I spoke with a guy in the TI booth at CES and he told me that the color wheel spins so fast now that no human can see rainbows. What I don't quite recall was if that was only true when LED is the light source.

What are your thoughts?

~Jay

nope still rainbows.

I think they need 32x color wheels to completely eliminate it. We are at 7x now.

I cant imagine what that rate of spin would sound like as some of the current crop sounds like a turbine engine now:(

I spoke with a guy in the TI booth at CES and he told me that the color wheel spins so fast now that no human can see rainbows. What I don't quite recall was if that was only true when LED is the light source.

What are your thoughts?

~Jay

I see RBE on some mono DLP PJ's even models with 10800 rpm wheels. So I would suggest the comment from the TI isn't currently accurate.

D

No, rainbows are still there and probably will never go away (unless you're talking 3-chip DLP). Faster spinning color wheels introduce other problems.

Some manufacturers are now using color wheels with primary and secondary colors to make the image brighter at D65. While this is good for more light output, it increases rainbows as well. So a CW with primary and secondary colors is a step forward and backward at the same time.

i see rainbows on all the new dlp televisions. I have not seen any new dlp pjs lately.

I see dead rainbows.

I have a friend who can't watch my HD70. He sees rainbos all over. I told him to take a nother hit of acid and calm down.

My Wife still sees RBE on the newer DLP pjs. I think she is human... :eek:

I spoke with a guy in the TI booth at CES and he told me that the color wheel spins so fast now that no human can see rainbows. What I don't quite recall was if that was only true when LED is the light source.

What are your thoughts?

~Jay

Not a single guest I have had over has ever mentioned "rainbows". Then again, almost all of them know nothing about AVS. AVS members I have had over must simply be too polite. I keep looking and dang if I see them.
BTW, my neighbor has an RS1 for comparison.

I spoke with a guy in the TI booth at CES and he told me that the color wheel spins so fast now that no human can see rainbows. What I don't quite recall was if that was only true when LED is the light source.

What are your thoughts?

~Jay

He is absolutely incorrect. Every single chip DLP I've seen, including the latest models I've been able to see, have shown obvious rainbows. (Obvious to folks like me, that is, who sees them). In fact, in some instances manufacturers, in tring to cut some costs and prices, have actually lowered the speed of their color wheels (I believe Marantz is one example...someone can correct me if that's wrong).

I only wish Mr. T1 Booth was correct. If he was, I would possibly choose one of the new DLP models.

Coming from a tech company, that is typical of sales talk. That person receives sales literature that states that the projector vendors are developing higher segmented faster spinning color wheels that virtually eliminate rainbows. So he takes it as the bible and as any good sales person does, repeats it with a dash of exaggeration. Events like CES with booths full of LCD manufacturers and the DLP vs LCD wars going strong, those types of comments will always occur. What would surprise me is if he actually said "we are working with projection manufacturers to reduce the rainbow effect and are getting better at it, but some will still be susceptible to it until we get rid of the color wheel altogether." :)

...and even 3-DLP.
I see occasional rainbows on my front projector, but am not overly succeptible to them. I have seen 3 chip DLP and have never seen rainbows.

Rainbows are an artifact of how single chip DLP uses the persistance in your eye/brain to construct colors. A single chip DLP doesn't display the 3 primary colors at the same time. 3 chip DLPs don't have this problem. In a 3 chip system, there is one DLP chip for each primary color. This is how LCDs and LCOS display color, and you've never heard anyone complain about rainbows, have you?

I see occasional rainbows on my front projector, but am not overly succeptible to them. I have seen 3 chip DLP and have never seen rainbows.

Rainbows are an artifact of how single chip DLP uses the persistance in your eye/brain to construct colors. A single chip DLP doesn't display the 3 primary colors at the same time. 3 chip DLPs don't have this problem. In a 3 chip system, there is one DLP chip for each primary color. This is how LCDs and LCOS display color, and you've never heard anyone complain about rainbows, have you?

Actually, I along with a few others have noted a rainbow effect - a similar type of image break up (without the color fringes) on some 3 chip DLPs. I thought it couldn't be, for just the reasons you state. But when I brought it up others mentioned seeing the issue as well and when some of the more technically knowledgeable joined in they said it was indeed a possibility. Not exactly the rainbow effect, but similar. I can't remember the exact technical details though, unfortunately.

Do not try to see them. If you do not see them you are blessed! ;)

"Rainbows are an artifact of how single chip DLP uses the persistance in your eye/brain to construct colors. "

It uses persistence for both color *and* brightness.

"A single chip DLP doesn't display the 3 primary colors at the same time. 3 chip DLPs don't have this problem. In a 3 chip system, there is one DLP chip for each primary color. This is how LCDs and LCOS display color, and you've never heard anyone complain about rainbows, have you?"

That is not an accurate comparison. DLP pixels are binary. They are either on or off. To control brightness levels of the pixel they dither... flash on/off very very rapidly and uses the persistance in your eye/brain to construct relative brightness levels. So even with 3 chip DLP the possibility of rainbows exists as you can have times when one color is on and another is off due to the dithering of the pixels for the various color/brightness components on the different DLP chips. Should be much less obvious then on single chip but still a possibility.

Traditional 3 chip LCD and LCOS are not binary pixels. The pixels are not simply on or off. They can directly modulated the brightness of the pixel as such they do not have to dither and rely on persistence of vision for brightness levels which means you won't have the situation of one color being on while another is off and getting RBE.

Shawn

"Rainbows are an artifact of how single chip DLP uses the persistance in your eye/brain to construct colors. "

It uses persistence for both color *and* brightness.

"A single chip DLP doesn't display the 3 primary colors at the same time. 3 chip DLPs don't have this problem. In a 3 chip system, there is one DLP chip for each primary color. This is how LCDs and LCOS display color, and you've never heard anyone complain about rainbows, have you?"

That is not an accurate comparison. DLP pixels are binary. They are either on or off. To control brightness levels of the pixel they dither... flash on/off very very rapidly and uses the persistance in your eye/brain to construct relative brightness levels. So even with 3 chip DLP the possibility of rainbows exists as you can have times when one color is on and another is off due to the dithering of the pixels for the various color/brightness components on the different DLP chips. Should be much less obvious then on single chip but still a possibility.

Traditional 3 chip LCD and LCOS are not binary pixels. The pixels are not simply on or off. They can directly modulated the brightness of the pixel as such they do not have to dither and rely on persistence of vision for brightness levels which means you won't have the situation of one color being on while another is off and getting RBE.

Shawn

I have noticed a similar type of "color break up" on 3 chip systems. I work in a post production facility, and our film colorists (3 of them) each have a Christie 3-chip dlp projector in their suite, and I can see it on the screens. I also noticed it at a dlp cineplex. Not anything nearly as bad as your run-of-the-mill 1 chip home theater projector, but something quite similar is definately there....

It seems to me that some of the latest DLP projectors have more RBE than recent past products. Maybe it is the whole secondary color wheel thingy that was mentioned above.

One other thing....

I wonder if people that have had Lasik see more RBE?

97% of all people who see DLP rainbows are on this forum.

TA

97% of all people who see DLP rainbows are on this forum.

TA

I think that's because 97% of people who see them had to look for them ;)

That's the case with me, and now I am learning to ignore them.

Right now I have in my house a h710 DLP, and a VW60. Except for rainbows I think DLP is better in every way.

I was hoping that new DLP projectors had solved the problem.

the guy at CES was not a sales guy, he was an engineer working with LED. I think he was saying that the LED can turn on and off so fast that it can be in sync with the color wheel to eliminate rainbows. Maybe he was wrong.

~Jay

97% of all people who see DLP rainbows are on this forum.

TA

Maybe that's because they have the knowledge to identify the issue and know it should not be there. Many of the public at large have no idea of the myriad of artefacts that plague displays. I also notice more RBE when using 24hz source rather than 60hz which means a trade off between RBE or 3:2 judder.

A similar question was raised on the UK forum recently -

http://www.avforums.com/forums/showthread.php?t=684507&page=3&highlight=fusion+flicker

Maybe ignorance is bliss. :)

D

Maybe ignorance is bliss.

I concur with that, Dazzer.

I dont see rainbows on my DLP and I certainly wont go looking for it. Then again, you didnt see any rainbows on the recent Marantz DLP you demoed, either .. and very recently you were witnessing rainbows on DLP machines.

With my recent, and very substantial, experience with the HD1/RS1 if anyone has issues with seeing rainbows I dont think you can go wrong with the RS1 or RS2. I havent yet seen the VW200, but I also owned a very superb VW100 and I can say that I felt that the RS1 was a huge upgrade from it.

Then again, you didnt see any rainbows on the recent Marantz DLP you demoed, either .. and very recently you were witnessing rainbows on DLP machines.



Cat

That's right. I viewed the 11S1 for around 3 hours with a variaty of material and didn't observe any RBE. I spent several hours yesterday with the Sim2 HT3000e and at 60hz the image was stable but with 24hz RBE started to appear. :(

I've spent time with several current mono DLP PJ's and for me the only machines that I see RBE is the Sim2 HT380 and 3000e. Both of these PJ's are relatively bright which may affect my RBE trigger point. YMMV. :)

D

I think that's because 97% of people who see them had to look for them ;)



~Jay

Before I ever knew what it was I'd be strolling the isles of, say, a Best Buy looking at the TVs. This was when there were actually RPTVs still on the shelves :)

Anyway, as my eyes would move from display to display I'd keep seeing these weird flashes. I didn't know what it was but noticed it seemed to happen only when passing by certain RPTVs. They were of course the DLP RPTVs, as I later learned about the artifact. But it was very distracting even before I'd heard about it.

Now the same thing is happening with the new LED vehicle tail/brake lights. Most of our buses now use those lights (which pulse on and off so quickly you aren't supposed to notice). If a bus is in front of me, especially at night, as my eyes scan the traffic the tail-brake light breaks up into stuttered trails in my vision. Very annoying. A few cars have those break lights too. I can only hope for my sanity that LED tail lights don't become the norm. Oh man...whole roads filled with the rainbow-like effect....:(

Rich,

"Now the same thing is happening with the new LED vehicle tail/brake lights. "

Yes, those are really annoying. Cadillacs have them and when the parking lights are on they drive me nuts. Same thing with them as DLPs... they dither the LEDs to control brightness. On the Cadillacs they are annoying when it is parking lights (dimmer) but I don't notice it when the brake lights (brighter) are on. They may not be dithering with the brake lights on.

I would assume the same people that see rainbows on DLP are the same that would notice this on cars too.

Shawn

"Rainbows are an artifact of how single chip DLP uses the persistance in your eye/brain to construct colors. "

It uses persistence for both color *and* brightness.

"A single chip DLP doesn't display the 3 primary colors at the same time. 3 chip DLPs don't have this problem. In a 3 chip system, there is one DLP chip for each primary color. This is how LCDs and LCOS display color, and you've never heard anyone complain about rainbows, have you?"

That is not an accurate comparison. DLP pixels are binary. They are either on or off. To control brightness levels of the pixel they dither... flash on/off very very rapidly and uses the persistance in your eye/brain to construct relative brightness levels. So even with 3 chip DLP the possibility of rainbows exists as you can have times when one color is on and another is off due to the dithering of the pixels for the various color/brightness components on the different DLP chips. Should be much less obvious then on single chip but still a possibility.

Traditional 3 chip LCD and LCOS are not binary pixels. The pixels are not simply on or off. They can directly modulated the brightness of the pixel as such they do not have to dither and rely on persistence of vision for brightness levels which means you won't have the situation of one color being on while another is off and getting RBE.

ShawnThanks Shawn for this lucid explanation; best I have seen. I'm cursed with vision that detects, and a brain that responds, to all these issues.

Now that I'm aging though the "rainbow effect" competes with other problems--including "floaters"!!! :D

Now that I'm aging though the "rainbow effect" competes with other problems--including "floaters"!!! :D

Betcha I can beat you in the "floaters" department. They suck...

What are floaters?

"Rainbows are an artifact of how single chip DLP uses the persistance in your eye/brain to construct colors. "

It uses persistence for both color *and* brightness.

"A single chip DLP doesn't display the 3 primary colors at the same time. 3 chip DLPs don't have this problem. In a 3 chip system, there is one DLP chip for each primary color. This is how LCDs and LCOS display color, and you've never heard anyone complain about rainbows, have you?"

That is not an accurate comparison. DLP pixels are binary. They are either on or off. To control brightness levels of the pixel they dither... flash on/off very very rapidly and uses the persistance in your eye/brain to construct relative brightness levels. So even with 3 chip DLP the possibility of rainbows exists as you can have times when one color is on and another is off due to the dithering of the pixels for the various color/brightness components on the different DLP chips. Should be much less obvious then on single chip but still a possibility.

Traditional 3 chip LCD and LCOS are not binary pixels. The pixels are not simply on or off. They can directly modulated the brightness of the pixel as such they do not have to dither and rely on persistence of vision for brightness levels which means you won't have the situation of one color being on while another is off and getting RBE.

Shawn
Thanks for the clarification. I hadn't thought about binary nature of DLP vs LCD before.


What are floaters?
Little pieces of "goober" that somehow "break off" and float around in the liquid inside your eye.

Thanks for the clarification. I hadn't thought about binary nature of DLP vs LCD before.



Little pieces of "goober" that somehow "break off" and float around in the liquid inside your eye.

Crap. Another thing to not look forward to in the future.

I spoke with a guy in the TI booth at CES and he told me that the color wheel spins so fast now that no human can see rainbows. What I don't quite recall was if that was only true when LED is the light source.

What are your thoughts?

~Jay

LED light source projectors (like Phlatlight in Samsung LED RPTV's) do not have a color wheel- the LED's pulse on and off. The speed the pulsing is capable of is currently limited by software controling it and the refresh rate of the DLP chip.

For those of you bothered by LED's in brake lights, I would anticipate that newer models would pulse faster to dim output (I believe they pulse at 30-60hz currently). If given steady dc current, they do not pulse. They do offer a number of benefits from a safety, reliability and energy consumption perspective. Anyway, you had better get used to them, DOE projects LED's to take over general illumination in the next 15 years.

Before I ever knew what it was I'd be strolling the isles of, say, a Best Buy looking at the TVs. This was when there were actually RPTVs still on the shelves :)

Anyway, as my eyes would move from display to display I'd keep seeing these weird flashes. I didn't know what it was but noticed it seemed to happen only when passing by certain RPTVs. They were of course the DLP RPTVs, as I later learned about the artifact. But it was very distracting even before I'd heard about it.

Now the same thing is happening with the new LED vehicle tail/brake lights. Most of our buses now use those lights (which pulse on and off so quickly you aren't supposed to notice). If a bus is in front of me, especially at night, as my eyes scan the traffic the tail-brake light breaks up into stuttered trails in my vision. Very annoying. A few cars have those break lights too. I can only hope for my sanity that LED tail lights don't become the norm. Oh man...whole roads filled with the rainbow-like effect....:(

Damn! Thats what that was!:mad:

My wife and I noticed this on our last trip to the Mainland, while driving on the 105 Freeway. I just thought we were jet-lagged and or going nutty. This would suck big time if all cars change to LED brake lights.

By the way, I do see RBE--wife does not. However, I almost never see RBE on my Marantz VP4001. I did see it on every other DLP I have owned.

Cat

That's right. I viewed the 11S1 for around 3 hours with a variaty of material and didn't observe any RBE. I spent several hours yesterday with the Sim2 HT3000e and at 60hz the image was stable but with 24hz RBE started to appear. :(

I've spent time with several current mono DLP PJ's and for me the only machines that I see RBE is the Sim2 HT380 and 3000e. Both of these PJ's are relatively bright which may affect my RBE trigger point. YMMV. :)

D

The brighter the projector the higher the percentage of people who will see rainbows. I don't see rainbows but I get tension headaches from eyestrain.

I get eyestrain too. I don't see rainbows, but I get eyestrain.

What is the difference between RBE and the eyestrain that many dlp owners experience?

edit: NM, I found some interesting articles including one by ProjectorCentral on this. It doesn't get down to the nitty gritty, but basically states that the RBE and eyestrain are caused by similar forces.

I have had a sharp 2K and never saw rainbows... but since I switched to an AE2000 I have not had the eye strain I used to get with the DLP. Not sure if it's because the AE2000 is brighter or if it's because of LCD vs. DLP.

However that being said, I must have had 100's of people over to see movies on the DLP and no rainbow comments. VERY few see them.

I look forward to seeing a LED driven DLP FP. As of now, I'd say no, rainbows / eye strain are not a thing of the past but they are reduced.

This was a, pardon the punn, eye opener of a thread. So DLP's can cause eye strain watching even if you don't see the rainbows? I see rainbows so it's moot to me but I thought they were great so long as you don't see rainbows. I was also a bit surprised that 3 chip DLP still has RBE. Essentially everything I've read on this forum stated as fact that 3 chip eliminates rainbows. I'd be beyond irate if I'd spent $30k to get a high end 3 chip DLP and then discovered I could see rainbows and got eye strain from watching it! Why do people on the forum repeat that 3 chip eliminates rainbows if it doesn't?

I don't think but a small percentage of the forum has ever viewed a 3chip dlp. Most information out there says that the RBE is due to the color wheel so naturally eliminating the color wheel, most would assume the RBE would go with it. I personally was not aware that the RBE could still be present even without a color wheel so this is an eye opener of a thread for me as well.

"Essentially everything I've read on this forum stated as fact that 3 chip eliminates rainbows."

Someone awhile back posted a picture showing color separation (RBE) on a three chip DLP.

Shawn

"Essentially everything I've read on this forum stated as fact that 3 chip eliminates rainbows."

Someone awhile back posted a picture showing color separation (RBE) on a three chip DLP.

Shawn

Was it not MC?

I do notice a difference with my VW200 and before that VW100. When we had the Optoma H79 up there was considerably more eye strain. With the sony sxrd pjs and my outdoor sanyo Z5 it is stress-free viewing...

"Was it not MC?"

No. They were moving a pencil or something rapidly in the light path and photographed that on screen. You could see the color separation over the area the pencil was moving. If it was MC it would not have been localized like that.

Shawn

RBE does not happen with 3 chippers. Its a color wheel issue caused by sequential color application and relys on the users temporal averaging. 3 chip is non sequential and therefore does not rely on temporal averaging at all.

Any other artifact seen is not RBE.

Also the explanation of color separation due to the mirrors switching is incorrect. The panel response is up to 500 times faster than LC based panels.

Due to PWM you can still get some colour separation artifacts with 3 chippers apparently. You won't necessarily get an identical overlay of RGB per frame reproduction (each RGB chip may do something slightly different to each other), so some people say they can still see some form of RBE. Then again, some people see a similar kind of artifact on CRT tv and plasma too. :)

Due to PWM you can still get some colour separation artifacts with 3 chippers apparently. You won't necessarily get an identical overlay of RGB per frame reproduction (each RGB chip may do something slightly different to each other), so some people say they can still see some form of RBE. Then again, some people see a similar kind of artifact on CRT tv and plasma too. :)

They will have differing mirror positions instantaneously, but switching speeds vastly exceed temporal averaging thresholds.

There's temporal and spatial dithering, but some people say still see something like RBE on 3 chippers. I don't know if the pic that was shown some time back (maybe before you joined here) was a genuine RBE image from a 3 chip DLP (I forget the details), but something like it was visible apparently.

Gary

There's temporal and spatial dithering, but some people say still see something like RBE on 3 chippers. I don't know if the pic that was shown some time back (maybe before you joined here) was a genuine RBE image from a 3 chip DLP (I forget the details), but something like it was visible apparently.

Gary

Im aware of the dithering types and mitigating factors.

As to the photo, i cant say what it was of or what it showed or if it were genuine, but it certainly wasn't RBE on a 3 chip.

"RBE does not happen with 3 chippers. Its a color wheel issue caused by sequential color application and relys on the users temporal averaging. 3 chip is non sequential and therefore does not rely on temporal averaging at all."

That is not accurate. Brightness on a DLP relies on the users temporal averaging too. What happens when the red pixel is at 100%, green is at 10% and blue is at 50%. At that pixel on screen you will not have all colors on at the same time.

"The panel response is up to 500 times faster than LC based panels."

And it is 500 times faster (or whatever) because it *has* to be to work at all. If DLP pixels were as 'slow' as LC panels they would have very poor control over brightness levels.

LC based panels do not need to dither to control brightness levels and as such there is no need for them to be as fast as DLP. DLPs speed is an absolute requirement to work at all because of their binary nature. Without that speed you would have very limited brightness control in a DLP.

Shawn

When I've noticed the motion artifact on the 3 Chip DLPs it has been in the form of flashes or trails, just like the rainbow effect, but with no rainbow colors. I remember I was watching Star Wars ROTS, the opening space battle. As my eyes roamed the scene the high-contrast high-lights on the edges of the ships etc seemed to flash or brake up, in to stuttering trails. Pretty much like those new LED lights on some vehicles. It really took me by surprise as I wasn't expecting, or looking for any such artifacts from a 3 Cipper.

FWIW.

"
That is not accurate. Brightness on a DLP relies on the users temporal averaging too. What happens when the red pixel is at 100%, green is at 10% and blue is at 50%. At that pixel on screen you will not have all colors on at the same time.


The brightness is modulated but the colors are non sequential. All three are combined via the wedge.The mark space ratio determines the brightness.Each color will have an instantaneous but varying ratio. The switching frequencies are way way above resolvable limits.

When I've noticed the motion artifact on the 3 Chip DLPs it has been in the form of flashes or trails, just like the rainbow effect, but with no rainbow colors. I remember I was watching Star Wars ROTS, the opening space battle. As my eyes roamed the scene the high-contrast high-lights on the edges of the ships etc seemed to flash or brake up, in to stuttering trails. Pretty much like those new LED lights on some vehicles. It really took me by surprise as I wasn't expecting, or looking for any such artifacts from a 3 Cipper.

FWIW.

You may have seen some artifact or other caused by any number of factors, not limited or exclusive to, the PJ. That actually sounds like a VP issue.

You didn't see RBE, which is what were talking about.

quick questions :

-I see green trails all the time on plasma TVs (but it doesn't bother me at all), am I going to be prone to see RBE?

-I absolutely can't live with the slow refresh rate of LCD flat panel displays, is it going to be the same with a LCD front projector?

-Overall, I don't have problem with 24fps content telecined to 60fps but I like fast action sports like hockey and football. I also like movies on blu-ray. Given what I've said, would DLP still be the best option for me? I would like to buy a 720p pj this summer (after school is over...), but I want have the time to spend a lot of time shopping for pj live. I'll buy on the web and will likely never see the pj in action before receiving it. Thanks.

You may have seen some artifact or other caused by any number of factors, not limited or exclusive to, the PJ. That actually sounds like a VP issue.

You didn't see RBE, which is what were talking about.

Sorry, I thought it was pretty clear: I wasn't talking about the rainbow effect on the 3 chip DLPs but rather a rainbow-effect-like artifact: the same sort of image breaking up, stuttered trails, minus the rainbow color break up.

Not a single guest I have had over has ever mentioned "rainbows". Then again, almost all of them know nothing about AVS. AVS members I have had over must simply be too polite. I keep looking and dang if I see them.
BTW, my neighbor has an RS1 for comparison.

I can see the very brief and rare RBE on my dVision 1080p, but on this machine at least, it is a non issue, and no one else has ever mentioned it. As for eye strain, we're watching a 118" x 50.25" screen from about 11 feet away. Just the size of the screen could cause eye strain, but it isn't a problem unless we watch 2 movies in a row after a big dinnner party with 12 bottles of wine.............:eek:

The screen size shouldn't cause eye strain, at least it doesn't at my ratio, 159" diag at 15 feet on an LCOS. I wasn't really sure what eye strain felt like till I tried to watch a couple movies at the office on a business type DLP projector. Without this forum I never would have connected it to that pj type. I would have blamed it on needing to get some sleep or some other environmental factor.

I kind of figured coldmachine would show up to claim 3 chippers are infallable and have no faults but I'd be interested to hear from someone other than cm or Bob sorel about the issue. I don't intend to buy a new pj anytime soon but I am trying to push my father to set one up and he may need something brighter than a JVC.

The screen size shouldn't cause eye strain, at least it doesn't at my ratio, 159" diag at 15 feet on an LCOS.

Of course it can. It's the " courtside seat at the tennis match " syndrome - with your eyes darting back and forth to follow the action. Unless you always sit in the first row at a commercial movie theater, then it's no problem!

I am posting these pictures again, since some people still believe that 3-chip DLP cannot have rainbows. The projector is an Electrohome VistaPro Plus, which was made in 1997, one of the first 3-chip DLP projectors.

Projecting dark yellow onto a rapidly-waved white pen, which reveals bands of red, green, and yellow:
http://home1.gte.net/res18h39/images/pen3.jpg

Projecting dark yellow or black, alternating at 60Hz, which reveals that the timing of the bands is similar to a 3X color wheel:
http://home1.gte.net/res18h39/images/bands.jpg

Projecting 50% gray, which reveals bands of cyan, magenta, green, and dark red:
http://home1.gte.net/res18h39/images/gray50.jpg

Projecting white, which reveals large bands of white and small bands of magenta and gray:
http://home1.gte.net/res18h39/images/white.jpg

The newer 3-chip DLP projectors have supposedly improved with regard to rainbows, but I have not seen any pictures of them.

"The brightness is modulated but the colors are non sequential. All three are combined via the wedge."

The whole point is that all three colors are not always on at the same time. That is due to the dithering required to modulate brightness. If only one or two of the three colors are on at any instant in time that is color separation.

Shawn

All 3 colors are on at the same time when the video color profile requires it.The mirror switching of an individual DMD determines the brightness of an individual pixels primary color, the mark space ratio will determine the brightness of that primary. The color profile of an image is provided by the recombination of these individual primary profiles simultaneously.The switching frequencies are vastly beyond any possible separation threshold. RBE is not possible.

The switching frequencies may be fast but we can still sometimes see the consequences in the form of dithering, so it's feasible that 3 DLP can still allow those that are susceptible to see a form of RBE (or whatever you want to call it since RBE is just a name given to the color separation artefact regardless of how it's generated). It's obviously happening because people have seen it and it's even been photographed so it seems pointless to say it's not happening.

I wonder if the same test Eric showed us has been done with 3 LCD/LCoS technologies and what the results were, in which case it's not just a DLP phenomenon.

Gary

I'd rather not address the images on an open forum, so I'll no longer debate the issue.

I could post a photo of the Loch Ness Monster.

I guess you could, but would it stop people seeing RBE (or whatever you want to call it) on 3 chip DLP?

Gary

I guess you could, but would it stop people seeing RBE (or whatever you want to call it) on 3 chip DLP?

Gary

Sorry, i deleted that Garry. I felt you deserved less flippancy.

Those images do not derived from a non sequential technology supplied with non varying video signal.


Maybe not, but if the 3 chip RBE isn't RBE (or whatever you want to call it) we may see it on other technologies. If we don't it's purely a DLP artifact.

Gary

I'd rather not address the images on an open forum, so I'll no longer debate the issue.

I'll take that as a 3 chips still get something like RBE and eye strain but we're just not calling it RBE.

Is there a bright LCOS or non-DLP out there? Is the VW200 on par with the DLP's?

I'll take that as a 3 chips still get something like RBE and eye strain but we're just not calling it RBE.


No, that was not what i meant at all.

It doesnt exist.

So what is it and what are people seeing?

I remember it being discussed quite some years ago on newsgroups as well as here, but I guess we could have some people making up stories and inventing pictures just for the fun of it (just like the Loch Ness Monster), but considering that the vast majority of people don't see them on 3 chip it seems pointless getting upset about it. If it is genuinly visible for those few people what's the problem (and what is really happening?).

Is it really just some urban myth after all?

Gary

Maybe not, but if the 3 chip RBE isn't RBE (or whatever you want to call it) we may see it on other technologies. If we do it's purely a DLP artifact.

Gary

I think you misunderstood what i wrote there.

Sorry, i deleted that Garry. I felt you deserved less flippancy.

Thanks.

I'm just curious as to what it is people are seeing.

So what is it and what are people seeing?

I remember it being discussed quite some years ago on newsgroups as well as here, but I guess we could have some people making up stories and inventing pictures just for the fun of it (just like the Loch Ness Monster), but considering that the vast majority of people don't see them on 3 chip it seems pointless getting upset about it. If it is genuinly visible for those few people what's the problem (and what is really happening?).

Is it really just some urban myth after all?

Gary

Isn't the percentage that see it on 1 chip DLP small to begin with? So if there was some improvement on it maybe even a smaller percentage see it now. It would interesting to get feedback from people that, without question, see RBE's on single chip DLP and see what percentage of them see it on 3 chip if any.

Out of the whole RBE issue at this point I'm most interested to know how many people get eye strain but don't see the RBE's and don't realize it might be the PJ causing it. On DILA I can go an entire day, 10 hours watching football without any eye strain, my butt gets sore but I digress.

I think you misunderstood what i wrote there.

I assumed you meant the other technologies are more analogue whereas DLP is purely digital - the mirrors being either on or off.

Gary

I'd be interested in someone that sees artifacts on 3 chippers to do a "persistance of vision" test. IOW stare at a bright, high-contrast object for a few seconds and then close your eyes and tell us how long the image remains visible.

I think the people still seeing "RBE" on 3 chippers must have extremely low persistance.

I can see RBE on even the fastest 1 chippers, but I can't see it on the 3 chippers.

I cant really go any further here.

Im happy to continue with you via pm

I cant really go any further here.

Im happy to continue with you via pm

I would like it if this continued openly so that we can all learn here. Is there some reason that this should be kept private? Is there industry knowledge that can't be shared openly but can be shared via pm?

So far, I've seen good arguments for the existence of rainbow-like-effects on 3 chippers as well as some photo evidence. I haven't seen anything to refute that and if there is something, I would like to see it. Even Tom Stites from JVC has said you can get these separation artifacts on 3 chippers, although he does represent a competing technology.

I would like it if this continued openly so that we can all learn here. Is there some reason that this should be kept private? Is there industry knowledge that can't be shared openly but can be shared via pm?

So far, I've seen good arguments for the existence of rainbow-like-effects on 3 chippers as well as some photo evidence. I haven't seen anything to refute that and if there is something, I would like to see it. Even Tom Stites from JVC has said you can get these separation artifacts on 3 chippers, although he does represent a competing technology.

I have nothing further to contribute publicly, thats all.

There is no industry knowledge being discussed.

I don't fully understand, nor do I have to time to even attempt to fully understand how DLP functions but am I correct in surmising from what I've read here that DLP, in part, takes advantage of perception anomalies with the brain instead of producing an actual full image? Obviously even CRT's do with scan lines being too fast to perceive but does DLP push further into exploiting perception deficiency? The reason I'm curious is the brain has a tendency to "fill in the blanks" to complete an image that should be there and my guess is if you push that requirement for the brain to complete the image it might be what's leading to eye strain / headaches. I wonder what the difference is in those that do and do not see RBE's. Am I simply a dullard and can't think fast enough to perceive the DLP image illusion?

Isn't the percentage that see it on 1 chip DLP small to begin with? So if there was some improvement on it maybe even a smaller percentage see it now. It would interesting to get feedback from people that, without question, see RBE's on single chip DLP and see what percentage of them see it on 3 chip if any.

There have been many polls carried out both here and on other forums and the results have varied from as little as 2% to much higher, so it depends on those who take part in the poll. Having said that I do feel it's a very small percentage. As for the 3 chip effect being discussed here I would say that was pretty much confined to a very small group of people.

Like I said earlier, some people have said they have seen a similar RBE effect from CRTs as well as plasmas but it doesn't make the technology inferior, it just means that those people are seeing something that pretty much nobody else is.


Out of the whole RBE issue at this point I'm most interested to know how many people get eye strain but don't see the RBE's and don't realize it might be the PJ causing it. On DILA I can go an entire day, 10 hours watching football without any eye strain, my butt gets sore but I digress.

I don't see rainbows at all (except once) and I've had DLPs since 2000 when I had a Davis DLS8 and I have watched more than one movie at a go from time to time so for me it's not an issue. My chairs are quite comfy too. :)

Gary

"There have been many polls carried out both here and on other forums and the results have varied from as little as 2% to much higher, so it depends on those who take part in the poll. Having said that I do feel it's a very small percentage."

RBE polling is one of those funny things.

I had a Sharp 12k before the RS-1 and RBE drove me batty on it. Asking family/friends if they saw 'rainbows' on it and they looked at me funny. I think they were expecting to see literal rainbows when asked that question.

While on the flip side if I took the same people played a 2.35 movie trailer on the PS3 and paused the image and asked them to look at the top right of the screen and then down to the white pause icon (lower left) and back everyone but one saw color seperation at the pause icon. When I mentioned that was seeing rainbows to my wife she correctly pointed out it isn't a rainbow, rainbows aren't just red green and blue. ;) After that they were seeing RBE during moves too, though they didn't find it nearly as distracting as I did.

As far as polling goes I think the bottom line is those that don't see rainbows think the number is very low. Those that do see them think it is higher.

Shawn

Rich,

"Now the same thing is happening with the new LED vehicle tail/brake lights. "

Yes, those are really annoying. Cadillacs have them and when the parking lights are on they drive me nuts. Same thing with them as DLPs... they dither the LEDs to control brightness. On the Cadillacs they are annoying when it is parking lights (dimmer) but I don't notice it when the brake lights (brighter) are on. They may not be dithering with the brake lights on.

I would assume the same people that see rainbows on DLP are the same that would notice this on cars too.

Shawn

My experience today was in line with the explanation above. I was driving at
night and suddenly noticed red trails in my vision. I realised the car in front of me had the LED brake lights. Just as you said Shawn, the lights only broke up into trails when they were dimmer (brakes not engaged). Every time the guy put on the brakes and the lights went brighter I could no longer see trails from the lights. I even tried to sweep my eyes around the bright brake lights and couldn't get any trails. But as soon as the lights went dim the trails occurred again (no effort needed to see them - they were unavoidable, unfortunately).

I suppose this might go some way to explaining why I've seen some image break-up on the 3 chip DLPs, if it is a similar dithering issue.

I wonder if the same test Eric showed us has been done with 3 LCD/LCoS technologies and what the results were, in which case it's not just a DLP phenomenon.
I did the waving pen test on an LCoS projector (Sony VPL-VW50), and it did not show any color separation.

Whenever I drop acid, I see rainbows. Does this make it more likely I will see RBE (flashback)?

Hey Rob, how is the new projector?

Except for seeing RBE when I drop acid and watch it, like everyone else who has one, great. Most importantly, it is still working.

I did the waving pen test on an LCoS projector (Sony VPL-VW50), and it did not show any color separation.

Thanks Eric.

Gary

I did the waving pen test on an LCoS projector (Sony VPL-VW50), and it did not show any color separation.

Of course it didn't, because LCoS works differently. Now if you call this RBE on a 3-chip or not, is up to your definition. When most people speak about RBE, they mean color separation artifacts caused by the color wheel of 1-chip DLPs. A 3-chip does not have a color wheel, hence no color seperation artifacts caused by a CW... so for most this means no RBE.

Of course 3-chip DLP has color seperation issues as well, after all it's a reflective, pulsed technology. I'm very sensitive to RBE (1-chip), but have no issue with 3-chips. Of course color seperation artifacts can be made visible with some tricks, such as the pen. Now, if you call these 3-chip color separation artifacts RBE or something else, is entirely a question of the definition of RBE. I don't think anyone claimes there are no artifacts on 3-chippers.

The artifacts you can see on 3-chippers are also visible on 1-chips, but they're by far outweighted by the CW-RBE problems.

...I don't think anyone claimes there are no artifacts on 3-chippers...

Uhh actually there are a few of them and at least one that's pretty certain they're perfection

For comparison, here is a picture of a 1-chip DLP (Compaq MP1600), projecting white onto the pen, which reveals bands of red, green, blue, and a thin band of white:

http://home1.gte.net/res18h39/images/pen1.jpg

Except for seeing RBE when I drop acid and watch it, like everyone else who has one, great. Most importantly, it is still working.

Jason mentioned that the last time you called you had to hang up because your chairs had come alive and were eating your carpet. Now it all makes sense... :D

Uhh actually there are a few of them and at least one that's pretty certain they're perfection

Anyone who claims there are no problems, hasn't done the homework. :p
There is no such thing as a perfect projector, period. They all have issues and drawbacks here or there.

Personally, I think it comes down to two types of projectors to choose from. 3-chip DLP and LCoS/SXRD. Each have pros and cons.

I like the natural, film-like look of LCoS/SXRD, better fill factor and CR.
What I miss are the pros of 3-chip DLP. Brightness (for bigger screens, say 20' or so) and higher Ansi CR.

I don't think one is really better than the other.

Anyone who claims there are no problems, hasn't done the homework. :p
There is no such thing as a perfect projector, period. They all have issues and drawbacks here or there.

Personally, I think it comes down to two types of projectors to choose from. 3-chip DLP and LCoS/SXRD. Each have pros and cons.

I like the natural, film-like look of LCoS/SXRD, better fill factor and CR.
What I miss are the pros of 3-chip DLP. Brightness (for bigger screens, say 20' or so) and higher Ansi CR.

I don't think one is really better than the other.

Exactly. No-one said they were perfection. I ve been more than critical of certain recent 3 chip DLP implementations

I have 20/15 vision and can't see it at all on my samsung 61" led dlp, or my infocus 7210 projector. I guess all the mushrooms during high school may have disabled my rainbow receptors.......just joking. But really, no rainbow apparent here and I've tried to find them several times.

rainbows are almost gone now
dithering is still here though, even with 3DLP (mirrors dithering)
but dithering has been improved a lot also
the BENQ W5000 is amazing in its absence of rainbows almost

rainbows are almost gone now
dithering is still here though, even with 3DLP (mirrors dithering)
but dithering has been improved a lot also
the BENQ W5000 is amazing in its absence of rainbows almost

What's the speed of the color wheel???

Isn't the percentage that see it on 1 chip DLP small to begin with? So if there was some improvement on it maybe even a smaller percentage see it now. It would interesting to get feedback from people that, without question, see RBE's on single chip DLP and see what percentage of them see it on 3 chip if any.

Out of the whole RBE issue at this point I'm most interested to know how many people get eye strain but don't see the RBE's and don't realize it might be the PJ causing it. On DILA I can go an entire day, 10 hours watching football without any eye strain, my butt gets sore but I digress.

I get eyestrain from single chip dlps but I do not see rainbows. I finally demoed a sim2 ht380 and really like it but with a brighter image I feel the eyestrain even more than I did with the SIM2 ht300e which I have owned for 3 years.

I like the look of the DLPs. I do get the eyestrain from the single chippers. I see dead people erm I mean rainbows, but that bothers me less than the eyestrain. I haven't watched the 3 chip DLPs enough to get eyestrain yet, but they look nice.

I am glad that the DLPs are much better with the dithering. That really bothered me on older units.

After hearing "rainbows are pretty much a non-issue these days" many times over, only to find that the newer models still drive me mad with rainbows, I take any such claims with a very large grain of salt. I'm definitely in "I'll believe it when I (don't) see it" mode.

I like Dlp pictures but due to RBE kill all of my joy, I think RBE is still present in today PJ
for sure, I saw rainbows on HD65 and VP12s4. I used an lcd technology since a month
and what I see is I could watch the PJ without eyestrain... But still like DLP POP.
It is a kind of love/hate with DLP

I see non-moving rainbows on my neighbors LCD projector....oh wait, thats just the polarizers degrading :)

I see non-moving rainbows on my neighbors SXRD projector....oh wait, thats just the polarizers degrading :)

Please expound more. I haven't seen any of this polarizer degrading even on old LCOS projectors and RPTVs. What are you talking about?

Oops, meant LCD, now fixed.

I personally don't see rainbows but if I did I would just turn them off.

Rainbows are a flicker fusion threshold problem (read some Wikipedia articles). Flicker fusion thresholds are lowered by alcohol (in you, not the projector).

If you see rainbows - have a Martini. Still see them? Have another.