304 North Cardinal St.
Dorchester Center, MA 02124
304 North Cardinal St.
Dorchester Center, MA 02124
The Celluon PicoPro projector has been out for a few months now for about $359. I have read a number of so-called “reviews” that were very superficial and did little more than turn on the projector and run a few pictures and maybe make a video. But I have not seen any serious technical analysis or review that really showed the resolution or measured anything beyond the lumens. So I am going to be doing a multi-part technical analysis on this blog (there is just too much to cover in one article).
In the photo at the top, I took a picture with the lasers on to more clearly see the various light paths. A surprise to many is that they used 5 lasers and not just three which adds to the cost and complexity of the design. They use two red and green lasers to get to the spec’ed (and measured) brightness of 32 lumens. In future articles, I will get into more details on the optical path and what is going on (there are a few “tricks” they are using).
It is no secret by now that the Celluon engine uses a beam scanning mirror from Microvision and the optical engine and electronics are from Sony (the engine looks identical to the one Sony Announced February 20, 2014) . Below I have taken the cover off the electrical part so you can see some of the chips. If you look carefully at the red arrows in the picture below, you can see the 3 clearly identified Sony ASICs used in the driver board (the 4th large chip is a Samsung SDRAM and the smaller device is a Texas Instruments power supply chip — there are more power supply chips on the backside of the board).
I have used test charts to measure the resolution, check the color control , and measured the power consumption. I have also taken a look inside to see how it is made (per the pictures above). I have collected data and many images so the biggest problem for me to boil this down into a manageable form for presentation on this blog. I decide to start with just a bit about the resolution and a summary of some other issues.
Celluon claims the resolution is “1920 x 720” pixels and not that is not a typo on my part, they really claim to have “1920” horizontal resolution with as claimed by Sony in a press release on the engine. It is easily provable that the horizontal resolution is much less than 1920 or even 1280 pixels and the vertical resolution is not up to fully resolving 720 lines. In fact the effective/measurable resolution of the Celluon engine is closer to 640 by 360 pixels than it is to 1280×720.
PC Magazine’s April 22, 2015 article on the Celluon PicoPro made the oxymoron statement “the image has a slight soft-focus effect.” To me “soft-focus” means blurry and indeed the image is in fact both blurry and lower in resolution. The article also stated “I also saw some reddish tinges in dark gray areas in some images, a problem that also showed up in a black-and-white movie clip“. The image is definitely “off to the red” (white point at about 4000K) and it has very poor color control in the darker areas of the gray-scale.
Resolution is a big topic and I have a lot of photos, but to get things started, below I have taken a center crop of 1280×720 HDMI input into the Cellulon projector. Below this image I have included the same crop of the text pattern in put zoomed in by 2X for comparison. In the photo you will see a yellow measuring tape that was flush against the projection screen, this both shows the size of the projected image AND proves that the camera was focused well and had enough resolution to show pixels in the projected image.
You might want to look at the various areas indicated by the red ovals corresponding to the same areas of the projected image and the test pattern. What you can see is that there is effectively no modulation/resolution of the sets of 1 pixel wide vertical lines so the horizontal resolution is below 1280 (more like about half 1280).
There is some modulation, but not as much as you should get if this were truly 720p, of the horizontal lines center of the of the image but this will fade out towards the left and right side of the projected image (I will get into this more in a future article).
You may also notices that the overall Celluon image is blurry. Yes, I know lasers are supposed to “always be in focus,” but the image is definitely out of focus. It turns out that at the size of this image (12 inches vertical or 24 inches diagonal which is moderately big, the width of the scanned laser beams are wider than a pixel and thus overlap.
The image is even more blurry if the image is say about 7-inches high projected on a standard letter size sheet of paper (the image is very blurry). The blurriness goes down if the image gets bigger but it is NEVER really sharp even with a 72-inch diagonal image. In a future article I will post the same test pattern at different image sizes to show the effects of image size and blurriness/focus. I have started to call this “never in-focus technology.”
Some summary observations (more to come on these subjects):
I have seen Microvision laser scanned projectors since the Microvision ShowWX came out in 2010 or 5 years ago and the Celluon unit has many of the same issues that I found with the ShowWX. While the Celluon is much improved in terms of brightness and speckle, has better resolution (but not as near what is claimed) and it delivers about 3X the brightness for the about the same power (much of this is due to laser improvements over the last 5 years) the progress is very modest considering that 5 years have passed.
Frankly, I still consider this technology far from ready for “prime time” high volume and sill has some major and in many ways fatal flaws. Being laser safety class 3R at only 32 lumens is chief among them. The flicker I also consider to be a fatal problem for a consumer product but this perhaps could be solved by going to a higher refresh rate (which would require a much faster scanning mirror). The power consumption is far too high for embedding into small portable products.
And then we come back to the issues with the “use model” that still exists with Pico Projectors (see my discussion from way back in 2011 about this).
On a final note, I know that Laser Beam Scanning has a very dedicated following with some people that vigorously defend it. I will be providing test patterns and other information so people can duplicate my experiments and verify my results. I am more than happy to discuss the technology and respond to dissenting opinions, but I won’t tolerate rude comments or personal attack in the discussion.
Below are some test patterns stored in lossless PNG format to try out on the Celluon or other 720p projector to see for yourself.
Right-Click on the given pattern download the original full size pattern. Note, they should be view at “100%” if not on a 720p monitor and should totally fill the screen on 720p projector.
The first one below is a resolution test with 9 “zone patterns” has well as sets of 1 pixel wide black and white horizontal and vertical lines.
Simple horizontal gray ramp. This is totally neutral gray from 0 to 255.
Below may look dark gray or even black but it a totally flat R=B=G=16 everyone (a flay gray of 16/255). See how it looks on the Celluon.
1) As I stated long time before: The maximum input resolution is 1280×720 pixels and Celluon denied to support 1920 pixels. Even 1280 pixel resolution is not usable for Powerpoint presentations because text in normal font sizes are difficult to read. But I do not think that the conclusion must be that the projector cannot display 1920 pixels horizontally. Sony has surely measured it. The problem could be the algorithm that is upscaling the low resolution input to the higher resolution. Can you please think about that possibility that the poor details (e.g. the lines) is caused by the upscaling and not the projector engine?
2) For videos with 1280 pixels the image is quite good, sufficient for a portable projector. In dark rooms you can put the projector on the floor and watch a 2 meter diameter video on the celling. That is quite impressive. Cool.
3) Power consumption: There are five lasers. For mobile devices I think it could be reduced to three which will reduce the power consumption (see the Lenovo design). Of course, you have with every smartphone problems with the power consumption. See the Apple watch.
Important: A s I mentioned before already, everyone should take in mind that the Celluon projector is based on the first Sony design from 2014. I think, Sony will have improved many details until now and will improve them until they release final products.
I see, you found some same details figured out as I am mentioned before, especially about the resolution.Of course, your analysis is much better than mine, I did no detail research. Can you please research if the resolution problem is based on the upscaling process or by the engine? One point that leads to that it is based on the upscaling algorithm is the initial picture of the projector. There, in small font the version is displayed. It looks like higher resolution than when connected to Windows.
The product is okay, but not good enough at the moment for Windows. I hope Sony improved the final design.
I’ve personally used it via hdmi with an Asus NX500 Windows 8.1 and Windows 10 tech preview that has output resolution of 3840×2160 with no problem. I’ve personally used it with no problems on Miracast with HTC One M8 that has output resolution of 1920×1080.
It is true that if you have Miracast problems that Celluon will tell you to “try” a 1280×720 output if you contact them for help, but it doesn’t seem to be a requirement but rather a trouble-shooting tip that might work if your particular device has a Miracast issue.
Could not be possible, please look into the Windows settings for the Celluon device. Celluon has updated the specification some wweks ago that clearly says that the maximum Input resolution is 1280 x 720 pixels. From all devices, it has nothing to do with Windows.
You got a link for that change in specs? Not seeing it.
Celluon can’t be changing my laptop resolution from 3840×2160 “on the fly” without me noticing. That’s a BIG change. Obviously they are downscaling it to the native resolution of the projector, but that’s industry standard. The input resolution to the projector across HDMI is still 3840×2160.
Sure, Celluon updated most specs even at Amazon. Go to the PicoPro specification on the Celluon homepage:
“1920 X 720p at 16:9 wide screen, offering amazing clarity by digitally upscaling 1280x720p Input”
They support only 1280 input and upscaling it. It is impossible that you connected with a higher resolution.
You can check it in Windows by looking into the Display Settings. You cannot select a higher Resolution than 1280×720 Pixels for the Celluon projector there.
Actually, Windows let me select 1920×1080 and I tried it but the image quality was worse (looked like it was skipping data horizontally).
Frankly the whole 1920 horizontal resolution is marketing trying to confuse everything, probably to get people to believe that it is higher in resolution. This is a scanning projector so horizontal resolution is based on turning the lasers on and off as the laser scans. Clearly the system is not up to 1280 pixels horizontally.
I will be putting more images up in the coming days and it is very clear this engine is not up to 1280x720p resolution. The somewhat surprising thing is how blurry the image is on top of the lack of resolution.
@Karl: Maybe there is a Little difference between PicoAir (I have it, connected with Miracast) and PicoPro when connected using a HDMI cable). Can you select for the PicoPro 1920 x 720 pixels using Miracast?
I haven’t tried Miracast and don’t have an Android based platform in my hands. I’m not sure what it would prove. Have you tried displaying the test patterns on your system?
Do you think that the STM design used by Lenovo will produce better quality pictures?
I would not know its image quality without seeing it. The STM/Lenovo uses a two mirror (one horizontal and one vertical) versus a single due axis mirrro). In theory, the dual mirror approach should have some image quality advantages. It certainly appears to be smaller. A lot will also depend on the scanning process and whether STM turns the laser on in both directions as this looses effective resolution. My educated guess is that the STM design will also have lower measurable resolution than it claims.
(I had multiple people tell me to turn it off as it was painful to even be in the room with it).
I have shown this to dozens of people and not a single complaint. Just answer one question. Why did you take the time to write this review?
The issues with 60Hz interlaced scanning (and with zero persistence) is well know from the days of CRTs. Back in the 1980’s the EU even pass regulations requiring high refresh rates. The issue with flicker is a real issue and it does seriously bother people whether you want to believe it or not.
I wrote the review because I like to understand how things work and analyzing them, it goes hand in hand with being an engineer.
Thanks Karl. Very interesting so far, as usual. Looking forward to reading the rest of it.
It’s difficult to be sure from looking at your picture but it seems like the two red and two green lasers have different wavelengths. If so, this would contribute to an observable reduction in speckle.
It wonder whether this the reason for additional cost and complexity of using more than one of each primary?
Could you elaborate on the alignment issue?
You said, “Additionally there are issues with aligning the 5 lasers into a single “beam””
Ideally, the 5 laser would all hit the mirror in the same place and at exactly the same angle (in every direction). A spatial/location error is not as critical as it will remain constant over the scanned image. But a small angular will be multiplied by the distance of the screen from the mirror. Then consider that the angular error could be in different directions by different amounts with each of the 5 lasers. Therefore it becomes very important to have the 5 lasers as “coaxial” (moving parallel to each other over 360 degrees) as possible.
The Celluon projector (and the ShowWX) before it support “digital alignment” that tries to correct for the spatial error in alignment. The error can be several pixels. Digital alignment resamples the image and does not physically move the laser beams so it does tend to blur the image but it keeps the colors from being more than 1 pixel off.
I hope this is clear. If not let me know,
To add to Karl’s comments. Heat affects lasers as well. Once the laser heats up, it can shift its focus. Horizontally, this can be fixed by retarding/advancing the pixels. That is more difficult to do vertically where some dithering or line buffering can help. Soft focus occurs when the pixels spread due to heat-caused distortion as well as using multiple lasers that are not perfectly aligned.
When watching video, this is a bit more acceptable. Using PowerPoint/text/graphics….not so good.
I can understand the lasers changing shape, but not focus as this is scanning the beam.
I have taken some pictures demonstrating how the various lasers intensities change with heating. For example if you put a black rectangle in the image to cool the lasers, the colors will change on either side of it.
Karl, thank you for your recent review on the Celluon laser-based projector. I’m curious about a few things you mentioned.
What’s special about the Sony engine or Sony electronics? So Sony and Microvision got together in the past couple years and integrated their technologies. But it puzzles me as to what technical mechanism is Sony offering to the Microvision platform? Graphic processors?
How might the laser scanning projectors improve their brightness in the future? For there must be an ever present government mandated safety constraint on the power of the visible laser(s) being used in image projectors.
What are some solutions to increase brightness of future laser projectors? For example, future laser projectors may 1) use an optical “spreader” like a lens to increase the beam diameter and, thereby, reduce the intensity of the beam; 2) use more laser beams (again to spread the light source intensity out across a larger light source region); or 3) increase scan rate and thereby increase light intensity on projection surface, etc.
This is a curious problem where I’d love to hear other readers ideas.
Frankly, I a bit baffled why Sony is dealing with Microvision. My best guess (no inside information) is that there is an R&D group working on blue and green lasers pushing it. Nothing I have published or am about to publish is unknown to Sony (they definitely have people that know all these issues, not the least of which is the Class 3R laser problem).
I don’t see where the Sony chipset is doing anything that special. There is a chip to control the mirror which I would assume came largely from Microvision. There is a chip to control the lasers and perhaps Sony has some better IP for doing this. Then there is a “graphics chip” with DRAM that is resampling the image from a rectangular grid to the laser scanning process, but this is very old stuff. If anything Sony just has a better position and motivation to get the chips made.
I don’t know of anyway around the laser safety issues for Laser Scanning. They become Class 2 above about 1 lumen and become class 3R above about 20 lumens (some variability due to resolution and scan angle). As a laser safety expert told me (paraphrasing), “there are 3 different types of laser safety, 1. what the safety standards say, 2 what is actually safe, and 3 what witnesses will say is safe in court if someone gets hurt.” Scanning a tight beam really is dangerous to eye safety. Most of the techniques you talk about would only have a modest improvement in the safety.
In theory one could have two sets of beams with a scan starting at say the top and one at the middle that would then halve the intensity at a point, but this would increase complexity could not be repeated very often maybe quadrants. Taken to its extreme, there was a the was the GLV (http://www.sony.net/SonyInfo/News/Press_Archive/200206/02-023E/) which had a linear array of light modulator the vertical resolution of the display are then scanned horizontally by a rotating mirror.
Panel based projectors (LCOS and DLP as well as “holographic” projectors such as Two Trees Photonics for example) avoid this problem by spreading the light over a the panel and then further spread the light with the projection lens, in this way they are no more dangerous than “normal” projectors (a panel projector can be Class 1 up to about 100 lumens — once again there are a number of variables). You can see groups like LIPA (Laser Illuminated Projector Association) distancing themselves from laser scanning projectors.
Sounds like the “laser scanning” projectors will bump a wall related to image brightness as you point out. Just in the cursory thoughts we shared today to increase brightness (e.g., multi beam projection), there appears to be severe limitations of cost and complexity for laser scanning.
I’m much more optimistic the “panel based” laser projectors using LCOS, DLP, etc. will be able to use more power efficient and powerful lasers at a lower cost and complexity than laser scanning.
But I suspect LEDs are (and perhaps forever) will have lower cost than laser diodes. So power efficient laser projectors will cost the consumer more moolah. But who knows, with all of the R&D going on and LED production (from Cree, Nichia, etc.) ramping up for home/office lighting, perhaps LED efficiencies will vastly increase in the future.
Edit: ‘per say’ should be ‘per se’. Latin phrase meaning by or in itself or themselves; intrinsically. Nothing to do with english ‘say’.
Thanks, I think I fixed it.
Nice detailed review. Interesting to see new products still coming out with this technology. It was quickly identified with the first generation of laser scanning picoprojectors that laser safety was going to be a problem. The power ceiling has already been hit so I don’t see many obvious technology developments to help this.
In particular, it’s surprising to see 5 lasers in use. My colleagues and I published an idea of using 4 lasers (RGB and an additional yellow) which could increase the flux by something like 70% without increasing total laser power output. So, something like 30 lumens could still be Class 2. The idea is playing with how wavelengths mix to produce white and how efficiently that can be done. We recognized that yellow laser diodes are not readily available (but who knows in the future). The second problem, which came from other people, was that no one ever would want to put 4 lasers in a projector. Far too expensive, complicated and consumes space!! And here we see 5…
As for resolution, I can only assume that they have scan characteristics and laser modulation compatible with 1920 and 760. If they want to project to, say, an A4 width with 1920 pixels, a pixel will be less than 200µm. The laser spot won’t be much less than 1mm! It’s all “specmanship”. It’s a bit like an LCoS or DLP picoprojector quoting the chip resolution but ignoring the projection lens. I measured various picoprojectors (LCoS, DLP and LBS) who don’t achieve the resolution specified.
Thanks for the detailed analysis. It would be great to see some side-by-side images of similar pico projectors along with a similar techical analysis (DLP vs. LCOS vs. other).
I don’t have a other projectors on hand at the moment to compare against, but it would be good to compare. The pixel type displays (DLP and LCOS) can certainly resolved down to the pixel. I would make an exception for the DLP “diamond pixel” designs which have to rescale everything (https://www.kguttag.com/2012/02/09/ti-dlp-diamond-pixel/) but their so call “Manhattan” pixels are ok.
If you are interested, I have a report on different picoprojector technologies (colour filter LCoS, colour sequential LCoS, DLP and LBS). It’s about 2-3 years old now. We studied resolution, colour gamut, contrast, power consumption etc.
Sure I would be interested. Back when I was a Syndiant, I did similar internal studies.
Thanks for this article. According to your experience and the tests you made which technology would you consider as the most effective and reliable (in terme of costs, quality, resolution, brightness and sturdiness) to launch a new portable pico projector which could be used indoor and outdoor.
Currently the most common technology for pico projectors is DLP followed LCOS, both of which with LED illumination. Which is best depends, among other factors, on your brightness and cost goals as well as say environmental (temperature, humidity) requirements.
[…] scanning process and why the resolution is so poor. Other good back articles from this blog include Celluon Laser Beam Scanning Projector Technical Analysis – Part 1, Celluon Laser Beam Steering Analysis Part 2 – “Never In-Focus Technology,” […]
[…] of articles Sony’s LBS engine using Microvision mirrors in the Celluon LBS projector (see: Celluon Laser Beam Scanning Projector Technical Analysis – Part 1, Celluon Laser Beam Steering Ana…2B – “Never In-Focus Technology” Revisit, and Celluon/Sony/Microvision Optical […]