Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Celluon Laser Beam Steering Analysis Part 2 – “Never In-Focus Technology”
June 6th 2015 – Note, I am in the process of updating this analysis with new photos. The results are not dramatically different but I was able to improve the horizontal resolution slightly and now have some better pictures.
One of the first things I noticed when projecting text pattern images with the Celluon PicoPro was that the images were very blurry. I later found out that the smaller the image the blurrier it became.
To the left are high-resolution center crops of images taken with a 12-inch diagonal (about as big as you can get on a letter size sheet of paper, a 24-inch diagonal image (about as big as fits on a standard “B” size sheet of paper, and a 72-inch diagonal image I project on a wall. For reference I have also included a the same portion of the source 3x magnified.
As you should notice the 12-in diagonal image is completely blurry even at 1/2 the stated resolution. With the 24-inch diagonal you can start to see some “modulation” of the single pixel size lines horizontally but not vertically. With the 72-inch diagonal the horizontal lines are pretty clear but still the vertical lines are still pretty much a blur (on close visual inspection there is a little modulation of the single pixel wide lines).
What is happening is that size of the laser beams is larger than the pixel size for small images. The size of the beam diverges but at a slower rate than the size of the image grows so eventually the laser beam size is smaller than a “pixel” and you start to see separation between horizontal 1 pixel wide lines.
As for the horizontal resolution, whatever is driving the lasers in their horizontal sweep is not able to fully modulate them at single pixel resolution.
For the next set of 3 images (plus a 2x Magnified source) I have scale the images down so you can see more area. Note you need to click on the image to see it at its intended size and to see the detail. In these pictures you can see the ruler with both indicates the size of the image and shows that the camera was in-focus and could see the detail if it was in the projected image.
On the 24-inch diagonal and 72-in diagonal image I have drawn 3 ovals. The left oval is around a set of 4 line pairs (see source image) of horizontal and vertical lines. The middle and right ovals are each around 4 line pairs of vertical lines and two sets of 4 pairs of horizontal lines and where the horizontal and vertical lines cross is a set of 9 white pixels (never visible in any of the projected images).
Looking at the 72-inch image you may notice that you can barely make out the horizontal line pairs in the center oval but that they become blurry in the right oval. This is due to the interlaced Lissajous scanning being done (for more detail on the Microvision interlaced scanning process see: https://www.kguttag.com/2012/01/09/cynics-guild-to-ces-measuring-resolution/). The net effect of this scanning process is that vertical resolution is reduce from the center to the left and right sides.
The 5 year old Microvision ShowWX having this blurring issue with small images. In looking inside at the optics with the lasers on, I notice that the laser spot sizes were larger than expected. I’m left wondering if the larger laser spot sizes were at least in part cause by efforts to reduce speckle or for some other reason.
Next time, I plan on giving a little “tour” of the optics.
Addendum – How the pictures were taken, full resolution images, and source pattern used
All the pictures were taken with a Canon 70D (5472 by 3648 pixel) DSLR. By framing the pictures so that filled roughly 90% of the width, this meant there were roughly 4 camera pixel “samples” per pixel in the output image. The ruler in the picture was both to keep track of the size of the image and to make sure the camera was in-focus and could resolve single pixels (if they were there).
I did selectively zoom in with the camera on smaller regions to see if it made any measurable difference in resolving features in the images and it did not. I have included the test pattern I used and would welcome anyone using it to verify what I have shown.
By clicking on the thumbnails below you will bring up the full size image (depending on your browser it may not display full size until after you click on the magnifying glass). You can then right click to download the images. Each image is about 8 to 9 Megabytes and is stored in a high quality (low compression) JPG format. The source test pattern is stored in loss-less PNG.
Thank you very much for your excellent investigation!
Can you tell me what frequencies the lasers are? It appears they are using 5 lasers to correct/complete the color gamut as mush as for brightness.
Tom,
I understand where you are going with this but I think the two reds and two greens are the same wavelength. The warning label list the wavelengths as Red 639nm, Blue 445nm, and Green 522nm. Unlike LEDs there are not the wide range of colors available with lasers. In fact for efficiency/brightness they would rather have the Red be closer to 625nm, the blue closer to 465nm, and the green closer to 530nm, but these lasers are harder/more expensive to make and/or have short lifetimes.
Actually the color gamut with these lasers is wider than they want. In fact when they put a a “pure red” the Celluon projector actually mixes blue and green with it to move the color point. Similarly for “pure green” they mix blue and red with it and for “pure blue” they add green and red.
Karl
Karl, it’s been pointed out on the MVIS Yahoo message board that the lasers in your Celluon projector appear to be horizontally misaligned. That’s why we see red on the left side of the vertical lines in the image and blue on the right side.
I don’t see that chromatic aberration in the images projected from my PicoPro.
I think you’ll still see blurry pixels once you’ve corrected that, but I also think you need to align the lasers and/or repeat with monochromatic test pattern images if you want to fairly represent the projector’s output.
I did go though the alignment procedure before starting, but it is possible that they have drifted since as I did a number of tests and then redid them to get better pictures. I will take a look at redoing the alignment and see if it helps.
Karl
Karl,
Don’t you think it would be appropriate to pull your last two entries until you determine if the lasers were actually aligned correctly?
Bob,
No, I have check the alignment and while I can change the effects, it does not substantially change the results with the exception that in some alignments I can now see some resolving of vertical lines (but then the horizontal line detail disappeared). It get some unpredictable results based on how I align the lasers I will be posting updates with my findings.
No matter how I align the lasers (and I was using some extreme techniques to get them aligned as accurately as possible) there is still the “chroma aberations” that are seen in the pictures. It should noted that the camera being “objective” may make these color fringes more noticeable than might be seen by the human eye and because my pictures in effect give you an extreme closeup view you are seeing detail you would not see.
BTW, after messing around with Celluon’s laser alignment (poor/minimal user interface), using a focus wheel is a piece of cake by comparison.
Karl
Haha, true to your DLP roots to the end I see! I happen to disagree with you. The majority of users will not need the detailed analysis you did nor will they need to align the lasers.
I am far from anything even resembling a scientist. However, I have bought three Celluon units. I gave 2 away for graduation gifts. They work really well and I have been very pleased. The only situation I have found the unit to struggle is in very low light scenes. While watching Game of Thrones on the unit through HBO Go, the really dark scenes were very difficult to see and had a some reddish hue to them.
If I had to use a focus wheel for on the go viewing, I would not be a buyer. However, that is just my preference.
Bob,
I don’t think you know my history as I don’t have “DLP roots.” I was a TI Fellow but in the graphics and image processing groups that were totally separate from DLP.
I had people suggesting that I didn’t aligned the lasers so I went back an aligned them better than probably 99% of the users would. I have now posted my updated results.
As you have found the color control of the Celluon unit is not very good, particularly in the darker areas. This will be a subject of a future article.
I’m glad you are pleased with your purchases. But by any objective measure, the Celluon unit does not have 720p (1280×720 pixel) resolution that it claims.
Karl
I stand corrected on your history. But please don’t put words in my mouth. I never stated the color control was not good. I simply said the only time I found the unit to struggle was in exceptionally dark scenes.
I was not able to allign the laser so I see white only lines. I have still red on the left and right side of the line.
I have gone through and re-align the lasers and shown the results in part “2B.” It really does not change things that much and there is a chroma “run-out” that gets worse on the left hand side of the image (there is a picture in the the “2B” article showing the left versus right side). I would be curious if you did a closer inspection of your projector if you don’t have it too. I’m thinking the chroma aberration may be caused by the lens on the output of the projector.
Karl
The 72″ picture actually looks very good to me. Sure, a little soft and inaccurate but hey – better than looking at the grey mess any other projection technology produces that they dare to call “black”.
What worries me is that grain/speckle/whatever you’ve captured here. But considering you’ve taken these at ISO800… might just be from the camera. Why didn’t you use ISO100? At 1/3 sec. exposure time, you clearly didn’t shoot this handheld anyway.
The contrast of the Celluon projector can be very good in a dark room (but the 80,000:1 claim is marketing hype that is not achievable in the real world).
The ISO800 in the previous 72-inch shooting mistake as I usually try and take these pictures at ISO100. I corrected this in the “2B” article posted today. Still, the camera use as a Canon 70D DLSR that still has pretty low noise at ISO800.
There is definitely some laser speckle (or grain) in the projected images but it is much better than the ShowWX from 5 years ago. It is impossible to fully represent speckle with a still image in a camera. The camera does not exactly work like the human eye and the eye itself is a factor in the speckle that is seen (what is known as “Subjective Speckle”). I think, however, the pictures I took at F6.3 at least comes close to capturing the level of noise in the image from speckle.
Karl
What i have figured out about celluon picopro:
It can project 720 lines with no problem (and i will show all of you pics made with 5Dmk2). But it can’t show 1280 columns. I assume it happens because of internal upscaling. As you remember they said about capability of displaying 1920*720 with input of 1280*720. SO there will never be true 1280 pixel to pixel resolution when talking about columns not lines.
And, of course, there is laser misalignment causing color halos.
The lasers are of course not perfectly aligned. I’m pretty sure there are also chroma aberrations (color halos) cause by the lens on the output of the projector.
The 5Dmk2 is a good camera but it is not going to see what is not there. BTW, I have taken shots an 85mm prime lens covering 1/3rd the area to get 3x the resolution to see if my 18-135 STM lens was missing anything and there was not a significant different in what can be seen. I will happily publish your pictures if you want me to, particularly if they show something different/better than what I have shown. My goal is to be accurate.
I know for a fact that the Celluon projector cannot resolve 720 lines throughout the image due to the scanning process (please see the article I wrote: https://www.kguttag.com/2012/01/09/cynics-guild-to-ces-measuring-resolution/ that details the Microvision scanning process). You can more or less resolve sets of parallel horizontal lines (they are a bit blurry and not well modulated but you can at least tell they are there) in the center 1/3 to 1/2 of the screen, but the resolution drops off on the left and right side.
There are 9 “resolution targets” spread around my test image. There is no way you will see full resolution on the left and right side targets.
I did find that by taking extreme measures with color filters to align the lasers (really this just moves their effect digitally) I was able to see some modulation of vertical parallel lines but only in a small part of the image (see the 72-inch diagonal image in my “2B” post). The problem with the way the scanning process works and they way the have to scale a rectilinear image onto a sinusoidal pattern with variable distance between lines and interlaced scanning there are regions of the display are are at about 1/2 the claimed resolution.
Or more simply, there may be small areas were you can detect 1280 by 720 pixel resolution (although never fully modulated) but there will always be areas where it can’t be resolved. Nothing your camera can do will change this.
i will make photos asap. I have made Interesting experiment. +4 diopters close-up lens infront of a projector drastically improve focus. So i’ve made lines pattern. It can be clearly seen.
here you can donwload it http://www.eonmail.ru/1/clip1093_6Kb.png
Oleg,
Thanks, but I think what you are going to find is that you will see some modulation (the lines will be neither black or white but you will be able to see them) in the center of the screen but they will fade/blur as you go to the left or right side of the screen.
BTW, I also tried also a screen full of 1 pixel wide vertical lines and found as serious of ovals and waves that moved around the screen in various places. I’m assuming this has something to do with laser heating and/or control. Below is a link to the 1 pixel vertical line test patter. http://wp.me/a20SKR-hE
Also try putting up a screen of about 160/255 flat gray or 100% white. If you blow it up large (say 50″ to 80″ diagonal) and move you eyes up and down slowly you will see lines various appear and disappear particularly on the outside left or right of the projected image. If you move you eyes at the “right” rate, the lines will appear longer. This is cause by the fact that the image is interlaced and on the outsides it only retraces the same location at 30Hz which can really bother people sensitive to flicker.
Also in case you are interested, I just added to the “2B” article some pictures I took with my 85F1.8 lens and covering a smaller area of the projected image to more than double the horizontal and vertical resolution. This is meant to take out all excuses that my lens and/or camera are causing any of the effects.
Karl
Karl
Karl, you are right. Lines can be seen, but they are not black and white. Contrast levels are low but you can distinguish the lines.
[…] from this blog include Celluon Laser Beam Scanning Projector Technical Analysis – Part 1, Celluon Laser Beam Steering Analysis Part 2 – “Never In-Focus Technology,” and Celluon/Sony/Microvision Optical […]
[…] Because they turn the laser on in both direction of the scan, unlike an old CRT that only turns on in one direction, the result is a zig-zag scan. Following just the blue field, you should see that in Column 5 (C5) crosses through the very top of R0 and the very bottom of R1, but then it crosses the very top of R2. So the “pixels” in C5 are in pairs. The red field has the same effect but shifted down one row. So they are counting on the interlaced fields compensating for each other. From experience with prior Microvision projectors, the effect is that left and right side tend to be blurrier than the middle, but even the middle is not sharp (see for example my study of the Microvision based Celluon projector). […]