or Connect
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › 3LCD vs DLP Here we go again...
New Posts  All Forums:Forum Nav:

3LCD vs DLP Here we go again... - Page 2

post #31 of 46
Quote:
Originally Posted by stanger89 View Post

It seems to be based on comparing their projectors to business-class DLPs, the ones with white segments or the Brilliant Color (RGBCMYW) colorwheels which can generate much higher light output for white, but at the expense of color saturation. I don't believe

Quote:
Originally Posted by HogPilot View Post

This is a blatant 3LCD marketing fluff piece that uses the word "study." Not even worth discussing.

Single chip DLP projectors that have other than just RGB colors in their color wheels, and especially those with a white (clear) segment, do produce an overall image brightness, with real video, that is less than what is indicated from a full white test pattern measurement (sometimes a lot less). However, when the DLP only uses RGB in the color wheel the lumens measurements, taken with a 100% white test pattern, will reflect the actual lumens output for normal video just as it does for all 3-chip projectors (be they LCoS, LCD or DLP).
post #32 of 46
Quote:
Originally Posted by Christian Bergh View Post

Here's a link to an article that claims 9 of 10 prefer 3LCD

Hopefully not already posted if so please delete.

http://www.aboutprojectors.com/news/2013/10/03/study-shows-consumers-prefer-3lcd-over-dlp-projectors/


/Christian

9 out of 10 ppl are idiots, i'm not surprised at all.

I bet 9 out of 10 ppl think edge lit LED's are the best TV's ever made and BOSE the pinnacle of high end audio
post #33 of 46
Quote:
Originally Posted by fierce_gt View Post

9 out of 10 ppl are idiots, i'm not surprised at all.

I bet 9 out of 10 ppl think edge lit LED's are the best TV's ever made and BOSE the pinnacle of high end audio

That assumption is probably fairly accurate.
post #34 of 46
Quote:
Originally Posted by Ron Jones View Post


This is taking one instance and trying to apply to all projectors, hence a blanket statement.

Single chip DLP projectors that have other than just RGB colors in their color wheels, and especially those with a white (clear) segment, do produce an overall image brightness, with real video, that is less than what is indicated from a full white test pattern measurement (sometimes a lot less). However, when the DLP only uses RGB in the color wheel the lumens measurements, taken with a 100% white test pattern, will reflect the actual lumens output for normal video just as it does for all 3-chip projectors (be they LCoS, LCD or DLP).

Measuring the white point of an LCD or LCOS in MAX lumens mode does not guarantee you that the colors are properly luminated at different saturation points, I'm not sure how you are getting to that.

It's not a relative comparison because the gamma and/or gray-scale of all these projectors is so far off in those max torch modes that the brightness in the mid-tones and the actual colors you see are so distorted that any such comparisons need to made individually between each projector. LCD's share many of the same issues at max torch modes as a DLP does that has a white segment in its color wheel, have you measured the Gamut and Gray-Scale of your Epson at 2200 lumens? The Epson also does not produce the correct saturation and luminance in its absolute brightest mode.

In fact, all these projectors have their own individual issues at or near max torch mode, including some that have gamuts that are so wide that the comparison isn't even relative. Although it is true the DLP with a white segment experiences the issue more severely than a DLP without a white segment, the effect is not visible to the entire spectrum of colors and it does not mean that an LCD will have better color as a % of its near-brightest mode to absolute brightest mode.

The white segment in the color wheel is not a ZERO sum gain, there are gains at some points and losses at others. For instance, the best projectors for projecting the most accurate color at the highest lumens are in fact the Benq sh910, Optoma eh505, Viewsonic pj7820hd and similar projectors. It doesn't mean they are perfect at these high lumens, it just means they have decent modes which are far brighter than a regular HT projector. If what you are saying is true, you are saying that projectors that can do 3000+ lumens that have white segments are no better than projectors that have 2000 lumens without a white segment at fighting ambient light in video, and this is incorrect.

What is CORRECT to say is that the best modes of projectors with white segments are not as clean as the best modes of projectors without white segments, but the white segment actually is an advantage when it comes to video brightness as long as you realize you are giving up some accuracy for the lumens bump.
Edited by coderguy - 10/6/13 at 4:48pm
post #35 of 46
Quote:
Originally Posted by coderguy View Post

Measuring the white point of an LCD or LCOS in MAX lumens mode does not guarantee you that the colors are properly luminated at different saturation points, I'm not sure how you are getting to that.

It's not a relative comparison because the gamma and/or gray-scale of all these projectors is so far off in those max torch modes that the brightness in the mid-tones and the actual colors you see are so distorted that any such comparisons need to made individually between each projector. LCD's share many of the same issues at max torch modes as a DLP does that has a white segment in its color wheel, have you measured the Gamut and Gray-Scale of your Epson at 2200 lumens? The Epson also does not produce the correct saturation and luminance in its absolute brightest mode.

In fact, all these projectors have their own individual issues at or near max torch mode, including some that have gamuts that are so wide that the comparison isn't even relative. Although it is true the DLP with a white segment experiences the issue more severely than a DLP without a white segment, the effect is not visible to the entire spectrum of colors and it does not mean that an LCD will have better color as a % of its near-brightest mode to absolute brightest mode.

The white segment in the color wheel is not a ZERO sum gain, there are gains at some points and losses at others. For instance, the best projectors for projecting the most accurate color at the highest lumens are in fact the Benq sh910, Optoma eh505, Viewsonic pj7820hd and similar projectors. It doesn't mean they are perfect at these high lumens, it just means they have decent modes which are far brighter than a regular HT projector. If what you are saying is true, you are saying that projectors that can do 3000+ lumens that have white segments are no better than projectors that have 2000 lumens without a white segment at fighting ambient light in video, and this is incorrect.

What is CORRECT to say is that the best modes of projectors with white segments are not as clean as the best modes of projectors without white segments, but the white segment actually is an advantage when it comes to video brightness as long as you realize you are giving up some accuracy for the lumens bump.

I was really meaning that when operated in a reasonably accurate video mode DLP projectors with white segments (that cannot be turned off) may produce high lumens values with a 100% white test pattern but still not produce an overall video image that is as bright as a pure RGB priojector that measures lower with a white test pattern. White segment DLP models have utility for a business projector for presentations in a conference room, but are not well suited for HT use (where accurate color are desired). I was trying to point out that when seeing the lumens specs for a single chip DLP projector you should check what type of color wheel is being used.
Edited by Ron Jones - 10/6/13 at 5:56pm
post #36 of 46
Yes, but you are then just re-enforcing what we have been saying all along, that it only matters in a calibrated mode, and the numbers spec'd do not show that number.

I don't mean to hammer you, but you should have made that distinction because Epson is claiming 2400 lumens CLO and WLO, insinuating the projector can do good color at 2400 lumens, which is rigatoni with baloney. Therefore this number is useless as we have been saying all along.

Also, I think it is going too far when stating having a white segment these days ruins the color so much that it is not suitable for HT, I think the better way to state this is that it is not suitable if you have other options and do not ABSOLUTELY need the extra lumens in a torchy mode. For 3D and some 2D setups, some people just need those lumens and are not a color purist.

I've seen some of the DLP's with white segments in their best mode, and though you could certainly see the color isn't perfect in an A/B, it definitely was suitable for HT for the average budget projector buyer. It might not be suitable to people like us that stare at the color all day and can see +5 dE, but it's suitable for most people.

Also, IMO the JVC RS-45's uncalibrated color isn't any better than a DLP with a white segment (though I hate to say it), because the gamma and SAT tracking is so messed up that it almost equals itself out. So if the most accurate DLP with a white segment doesn't have accurate enough uncalibrated color in best mode, neither does a JVC RS-45...

The JVC color beats it, but only after an intense calibration.

Most people buying DLP's with white segments aren't even calibrating their projectors anyhow...
Edited by coderguy - 10/6/13 at 6:13pm
post #37 of 46
Coderguy. You should be on a rock chain gang. You live to hammer people.
post #38 of 46
Haha, I am usually trying to be polite, it just doesn't come off that way sometimes. I can be a griefer...
You are a griefer as well...

I claim self defense, I am still alive because of all the blood on my hammer is not my own blood smile.gif
post #39 of 46
Quote:
Originally Posted by coderguy View Post

Yes, but you are then just re-enforcing what we have been saying all along, that it only matters in a calibrated mode, and the numbers spec'd do not show that number.

I don't mean to hammer you, but you should have made that distinction because Epson is claiming 2400 lumens CLO and WLO, insinuating the projector can do good color at 2400 lumens, which is rigatoni with baloney. Therefore this number is useless as we have been saying all along.

Unfortunately that's not what they are saying CLO, or I suppose technically Lcsw (Lumance, Color Signal White):
http://www.icdm-sid.org/downloads/
Quote:
5.4 COLOR-SIGNAL WHITE
DESCRIPTION: We measure the center luminances of three patterns
(nonatile trisequence patterns) and add those luminances to give the
luminance of color-signal white. Note the application below. Units:
cd/m2 for luminance. Symbol: LCSW.
APPLICATION: For color displays in which the input signals
conform to a standard set of RGB voltages or digital values, this
method is used to determine and to report if the luminance of the fullscreen
white (LW, the previous section, § 5.3) approximately equals
the combined luminances of the individual R,G and B primaries,
where the RGB colors are created by supplying the display with the
maximum signal for each of the R, G and B inputs independently. This
test verifies additivity of the color signal primaries. If it is the case
that:
LW != LR + LG + LB, (1)
then such a display may exhibit characteristics, employ features, or
provide processing to increase or decrease the luminance for portions
of the display color gamut. These displays may produce nontrivial errors in colorimetric reproduction and color appearance
from the intention based upon the color signal. If the relationship of the input signal primaries to signal white is properly
maintained, as in
LW ~= LR + LG + LB, (2)
then any colorimetric or color-appearance errors due to non-additivity should be minimized.
This equation may not hold
precisely for systems with nontrivial anomalies such as a lack of color-channel independence and power supply limitations.

PROCEDURE: Measure and record the luminances of the center rectangle of all three of the nonatile-trisequence patterns as
shown in Fig 2.
ANALYSIS & REPORTING: Calculate the color-signal white luminance
LCSW = LR + LG + LB. (2)

If the full screen white luminance is not approximately equal to color-signal white luminance, LW !~= LCSW, then report the
color-signal white luminance LCSW to three significant figures.

As noted, this is a completely contrived specification, designed, 100% IMO to show superiority of LCD to DLP in the case of business class DLP machines which utilize a white segment to boost lumen output at the cost of saturation. You'll note that nowhere in it is the color normalized to D65, so it's completely useless for home theater.
post #40 of 46
Quote:
Originally Posted by coderguy View Post

Yes, but you are then just re-enforcing what we have been saying all along, that it only matters in a calibrated mode, and the numbers spec'd do not show that number.

I don't mean to hammer you, but you should have made that distinction because Epson is claiming 2400 lumens CLO and WLO, insinuating the projector can do good color at 2400 lumens, which is rigatoni with baloney. Therefore this number is useless as we have been saying all along.

Also, I think it is going too far when stating having a white segment these days ruins the color so much that it is not suitable for HT, I think the better way to state this is that it is not suitable if you have other options and do not ABSOLUTELY need the extra lumens in a torchy mode. For 3D and some 2D setups, some people just need those lumens and are not a color purist.

I've seen some of the DLP's with white segments in their best mode, and though you could certainly see the color isn't perfect in an A/B, it definitely was suitable for HT for the average budget projector buyer. It might not be suitable to people like us that stare at the color all day and can see +5 dE, but it's suitable for most people.

Also, IMO the JVC RS-45's uncalibrated color isn't any better than a DLP with a white segment (though I hate to say it), because the gamma and SAT tracking is so messed up that it almost equals itself out. So if the most accurate DLP with a white segment doesn't have accurate enough uncalibrated color in best mode, neither does a JVC RS-45...

The JVC color beats it, but only after an intense calibration.

Most people buying DLP's with white segments aren't even calibrating their projectors anyhow...

Of course most projectors offer some vivid or dynamic mode that has max. light output but poor colors. Actually a DLP w/clear segment perhaps does more to distort the lumens spec. when compared to similar DLP projectors without a white segment. So which really puts up a brighter video image with reasonable colors: a projector with a 6-segment RGBRGB color wheel or a similar projector with a 7-segment RGBWRGB color wheel. The latter will have substantially higher lumens specs. but the former may put out higher lumens when calibrated. I guess we do agree that in the ideal world manufacturers would put out specs for lumens in a calibrated mode, but with the current way projector's are spec'ed we need web sties such as projector reviews or projector central to lets us know what the 'real' lumens are for a HT application..




.
Edited by Ron Jones - 10/7/13 at 6:59am
post #41 of 46
Quote:
Originally Posted by Ron Jones View Post

but with the current way projector's are spec'ed we need web sties such as projector reviews or projector central to lets us know what the 'real' lumens are for a HT application.

Yet they missed the fact that a projector they stated was razor sharp was actually only allowing a 720p image that it then upscalled to 1080p. Call me silly but I don't trust reviews very often especially when they don't bother to truly test the unit with some basic test patterns (to me their method seems to be the equivalent of a movie critic who never watches the first or last ten minutes of a film). I'm sure there are some excellent reviewers out there, but there is no way of knowing who can be trusted in the slightest.


More on topic, I can't be the only one who is completely unable to see rainbows on dlp projectors can I?

Quote:
Originally Posted by coderguy View Post

Most people buying DLP's with white segments aren't even calibrating their projectors anyhow...

I imagine the vast number of people don't calibrate there display at all. A very small number bother with some basic patterns and a speck bother with a proper calibration. A few years ago I would have shaken my head at not calibrating a display but from what I have seen what is being released is increasingly more accurate then has been the case in the past (for example now as apposed to 6 or so years ago). Odd question, does anyone know or think that projectors in cinemas are calibrated? I very much so doubt they are, and given that is what films are made for it is hard to provide an excellent argument for why calibration is a must. (I personally am in favor of calibrating but I don't see it as an absolute must for the overwhelming majority of people, though if one bought an x95 and never bothered calibrating it I would have to scratch my head).
Edited by FilmReverie - 10/7/13 at 7:12am
post #42 of 46
Quote:
Originally Posted by Christian Bergh View Post

Yes it's too bad they don't disclose which projectors that were used in the comparison.
eek.gif

This link has some more information.

http://www.sacbee.com/2013/10/03/5790671/study-reveals-9-out-of-10-people.html

Did you notice the bottom of the article you linked (my local paper), on where it came from?
Quote:
SOURCE 3LCD
post #43 of 46
Quote:
Originally Posted by stanger89 View Post

Unfortunately that's not what they are saying CLO, or I suppose technically Lcsw (Lumance, Color Signal White):

How can it not be what they are saying, it's what they listed in their specs and from that other page Zombie posted that PDF also lists the pattern they used.
Color Brightness = Color Light Output (on the Epson page), and then Epson says 2400 lumens for both white and color...
post #44 of 46
Quote:
Originally Posted by Ron Jones View Post

So which really puts up a brighter video image with reasonable colors: a projector with a 6-segment RGBRGB color wheel or a similar projector with a 7-segment RGBWRGB color wheel. The latter will have substantially higher lumens specs. but the former may put out higher lumens when calibrated.
.

The white segment one will not calibrate as well, but calibrated to its own best spec it is actually likely to still put out more lumens given the same lamp and internal light path design. The RGBRGB one's Best Mode will put out a number closer to its own max lumens as a percentage of overall light output, but the one with the white segment will still likely be FAR brighter in best mode.

I don't think these white segment designs are quite as intrusive as they were in the past, the color tables are better now.

Here is what PJC had to say about the Optoma hd25-lv (white segment variant) of the hd25... Notice that even the best mode is doubled and he states that there is not a lot of color quality sacrifice.

"The HD25-LV produces just over twice as many lumens as the HD25. Bright mode measured 2420 lumens on the HD25-LV to the HD25's 1185 lumens. Cinema mode measured 2328 lumens to the HD25's 1146. Reference mode measured 2000 lumens (vs 982), while Photo mode measured 1993 lumens. Color balance in these image modes is roughly equal to their equivalents on the HD25, so you are not sacrificing color fidelity in order to gain the extra brightness." - Bill @ PJC
Edited by coderguy - 10/7/13 at 4:11pm
post #45 of 46
3LCD could be quite good, but until they get the color uniformity down, I don't see it as a value proposition for anyone that REALLY cares about their image quality. I've suffered through too many projectors with varying tints across the projected image, and I'm just tired of it. No more 3LCD for me, that's for sure.
post #46 of 46
Quote:
Originally Posted by coderguy View Post

How can it not be what they are saying, it's what they listed in their specs and from that other page Zombie posted that PDF also lists the pattern they used.
Color Brightness = Color Light Output (on the Epson page), and then Epson says 2400 lumens for both white and color...

What I meant is that while "insinuate" that Lcsw is with good color, what they actually say is nothing along those lines. Basically I was agreeing with you, and trying to add specifics on exactly how they are trying to have it both ways.
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › 3LCD vs DLP Here we go again...