will be switched on and at maximum. This will achieve very close to the advertised lumen and CR figures but is a meaningless number. They might as well just measure the bare lamp on a table in a darkened room IMHO. :)

    Colour correction filters work by optically cutting the boosted green and/or blue so you retain the high ratio between black and white (as the filter will also dim the image equally from black to white, but your white level is now comparatively brighter). Doing this will give you a higher contrasty ratio at the expense of reduced overall brightness.

    Currently manufacturers will digitally reduce the green and blue contrast so in effect they reduce image brightness and contrast since it doesn't alter the black level.

    ANSI contrast is the measure of contrast when displaying both white and black (usually 8 black 8 white checkerboard) at the same time. I'm guessing, but it's possible that the LCD panels allow more light from the white to bleed into the black which is why the DLP is achieving better results. ANSI is also a measure of the room as well as the pj - if the walls are light the projectors light will reflect from the screen, onto the walls and back onto the screen, washing out the blacks. Not that it makes any difference to the numbers in this case as the test was probably done in the same room so it would be a level playing field.

    Gary.
     
  8. PJTX100

    PJTX100
    Well-known Member

    Joined:
    Dec 21, 2004
    Messages:
    8,132
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +563
    It's all part of the global plot Cybes. Keep chipping away, you'll get to the mastermind behind it all eventually, then this LCD insurgence can be wiped out once and for all...PJ :D
     
  9. gingerone

    gingerone
    Well-known Member

    Joined:
    Dec 10, 2002
    Messages:
    2,418
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Ratings:
    +305
    Personally the stated contrast ratio does'nt mean a great deal to me, if it looks good and is cinematic I'm a happy man.
    Who cares about DLP vs LCD.
     
  10. inzaman

    inzaman
    Moderator

    Joined:
    Dec 16, 2002
    Messages:
    7,888
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Location:
    Derby
    Ratings:
    +941
    Does anyone know why manufacturers dont provide colour correction filters with projectors with a preset setting for their use. This way we could then get the optimum contrast ratio's or closer to the optimum contrast ratio's that are quoted.
     
  11. gingerone

    gingerone
    Well-known Member

    Joined:
    Dec 10, 2002
    Messages:
    2,418
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Ratings:
    +305
    Or why they don't just sort it out in the first place?
     
  12. Maff et1

    Maff et1
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Because the filters cost money and wouldn't increase the brightness/CR they can claim.
     
  13. cyberheater

    cyberheater
    Well-known Member

    Joined:
    Apr 3, 2003
    Messages:
    3,190
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Scotland
    Ratings:
    +175
    Yes but these are home cinema projectors. I'm sure we'd all fork out a extra tenner for a proper calibrated system with filter.
     
  14. gingerone

    gingerone
    Well-known Member

    Joined:
    Dec 10, 2002
    Messages:
    2,418
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Ratings:
    +305
    I agree, I'd pay a bit extra for a pj pre-calibrated to d65 and set up for the best possible display.
    I don't know what dlp's are like but it appears you have to tweak LCD's quite a bit to get the best out of them.
     
  15. Gary Lightfoot

    Gary Lightfoot
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    12,478
    Products Owned:
    1
    Products Wanted:
    1
    Trophy Points:
    166
    Location:
    Surrey. UK.
    Ratings:
    +2,120
    Each pj is different (even of the same make and model) and the colour balance changes to some degree as the lamp ages. I think it's more noticable after the first 100 hours or so. If they used xenon lamps instead of UHP lamps that might help a little, and some manufacturers have larger red sections in the colour wheel to help counteract the deficiency in the lamp.

    My HT1000 was about 1300:1 CR out of the box, and 1250:1 at D65. Using a Hoya FL-Day filter allowed me to increase the green and blue contrasts quite a bit and I managed to get 2000:1@D65. I could only manage an extra 300/400:1 with the Optoma due to it's better colour balance (I assume that's due to the CW).

    CR is improving all the time with newer models, and the Benq 8720 is said to use two irises instead of one to achive over 6000:1. It'll be interesting to see how much that genuinly has both out of the box and at D65. Something over 4000:1 would be a step in the right direction for D65.

    Gary.
     
  16. monopole

    monopole
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    Most manufacturers will quote on/off contrast ratio. i.e. measure brightness of a white, iris open image, then measure the brightness of a black, iris closed image and calculate the contrast ratio from that. And as stated above, some manufacturers will go to extremes, going to high lamp mode/low lamp mode with maximum contrast levels (which does not produce a pretty image). I guess some manufactureres are more honest than others.

    In reality though, on/off contrast pretty much mimics the human eye (which has a really cool dynamic iris which responds to light ;) - comparing my AE700 (with low ANSI contrast) to a friends BenQ PB6100 (which has higher ANSI contrast) was interesting. The perceived contrast was very much in favour of the AE700 - blacks were blacker, shadow detail was better etc... How come? Well, for a start, most of us don't have 100% blacked out rooms with dark walls, so a lot of light from the projector goes to lighting up the room - this destroys ANSI contrast in one fell swoop. Secondly, the dynamic iris coupled with dynamic gamma adjustment leads to great shadow detail on the AE700. Thirdly, the AE700 (sorry, this isn't meant to be an AE700 vs PB6100 review, just an example on how environment and other factors can affect the quality of a projected image) is 16:9, the PB6100 is 4:3, so when watching widescreen material, lumens are lost in the "grey" bars above and below the 16:9 frame - lumens which light up the room.

    So there's much more to how good a projected image looks than contrast ratio alone, other factors like the colour of the room become much more important in the common, home environment.
     
  17. cyberheater

    cyberheater
    Well-known Member

    Joined:
    Apr 3, 2003
    Messages:
    3,190
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Scotland
    Ratings:
    +175
    Some interesting points raised there.

    It was for those reasons raised that I decided to get a anamorphic lens for my PB6100. It does allow you to use the full vertical panel resolution for 16:9 material and makes a huge difference.

    I also modded my PB6100 with a iris. This reduces lumens by around 20% and really improves the contrast ratio. Because there is less light hitting the screen and scattering around my room, contrast is improved (or preserved).

    The difference between the before and after is dramatic.

    I wish I could test the contrast ratio myself. I'd like to run the numbers.
     
  18. PJTX100

    PJTX100
    Well-known Member

    Joined:
    Dec 21, 2004
    Messages:
    8,132
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    136
    Ratings:
    +563
    As you say interesting stuff. I can't help thinking though of that optical illusion recently posted which proved that the brain's automatic perception of that's black and what's not black by relative comparison of the pic as a whole has a lot to do with how much people perceive "blackness".

    Perhaps somewhere there's a base contrast ratio above which the brain takes over and say's "yes, can cope with this, leave it to me". People have come on here saying that older PJs with lower CR eg 300:1 etc look grey. When you get (say) into the 800's and above people seem to be OK, with comments such as "well it looks black to me". Increasing the CR doesn't do any harm of course, and to the trained eye with experience of these things it does make a difference. But whether it's the "be all and end all" I'm not so sure.

    Just IMHO of course...PJ :)
     
  19. cyberheater

    cyberheater
    Well-known Member

    Joined:
    Apr 3, 2003
    Messages:
    3,190
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Scotland
    Ratings:
    +175
    I know what you mean. I've got a few friends with PJ's now and I asked them about the black levels and they said that it looked black to them.

    I then explained how to 'look' for black and they reported that the blacks were no longer black!

    Cinema screens aren't black either.

    To me though. The higher the contrast ratio. The more 3-D the image is. But then I'm probably more bothered by black levels then most folks and probably has something to do with my film selection. Very few of them are bright sunny day comedys :)
     
  20. cyberheater

    cyberheater
    Well-known Member

    Joined:
    Apr 3, 2003
    Messages:
    3,190
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Scotland
    Ratings:
    +175
    BTW. If I get myself a light meter. How do I go about checking the contrast ratio?

    Also. What light meter would be suitable. How small a lux would it have to measure?

    Ta.
     
  21. monopole

    monopole
    Guest

    Products Owned:
    0
    Products Wanted:
    0
    Ratings:
    +0
    If you're just measuring contrast and you have access to a digital multi-meter (i.e. a digital volt-meter thingy - even a cheap one) that can measure resistance, you could simply buy a light dependent resistor from maplin (www.maplin.co.uk - search for LDR, cost is about £1.50), pop it in a cardboard tube, point it at your screen then measure the resistance of an all white image, then measure the resistance of an all black image, take the ratio and you have a rough contrast measurement (assuming the LDR response is linear).

    You need an LDR that has maximum sensitivity around 555nm, which is pretty much the wavelength of light which the human eye is most sensitive to.

    You can find more info here:

    http://home.pacbell.net/steve367/buildit.html
     
  22. cyberheater

    cyberheater
    Well-known Member

    Joined:
    Apr 3, 2003
    Messages:
    3,190
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    86
    Location:
    Scotland
    Ratings:
    +175
    Magic Paul. I've got a DMM and AVIA so will give this a go.

    I haven't got the spreadsheet though.
     
  23. Gary Lightfoot

    Gary Lightfoot
    Distinguished Member

    Joined:
    Apr 6, 2001
    Messages:
    12,478
    Products Owned:
    1
    Products Wanted:
    1
    Trophy Points:
    166
    Location:
    Surrey. UK.
    Ratings:
    +2,120
    I use an Extech light meter that came with SMART III but Robin/Fluke make an identical meter that can be bought on ebay for £20 or less - I got one for £14. It's an RT24. You should measure from around 2 or 3ft from the lens as you will get innacurate black level readings if you go too far away and you can have infinite as your CR. :) Make sure the meter head is in a fixed place when you measure, so use a tripod or similar. You can change the range from the meter as it's on a lead so you don't alter the reading position at all. Let the readings settle down, especialy the black level as it takes a bit longer to sample the reading. Simply divide the black reading into the white reading. Make sure the room is as dark as possible.

    I use Avia and generate a full white field using title 1 chapter 33 (100ire) and then measure full black from title 1 chapter 12 (0ire). I only do this after setting white and black levels and as my pj is calibrated to D65 as well, the results are pretty accurate I think.

    A little simple maths - if you just double your black level, you halve your contrast ratio. That's why any light in the room will not only destroy your ANSI but also devestate your on/off CR. For example, if you have 12ft lamberts at full white, and 2000:1 measured CR, you black level is 0.006 ft lamberts. If you double the black to 0.012, then the CR drops to 1000:1. 0.006ft lamberts isn't a great deal of light as you can imagine, and most ambient would be greater than that.

    To calculate for ft lamberts, you can use the ft candles result directly off the meter IIRC provided the meter is flat against the screen. You will have to multiply by the screens gain if it has any. Another way to do it is to use the lux reading, then multiply by the area of the screen in square meters (i.e. 7ft wide 16:9 screen is about 2.56 sqr meters). So 200 lux at the screen is 512 lumens. Divide the lumens by the area of the screen in square feet (27.56) will give you ft lamberts (in this case over 18).

    I think a light meter is a very usefull tool to have. You can even use it to find your best settings for white and black - set the white level write down the reading, then set the black level and write it down. You now what what to aim for when setting them both up, as setting one will effect the other, so you know what your best white should read and what your best black should read. I do this now as I once found when setting WinDVD 6 up that the CR was a lot less than it was on PowerDVD 6. I went through the above process and found that I could get far better results than just doing it by eye.

    I prefer having around 12ft lamberts of reflectance (and I don't mind less), so having a light meter to help calculate it is handy too. I use lens filters to help with colour correction and to improve CR but try not to drop the image below 12ft lamberts too much, so a light meter is an invaluable tool IMHO.

    HTH

    Gary.

    Gary.
     

Share This Page

Loading...