The Ultimate 3D projection system: A Practical Discussion Thread - Page 52

Any idea what the difference is between 8 and 10 bit processing?

Doesn't sound like it has 1080p120 input support. So my thought is this. I want both 3D BR support and PC 3D gaming support. The advantage of dual 3DXL's is that the BR3D can go to input one of each box and they will split that into passive left and right, and the PC can go to the input two of each box. nVidia 3D Vision would't work, but using Tridef drivers I could output 1080p60 to each eye and the 3DXL boxes will pass that straight through to each eye. Or did I miss something?

No unfortunately the 3d xl take 1.4a hdmi spec as input so 1080p 24hz 3d at max splitting that to L and R channel. No higher hz than that at 1080p

I believe the G-501 was developed for professional applications including cinema use. 10 bit processing allows expanded color space and reduces artifacts such as banding and aliasing, which is the most likely reason they went to the trouble of using 10 bits.

8 bits is usually adequate for most video, as long as that video doesn't undergo further processing downstream. You might see differences under critical viewing situations, but normally, you wouldn't notice the differences between these devices (due to 8 versus 10 bits). I'm sure when these units become more available, we'll see some objective comparisons in print.

On another note, these devices are more suited to BD use (rather than HTPC or gaming use) as wnielsenbb noted, but for 3D BD playback, they may be the demultiplexer of choice when prices and features are compared to currently available options.

I think what he meant was he could either use hdmi 1.4a to play blu-ray using a splitter and the first input of the optoma box
or
dual-outs for games using the second input of the optoma box

Allowing to switch between PC gaming and PC BluRay without having to switch between cables.

exactly. The 3DXL would be in passthrough mode for pc gaming.

not to change subject as im very interested in all the knowledge about these demux units. and better info here about the 300 or 301 box from vns than ive found.
but the issue with color on the omega filters is much more clear. we are making a corrected set for types of LCD units that differ from what we made for DLP.
here is a scan i took today of the spectral output from the infocus projectors from my demo on youtube. these are business models but show why i am not getting the same color issues on these ones at least.
they still lack red performance but are not effected like LCoS units the same. i can fix the issues.
i will make all the data more clear later but you can see the native white and the effect of the pro filters dont move white much as a pair. these are only the projector filters and not the combo with the eye filters. the combo blances the colors better.

Will there be an option for those of us using the original kit with LCD projectors to swap out th old kit for the new kit?

yes i will make it right by anyone who got a kit. im working now on the right alteration to the projection filter and collecting spectral performance for the variations out there. it will be focused on the higher end units that most use for home theater. i apologize for the error.
I have run into this before working on the only large cinema LCD but did not anticipate that the same variation would be in the home theater units.

Sounds good, appreciate the support. Will the glasses change as well? I'm using the the Panasonic AR100U units, which I believe uses Epson LCD panels.

Motorman, Wolfgang posted his Lux readings comparing the new Infitec filters to your Omega filters. Anyone reading this that doesn't already know, the Infitec does not have a consumer product and their filters cost around 4K. He is using two 4K Barco projectors with xenon lamps. He reported 970 lux with the Omega filters and 1170 with the newest Infitec (Dolby was somewhere between those). Could the difference be the lamp or do you think there would be a % increase in lux no matter what projector is used?

Hello, I guess I should have waited but I just ordered your kit off of Ebay yesterday and I am using two Panasonic PT-AE4000U LCD projectors in my system. Do you think I am going to need theses new reworked lenses?

My projectors specs. are, 170W UHM Red-Rich Lamp, 0.7" 3 LCD

Would both glasses and projector lenses need to be changed?

Thanks,

Great customer service!

Ok that explains it. I was looking at 16x9 screen previously and put off By the relatively small size limit on a seamless screen.

Motorman if you get the chance to test your filters on a BD screen (or a non white screen). I would be interested to hear feedback as I'm keen to hear if your filters will be affected.

Thanks I'll need a change when ready too as I'll be primarily using mine for LCD. Will these "consumer" versions of your filters also be suitable for non xenon dlp?

the glasses will not need to change. the AR100U will work wll with the new LCD filter.

the regualr filter will work the best with DLP . especially the home theater type of DLP. the business models trade brightness for lumens and the home theater models having better color and contrast will perform best with the regular omega filter. the LCD corrected one will also work well but it has to be different becuase of the ouput spectrum is altered by all the internal dichroics that split and recombine RGB. DLP wheels have less loss and color separation.

you would just need the newer filter for LCD. try the set out and see. with some adjustments you may be satified with the set. if not i can exchange the projection filters. the glasses are the same so no need to change that.

I don't understand.
What do the corrected filters do ? Since the problem is a wrong spectrum, I thought the new filters would work with slightly different spectrums than the original filters, and thus needing different glasses.

How will the new filters affect the output compared to the original filters ?


I just had an idea for an other possible spectrum problem. There's the 3-DLP projector type, it's rare because it's much more expensive than single-chip DLP with colour wheel but you might find customers using these types of projectors. I hear the dichroic colour separators are different from the ones found in 3-LCD projectors. Do you have spectum measurements for those too ?

I do have spectral data from a lot of these type as that is the most common cinema system. the filter system works best with 3 chip DLP and there are even internal wheels we make for these higher end systems.
that is what was refined as the panavision 3D system. there is little differance between barco, christie, nec and runco in terms of how these function with the Omega 3D system.

I did see his post and i dont doubt his measurements. but he also states he did not feel any needed color correction. all the testing we ran with xenon cinema systems had some level of correction. the dolby especially and they rent a color process server to do this. if color correction was done to fix the color issues with infitec and dolby the %efficiency would drop of course as you attenuate bands to balance things. ours needed far less correction to meet DCI spec leading to ours being higher % after the calibration.
we also have higher extiction than the other 2 but that is a matter of choice. i could create more light on the screen at the cost od extiction but we opted not to.
after all home cinema systems trade lumens for color precision. I took that same thought in creating the system but also in extiction as well.

Motorman,
I don't know if you would be allowed to talk about it BUT have any of the major manufactures of home theater projectors contacted you about incorporating your system into one of their projectors?

yes they would be in the same category. they have similar spectral performance that is the issue with the method of spectral filtering with ten bands

you know they have not but i have a possible connection with one and i am looking forward to developing something with one or any projector maker. i know form all ive worked on with 3D that given a fixed system i can optomize near perfect performance. idealy it would be a dual light engine system like the LG 3D system in one package.

I finally got my setup going. I believe I got the second last kit that was offered on ebay, but it's taken me a while to get everything in place. At first, I got 2 Panasonic PT-AR100Us, but when I tried the first one, I thought the fan was way too noisy, so I sent them back.

Then I got 2 refurbished Epson 8350s, but one of those had a weird convergence problem where there was a double image in magenta about a quarter of an inch from the real image. So I returned that one, and this time Epson sent me one that works.

So I have 2 Epson 8350s and 2 3D-XLs, I'm shooting to a 125" screen from 12.5 ft. Got everything hooked up and watched Avatar on Sat. night and some various 3D programs on DirecTV, including the last couple hours of the Masters yesterday.

Avatar was excellent, MUCH better than my Epson 3010, which has a lot of ghosting. I noticed very little ghosting with the Omega kit, and overall the image just looked very sharp and clear. The Masters was not quite as clear, but I attribute that to the half resolution side by side and whatever else DirecTV does to the signal.

There are just 2 small issues - the reds are very bright and overpowering - they practically shimmer at times. Looking back in this thread at a post with the spectral graphs, I can see where the LCD red spike is pretty tall compared to the DLP one, so maybe that's the cause. I haven't tried adjusting any settings yet, but I will. But it also sounds like I may be a good candidate for the corrected filters.

The other thing I noticed is that sometimes when there is some slight ghosting like with words on the screen, if I turn my head just a little bit away from where the words are, the ghosting goes away (the letters seem to re-converge). I'm quite sure my projectors are square with the screen, but maybe this indicates otherwise? Or is it something to do with the filters? Do they have to be exactly parallel? Has anyone else noticed this?

Overall, though I'm pretty pleased with how this kit performs, I'm actually looking forward to watching more 3D again.

Since my projectors are close to yours (Epson EH TW3500 = Epson 8100), we should have very similar results.
My reds are not that strong but they are very different between the eyes, the left red eye is very bright but looks washed out while the right red eye is extremely saturated but very dark : producing eye rivalry, which I believe is what you seem to be describing as this shimmering effect.
I also have a similar issue issue with cyan : I get lots of rivalry in shots where you see the sky, my green depends on the colour mode, in dynamic and living room modes the right green eye is almost yellow, in the other modes the colour is better but the picture is way too dark.

Ghosting : I do get ghosting if I turn my head away from the screen : ghosting appears first in the corner of the glasses.

The ideal position of filters is perpendicular to the light path, that's where you get the best filtering and max brightness (regardless of the screen position)
If you use lens shift, then the filter should be slightly tilted according to how much shift you are using. If they are not perpendicular to the flow of light, you'll get crosstalk and serious colour problems.
On my projector which is on top of the stack, I am at max shift downwards and to the left (I hit the diagonal shift limit), but my filters are simply suspended and remain vertical (parallel to the screen). I don't see any ghosting on most of the picture but the bottom left corner of my screen (the size of my windows start menu) does leak a little (and is slightly too blue).

The problem with the optimal position (perpendicular to the flow of light) is that there is a significant amount of light reflected back into the projector (increased heat, possible reduced lifespan of the LCD panels and/or the fan must turn faster to keep them cool).
As a compromise, the filter should be very slightly off angle but not too much, just enough to make sure the reflected light doesn't hit back on the LCD panels.

Sorry i didnt answer one of your questions before. the correctors alter the peaks of certain bands some to move the white point of each eye to balance the overall white for a given projector spectrum. and yes this will impact throughput some but in testing on this before for the Sony sxrd it was less loss than the loss you get from adjusting the projectors color balance. so the bands do not move or change in widths just peaks. the glasses remain the same.
with the Epson spectrums ive been able to find as well as good data for JVC and a couple of other LCD's ive been able to test the color gamut and white points for each to see how the new LCD filters will function. they all are similar enough that the correction should be a real improvement for those.
its not going to be perfect for all as some viewers have a different take on things and its hard to make perfection for a wide range of projectors. from what i see so far a majority will have good left and right white points while maintaining the extiction ratios and decent througput.

I know you guys have been doing a lot with the motorman's new lens, but I was wondering if I could get some advice for a polarizing system. I just got two Barco Sim6 projectors which were built for stereoscopy and have decided to go with a realD system because it will be outside (and the high gain silver screen should be beneficial in such an environment), and a lot of kids will be viewing it so I wanted super cheap glasses since they will likely get broken and stolen etc... What would you guys recommend for polarizers on the projectors? They are linearly polarized (the same way on every panel) so I thought I could get a quarter-wave plate (I was thinking of the high temp ones for sale at www.polarization.com). Thanks for any advice you can provide.