Können Sie mir bitte einen 1080p oder höher 3 Chip DLP mit ordentlichem Kontrast empfehlen??

  • Hello, I'm new to the forum and google helps me to speak German. : D Can you please recommend a 1080p or higher 3-chip DLP projector, either a home cinema projector with normal noise and heat levels or less, or a cinema projector that is very quiet and cool, that meets these criteria? Most likely it needs to be "used." It can be released yesterday, or 10 years ago, so long as it is excellent. The criteria are:


    - 3-chip DLP

    - 1080p or higher resolution

    - As much color space as possible

    - At least 1,750 lumens after calibration

    - 5,000: 1 or higher native contrast plus dynamic contrast 20,000: 1 or higher

    - Frame interpolation

    - Cool and quiet enough to work without a hushbox or dedicated projector cabinet

    - Maximum weight of 80 pounds.


    Thanks very much. I already know the Sim2 Lumis, so I'm looking for recommendations next to this model. If nothing exists that meets these criteria, then take away dynamic contrast and frame interpolation. But we prefer to find something in them too ... Don't drag them off the list just because it's easier, only take them off if there is really nothing apart from a few Lumis models that meet these and the other criteria: D Thank you very much!!

  • macelman

    Hat den Titel des Themas von „CKönnen Sie mir bitte einen 1080p oder höher 3 Chip DLP mit ordentlichem Kontrast empfehlen??“ zu „Können Sie mir bitte einen 1080p oder höher 3 Chip DLP mit ordentlichem Kontrast empfehlen??“ geändert.
  • Hi there and welcome!


    The only projector that checks all your boxes is the one you already found by yourself, the Lumis. Your point "cool and quiet enough to work without a hushbox..." renders pretty much every professional machine inappropriate. Secondly, your contrast requirements are also too high for professional projectors - at least in stock form. If you fancy some diy iris tuning and the like, 5000:1 native are achievable with the Christie HD6K-M or HD10K-M - but NOT stock! Good thing about the Christies, they bring the dynamic iris function which is very efficient and widely customizable.

    If you really need the frame interpolation, it all comes down to a few Panasonic units (in the pro sector). Again, both Christie and Panasonic are too loud to use within the seating area without a hushbox!

    Regarding frame interpolation, I also recommend you to read the full "Profi-Beamer-Thread" which Mankra has linked. I gave some advice there about a PC-based frame interpolation which works very good, even together with MadVR. So you can at least remove this requirement from your list.


    May I ask what our plans are? How come you have such special requirements (3-Chip DLPs are (unfortunately) not exactly very common in the home theater scene)? Have you already owned one? You may want to introduce yourself and your project here, then you will receive more feedback and support from the community.


    Best regards, Martin

  • One more recommendation would be the Barco RLM W Series. W8 for example, or even W12. These are not too loud (in professional scales) and rather cost-efficient. The Gamut is not the largest, but can be increased decently beyond Rec.709 by means of a didymium filter. Contrast would also have to be improved by adding one or two irises. Dynamic contrast and frame interpolation is not available.

    But the basis is good to improve upon - in the end it's all about the question "which model can you find second hand and how much are you willing to spend on it". This is usually a tad more difficult than walking into a pro shop with 100 grand and picking the rig of your choice - believe me, I know what I'm talking about :sbier:



  • One more recommendation would be the Barco RLM W Series. W8 for example, or even W12. These are not too loud (in professional scales) and rather cost-efficient. The Gamut is not the largest, but can be increased decently beyond Rec.709 by means of a didymium filter. Contrast would also have to be improved by adding one or two irises. Dynamic contrast and frame interpolation is not available.

    But the basis is good to improve upon - in the end it's all about the question "which model can you find second hand and how much are you willing to spend on it". This is usually a tad more difficult than walking into a pro shop with 100 grand and picking the rig of your choice - believe me, I know what I'm talking about :sbier:


    Thanks very much. Good idea to introduce myself. You can call me Art. My project is a bit complicated as far as projection... just a little bit... I actually need to make post about some of the most complicated aspects having to do with 3D, which it might be a big help if you guys took a lot at it since you seem to know what you are talking about. Actually I am impressed coming from a very popular english HT forum to see great answers on the best models that come closest to meeting my criteria right away, instead of going in circles for dozens of pages of arguments and questions "that's stupid why do you want that" while no one knows the answers. :D So I really appreciate the knowledge here, which is why I joined after reading the threads and seeing it, but let me get to describing my project.


    I will be sitting 15 feet from a 130 inch 2.35:1 screen for 2D. I want one bright, high quality 3 chip DLP projector for 2D HDR. I would prefer 4K of course but I dont have $100,000 to spend on a projector so it will probably be 1080p combining with madvr, although I do have a curiosity what more informed people think pf projectors like, for instance, the BenQ HT9060/x12000H, and if, despite very poor contrast, it can still perform as good or better than, for instance, a Lumis overall, simply because it has DLP sharpness combined with four times as many pixels on screen. Maybe clarity can outweigh everything else? Or is it wrong to assume it would have more clarity than a Lumis even despite 4x as many pixels?


    It sounds like there are at least some things to look into. But why 3 chip DLP? It hasnt really been a problem, but I have seen rare rainbows on 1 DLP projectors with RGB LED lightsource, specifically the Sim2 M-150. So for peace of mind, 3 chip DLP would be better for that reason. I am also very sensitive to motion blur. I compared the M-150 to a Lumis and noticed zero difference in the motion, but for peace of mind, I know 3 chip DLP cant mess it up, whereas sequential color, maybe it could even if I am not noticing? Or maybe it will for content I havent tested yet? Probably not but 3 chip DLP is to me a "sure thing."


    And except for color wheel models, which I want to avoid, and the BenQ HT9060 (which has low contrast and I think cannot output 24fps sources at 24fps), 3 chip DLP are the only bright DLP projectors I can find.


    Then for 3D, I want to double-stack projectors for passive 3D. Why? I have sensitive eyes and 3D is sort of all or nothing. Either it works for you and doesnt bother you, and you can enjoy it immensely, or it doesn't and you can't enjoy it at all or even watch it. So I want to build the best 3D setup I can afford and give myself the best chance to be able to enjoy it.


    3D stacking is the big reason why I care about frame interpolation. I have no way to know if it's true or not until I do it, but I have seen people say that the artifacts from frame interpolation are less visible in 3D, but meanwhile the benefits are a bigger deal, and that frame interpolation really enhances 3D much more than 2D. I already have two Sim2 M-150's for stacking, which should be bright enough on close to a 3 gain screen (at least, with linear polarization filters and glasses), however I am worried about hotspotting. I've seen people say it is less visible in 3D, but I dont want to take that chance given the cost of the screens I am looking at.


    Can it work? Yes. With a 2 gain screen, with polarization, I can get 15 foot lamberts after 3D filters and glasses. However, I dont know if that is actually the ideal brightness for 3D, or just the standard at a time when most theaters were not capable of achieving higher brightness than that. Additionally, I am sensitive to eye strain, and linear polarization has more crosstalk than color bandpass filters. I have the equipment necessary to do 3D luts for both projectors in a stack, so I would like to use color bandpass filters. A privilege of being in the U.S., I was able to find the M-150's at good prices, and believe I can unload them without much hassle or losing any money. If I could find 3 chip DLP projectors with frame interpolation, or at least brighter 1DLP projectors with fast sequential color like the M-150's, it should work a lot better and allow me to use color bandpass filters with less crosstalk, and a lower gain screen with less hotspotting and sheen. And if I already see rainbows sometimes, albeit very rarely, in 2D, I wonder if it could be a problem, or cause extra eye strain, to have stereoscopic sequential color. With a double stack of 3 chip projectors, there is peace of mind on that issue also.


    I was also told DLP projectors would work the best with color bandpass compared to lcos and LCD at the time I bought the M-150's, but did not know that did not apply to DLP projectors with LED lightsource. It might still work but it will require an even higher gain screen.


    Maybe the sheen wont be noticeable from 15 feet away and through 3D glasses, so if I could find a 5 to 10 gain curved screen that will not cause homogeneity issues with a projector stack, maybe I can go forward with the M-150's in a stack.


    There are many elements involved which makes it very complicated on the surface. On the other hand, it is definitely something that can be figured out, but only once you have all the information, like how bright, after the 3D filters and glasses, is the ideal brightness for 3D? What number do I want to hit for the best experience? As well as, from 15 feet away on a 135" (diagonal) 16:9 screen, how much gain before there will be hotspotting? For reference the gain screens I found so far around 3 gain have a 40 degree half gain angle. Throw distance will be 17.5 feet from lens to screen. There is also a possible "prism" solution but I will make a dedicated topic for this.

  • Your point about frame interpolation is something I wanted to ask more about. I am guessing you are talking about SVP frame blending? So the first question would be, is there any way to tell how good quality that is, which converts 23/24 frames per second movies into 60fps and has to use "blending" because 23/24 do not fit into 60 evenly, compared to frame interpolation on a Lumis 3D-S or Sim2 M-150 which evenly doubles the frames from 23/24 to 47/48? Not to mention even if 23/24 could fit evenly into 60, 60 is higher than 48 so maybe more soap opera effect? But also smoother so... it depends how good the "blending" is compared to regular interpolation? In any case, the main reason I want frame interpolation is for double-stack 3D, and unless I am mistaken, I do not think SVP works with 3d blu-rays at all. But if it does then it would be a great option.

  • One more recommendation would be the Barco RLM W Series. W8 for example, or even W12. These are not too loud (in professional scales) and rather cost-efficient. The Gamut is not the largest, but can be increased decently beyond Rec.709 by means of a didymium filter. Contrast would also have to be improved by adding one or two irises. Dynamic contrast and frame interpolation is not available.

    But the basis is good to improve upon - in the end it's all about the question "which model can you find second hand and how much are you willing to spend on it". This is usually a tad more difficult than walking into a pro shop with 100 grand and picking the rig of your choice - believe me, I know what I'm talking about :sbier:

    Do you know what contrast it can get after those modifications? Does it have dynamic contrast feature?


    Also, does anyone know the noise levels on these and what kind of contrast these can get? https://www.projectorcentral.com/Barco-F85_1080p.htm or the F82 version?


    If you take out my "frame interpolation" requirement, are these, the Barco RLM W8 and W12, and the Christie PT-RQ13 and Christie HD10K-M, the best options, or are there any others to add to the list now that we've eliminated frame interpolation?


    I notice some of these appear to have nicer lenses than the Sim2 Lumis line, but the Sim2 Lumis line may have better native contrast, as well as a decent dynamic contrast multiplier. It raises a question, how important is the lens? What effect does it have from seating distance? Up close, sometimes I can see chromatic aberration on these HT projectors, but from seating distance, you can't consciously see it anymore. My question is, even if you can't consciously see it anymore, is it still causing some sort of subconscious blur, or strain on the eyes, where the eyes are still picking up the light particles from the chromatic aberration even if you cannot consciously make them out through all the other "correct" light particles making up the image?


    Or is it the case from seating distance, that if you cannot consciously see the chromatic aberration on a worse lens anymore, then if you view an image with a better lens side by side, the images will look identical, and feel identical, because the benefits of the greater lens are not visible from your viewing distance?


    It reminds me of how, out of the three people I have seen online who have owned both the Sim2 Lumis, and the Runco LS-10, two seemed to prefer the LS-10. This always confused me because both projectors seem very similar other than the Lumis is a better version on paper. The Lumis is 3-chip with 0.95 DMDs, and has higher contrast. The Runco is 3-chip with 0.65 DMDs, and lower contrast. it is just probably just randomness, or small sample size, but the only other explanation I could think of is, what if none of the lenses on models like that are big enough to fully resolve the 0.95 DMDs, and therefore they are creating chromatic aberration? Whereas the 0.65 DMDs, while worse DMDs, are taking a load off the lenses, and able to work with those lenses without as much chromatic aberration. So maybe the worse chip is actually able to create a sharper image than the better chip because the worse chip can interact with these HT-sized lenses with less negative effects?


    I have never had the opportunity to try a Christie or Barco type of lens, so I dont know what is better, 20,000:1 contrast with a Sim2 lens, or 10,000:1 contrast with a Christie lens, and stuff like that. What do you think?

  • Hi Art,

    sorry for the late answer, but your very, very long statements and thoughts spread over different topics, combined with my very, very little amount of free time makes it quite hard to give you an appropriate answer to your questions. I will try and start now :)

    First of all - you are not alone :sbier:

    I went through almost everything that you are thinking about at the moment, and I am pretty pleased with the outcome in my modest "Belle Cinema". Maybe you wanna check out my build thread : Belle Cinema

    In the first post you also find a link to the VISATON forum, where I "originally" come from and where much more details can be found. However, I haven´t updated anything in both threads during the last 2 years due to time constraints...


    The similarities and parallels are quite astounding: I also use a passive 3D stack with color-bandpass-filters from OMEGA OPTICAL, a Geobox for signal splitting and warping and 2 DLP Projectors. I have developed and built my own double curved / torus screen with gain 2.0, and I also own a SIM2 M.150 Projector :respect:. All projectors equipped with anamorphic lens.

    So here´s my first advice: the 3D stack with Omega filters works very good, I am very pleased with it. But as you already assumed, I highly doubt that it will give you good results with the SIM2 M.150. The color coordinates of red and even more green are so far from "normal" projectors (hence the huge gamut), I don´t expect the pass- and stop-bands of the filters to match with them and give you satisfying results in terms of hue shift and crosstalk.

    I would recommend to ask Bob from Omega directly what he thinks about the issue, but not sure if he knows the gigantic gamut of the M.150...

    If you ask me, I would not take the risk of combining the projector with the largest gamut and narrow-bandwidth spectrum with the Omega filters. A standard bulb-lit device is definitively the safer way to go here.

    Staying with the projector choice: I wouldn´t consider laser-phosphor single-chip DLPs, especially with RGBW color wheels as your linked Christie above. Those have very limited gamuts, nothing for home cinema use.


    Now regarding your curved / high gain questions. I answer in this topic, as your questions and considerations about projectors and screens are all related to each other, and I find it rather difficult to have two parallel threads here.

    First of all, as you obviously struggle with envisioning the multiple impacts on geometry and light distribution that come with a curved screen, I also strongly recommend you to get familiar with the hotspot simulator from FoLLgoTT ! It is a great and very helpful tool that shows you very precisely what happens to your light distribution depending on the curve radius, projection distance, viewing distance and material properties. I believe the 40 days trial issue has nothing to do with the simulator itself. It is only the WinRar which wants you to buy it! The hotspot simulator is only an excel, and it is free of charge.

    That said, your preconditions are pretty unproblematic in terms of hotspotting, as your distances are rather long and your screen is very small. Nontheless, there is no gain without hotspotting, and even under your circumstances, a flat screen is not an option once you go higher gain than say 1.5. On the other hand, I am wondering about the gain figures that you undiscerningly talk about. Gain 5, gain 8, gain 10... As if you could pick whichever gain you choose off the shelf somewhere :shock:. Same goes for single or double curved. Other than the aforementioned Couchscreen, which is a very special case, I am not aware of one single manufacturer of double curved screens for home cinema use. So as long as you don´t go the "hard way" and construct one by yourself as I did, single curved is the only option for you. And then again, which company offers curved screens of the desired size with more than gain 2 (or something the like)? Before you dig deeper into pro and cons for higher gain, you should maybe check out if these considerations are only of academic nature, or if they can be put into reality somehow.

    As it is very hard to find suitable screen materials with a gain higher than 2, I on my side opted for the Opera High Gain from Gerriets and built my torus screen with this sheeting. It is a good compromise between higher light output, but still manageable side-effects. Due to the shape of my screen, I can see no hotspot at all, never ever, even though the curve is not narrow enough in theory. But such a high curvature as theoretically needed to fully avoid hotspotting is just not feasible in terms of focus depth and warping. And to obtain the theoretic ideal, one would have to place the projector "in your head", or at least on the same horizontal line of your head, which is nothing but impossible. That said, going "very" high gain, say 4, 5 or more, even if you could find a screen material that does it, brings you problems that you cannot mitigate any more, even with the best possible curvature. In addition, such a high gain will not come without glitter issues.

    Also, very important point, "very high gain" can only work for one master seat, and even there only in theory. You ask about projector placement issues and want to use prisms to combine the two projectors (which is absolutely not feasible in practice by the way). Why? Because your projectors are like 20 cm apart from each other. But in the next chapter you talk about your complete seating row and viewing positions 100-150 cm apart from the master seat?!?! :dry:

    Seating postition and projector position are to be considered exactly the same way! There is no difference technically. So as long as you want a decent image uniformity for a complete row of seats, don´t waste any more thoughts on gains higher than something around 2! And if you stick in the region of around gain 2, you also don´t have to waste a single thought on the slightly differing projector positions, as you will never be able to notice it on your screen!


    Best regards so far, Martin

  • Your point about frame interpolation is something I wanted to ask more about. I am guessing you are talking about SVP frame blending? So the first question would be, is there any way to tell how good quality that is, which converts 23/24 frames per second movies into 60fps and has to use "blending" because 23/24 do not fit into 60 evenly, compared to frame interpolation on a Lumis 3D-S or Sim2 M-150 which evenly doubles the frames from 23/24 to 47/48? Not to mention even if 23/24 could fit evenly into 60, 60 is higher than 48 so maybe more soap opera effect? But also smoother so... it depends how good the "blending" is compared to regular interpolation? In any case, the main reason I want frame interpolation is for double-stack 3D, and unless I am mistaken, I do not think SVP works with 3d blu-rays at all. But if it does then it would be a great option.

    Sure there is a way to tell how good the quality of any FI is - watch it, test it, and then judge for yourself. Frame interpolation ist a difficult topic, there is no absolute right or wrong or better or worse. In the end it even comes down to a question of personal taste. I for myself can not accept the terrible judder of 24p, as with my screen size and viewing distance, it is absolutely unbearable. And yes, I can confirm that a very good FI is even more important for 3D watching. This is why I chose Vivitek 1188 for my 3D stack, as these sport the best, flawlessly working FI that I have seen until today - and I have tested and compared many of them.


    In order to be able to use also professional projectors (e.g. 3-chip DLPs) in my cinema, which hardly ever come with integrated FI, I extensively tested and compared the 3 PC-based frame interpolation solutions available: SVP, AMD Fluid Motion and Dmitri Render. SVP is getting better and better and has many tuning options. It makes use of the CPU instead of the GPU, which is also helpful as one needs all the available GPU power for MadVR. However, up to now there is no possibility to get SVP running together with MadVR tonemapping, so it is basically useless. Luckily, not so with AMD Fluid Motion. I can run fluid motion on an older AMD graphics card (newer ones don´t support it any more), but still have another, way more powerful GPU for MadVR rendering. So it is basically a dual GPU setup with one GPU only for the FI, and the other one (can then also be a NVidia) for all the other stuff. This solution gives very satisfying results in terms of frame interpolation. My conclusion after many long comparisons is that the AMD FI works very well, absolutely smooth and fluid, only little artifacts from time to time. It doesn´t fully reach the perfection of the Vivitek or some other good DLP solutions such as Optoma HD83 or the like, but it is fairly close. In any way it performs noticeably better and more flawless than the JVC FI in the X series.

    But coming to the downside: regardless of which PC-based FI we are talking about: they cannot work with 3D BluRays, as the frame packed format doesn´t support passing 60 Hz on to the projector. This is why, for my 3D stack, I have to do the frame rate conversion after the splitting in right and left image (GeoBox 601), within the projectors.

  • If I were you, I would go and search for two Sim2 Lumis for your 3D stack, and keep one of your M.150 for 2D. They all have decent frame interpolation, so you don´t have to worry about that. The lumis is brighter, but with a "normal" light source, so optimal for a passive stack with interference filters.

    And finally, for 2D, the brightness of the M.150 is more than enough on a gain 2 screen of your size, and all the other qualities of this machine are impeccable.


    Btw, concerning 2k vs. 4k: NEVER would I go for a 4k projector with less quality in your case! Your viewing ratio (viewing distance divided by screen width) is so huge (must be in the region of 1.5:1, right?), you will never ever benefit from the higher resolution as much as you do from the other qualities of your SIM2s.

    I have a viewing ratio of around 0.6-0.7:1 (appr. 2.7 m seating distance, 4 m visible screen width), which is absolutely borderline in terms of resolution (by means of an anamorphic lens it is acceptable). And still I haven´t found a projector which gives me a better overall performance than my beloved Sim2.

  • This is why, for my 3D stack, I have to do the frame rate conversion after the splitting in right and left image (GeoBox 601), within the projectors.

    Is there a way to put the separated left and right eye images, after the Geobox, back into the PC to use PC frame interpolation on them, without losing the sync of left and right eye signals? Maybe with this or something? https://www.avsforum.com/threa…or.3206050/#post-60846896

    If I were you, I would go and search for two Sim2 Lumis for your 3D stack, and keep one of your M.150 for 2D. They all have decent frame interpolation, so you don´t have to worry about that. The lumis is brighter, but with a "normal" light source, so optimal for a passive stack with interference filters.

    And finally, for 2D, the brightness of the M.150 is more than enough on a gain 2 screen of your size, and all the other qualities of this machine are impeccable.


    Btw, concerning 2k vs. 4k: NEVER would I go for a 4k projector with less quality in your case! Your viewing ratio (viewing distance divided by screen width) is so huge (must be in the region of 1.5:1, right?), you will never ever benefit from the higher resolution as much as you do from the other qualities of your SIM2s.

    I have a viewing ratio of around 0.6-0.7:1 (appr. 2.7 m seating distance, 4 m visible screen width), which is absolutely borderline in terms of resolution (by means of an anamorphic lens it is acceptable). And still I haven´t found a projector which gives me a better overall performance than my beloved Sim2.

    Do you have the M-150? Do you think the single chip native contrast is a problem? (like 3,000:1 I have heard. I do not have my measurement equipment set up yet to measure mine, or my black curtains yet). The Lumis lineup has not just 5,000:1 or 6,000:1 native contrast, but dynamic contrast multiplied to around 20,000:1. Even the Mico 50 has dynamic contrast multiplied to I think 14,000:1 or so (although this model has no frame interpolation, cannot make near black shades like the M-150, coil wine at full LED power, etc, apparently). The M-150 is the only top Sim2 model where, I have heard, the dynamic contrast does not reach above 5,000:1. That is my biggest concern using it as my main projector, along with rainbow effect, although I have only seen maybe 1 every five hours. It has been a non-issue so far, but just knowing it is a possibility gives me "commitment issues," hehe.


    How do you think the M-150 frame interpolation compares to the Optoma 83 and Vivitek? I think it is very good except I dont know if my mind is playing tricks, but after bright flashes in a scene, it feels like it might drop a frame, and also after camera cuts sometimes to a new scene, it feels like it drops a frame. The whole point of frame interpolation is that it's smooth. Even with some artifacts, it should be smooth. Any dropped frame here or there is ruining the whole thing for me, at least in 2D. Maybe dropped frames will be less visible in 3D? I hope? I tested two M-150's and I am seeing the same effect at the same time in the same movies on both of them. I would much rather listen to a fast song where every 20 seconds, one of the fast drum beats creates a slight distortion, that a fast song where every 20 seconds, the entire song cuts out to zero volume for a split second before coming back. I can't stand dropped frames (at least in 2D) in frame interpolation. I can't believe Sim2 projectors would have that, because they are high end. I thought "at least they won't drop frames." And maybe they are not, maybe it is something else, but I also tested a Superlumis frame interpolation and its frame interpolation does the same thing at the same points in the same movies as the M-150. As for your suggestion, it could be 2030 before I find two of them used. The less bright models with frame interpolation would still take awhile, but not as long, however they are less bright and would still require a 2 gain screen as you said, which may defeat the purpose?


    It really depends, from 1.5 distance away, how much gain is too much on a flat screen before the negatives of high gain screens, like hotspotting and sheen, become visible? Because if high gain screens are acceptable, then two M-150's can possibly do the same job as two Superlumises, provided you raise the gain high enough (and provided the filters will work with the LED lightsource if you have enough brightness headroom to manipulate the colors with 3D lut). And then there is also the question about removing hotspotting by using a curved screen, but will this prevent hotspotting for the center seat only, or also seats that are let's say halfway between the center seat and the edges of the screen, or maybe even 65/35 closer to the edges than the center?


    And a curved screen requires less warping to fit the image on it if you use two anamorphic lenses, but is double-stacking with two anamorphic lenses viable even with warping, or will it be impossible to align two projectors, even with warping, when using anamorphic lenses? I always figured "if warping can correct it without anamorphic lenses, then it can correct it with anamorphic lenses. Once the anamorphic lenses are on, both projectors will be throwing images where they are, and then you just warp them back to the proper place, same as if you weren't using anamorphic lenses." But other people have warned me not to attempt it with two anamorphic lenses. At the same time, they have never attempted it (but neither have I).


    There are so many variables involved. One variable leads into another. The one "sure thing" is two Superlumis, with the rarer "high brightness" lenses to maximize brightness, shooting onto a flat 1.0 or 1.3 gain screen. But like I said, it could be years before I find that, and the prices seem to be all over the place too, sometimes workable compared to the price of a used M-150, and sometimes pretty crazy. At least, that's what I was told, but really I dont know. You cant even find records of any past sales in the U.S. to see the prices. But if there is a way to do frame interpolation for both right eye and left eye signals on a PC, that would give me many more options of bright 3 chip projector models that dont have frame interpolation. That would also open up yet another variable, "how does PC frame interpolation from 23/24fps to 60fps compare to frame interpolation from 23/24 to 47/48fps, where it fits evenly?" And what framerate is better for 3D in general, 48, or 60? But at this point, I would take it, regardless...


    Otherwise if not, stacking M-150's is probably the best direction for me, short of just getting two Optoma 83's or Vivitek's with bulb lightsource, like you mentioned, where I now know the frame interpolation is reliable, except now I am definitely having a rainbow effect problem (unless it disappears in 3D or something?). But assuming I will get better results in that way, and maybe overall picture quality, with double-stacked M-150's, I'd definitely appreciate your advice about how high gain I can go, on a flat screen, without problems, from 1.5 distance away, as well as whether doing a curved screen so I can go higher gain than whatever that is without issues is a good idea, as well as whether a curved screen will eliminate hotspotting only for the center seat or also the side seats, as well as whether to use horizontal curves only or horizontal + vertical curves, as well as whether to use double anamorphic lenses or not.


    *Lets out deep breath.* I very much apologize for the complexity of it all, but, this is what I am dealing with, lol, so you can see why I need the advice so badly.

  • Sure there is a way to tell how good the quality of any FI is - watch it, test it, and then judge for yourself. Frame interpolation ist a difficult topic, there is no absolute right or wrong or better or worse. In the end it even comes down to a question of personal taste. I for myself can not accept the terrible judder of 24p, as with my screen size and viewing distance, it is absolutely unbearable. And yes, I can confirm that a very good FI is even more important for 3D watching. This is why I chose Vivitek 1188 for my 3D stack, as these sport the best, flawlessly working FI that I have seen until today - and I have tested and compared many of them.


    In order to be able to use also professional projectors (e.g. 3-chip DLPs) in my cinema, which hardly ever come with integrated FI, I extensively tested and compared the 3 PC-based frame interpolation solutions available: SVP, AMD Fluid Motion and Dmitri Render. SVP is getting better and better and has many tuning options. It makes use of the CPU instead of the GPU, which is also helpful as one needs all the available GPU power for MadVR. However, up to now there is no possibility to get SVP running together with MadVR tonemapping, so it is basically useless. Luckily, not so with AMD Fluid Motion. I can run fluid motion on an older AMD graphics card (newer ones don´t support it any more), but still have another, way more powerful GPU for MadVR rendering. So it is basically a dual GPU setup with one GPU only for the FI, and the other one (can then also be a NVidia) for all the other stuff. This solution gives very satisfying results in terms of frame interpolation. My conclusion after many long comparisons is that the AMD FI works very well, absolutely smooth and fluid, only little artifacts from time to time. It doesn´t fully reach the perfection of the Vivitek or some other good DLP solutions such as Optoma HD83 or the like, but it is fairly close. In any way it performs noticeably better and more flawless than the JVC FI in the X series.

    But coming to the downside: regardless of which PC-based FI we are talking about: they cannot work with 3D BluRays, as the frame packed format doesn´t support passing 60 Hz on to the projector. This is why, for my 3D stack, I have to do the frame rate conversion after the splitting in right and left image (GeoBox 601), within the projectors.


    Also, for 2D, I dont know how to do it, but I think I saw people using SVP and madvr together, without the use of "VideoProcessor" also. Not 100% sure but pretty sure there is some way to do it. It would be buried somewhere on that same forum I linked.


    Edit: I went through my old searches for SVP and madvr. This mentions madvr: https://www.svp-team.com/wiki/SVP:MPC-HC


    This has more instructions halfway down the page: https://www.svp-team.com/wiki/SVP:4K_and_HDR


    More: https://www.svp-team.com/forum/viewtopic.php?id=4448, https://www.reddit.com/r/softw…rpolation_svp4madvrmpchc/, https://www.reddit.com/r/Windo…hcbe_with_madvrsvp_guide/, https://www.svp-team.com/forum/viewtopic.php?id=5353, Madvr also has smooth motion. Does top response say it's better? https://www.svp-team.com/forum/viewtopic.php?id=1287

  • Is there a way to put the separated left and right eye images, after the Geobox, back into the PC to use PC frame interpolation on them, without losing the sync of left and right eye signals? Maybe with this or something? https://www.avsforum.com/threa…or.3206050/#post-60846896

    ...

    I would say yes. With two additional PCs! I have already had the same idea and thought about it. After the Geobox, you have two standard 1080p24 pictures for each projector. You could now build 2 identical PCs with grabber cards and AMD fluid motion, and send those two streams through those machines, 24p in, 60p out. In theory this should work. Don´t know which drawbacks would await you if you tried in practice, but I´m sure there will be some...

    Zitat

    Do you have the M-150? Do you think the single chip native contrast is a problem? (like 3,000:1 I have heard. I do not have my measurement equipment set up yet to measure mine, or my black curtains yet). The Lumis lineup has not just 5,000:1 or 6,000:1 native contrast, but dynamic contrast multiplied to around 20,000:1. Even the Mico 50 has dynamic contrast multiplied to I think 14,000:1 or so (although this model has no frame interpolation, cannot make near black shades like the M-150, coil wine at full LED power, etc, apparently). The M-150 is the only top Sim2 model where, I have heard, the dynamic contrast does not reach above 5,000:1. That is my biggest concern using it as my main projector, along with rainbow effect, although I have only seen maybe 1 every five hours. It has been a non-issue so far, but just knowing it is a possibility gives me "commitment issues," hehe.

    Yes, I own a M.150, just read my post above :zwinker2:

    To be honest, I don´t really understand your approach here. No offense, but if you are planing such a complex setup, you cannot only rely on somebody´s post in a forum or hope for some general rule of thumb - as there isn´t one in those regions we´re talkin about. You have to fully soak it up and figure the physical / technical basics out by yourself. After that (or parallel), you have to gather your own practical ecxperience, test, compare, find out what works and what doesn´t work.

    Coming to your question: you own two M.150s. Go ahead and play with them, test them, find out what you wanna know! I cannot tell you whether the native contrast will be enough for your demand and in your specific setup. I can only say, for me it does the job. And it does it (most of the time) better than my JVC X7000 with more than 10 times the native contrast. But that might not be valid for you. Same for rainbows. Watch some content with it and then judge if it works for you or not. Nobody can take that decision from you.


    Zitat

    How do you think the M-150 frame interpolation compares to the Optoma 83 and Vivitek? I think it is very good except I dont know if my mind is playing tricks, but after bright flashes in a scene, it feels like it might drop a frame, and also after camera cuts sometimes to a new scene, it feels like it drops a frame. The whole point of frame interpolation is that it's smooth. Even with some artifacts, it should be smooth. Any dropped frame here or there is ruining the whole thing for me, at least in 2D. Maybe dropped frames will be less visible in 3D? I hope? I tested two M-150's and I am seeing the same effect at the same time in the same movies on both of them. I would much rather listen to a fast song where every 20 seconds, one of the fast drum beats creates a slight distortion, that a fast song where every 20 seconds, the entire song cuts out to zero volume for a split second before coming back. I can't stand dropped frames (at least in 2D) in frame interpolation. I can't believe Sim2 projectors would have that, because they are high end. I thought "at least they won't drop frames." And maybe they are not, maybe it is something else, but I also tested a Superlumis frame interpolation and its frame interpolation does the same thing at the same points in the same movies as the M-150. As for your suggestion, it could be 2030 before I find two of them used. The less bright models with frame interpolation would still take awhile, but not as long, however they are less bright and would still require a 2 gain screen as you said, which may defeat the purpose?

    I find the FI of the Sim2 pretty good. On a scale from 0 to 10, where the Vivitek is a 9.5 and the Optoma maybe a 8.5, I would rank the Sim2 somewhere around 8. It does some blurring/smearing/tearing around objects in the foreground before a moving background, which is more visible and bothersome than on the Vivitek, but still very good. For comparison, the JVC X---- series FI would rank around 5 on my personal scale...

    That said, I haven´t experienced frame drops with the Sim2 yet. Should pay attention to this when I watch the next movie on it.

    I understand that it is very hard to come across 2 Superlumis! I just suggested it based on your statement that you are lucky to find preowned Sim2 for decent money in the US...


    Zitat

    It really depends, from 1.5 distance away, how much gain is too much on a flat screen before the negatives of high gain screens, like hotspotting and sheen, become visible? Because if high gain screens are acceptable, then two M-150's can possibly do the same job as two Superlumises, provided you raise the gain high enough (and provided the filters will work with the LED lightsource if you have enough brightness headroom to manipulate the colors with 3D lut). And then there is also the question about removing hotspotting by using a curved screen, but will this prevent hotspotting for the center seat only, or also seats that are let's say halfway between the center seat and the edges of the screen, or maybe even 65/35 closer to the edges than the center?

    As I also wrote before, I personally see the gain limit for flat screens at a maximum of 1.5. Wouldn´t go higher, even with your huge viewing distance. But like I said, there is no rule of thumb for this, you have to find out for yourself. And you really should play with that hotspot simulator for a while, as it will dramatically improve your understanding about the correlation of high gain, projection distance and viewing distance.

    Zitat

    And a curved screen requires less warping to fit the image on it if you use two anamorphic lenses, but is double-stacking with two anamorphic lenses viable even with warping, or will it be impossible to align two projectors, even with warping, when using anamorphic lenses? I always figured "if warping can correct it without anamorphic lenses, then it can correct it with anamorphic lenses. Once the anamorphic lenses are on, both projectors will be throwing images where they are, and then you just warp them back to the proper place, same as if you weren't using anamorphic lenses." But other people have warned me not to attempt it with two anamorphic lenses. At the same time, they have never attempted it (but neither have I).

    Why on earth should warping not work if you use anamorphic lenses? Have you looked into my build thread? I told you that I practice this for several years now with my 3D stack and the GeoBox 601. A proper warping tool like the GeoBox warps everything for you. You could project into a corner of a white room and then correct the image to look even and flat from your p.o.v., even though it is spread above parts of the ceiling, the front wall and one side wall. Correcting a curved screen is piece of cake for the GeoBox. So yes, you can of course bring 2 projectors with 2 anamorphic lenses into alignment, be it on a flat screen, on a single curved screen or on a torus screen like mine.


    Zitat

    There are so many variables involved. One variable leads into another. The one "sure thing" is two Superlumis, with the rarer "high brightness" lenses to maximize brightness, shooting onto a flat 1.0 or 1.3 gain screen. But like I said, it could be years before I find that, and the prices seem to be all over the place too, sometimes workable compared to the price of a used M-150, and sometimes pretty crazy. At least, that's what I was told, but really I dont know. You cant even find records of any past sales in the U.S. to see the prices. But if there is a way to do frame interpolation for both right eye and left eye signals on a PC, that would give me many more options of bright 3 chip projector models that dont have frame interpolation. That would also open up yet another variable, "how does PC frame interpolation from 23/24fps to 60fps compare to frame interpolation from 23/24 to 47/48fps, where it fits evenly?" And what framerate is better for 3D in general, 48, or 60? But at this point, I would take it, regardless...

    As described above, this would require 2 additional PCs only for the FI, and a lot of hassle will be involved to get this running stable, if even possible...

    I wouldn´t bother with that if I didn´t really have to. If you are interested in PC-based frame interpolation, start with one machine and get this one working. Then you see if it performs to your satisfaction and then you can think about where else you could go from there. Something of this complexity is not straightforward by following some cooking recipe any more! This is hard work, a steep learning curve, much time and efforts and last but not least, a whole bunch of setbacks!


    Zitat


    Otherwise if not, stacking M-150's is probably the best direction for me, short of just getting two Optoma 83's or Vivitek's with bulb lightsource, like you mentioned, where I now know the frame interpolation is reliable, except now I am definitely having a rainbow effect problem (unless it disappears in 3D or something?). But assuming I will get better results in that way, and maybe overall picture quality, with double-stacked M-150's, I'd definitely appreciate your advice about how high gain I can go, on a flat screen, without problems, from 1.5 distance away, as well as whether doing a curved screen so I can go higher gain than whatever that is without issues is a good idea, as well as whether a curved screen will eliminate hotspotting only for the center seat or also the side seats, as well as whether to use horizontal curves only or horizontal + vertical curves, as well as whether to use double anamorphic lenses or not.

    Buy those filters from Omega Optical and test it with your Sim2s. If you experience troubles related to the narrowband light source, then you can look for some alternatives. Regardless of which projectors you will use, I would opt for some sheeting with a gain of around 2, if you can find it, even a bit more (2.5 - 3), due to your large viewing distance. Then calculate the curvature matching your circumstances and see how it works out for you. My estimation will be that gain 2, max. gain 3 together with the right curvature will work out quite well for you, even for the two side seats. It´s a compromise for sure (like everything), but the benefits will outrun the detriments.

    Zitat

    *Lets out deep breath.* I very much apologize for the complexity of it all, but, this is what I am dealing with, lol, so you can see why I need the advice so badly.

    YESSSSS!!! Don´t get crazy about details of only academic nature. Start soaking up the basics, understand the coherences, and then work out a possible solution and start trying, testing, comparing.

    Again: what you are planning is NOT straight forward! You have to work it out for yourself.

  • I would say yes. With two additional PCs! I have already had the same idea and thought about it. After the Geobox, you have two standard 1080p24 pictures for each projector. You could now build 2 identical PCs with grabber cards and AMD fluid motion, and send those two streams through those machines, 24p in, 60p out. In theory this should work. Don´t know which drawbacks would await you if you tried in practice, but I´m sure there will be some...

    Someone has told me you can lose perfect sync with separate PCs, except if you use nvidia quadro cards, or did they mean using one quadro on the same PC, or two on the same PC? I do not know.



    Yes, I own a M.150, just read my post above :zwinker2:

    I think it said "my beloved Sim2" so I didnt know if it was the Sim2 M-150 or a different Sim2, although the context was 90% talking about an M-150.

    Coming to your question: you own two M.150s. Go ahead and play with them, test them, find out what you wanna know! I cannot tell you whether the native contrast will be enough for your demand and in your specific setup. I can only say, for me it does the job. And it does it (most of the time) better than my JVC X7000 with more than 10 times the native contrast. But that might not be valid for you. Same for rainbows. Watch some content with it and then judge if it works for you or not. Nobody can take that decision from you.


    That is helpful to know. Yes I have them but white walls right now, for another month or two while it takes to get curtains made for an entire 2,000 cubic foot room, so it's helpful to hear other people's opinions, especially ones with experience with other projectors. I do not actually own any projector that is considered high contrast not just for DLP, but overall, so it is helpful to learn from others' experiences. I wish I had the knowledge from testing many different projectors but it is too much to buy all at once with risk of not being able to re-sell, and too much work right now.

    I find the FI of the Sim2 pretty good. On a scale from 0 to 10, where the Vivitek is a 9.5 and the Optoma maybe a 8.5, I would rank the Sim2 somewhere around 8. It does some blurring/smearing/tearing around objects in the foreground before a moving background, which is more visible and bothersome than on the Vivitek, but still very good. For comparison, the JVC X---- series FI would rank around 5 on my personal scale...

    That said, I haven´t experienced frame drops with the Sim2 yet. Should pay attention to this when I watch the next movie on it.

    I understand that it is very hard to come across 2 Superlumis! I just suggested it based on your statement that you are lucky to find preowned Sim2 for decent money in the US...

    Thanks yes that was a great suggestion. It is definitely the easiest way to go if I can find it. There is also the issue of Superlumis samples, if one has slightly uneven misconvergence over here, while the other has it over there, and they do not match up. I wonder if that would be any issue for 3D? I have warping from the geobox, so I can align things correctly. The only thing I cannot align is misconvergence of red, green, or blue on a 3 chip projector. Whereas, the two M-150's are single chip, so they have no misconverged red, green, or blue. Is hypothetically let's say 1 pixel difference of only 1 color, between two Superlumises, something that will make any difference in the 3D, something that I should be factoring into this decision as an advantage for the M-150's, or is it "close enough," since it is usually only 1 color and not the entire pixel, that it's a non-factor in this case?

    As I also wrote before, I personally see the gain limit for flat screens at a maximum of 1.5. Wouldn´t go higher, even with your huge viewing distance. But like I said, there is no rule of thumb for this, you have to find out for yourself. And you really should play with that hotspot simulator for a while, as it will dramatically improve your understanding about the correlation of high gain, projection distance and viewing distance.

    Is 1.5 your maximum for 2D, or for 3D also? Because some people have said, as with frame interpolation's negatives, that the negatives of high gain screens are less visible in 3D. Well, sheen for sure. Hotspotting, the comments were not as clear, but I think even hotspotting was less visible in 3D. Have you tested > 1.5 in 3D or 2D or both?


    It's not letting me quote the next part, but no I didn't see your build thread? AHHHH now I see the problem. I missed your entire first response to me!! The first one I got notification for was "there is only one way to learn about FI. Test it. Etc." That is also why I only knew you had a "Sim2" not an M-150. Haha. Oops! I will read that.

    Buy those filters from Omega Optical and test it with your Sim2s. If you experience troubles related to the narrowband light source, then you can look for some alternatives. Regardless of which projectors you will use, I would opt for some sheeting with a gain of around 2, if you can find it, even a bit more (2.5 - 3), due to your large viewing distance. Then calculate the curvature matching your circumstances and see how it works out for you. My estimation will be that gain 2, max. gain 3 together with the right curvature will work out quite well for you, even for the two side seats. It´s a compromise for sure (like everything), but the benefits will outrun the detriments.

    I bought Infitec filters already, but they have different wavelengths than Omega. Omega also have the original "DLP" version plus also what they dubbed the "LCD" version, which is confusing because apparently it was first developed for SXRD Sony cinema projectors. I asked them which they would recommend for LED projectors and they have no idea. Whoever designed them does not seem to still be with the company, and they wouldn't answer how I could contact him either, in fact it was perplexingly difficult to get an answer about that. So it would require buying three different sets of filters, which are not particularly cheap, just to see if any matches the M-150's well. I will certainly test the Infitec filters once I figure out what stand to get for my xrite i1display pro plus sensor, and install the software and learn how to use it. I have gotten through the first 2,000 things on the list, and this is very high on the remaining 1,000 :D :D :( If the result seems hopeless then I will use linear polarization 3D with the M-150's, or find different projectors that will work with Infitec.


    That is another question for you. I know that linear polarization 3D degrades if you tilt your head, but, provided the viewer keeps his head upright, do you think there is any difference in 3D, eye strain, etc, the overall experience, with linear polarization 3D (i.e. IMAX 3D at least before they switched to lasers) compared to Infitec 3D (i.e. Dolby 3D, as you know)? Does having differently polarized light go to each eye at the same time do something weird to the eyes or brain, or is the effect the same as Dolby 3D except for the head-tilting drawback? I've heard stories about people getting nauseous from circular polarization 3D (realD) so I wondered if the differently polarized light to each eye could be the culprit, but I haven't heard stories like this about linear polarization 3D, in fact some people said they liked it while they didn't like circular. And on the other hand, some people get nauseous from any type of 3D period.


    But basically the question is, if I cannot get Infitec 3D to work, is it a "downgrade" to do linear polarization instead? Of course the head tilt issue is a downgrade, but is it a downgrade in any other ways, or should the experience be the same besides this?


    On a Stewart 3D screen (3.0 gain, 25 degree half gain viewing angle), or a Severtson 3D GX with 2.5 to 3 gain and a similar viewing angle, or a Severtson 3D GX with Wide Angle coating (can increase the half gain angle to 40 degrees on a 3.0 gain screen, but I have no idea whether that coating degrades the picture in other ways), I should be able to do linear polarization 3D with crosstalk at about 0.5% with the filters mounted in front of the projector lenses. Mounting them inside might make a small difference but since the major limiting factor is still the screen, not the filters, not a huge difference.


    With "gen 1" Infitec filters mounted in front of the lens, projectorreviews.com measured 0.4% crosstalk. I have "gen 2" Infitec filters, so I don't know if they are less affected by this, but I have to assume it's the same. However, mounting them inside the projectors, I'm guessing I could get crosstalk down to 0.1%, since in theory they should be performing at over 99.9% efficiency, and therefore one surmises that the only reason they measured as low as 99.6% in projectorreviews.com tests was that the projector beam was spreading out by the time it left the lens and so some of the light was hitting the filter at an angle instead of straight ahead. And this is what projectorreviews.com also said was the cause. However, mounting them inside the projector would be fairly complicated...


    So with the filters in front of the projector lenses, even the crosstalk is essentially the same (if you don't tilt your head). This adds back to the question of whether there is something innate in the way the filters operate with the eyes and brain that even when the crosstalk is identical, one would still be better to use if possible, or no? And then, once you get to 0.5% crosstalk or below, are the improvements visible, or will they impact eye strain and headaches? Going from 0.5% to 0.1% is a 5x improvement, which you could say is "five times better," and definitely worth it, or you could say, "it's probably not visible either way at 0.5% or below, so it's not worth modifying the projector." What do you think? Is 0.5% already low enough to not be seen or felt? Projectorreviews said that 0.5% is the threshold at which it is generally visible or noticeable on a test pattern or something like that, so 0.5% would be right at the "official" border of visible/not visible. But most people who used linear polarization set ups had way more than 0.5% and most didn't seem to notice any crosstalk in content, so... again it's very difficult to test without buying an expensive screen, modifying projectors just to compare them before and after, etc etc, and much easier to ask people in case they have tested it before themselves, or have learned the knowledge from someone else.


    My estimation will be that gain 2, max. gain 3 together with the right curvature will work out quite well for you, even for the two side seats. It´s a compromise for sure (like everything), but the benefits will outrun the detriments.

    Would you recommend 2.35:1 screen, using dual anamorphic lenses, in order to maximize the number of pixels being used for 2.35:1 movies, which make up most but not all of the 3D movies, or do you think from 1.5 viewing distance, about 26% less pixels, which are each 26% bigger, will not make any perceptible difference, and I am better off having the big 16:9 screen for 3D movies, since depth can be created with good height of a screen, and then maybe masking the same size for 2.35:1 that I would use for a 2.35:1 screen? I would get a better 16:9 experience, but less pixels, and bigger pixels, for 2.35:1 movies. Either way, it is a trade off, but the question is whether each tradeoff will be visible, or whether only the size of the 16:9 screen will be visible from 1.5 viewing distance, but not the extra and smaller pixels from using anamorphic lenses?

  • Staying with the projector choice: I wouldn´t consider laser-phosphor single-chip DLPs, especially with RGBW color wheels as your linked Christie above. Those have very limited gamuts, nothing for home cinema use.

    I only realized this after. I thought it was 3-chip like the others.

    First of all, as you obviously struggle with envisioning the multiple impacts on geometry and light distribution that come with a curved screen, I also strongly recommend you to get familiar with the hotspot simulator from FoLLgoTT !

    I couldn't get it to work. Thanks for explaining that it is a winrar problem. I will go back to it!

    Gain 5, gain 8, gain 10... As if you could pick whichever gain you choose off the shelf somewhere :shock: . Same goes for single or double curved. Other than the aforementioned Couchscreen, which is a very special case, I am not aware of one single manufacturer of double curved screens for home cinema use. So as long as you don´t go the "hard way" and construct one by yourself as I did, single curved is the only option for you. And then again, which company offers curved screens of the desired size with more than gain 2 (or something the like)?

    There is some company in China that had a 19 gain screen, and also, I dont remember, 4, 6, 7. But there was at least one option there. And couchscreen I was told is 6 gain? So I was thinking I have high gain options if needed, and that I might need them because the Infitec filters could take 75% of the light from a bulb projector, and maybe 85% from the M-150's (provided they will work at all). I am also unclear, let's say the Infitec filters do only leave 10% of the red color in one M-150, can I simply turn the red 90% down on the other M-150, and use a 10 gain screen, and now have 100% red on each projector? 10% on each multipled by 10 = 100%. Etc. Or even with a 10 gain screen, will turning red down so much with 3D luts etc prevent the M-150 from reaching rec709?


    The part I dont understand is, if the M-150's colors are turned down too low to reach rec709 measuring from the lens, does this also mean it cannot do it onscreen? Or if you measure it off the screen with a high gain screen, can the high gain screen increase not only the brightness, but the rec709 coverage as well due to reflecting back higher "color luminance" than is coming out of the lens?


    But if this were the case, how come a normal rec709 projector cannot reach 100% bt2020 color coverage by using a high gain screen? I have never heard of this being the case, so I will assume what I am asking will not work, but, just checking.


    Seating postition and projector position are to be considered exactly the same way! There is no difference technically. So as long as you want a decent image uniformity for a complete row of seats, don´t waste any more thoughts on gains higher than something around 2! And if you stick in the region of around gain 2, you also don´t have to waste a single thought on the slightly differing projector positions, as you will never be able to notice it on your screen!


    I see. Then couchscreen is ruled out I guess since it is 6 gain, unless it has special properties that will make it work. And using the M-150's in anything but linear polarization stack is also ruled out. :( Because the only way that would work is with probably, minimum, 4 gain. Maybe 3.5 depending on if it's true or not that using "LED overlap" mode will increase the projector's brightness by 25% and still allow it to reach rec709 calibrated. So my questions about linear polarization vs Infitec suddenly become much more relevant.


    I really really appreciate your help and advice! It is not just complicated about stacking in general, but because of determining what equipment to use, and to match with other equipment, most of which is not an option for me to test, so asking the people who already determined the info is the best way.

  • Someone has told me you can lose perfect sync with separate PCs, except if you use nvidia quadro cards, or did they mean using one quadro on the same PC, or two on the same PC? I do not know.

    I cannot tell you, as I haven´t experimented with capturing video streams yet. Wouldn´t wonder if you experience some sync issues (even with only one capture). However, capturing 2 strings on the same PC, and converting the frame rate on both of them, is - as far as my knowledge goes - not possible.

    Zitat


    Thanks yes that was a great suggestion. It is definitely the easiest way to go if I can find it. There is also the issue of Superlumis samples, if one has slightly uneven misconvergence over here, while the other has it over there, and they do not match up. I wonder if that would be any issue for 3D? I have warping from the geobox, so I can align things correctly. The only thing I cannot align is misconvergence of red, green, or blue on a 3 chip projector. Whereas, the two M-150's are single chip, so they have no misconverged red, green, or blue. Is hypothetically let's say 1 pixel difference of only 1 color, between two Superlumises, something that will make any difference in the 3D, something that I should be factoring into this decision as an advantage for the M-150's, or is it "close enough," since it is usually only 1 color and not the entire pixel, that it's a non-factor in this case?

    Definitively a non-factor! The projectors in a stack drift all the time with temperature (warm-up <-> cool down). That´s what makes it so hard (nearly impossible) to establish a long-term-stable stack (or edgeblend) for 2D. Because half a pixel off, and the image is completely blurred and your sharpness is gone :beat_plaste.

    Luckily, for 3D this is a completely different story. As one eye only sees the image of one projector, the only impact a slight shift has, is altering the perspective perception - but the sharpness remains untouched. This is tolerated and compensated by our brain easily, as long as it´s not abounding. My 3D stack shifts around in the region of 1, 2 or 3 pixels every time I fire it up - I don´t even bother compensating it with the GeoBox, as it has no visible impact on the 3D image. So please stop worrying about half-pixel-sized hypothetical convergence issues!!!

    Zitat


    Is 1.5 your maximum for 2D, or for 3D also? Because some people have said, as with frame interpolation's negatives, that the negatives of high gain screens are less visible in 3D. Well, sheen for sure. Hotspotting, the comments were not as clear, but I think even hotspotting was less visible in 3D. Have you tested > 1.5 in 3D or 2D or both?

    I do not distinguish here between 2D and 3D. Pleeease, these are some very very coarse estimations, based on my experiences and theoretical understandings. I will not tell you "1.5 is the max for 2D, in 3D you could even go 1.55 or 1.57" ...

    Zitat

    That is another question for you. I know that linear polarization 3D degrades if you tilt your head, but, provided the viewer keeps his head upright, do you think there is any difference in 3D, eye strain, etc, the overall experience, with linear polarization 3D (i.e. IMAX 3D at least before they switched to lasers) compared to Infitec 3D (i.e. Dolby 3D, as you know)? Does having differently polarized light go to each eye at the same time do something weird to the eyes or brain, or is the effect the same as Dolby 3D except for the head-tilting drawback? I've heard stories about people getting nauseous from circular polarization 3D (realD) so I wondered if the differently polarized light to each eye could be the culprit, but I haven't heard stories like this about linear polarization 3D, in fact some people said they liked it while they didn't like circular. And on the other hand, some people get nauseous from any type of 3D period.


    But basically the question is, if I cannot get Infitec 3D to work, is it a "downgrade" to do linear polarization instead? Of course the head tilt issue is a downgrade, but is it a downgrade in any other ways, or should the experience be the same besides this?

    I would not expect the filter functionality to have an system-inherent impact on our wellbeing when watching 3D. It´s more the general factors like flicker-free, ghosting-free and remaining luminance which separates "good" 3D from "bad" 3D. So whichever technology gives you the best results in these main disciplines should work best for you.


    Every optical filter is designed and thus works best with collimated light. So yes, best thing (especially for the color bandpass filters!) would be to mount them inside the projector, right after the lamp, or at least before the lens. I was not able to fit them inside my tiny little Viviteks, so mounted them outside, before the anamorphic lens. Still works to my satisfaction, so I don´t bother with it any further. For your huge throw distance it is even less critical, as your beam angle is way smaller than mine.

    However, are you fully aware of the fact that for any polarized 3D splitting, your screen material must be polarisation-preserving? This is the major downside of polarized 3D - you cannot use it with "normal" screens, wether they are gain 1 or 2 or whatever. In order to reflect the light with the same polarisation as the incoming light, the screen must have some sort of silverish coating - which on the other hand can impair overall picture quality.



    Zitat


    Would you recommend 2.35:1 screen, using dual anamorphic lenses, in order to maximize the number of pixels being used for 2.35:1 movies, which make up most but not all of the 3D movies, or do you think from 1.5 viewing distance, about 26% less pixels, which are each 26% bigger, will not make any perceptible difference, and I am better off having the big 16:9 screen for 3D movies, since depth can be created with good height of a screen, and then maybe masking the same size for 2.35:1 that I would use for a 2.35:1 screen? I would get a better 16:9 experience, but less pixels, and bigger pixels, for 2.35:1 movies. Either way, it is a trade off, but the question is whether each tradeoff will be visible, or whether only the size of the 16:9 screen will be visible from 1.5 viewing distance, but not the extra and smaller pixels from using anamorphic lenses?


    That´s the old and often enough discussed question "Cinemascope vs. 16:9" "Anamorphic lens yes/no" or "constant height vs. constant width"... I will not dive into these topics here, as pros and cons for each solution can be found enough in the forums and you have to decide which is the most bearable compromise for your demand. My favorite solution can be seen in my build thread.

    The only advice I want to give you here is: go waaaaaay bigger with your image and/or sit waaaaaay closer to it!!! 1.5:1 is better TV watching - that´s nothin to do with a really immersive cinema experience - point-blank. Especially since you are a fan of 3D. With 3D, you have to "fall" into the image, it has to fill all of your field of view, while the rest of the room is invisible (black). Only THEN will you experience the perfect immersion:woohoo:

  • Puuuh, you are dreadfully mixing up chromaticity coordinates, wavelengths and color luminance all together here :dry:

    This is way too complex to explain here within a few sentences. I recommend reading some literature about light spectrums, the CIE 1931 system, the physioligy of our vision, additive color-mixing and so on and so on...

    In short: No, a high gain screen will of course not alter the gamut of your projector (small unwanted hue-shifts due to the screen material´s properties neglected). Chromaticity coordinates and color luminance are two different pairs of shoes...

    And also, calibrating your Sim2 down to Rec.709 or using the LED-overlap mode will do nothing to the spectral composition of the emitted light. Yes, it will shift the balance between your three peaks (R, G and B), but it cannot alter the bandwidth of each peak, as this is "hard-coded" to your LEDs. So if those color filters cause problems because of the narrowband-lightsource, you can for sure try and compensate with some calibration, but you cannot ammeliorate the root cause of the mismatch.

  • Definitively a non-factor! The projectors in a stack drift all the time with temperature (warm-up <-> cool down). That´s what makes it so hard (nearly impossible) to establish a long-term-stable stack (or edgeblend) for 2D. Because half a pixel off, and the image is completely blurred and your sharpness is gone :beat_plaste .

    Luckily, for 3D this is a completely different story.

    But for 2D, if I want to use 3-chip DLP, does this mean if I buy a 3-chip DLP where the red pixel is off 0.5+ pixels from the others, throw it in the trash?


    And last question at the moment. Have you ever tried JVC NX5/NX7/NX9 aka RS1000/RS2000/RS3000 projectors "Clear Motion Drive," or the Sony VW885ES or similar models frame interpolation? Do you know if they ever skip frames?


    I would be extremely interested to hear, with your trained eye, if you notice the M-150 frame interpolation ever skipping frames. When a projector skips an interpolated frame every 30 seconds, it ruins the entire effect for me.


    I want to do my 3D stack with whatever the best projectors are I can afford that have frame interpolation that does not ever skip an interpolated frame, that keeps the smooth pace consistent like a clock. My preference was for DLP, so I thought either M-150's or Superlumises would be perfect for this, however, if their algorithm is sometimes skipping frames, which I believe it is (unless my settings are wrong and causing the problem, or something), then sadly it sort of ruins the whole purpose of buying them for 3D. For 2D, they are still great projectors to watch without frame interpolation, although even then, who is to say I would not prefer to watch 2D with frame interpolation on as well if it was not skipping frames? Whenever there is a rare scene where it doesn't skip at all, I do prefer the frame interpolation to having it off. So if it is skipping, that basically throws a wrench into everything I've spent months buying and setting up, not just for 3D but possibly for 2D too! :( True, at least for 2D, I can use PC interpolation, but that is not giving you a balanced number of original frames and interpolated frames, rather it is probably using all artificial frames, "blending" etc," in order to turn 24 into 48. But at least there is some option there).


    And if the Sim2's cannot do it without skipping frames, then I believe there are no other high end 1080p DLP projectors with frame interpolation besides them, which means I would have to look at lcos. Or I could copy your solution with Vitiek's or Optoma's, but I already tested two color wheel projectors and both bothered me. Maybe the rainbows would disappear in 3D though?


    Man, it is so difficult. Even after I found the perfect models on paper, both high quality projectors but also with the full features I needed, and even after I spent months to find them on the used market, their feature turns out to (seemingly) not perform correctly, something I could never have known until I bought them. In fact, the only person before you I found who had an M-150 told me that the frame interpolation was great...


    So now, I have to determine if it is truly the case or not that they are skipping frames, and if so, then I have to start over looking for something else. But the first step, if you can help me with your trained eyes, is to see if the M-150 is skipping frames. (For reference, whatever I noticed it doing, the Superlumis does in the exact same places. So either both are interpolating well, or both have the same problem). If it is, then unless you know of some Digital Projection model with frame interpolation, or some other 1DLP rgbLED or 3-DLP model with frame interpolation, to try, then Im guessing my best hope would be Sony or JVC 4K models. But I've at least heard rumblings that JVC may also skip frames, however I have not confirmed this. That would leave Sony. People always say Sony is known for good processing... maybe that will be the only one that can do it besides the color wheel DLP's? But step one, I need to figure out if the Sim2's are skipping interpolated frames at times, or not. Because of course I hope they are not, and that I can use them as I planned to after all.

  • But for 2D, if I want to use 3-chip DLP, does this mean if I buy a 3-chip DLP where the red pixel is off 0.5+ pixels from the others, throw it in the trash?

    No, of course not. Convergence mismatch is not as bad as whole-pixel-mismatch between 2 stacked projectors. So if you are talking about 1 single 3-chip projector for 2D use, which has one color half a pixel misaligned, you won´t even notice it from your viewing distance. Same goes for a 2D stack: if you have aligned the two 3-chippers pixel-perfectly, but one of them has .5 pixel misconvergence, you don´t have a problem really.

    The problem starts if one projector completely drifts away more than half a pixel from the other projector (in 2D!!!). And it will - believe me... Then your overall sharpness suffers visibly :(

    Zitat

    And last question at the moment. Have you ever tried JVC NX5/NX7/NX9 aka RS1000/RS2000/RS3000 projectors "Clear Motion Drive," or the Sony VW885ES or similar models frame interpolation? Do you know if they ever skip frames?


    I would be extremely interested to hear, with your trained eye, if you notice the M-150 frame interpolation ever skipping frames. When a projector skips an interpolated frame every 30 seconds, it ruins the entire effect for me.

    I haven´t seen the JVC NX line in person, so cannot say a thing about their FI. I only have a X7000 (and some older pro-JVCs without frame interpolation).


    Regarding the M.150, which level of frame interpolation are you referring to? I have been watching 2 full movies in the last days, and skipped through all the fi-levels on the SIM2 to watch out for possible skipping. Usually I have the fi on high, but put it to low and medium for quite some time to check on skipping. I must say I didn´t notice any skipped frames on the M.150, no matter which fi level I was using. I stay with my verdict that I find the fi on the SIM2 pretty good and usable. On low, it is still a bit too stuttering on my huge screen, whereas on high it´s perfectly smooth, but with some blurring around moving objects. But no lost frames as far as I am aware of.


    What is your image source? Can you check different sources/players? And which GeoBox do you use? Can it output 24 Hz? And have you activated the 24 Hz throughput? If not, the GeoBox will output 60 Hz, but with a 3:2 pulldown, so no good!

    Or do you use a HTPC and have low frame repeat times? The madVR info shows you how often a frame repeat / frame drop is expected. Could this be the root cause of your problem?

    Zitat

    And if the Sim2's cannot do it without skipping frames, then I believe there are no other high end 1080p DLP projectors with frame interpolation besides them, which means I would have to look at lcos. Or I could copy your solution with Vitiek's or Optoma's, but I already tested two color wheel projectors and both bothered me. Maybe the rainbows would disappear in 3D though?

    Don´t look for LCOS if you want the perfect motion handling! JVC´s frame interpolation is inferior to a good DLP´s fi, and Sony´s might be quite good (haven´t seen one in person), but can´t buy a Sony at all because of their panel degradation :blush:.

    Really, that Vivitek´s FI is absolutely gorgeous! I always run it on high. Smooth as butter, and almost no visible artifacts at all. I haven´t seen nor heard of any better frame interpolation. I luckily don´t have a problem with rainbows, but if you are sensitive to that, it might be a problem. If you see them in 2D, you also see them in 3D, that makes no difference. However, as these Viviteks have RGBRGB 6-speed colorwheels, it´s the best possible colorwheel solution (for rainbows as well as for Gamut).

    What type of colorwheel did the 2 DLPs have that bothered you?

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!