AVS Forum banner
Status
Not open for further replies.
1 - 6 of 6 Posts

·
Registered
Joined
·
275 Posts
Discussion Starter · #1 ·
Just got back aching and wining from my first DLP experience with a Sharp MX20.

What a big dissapointment :( After several minutes I started to feel eyestrain and eventually got a serious headache. Interestingly enough, did not notice any RBEs.

The setup included an acceptably clean composite cable signal (PAL), plenty of ambient light and a matte "Draper" screen. Switched to a SVID signal from a DVD with the same consequences.

I have had set my mind on Optoma H56 (prices in Tokyo have reached 3K$) which is a 4x color wheel DLP, only slightly faster than the MX20 3x.

I wonder if I am totally doomed to LCD (LPX500 is the next runner up), or there are other parameters that may have impact on eye strain.

I have heard the (2x color wheel) NEC LT240/260 have some dedicated circuitry that reduces the RBE.

Maybe H56 or the new Mits PLV-D1208 do a better job. Unfortunately there is no way I can see these PJs for myself.

The god damned DLP looked so good though...
 

·
Premium Member
Joined
·
24,995 Posts
Others have reported that if you stick with it for a week or so, the headaches stop.


I wonder if the people who get headaches are just on the edge of being able to see rainbows, so that their nervous systems strain to tell what's going on, but then after awhile it gives up and relaxes.


A 4X wheel might do it for you.


"I have heard the (2x color wheel) NEC LT240/260 have some dedicated circuitry that reduces the RBE."


I haven't seen mention of that, but I have heard lots of complaints about RBE from them.
 

·
Registered
Joined
·
165 Posts
The eye strain could be caused by the brightness as well. I use a Hoya FLD filter on my LT-150 (w/white segment disabled) and its very pleasent to watch. If I remove the filter I feel like someone is pushing on my eye balls during viewing.


"I have heard the (2x color wheel) NEC LT240/260 have some dedicated circuitry that reduces the RBE."


NEC's claims are baseless. I don't see how you can reduce rainbows without changing the color wheel.
 

·
Registered
Joined
·
275 Posts
Discussion Starter · #4 ·
"I have heard the (2x color wheel) NEC LT240/260 have some dedicated circuitry that reduces the RBE."


An excerpt from PJC review of the LT240/260:


"The light engine features a four-segment RGBW color wheel that rotates at 120 Hz, commonly referred to as a "2x" rotation speed. All of NEC's LT series projectors use this wheel. (Note: readers who automatically correlate color wheel rotation speed to rainbow artifacts should note that NEC engineers have incorporated new electronics to reduce rainbow artifacts as compared to earlier products like the LT150z.)"


I do not know how NEC may reduce RBE by video processing alone. A PJ screen has no memory and the scene will always flicker as fast and often as the color wheel turns. There may be some degrees of freedom controlling the integration time of each color segment versus the individual pixel brightness. I can also think about the actual position of the color frame within its time slot to be an alterable parameter as well. A slight modulation in time of the color position within its frame may do the trick. Maybe I'll test this someday ;-)


I speculate that a 4 segment RGBW may cause strain / RBE more than a 3 segment RGB spinning at the same speed, simply because of the higher time division factor - shorter integration time for the eye for each color.
 

·
Premium Member
Joined
·
24,995 Posts
"I speculate that a 4 segment RGBW may cause strain / RBE more than a 3 segment RGB spinning at the same speed, simply because of the higher time division factor - shorter integration time for the eye for each color."


Plus that fourth segment is bright white.
 

·
Registered
Joined
·
275 Posts
Discussion Starter · #6 ·
Noah,

I guess the (flickering) bright white enhances even further the strain sensation. I recall that during my disco dancing days I always got headaches after stroboscopic flash lights were applied.

I think the answer lies with the correct modelling of the impulse response of one's optical receptors and nerves. I wonder if TI has really put an effort into the definition of the human vision transfer function. The single chip DLP system relies heavily on the low pass (integration) filtering of human vision. The fact that TI does not supply it own set of companion chipsets and color wheels (like Intel or AMD chipsets) with the DLP chips does raise some questions.

I am aware that TI is lately pursuing an alliance with a taiwanese ODM in order to put up a reference design for low cost DLP PJs.
 
1 - 6 of 6 Posts
Status
Not open for further replies.
Top