Syndiant is working towards Field-Sequential-LCoS microdisplays for pico projectors, driven be either laser or LED. Syndiant's CTO, Karl Guttag, wrote an interesting blog post a few weeks ago, giving his views on different projection technologies. One of his claims was that Laser-Beam-Steering (LBS) is "not viable", which resulted in a "reply-post" in Microvision's own blog.
Karl has agreed to answer a few questions we had, and hopefully this interview will clarify things a bit. Obviously we'll have to wait a few years to find out who is right, and whether LBS projectors will be successful or not...
Karl is the CTO and co-founder of Syndiant, and has 32 years of experience in semiconductors. Prior to Syndiant, Karl worked at Texas Instruments and Silicon Display. Karl is named as inventor on 138 issued US patents, including some for LCoS microdisplays. Karl has an M.Sc from the University of Michigan.
Q. Karl, thanks again for this interview. Can you give us some background info on Syndiant and your pico projector solution and products?
Syndiant was founded in 2004 to develop Liquid Crystal on Silicon (LCOS) microdisplays with all-digital display processor on the silicon. Syndiant’s first devices were 1920x1080P with over 12-bits per pixel of color depth aimed at the rear projection television (RPTV) market. Seeing the decline in RPTV and having developed small pixels with a programmable architecture, Syndiant found that its technology could support small high resolution microdisplays for the emerging pico projector market.
Syndiant is the leader in high resolution microdisplays in small packages. We are currently sampling the SYL2010 with 854x600 true pixels, the highest resolution device that is less than 7mm tall. Syndiant is working on devices under 6mm tall and devices with higher resolution.
Syndiant’s small pixels microdisplays enable high resolution with small optics. Working with our partners we expect to see embedded cell phone optical modules with our microdisplays that are less than 5 cubic centimeters in volume, yet have true WVGA (854 by 480 pixel) resolution.
Syndiant develops the LCOS display devices themselves and the ASIC to control the display devices. Syndiant then works with ODMs and OEMs with optical design and system expertise to develop a complete projector.
VueG8 (pronounced View-Gate) is the name we gave our pico projector microdisplay family. The SYL2010 is the first device in that family. The SYL2010 has 854 by 600 pixels on it so that it can support either SVGA at 800 by 600 pixels or WVGA at 854 by 480 pixels.
Note that SYL2010 is just the first member of our pico projector family that was completed quite a while ago. You can expect to see both smaller devices and higher resolution devices from Syndiant in the near future.
Q. Are you selling the projector modules already? Are there any products out there that use your technology?
A. Syndiant sells the LCOS microdisplays and the ASIC to control them. While we have built some demo projectors for internal use and customer demonstrations, production products with our microdisplays are being developed by our optical partners and customers.
Syndiant is in the volume sampling and will soon be starting mass production. Announcement by our customers of end products based on our microdisplays can be expected when these products go into production.
Q. How will your products compare with Color-Filter LCOS projectors? In performance, size and price?
CF-LCOS has exhibited poor color quality that some people think is somehow characteristic of “LCOS.” CF-LCOS uses 3 (or more) color filtered sub-pixels form 1 pixel and is illuminated by a “white” LED. While CF-LCOS is being used in some early product, it has drawbacks in terms of the size required for the 3 or more sub-pixels and the color bleed between the sub-pixels causing poor color saturation and poor color control. CF-LCOS is discussed in more detail in the Competitive Advantage section of our website.
Syndiant’s approach uses a field sequential color which is totally different and has excellent color capability and we use on-display processing to give very fine color control. Syndiant microdisplays are much smaller (typically our pixels are 4 to 5 times smaller), have higher resolution (more pixels), and much better color quality than CF-LCOS devices. Size also plays an important factor in cost and with our devices being much smaller than CF-LCOS microdisplays of similar resolution, the high volume manufacturing cost will be substantially lower.
Q. I understand that your products can work with a LED or laser light source? how is that possible?
Syndiant has multiple customers using both LEDs and lasers. In the case of LEDs the illumination optics collects the light whereas with lasers they tend to spread the light to illuminate the microdisplay.
LEDs today have a major cost advantages (often less than 1/10th the cost of lasers) as well as being readily available so most of our customers will be using LEDs. The broad interest in LEDs for general illumination is driving rapid improvement in efficiency.
In the case of lasers, the beam has to be shaped and you have to reduce speckle which involves techniques that are proprietary to our optical partners. Laser’s unique properties enable focus free operation while supporting even smaller microdisplays and projection optics with higher optical light efficiency.
LEDs are cost effective today for the highest volume applications and we see our panels going into high volume production with LED illumination allowing us to ramp our manufacturing capacity and drive down our manufacturing costs. We currently see lasers with our microdisplay as has giving more performance but at a high cost. As laser light sources become more cost effective should see more applications take advantage of the laser light’s unique characteristics.
Q. You have raised 12.1M$ total so far, including 3.5M$ recently (March 2009). Will you need more funding?
We are currently raising money to help fund our ramp into mass production. We have multiple customers that are telling us they want well more than 1 million units in 2010 and we are in the process of growing to meet the market needs.
Q. You were granted a 1.4M$ grant from the Korean government - you are working with Korean companies on Pico Projectors. How is that project coming along?
Our partners in this activity have asked us not to give details on this project but the fact that we received funding from the Korus program is a matter of public record.
Q. In your web site you mention a full HDTV with 1920x1080P resolution in volume production in 2010. Will this be a pico projector? Are you still on track for that? Is it practical?
We have the technology to make a 1920x1080P device for pico projectors and this device would certainly be small and cost effective for a pico projector, but the exact schedule for this device may change. The huge demand we are seeing for embedded cell phone devices has us re-evaluating where we prioritize the development of the 1080P for pico projectors.
Q. In a recent blog post, you have made some claims, and I'd be happy if you can give some more info. You said that laser-beam-steering projectors (LBS) are 'just not viable'. Several companies were not happy with those comments...
Laser bean steering appears to be one of those things that may sound great in theory but has a lot of problems in practice. The cost and availability alone should rule out LBS out for more than 99% of the potential market and the image quality appears much worse than can be achieved for much less with other technologies.
Quoting from Forbes Magazine dated June 08, 2009, “Once Microvision is making large volumes of its component, Tokman [CEO of Microvision] believes, the devices will cost equipment makers $100 apiece.” My goodness, I don’t know anyone else that thinks pico projectors will ever go “large volumes” when “costing equipment makers $100 apiece.” The vast majority of this market is looking for projector modules (LEDs or Lasers, microdisplay, electronics and optics) in the $50 range as a starting point and then dropping to less than $25 for high volume; these are the price points we are targeting for projector modules using Syndiant’s microdisplays.
One other thing to note is that LBS requires fast switching lasers, on the order of 100MHz switching speed were lasers illuminating Syndiant’s microdisplays can use more efficient and more commonly available “Continuous Wave” (CW) lasers. CW-lasers are generally more plentiful, less expensive and more efficient that the high speed switching lasers required by laser beam steering. There are also many more well understood techniques for dealing with speckle with lasers illuminating microdisplays whereas most people I talk to say that despeckle for LBS is never going to be very good due to the need for a tightly focused pixel size beam.
The speckle with LBS systems creates a noisy looking image and worse yet as you move your head the speckle moves. The LBS demos I have seen to date seem to choose demo content with dark backgrounds that will make speckle less noticeable. I have yet to see a LSB demo of a bright white background such as a text document (as shown in our video), a spreadsheet, or a typical presentation slide. LBS seems to always want to show text on dark backgrounds.
LBS systems also have lower effective resolution. For example if you look at our video, we demonstrate 8-points MS-Word® Arial font (on a white background I should add) to demonstrate we truly have single pixel resolution which should be readable on a WVGA projected image. The so called “WVGA” LBS projector demos at SID 2009 only showed a 10-point font on a dark background. On top of this there was considerable image noise that makes reading text less pleasing with LBS displays.
Don’t get me wrong, I think showing videos and cartoons is fine as a part of any pico projector demo and we show this type of content as well. But I think that pico projectors are going to be showing all kinds of content including videos, web browsing, text documents and presentations. I found the demo content being show for laser beam steering to be more than a bit contrived to avoid showing image quality issues with LBS.
There are a lot of other serious issues with LBS including light output limits due to eye safety, image uniformity, distortion, and alignment of the laser beams that will take more time to explain.
Q. You mention eye-safety issues with lasers. Is this real? Isn't those lasers just like the laser-pointers we all use? Why is laser more dangerous than LED? (especially as it's a moving laser, so it will not stay on one point in the eye for long, and the brightness/energy on each 'pixel' is the same as in other projectors).
I would suggest you ask the LBS companies and laser makers how bright a LBS can legally go before they require a special performance permit; the answer I have been told repeatedly is somewhere between 10 and 20 lumens. With lasers illuminating LCOS, the laser light exiting the projector is much more spread out and thus higher brightness levels can be achieved within existing regulated safety limits.
Laser pointers only have to illuminate 1 spot and are typical over 20 to 50 times less bright than that required for a 10 lumen LBS projector. The issue with LBS is that all the light of the whole image is focused in a beam the size of a single pixel. So for example the at WVGA resolution the beam is about 400,000 times brighter per unit area than with a LCOS system illuminated by a laser. Even though an LBS system quickly sweeps the beam, there is the concern of the instantaneous light output and the danger if the beam sweeping mechanism becomes defective.
> Q. You say that LBS projectors will be more expensive, more power-hungry, with less resolution and colors. I have seen a demo of a laser projector against a Color-Filter LCoS, and I have to say that besides the speckle, the image is much brighter and the colors are better. Also a laser can be turned to a lower energy on less-bright pixels in the image... Will your LCoS based projector deliver much better quality than color-filter LCoS? Can you clarify those claims?
Pico projectors using Syndiant’s technology will have higher resolution, better color, and more brightness than then color-filter LCOS projectors in the market today. Anyone that understands the costs of the lasers and LEDs knows that laser beam steering is going to cost considerably more than LEDs with LCOS in pico projectors today and for at least the next several years.
In terms of power, with LBS you have to consider the power that is used to make the lasers switch fast enough to turn the pixel on and off in the time of 1 pixel. This requires high speed analog power drivers that can consume considerable power. Green lasers that switch fast enough to support LBS are typically less efficient than their slower switching counterparts. Then there is all the power associated with the control of the laser and mirror.
Using much lower cost LEDs, we expect to have projectors with greater than 10 lumens per Watt soon and with the improvements in LED, optics, and our microdisplays we expect to have greater than 20 lumens per Watt in less than 2 years. These numbers are for an image that can have bright or dark content in the scene. It is also possible in movie modes to use dynamic light control as is done with today’s TVs and monitors to reduce the power in darker scenes.
Lasers when used with LCOS can be very power efficient, in fact, today is possible to project over 20 lumens per Watt using LCOS and lasers. Even though our microdisplays work with lasers, we expect that for at least the next several years most of our devices will use LED illumination due to cost.
In terms of color, the picture on your web site was with a Color Filter LCOS (CF-LCOS) projector and is in no way indicative of the color quality of Syndiant’s LCOS (please see the demo video on our web site in HD mode). In fact LCOS has been used in some of the best looking projectors ever made for home and theater use.
Additionally, our resolution is higher and our images are vastly smoother without speckle and other noise I see in LBS systems.
Q. You mention that your solutions will be 'focus' free. How is that possible? If the laser beams are 'diffused', won't it require an optical output lens to re-focus it?
It is not only possible, but it has been done by many different companies. I can assure you that I have personally seen multiple different laser projectors with LCOS that were focus free. With laser light, even after beam shaping and reducing speckle, the illumination light has a very high F-number. It’s complicated to explain but because the illumination has such a high f-number, the depth of focus will be so large as to not need focusing. The projection lens is used to enlarge the image but the lens can be smaller with lasers and there is no need for a focus mechanism. Also, the fact that the laser illumination has a high F-number will also result in very efficient use of the light by the optics.
Q. Where do you see the Pico Projector market in 3 years? What role will Syndiant make?
I don’t want to give too much away, but Syndiant is the leader in resolution and size for pico projector microdisplays and we plan on maintaining that leadership. With our small pixel technology enabling high resolution in a small package and low manufacturing cost we are exceptionally well positioned for this market.
It looks like the sweet spot for cell phones initially is going to be WVGA (854 by 480 pixels) and I expect that we will soon see 720P (1280 by 720 pixels) in cell phone projectors. In the PC accessory, video game, media player, and camera markets, I see the resolution quickly migrating to 720P and 1080P (1920x1080).
The cost of pico projectors is going to enable their use in all manner of products. Automobile manufactures are considering projectors for dash board displays that will consolidate a number of gauges and expensive wiring into a single, more flexible information display. The smaller and more efficient we can make the projectors, the larger the volumes will be. Within 3 to 4 years we could see over 100 million pico projectors shipped per year.
Thanks Karl, for this interesting interview. Good luck to you and Syndiant!