For anyone shopping for a new DLP under $1000... - Home Theater Forum and Systems - HomeTheaterShack.com

Reply
 
Thread Tools
post #1 of 17 Old 04-09-14, 05:42 PM Thread Starter
Shackster
 
Join Date: Nov 2013
Posts: 85
My System
For anyone shopping for a new DLP under $1000...

There are a ton of choices for inexpensive DLP projectors, but there are very few lamp-based(count on one hand) that use an RGB/RGBRGB colorwheel. The overwhelming majority use an RGBCMY/RGBCYW wheel which leaves you with a difficult choice between:
(bright greyscale, dim colors, good contrast, unbalanced image)
OR
(dim greyscale, dim colors, lower contrast, balanced image)
OR
(some mix between in the cases where BC can be set OFF/ON1-10)

The only way to avoid those choices with an inexpensive new DLP is to find an RGB/RGBRGB model like the Benq W1070/1080 or Mitsubishi 7900 or an inexpensive LED-based model if you can settle with the lower brightness and 720p HD.


The following three live-action comparison shots show an RGBCMY-Optoma(top) and RGB-LG(bottom). The Optoma was set to "Film" gamma (appeared the smoothest and least exaggerated), "REC" color-preset (all values are middled and increases cause clipping, tint is not adjusted), white balance is "Warm". The LG LED model below it is set to "Warm" while its Gamma, and CMS/color is adjusted in attempt to mimic the Optoma. The LG is using its brightest backlighting while the Optoma is using ECO mode and is 1ft farther back to help balance out the starting image brightness so the camera doesn't freak out as the brightness difference increases.
ALL pictures should be taken with a grain of salt because your results may vary.

The LG(bottom) is the same in all pictures, no settings are changed.
1-The first set is using BrilliantColor at minimum (there is no "OFF" setting, unfortunately) and they are pretty well matched.
2-Turning BC up to halfway has a small cooling affect on the Optoma's whitebalance and increases the hightlights a touch.
3-Cranking BC to full shows a noticeable cooler whitebalance and increased brightness (look at the highlights on faces, shaved heads, and the coat/shoulder that is facing the camera). The shadow details get a nice lift as well. The stronger colors remain untouched however and appear dimmer by compare to the newly brightened highlights (look at the right side red sign on the jumbotron). Skin-tones also become a little washed out.

The difference is notable, but I would still consider this a very watchable image (forget my lack of camera skills).
Attached Thumbnails
For anyone shopping for a new DLP under 00...-1low-outake.jpg  

For anyone shopping for a new DLP under 00...-1med-outake.jpg  

For anyone shopping for a new DLP under 00...-1high-outake.jpg  


Last edited by Ftoast; 04-24-14 at 05:57 AM. Reason: Updated
Ftoast is offline  
Sponsored Links
Advertisement
 
post #2 of 17 Old 04-09-14, 05:48 PM
Plain ole user
 
lcaillo's Avatar
Tech Guru
 
Join Date: May 2006
Location: Gainesville, FL, USA
Posts: 11,121
Send a message via AIM to lcaillo
Re: For anyone shopping for a new DLP under $1000...

How about some colorimetry rather than screen shots. Sorry, but your equating these effects to the type of color wheel with just screen shots to show is not very convincing. It is easy to take color gamut and luminance measurements. Even if your assessment turns out to be correct, how do we know it is due to the color wheel and not some limitations in the color processing? Now if you measure a number of projectors with the same color wheel type across manufacturers and find a trend, that would be convincing.




Looking for me, just google my username. I have used the same one for most sites for many years.
lcaillo is offline  
post #3 of 17 Old 04-09-14, 06:11 PM Thread Starter
Shackster
 
Join Date: Nov 2013
Posts: 85
My System
Re: For anyone shopping for a new DLP under $1000...

The differences between RGBRGB and RGBCMY/RGBCYW become more noticeable with bright colored content such as most animated cartoons, anime, and videogames.

The first picture shows the frame with proper color-balance (brilliantcolor is off). The second picture is with "brilliantcolor" ON and the resulting greyscale/color imbalance makes color look dim or darker. Look specifically at the deep oranges and reds which have turned more brown looking. The colors are actually the same in both shots, the difference is that the CYW segments are now being used to brighten whites and near-greys and have left the colors looking much darker by compare. In person, the whites of those eyes seem to glow like neon and the effect can be very distracting. Glowing eyes and teeth can be seen to a lesser degree in live-action content because many skin-tones are close enough to greyscale that they are brightened at least a little bit which helps balance things out.
Attached Thumbnails
For anyone shopping for a new DLP under 00...-bconoff.jpg  


Last edited by Ftoast; 04-23-14 at 09:32 AM.
Ftoast is offline  
 
post #4 of 17 Old 04-09-14, 06:25 PM
Plain ole user
 
lcaillo's Avatar
Tech Guru
 
Join Date: May 2006
Location: Gainesville, FL, USA
Posts: 11,121
Send a message via AIM to lcaillo
Re: For anyone shopping for a new DLP under $1000...

Did you calibrate the set to its best performance in each mode to make this comparison or just swith this mode on/off? Again, screen shots like this don't tell us much of anything.




Looking for me, just google my username. I have used the same one for most sites for many years.
lcaillo is offline  
post #5 of 17 Old 04-09-14, 06:36 PM Thread Starter
Shackster
 
Join Date: Nov 2013
Posts: 85
My System
Re: For anyone shopping for a new DLP under $1000...

Quote:
lcaillo wrote: View Post
How about some colorimetry rather than screen shots. Sorry, but your equating these effects to the type of color wheel with just screen shots to show is not very convincing. It is easy to take color gamut and luminance measurements. Even if your assessment turns out to be correct, how do we know it is due to the color wheel and not some limitations in the color processing? Now if you measure a number of projectors with the same color wheel type across manufacturers and find a trend, that would be convincing.
I agree that proper equipment and testing would make this a true scientific test/result which it currently is not. As it stands all I have are visual tests and information from TI on how brilliantcolor works. The main points however are so straight-forward that they don't require much beyond deductive reasoning: Take two otherwise equal projectors, one with a HT RGBRGB CW and another with a business style RGBCMY CW and compare color brightness and white brightness. The secondary color segments come into play (along with RGB spoke mixes) with BC engaged, but aren't used with BC fully off. Even RGBRGB brilliantcolor using just the "spokes"/mixes between segments can give an additional 15%+ increase in peak white brightness, but no amount of secondary color can be used to increase the brightness of a pure primary. As colors get farther from greyscale they rely more on their primary...which the RGBCMY projector will have available for roughly half the time of an RGBRGB projector. I have only so far been able to personally compare the color/color+greyscale/greyscale brightness changes presented by BC on/off off/on1-10 on three very different projectors.

I'm sure you're aware of the countless tests and reviews confirming the differences, but I was more interested in putting out a simple visual comparison for anyone less interested in stats and numbers and more interested in "What does that all mean".

Let me know if I'm portraying anything too strongly or bias, I get overly passionate about this topic sometimes.

Last edited by Ftoast; 04-24-14 at 09:22 PM.
Ftoast is offline  
post #6 of 17 Old 04-09-14, 06:55 PM Thread Starter
Shackster
 
Join Date: Nov 2013
Posts: 85
My System
Re: For anyone shopping for a new DLP under $1000...

Quote:
lcaillo wrote: View Post
Did you calibrate the set to its best performance in each mode to make this comparison or just swith this mode on/off? Again, screen shots like this don't tell us much of anything.
The shots where all taken after calibrating blacklevel/peakwhite/peakcolor with their respective sources (USB/USB/HDMI). The BC on/off projector was displaying from a 0-255 (not 16-235) HDMI set for 0 to have no movement and 1 to show dithering, 253-254 clipped into white and 252/below showing grey, and colors set at maximum before clipping. This particular projector didn't change its black/white/color clipping points during calibration with BC being tested in on/off positions, so I didn't change any settings between shots. I did not set gamma/tint/CMS, and whitebalance (which was only available with BC ON) was only set to match the WB with BC off.

I should also note that this particular projector only had BC options for ON or OFF whereas many others offer differing intensities (1-10) when BC is ON.

Both the projectors I tested showed changing BC/secondaries on/off to not affect the lumen-output of heavily saturated colors (not even secondaries), somewhat affect mid-saturation colors, and heavily affect greyscale. Personally the color imbalance didn't bother me too much on live-action content, I just perceived it like an overly high gamma setting. I found animated content to be outright unwatchable with the imbalance, and the balanced image backed up what others' test results have shown where the necessary lumen decrease to properly match color/white brightness is only putting out around 600-700lm on a fresh bulb.

The realization that a swap to an RGBRGB colorwheel would cost the company $0 and roughly double the color brightness was a frustrating one. But by using an RGBCMY/RGBCYW colorwheel a manufacturer can increase the stated lumen output by 50-85% at zero cost by sacrificing half of the calibrated brightness (which they don't have any obligation to tell you). You, me, and anyone else who reads enough about HT probably understands this, but there's many a visual learner and first-timer that might not yet know how it affects them and why they should care. The inexpensive DLP circuit is largely copy/paste, CR varies a little between 1700:1-2500:1, output varies a little between 190-240W, native resolution is widely available at 720-1080p, and shift/zoom options are small to nill..the biggest differences are in colorwheels (and software) and manufacturers are oddly quiet on that particular front nowadays. It's one line and eight-ish digits of information, they aren't neglecting it to save ink. It pads a stat at the consumers picture's expense and they do it because they assume people won't notice or won't care and...I should probably calm down.

Last edited by Ftoast; 04-23-14 at 09:08 AM.
Ftoast is offline  
post #7 of 17 Old 04-09-14, 07:06 PM
Plain ole user
 
lcaillo's Avatar
Tech Guru
 
Join Date: May 2006
Location: Gainesville, FL, USA
Posts: 11,121
Send a message via AIM to lcaillo
Re: For anyone shopping for a new DLP under $1000...

It would be helpful in assessing your conclusions if we saw calibration data.




Looking for me, just google my username. I have used the same one for most sites for many years.
lcaillo is offline  
post #8 of 17 Old 04-09-14, 09:34 PM Thread Starter
Shackster
 
Join Date: Nov 2013
Posts: 85
My System
Re: For anyone shopping for a new DLP under $1000...

Quote:
lcaillo wrote: View Post
It would be helpful in assessing your conclusions if we saw calibration data.
I'll give what little I can. I was visually comparing (via test patterns) this Benq model with brilliancolor on/off both side-by-side and sequentially to a pair of dimmer projectors (Lg, Aaxa) that both do NOT feature BC and respectively output roughly 400-450lm and 130-150lm. The Benq is rated at 3500lm but its brightest preset is terribly green. More balanced presets showed peak white with BC on to be almost 4X brighter than the 400-450lm Lg, but only slightly brighter than the Lg when BC was off. Likewise, with BC on, as colors deviated away from greyscale they quickly dimmed down to nearly match the 400-450lm Lg. With BC off, colors were just slightly brighter than the Lg by a visually consistant amount through to grey.

On/off native-contrast with BC OFF was fairly low, under 1000:1, but its black was pleasantly neutral which made the most of the low ratio. Turning BC on/off does not appear to affect black level, but it does increase peak-white by over 3X. The perceived contrast increase doesn't seem to look as significant as the numbers suggest in practice (watching regular content), but that's probably due to color levels/contrast largely remaining unchanged and closer to 1000:1.

For all the negatives, I consider this to be one of the better implementations of BC. Its color/brightness transitions didn't poster or clip, and beyond the odd look of a near 4:1 white/color-brightness disparity it didn't seem to add any strange artifacts.
I also tested a Vivitek model with BC off/on1-10 which showed similar results as listed above, but to a lesser degree most likely due to it using RGBCMY which has a smaller effect than RGBCYW. The Vivitek, however, added jarring artifacts whenever BC was on (at any level 1-10) and most higher-contrast edges were given speckled color edges or halos.

I only compared color accuracy visually (which is to say, not at all), but the BenQ with BC off displayed pleasing colors. With BC on, and white/greyscale content onscreen to affect perception, colors looked darker and brown-ish.

That's all I can think of from a numbers standpoint, but if you have any questions I might have an answer or elaboration.

Last edited by Ftoast; 04-09-14 at 11:05 PM.
Ftoast is offline  
post #9 of 17 Old 04-09-14, 11:02 PM Thread Starter
Shackster
 
Join Date: Nov 2013
Posts: 85
My System
Re: For anyone shopping for a new DLP under $1000...

The end result is that most inexpensive DLP projectors have very low color-brightness and therefore very low calibrated brightness. If you're curious about a particular model, ProjectorCentral, ProjectorReviews, and many other sites (including this one) take measurements that compare "brilliantcolor" on and off. The HIGHEST listed output I remember seeing on an RGBCMY projector with BC off is only 675lm on a fresh bulb. Many people run projectors in eco mode to lower fan noise which can cut brightness by 15-25%, and an aging bulb can expect to lose 50% of its brightness by half-life. That "3000-3500lumen" claimed projector is likely hitting under 700lm, on eco that's 560lm..after a couple years it could be 350lm.

Most RGBRGB projectors only claim 2000-ish lumens. Many RGBCMY/RGBCYW projectors claim over 2500. Ironically if you're searching for brighter colors from a DLP, you can look for lower claimed brightness..then call the manufacturer and have them look up the colorwheel information from the SERVICE manual of that model. They list the information as 6 letters or 6 letter-and-number combinations. Any letter that isn't an R,G,or B means a segment of the wheel is NOT able to be properly used to create color.

I'll leave you with this comparison shot. They are all set for brightest whites before clipping (BC off). The bottom is noticeably closer to the wall while the top and middle are equally distant on a ledge with the top being rightsideup and the middle being upsidedown. The top is claiming 3500lm, middle measures 400-450lm, and bottom measures 130-150lm. I bought it because it claimed 3500lm..I returned it the next day because THIS is how much brighter it was in reality, and it would be HALF as bright in a year or two with my viewing habits.

If you're shopping around and can return what you buy, give it a try. If you can't return or will get stuck with a shipping/restocking fee, don't be fooled by manufacturer specs, look up a thirdparty review with test results and keep in mind the inherit differences in colorwheels.
Attached Thumbnails
For anyone shopping for a new DLP under 00...-pj3.jpg  

Ftoast is offline  
post #10 of 17 Old 04-10-14, 04:00 AM Thread Starter
Shackster
 
Join Date: Nov 2013
Posts: 85
My System
Re: For anyone shopping for a new DLP under $1000...

The Long story:

DLP works by strobing colored light SO fast that your eyes/brain perceives regular colors and images, even though a singlechip projector is actually only showing red/green/blue/black in its pixels.
Light goes through a colorwheel with red,green,blue glass and the red/green/blue light is bounced off a little DLP chip with one tiny mirror for each pixel.
Each tiny mirror has a hinge underneath that turns it to face one way or the other. When the mirror faces one way (on), the light goes out the lens and onto your screen, when the mirror faces the other way (off), light gets shined onto a heatsink inside the projector. Each mirror/pixel can spend all its time on for white, off for black, or strobing a particular number of flashes-per-second to make any shade of grey in between. Since the mirrors are actually flashing red, green, and blue colored light instead of white, each mirror/pixel is showing a mix of any shade of red with any shade of green and any shade of blue. JUST LIKE YOUR COMPUTER SCREEN right now is a bunch of red/green/blue pixels at varying brightness, the DLP is showing red/green/blue pixels at varying flashes-per-second to create the different brightness! Where LCD like your computer screen blocks some light in each pixel with liquid chrystal, DLP flashes its pixel/mirror half as many times to block light in each pixel.

Just like your computer screen, DLP has to make every color by mixing red, green, and blue. Red and green make yellow, green and blue make cyan, blue and red make magenta and so on while red, green, and blue all together make white. The perfect part about this is, red and green mixing together to make yellow causes the yellow to be brighter than either red or green is by itself because it's actually TWO colors shining together (in LCD) or twice as often (in DLP), and this is the proper brightness your eyes expect it to be. Just like white, being a mix of red, green, and blue is brighter than any single color by itself, yellow is brighter than either of the colors that are mixed to make it..and so are the other colors that are made by mixing. The whole system depends on these three colors being proper to create other colors properly, and the resulting brightness of all the other colors AND white are then perfectly balanced exactly how they should be. Since these colors are all standardized, film-makers can make a movie and it'll be decoded on your screen to look how they shot it to look. The problem comes when display-makers decide to tamper with or deviate from these standards. If for instance green on my screen is actually kind of yellow, everything that is supposed to look green or be made with any mix of green will now have more of a yellow tint to it.

So why would a manufacturer want to deviate from the standard anyway? What possible benefit could it bring when it's messing with the quality of the image? Brightness. If white pixels(LCD) or white color-segments(DLP) are added to pass more light through, it'll make the display brighter. Well my LCD screen is full of red/green/blue pixels and any other colored pixels would have to take up space that would normally have an R/G/B pixel. This means that while the extra pixels are increasing white's brightness, they're actually DECREASING color brightness by taking away some red, green, and blue pixels. In the same way, adding a white segment to a DLP colorwheel is shrinking the space that would normally be red, green, or blue.

At some point this became obvious enough to the average consumer that display makers decided to find a sneaky way around. Instead of adding a white pixel, add a yellow one. It's 92% as bright as white, but the consumer will think it ENHANCES color while it actually lowers color brightness. In the same way, instead of adding a big white segment on a colorwheel, add three smaller cyan, yellow, and magenta/white segments. Let them think this new 6-color wheel is adding colors when it's simply taking advantage of the fact that the C/M/Y half passes 2X more white light than an equal portion of R/G/B would. Even if the cyan/yellow/magenta sections were used to increase the brightness of C/M/Y colors, it would just make those colors 3X brighter than they should be, and STILL wouldn't be able to help red, green, or blue at all. In fact, their very presence on the colorwheel is taking away from space that could be used to make red, green, and blue brighter. And having R/G/B brighter means every color including white gets to be brighter in the proper proportion so nothing is wasted. When one or more color is brighter than the rest, it looks wrong and makes the image appear strange. Likewise, when white is much brighter than the colors, it looks wrong and makes white and near-white appear to glow while colors will look too dark and muddy.

Why would manufacturers do this, who buys a display that looks wrong? Many first-time buyers may have a hard time telling the difference, or assume it's their fault. The display might look passable, and they bought the one with the higher brightness number, so they're happy. Meanwhile the manufacturer has their money and the incentive to keep making inferior displays with higher claimed brightness. The first-time buyer doesn't realize they could've purchased a display that claimed a slightly lower number, but put out a properly balanced image that was actually 2X brighter. The worst part is, using a proper colorwheel or proper R/G/B LCD doesn't cost the display-maker any more money..in fact, adding a fourth/odd color to an LCD is an expensive endevor! The only thing they lose, is a tiny bit off the top of that claimed brightness number. But a consumer will buy according to that number without fully understanding what it means. The more informed you are as a consumer, the more you can shape the market you are buying from with your wallet.
Ftoast is offline  
Reply

Bookmarks

Tags
$1000 , dlp , shopping

Quick Reply
Message:
Options

Register Now




PLEASE COMPLETE ALL REQUIRED FIELDS BELOW... THANKS!

REQUIRED FIELDS ON THIS PAGE
YOU MUST COMPLETE ALL OF THESE

Username
Password
Confirm Password
Email Address
Confirm Email Address
Random Question
Random Question #2




User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.

Password:


Confirm Password:
Email Address
PLEASE READ BELOW PRIOR TO ENTERING AN EMAIL ADDRESS!

ATTENTION!

YOU MUST ACTIVATE YOUR ACCOUNT!

Activation requires you reply to an email we will send you after you register... if you do not reply to this email, you will not be able to view certain areas of the forum or certain images... nor will you be able download software.

AN INVALID EMAIL ADDRESS WILL CAUSE YOUR ACCOUNT TO BE DELETED!

See our banned email list here: Banned Email List

We DO NOT respond to spamcop, boxtrapper and spamblocker emails... please add @hometheatershack DOT com to your whitelist prior to registering or you will get nowhere on your registration.


Email Address:
OR

Log-in










Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page



Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML is not allowed!
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

 


For the best viewing experience please update your browser to Google Chrome