Archive for Karl_Guttag

Pico Projector “Use Model” (Part 3 – Raison d’être)

For this installment on the pico projector “use mode” I would like to step back and talk about the original reason for pico projectors in the first place,  their Raison d’être if you will.  I will also discuss why drifted away from that use and how they should return.

Pico projectors original premise was built around producing an image larger than the physical dimensions of an ultra portable device such as a cell phone.  Implicit in this premise is that the device must be able to work within the power, size, and cost constraints imposed by a cell phone.    We simplified this to it having to be “battery powered and easily fit in a shirt pocket.”

So the Raison d’être for pico projectors was to give a bigger, and I would add, higher resolution image than is possible with a portable device such as a smart phone.

Surveying the market a few years ago, cell phone as a group had a similar set of requirements

  1. Cost less than $30 for the projector including electronics.
  2. 1 to at most 1.5 Watts of total power (display and any electronics)
  3. No more than 7mm tall and  about 4CC of space

Less well understood was the brightness and resolution for this new product concept so these were traded for the requirements the cell phone makers better understood.    Early on nobody could meet any of the requirements above.

To try and even get close to the cost, size, and power goals, TI’s DLP® pico devices went with very low resolutions of half VGA (480×320) and nHD (640×360).   With LCOS it was easier to make small low-power pixels, so we focused on WVGA (854×480) which was 6mm tall and SVGA (800×600) which was 7mm tall with both devices having a 0.21-inch diagonal displays.   LCOS’s challenge was that it works with polarized light and with LEDs causes 25% to 50% of the light to be lost.    BTW, laser beam scanning (LBS), was a non-factor as the cost of the required lasers alone would break the budget, not to mention the size of the electronics and a myriad of other problems.

Looming over DLP and LCOS when used with LEDs is an optical physical property call “etendue.”  (for a good paper on the subject of etendue with LEDs see OSRAM’s “Projection with LED Light Sources”).   Without getting into all the physics, LEDs produce light that spreads out such that the optical efficiency goes down dramatically as the microdisplay size decreases and/or the size of the LED increases.   One of the virtues of lasers used with microdisplays that they have near zero etendue and so there is no light loss due to etendue no matter how small the display.   An additional big benefit for LCOS is that lasers produce polarized light which eliminates the polarized light loss associated with LCOS using LEDs.

With the constraints imposed by LEDs and small size, the result was some “gimmick” products with low brightness and low resolution.    You could project a relatively low resolution image in a dark room.  It is impressive when you first see a big image coming from such a small product, but the big question became “how are you going to use this on a regular basis?”  Samsung and LG both made some exceeding low volume test the market cell with projectors in them.    Nikon has made a couple of cameras with pico projectors in them.  The biggest volume of embedded pico projectors has been some low resolution color filter LCOS projector embedded in cell phones sold mostly in India where they are use as cell phone based televisions.

For round 2 of the pico projector market, DLP and LCOS drifted away from the embedded market which wasn’t selling well and backed up (got larger) to support portable media player projectors.  The projectors required more brightness and to go brighter with LEDs meant requiring bigger microdisplays and bigger optics.   Syndiant kept the 800×600 resolution and simply made the pixels bigger.  DLP keep the pixels the same size and added more pixels to make a bigger displays, first 0.31-inch diagonal 848×480 (WVGA) and later a 1280×800 0.44-inch.    These products require larger batteries or wall plugs and more for carrying in separate bag than in your pocket.   Interestingly, DLP also rotated the mirrors 45 degrees to what they call a “diamond pixel” which causes some strange artifacts and a loss of effective resolution (the issues with diamond pixels will be discussed in a future article).

I don’t think anyone believes that these portable projectors as they stand today are anywhere close to the vision of embedding projectors with cell phones.  The microdisplay makers found that at least there was a class of sellable products that could be made by going larger to support more light output with LEDs.

To get back to the original vision for pico projectors and the opportunity for very large volumes of products with the embedding in pocket sized portable products, my conclusion is that affordable lasers, particularly green, are required.   Lasers will enable the microdisplays to be made small without a large loss of light and this in turn will lead to smaller and less expensive optics.  Since the lasers naturally produce polarized like, LCOS will gain the additional efficiency benefit.

If I “plug into the equation” affordable lasers, then I see how to meet the requirements first outlined by the cell phone makers.   Additionally, the properties of laser light enable very short throw so that a image the size of a piece of letter paper (about the size of an iPad) can be projected from only a few inches above a surface in what I call “down-shooting.”    With a built-in inexpensive camera aimed at the projection image to capture user input and you have a ipad/tablet that fits in your pocket.

The concept art at the top is one I generated when I first started working on pico projectors several years ago.  I knew they would not be very bright if they were going to work on batteries so down-shooting onto a piece of paper seemed like the way to go and I think it will be coming back to this use model as the technology progresses.

One last thing:  While one is building a virtual iPad/Tablet product, it better have at least the resolution of an iPad which today is 1024×768 pixels and will likely be going higher in the future.   One reason will be is that more resolution is more useful, the other is that history has shown that consumers want more pixels whether they need them or not.


While there is talk about someday having OLED’s or LCD display that will roll up like a sheet of paper, it is believed that a practical display that can roll up is still many years in the future.   There are many serious technical problems including ruggedness/lifetime in having a near perfect air tight seal with something that is thin and flexible enough to roll (air will destroy OLED’s and LCDs) and the optical uniformity problems with something that is flexible.

There have been some “lab prototypes” which may work if carefully handled and with short lifetimes, but these are far from being ready for consumers.  So this opens the door to pico projectors that can be small, rugged and project onto things like a piece of paper.

Diode Green Lasers (Part 1, Wavelength and Efficiency)

Today’s blog is the first in a series about green lasers, and in particular about how wavelength, power, efficiency, and lumens relate to each other.  Also, I’m also going to write a little about the difference between and R&D announcement and what needs to be known to build a projector.

I believe that direct green lasers are the key to making very small embeddable pico projectors regardless of the display technology be it LCOS, DLP, or Laser Beam Scanning (LBS).   Unfortunately the “physics” of green lasers makes them hard to produce. Read more

Welcome to Karl Guttag on Technology

Welcome to Karl Guttag on Technology (KGonTech for short), a blog about technology with an emphasis on technology for image generation (including microdisplays, pico projectors, and video/image processors) and human-computer interfaces.  A big part of this blog will cover pico projectors technology, but I also plan to discuss the technical creative process, high tech startups, market strategy, and the early days of home computers and video games.  I will rotate the subjects, but there will definitely be more about pico projectors including the technology behind them and their applications.  While the header of this blog is kind of busy (so I have been told), it is kind of a word association with topics and devices I have been associated with to give you some idea as to what I might be covering.

Lately, I’m best known for my work in pico projectors and LCOS microdisplays as the former Chief Technical Officer (CTO) and one of the founders of the Syndiant.  Prior to becoming CTO of Syndiant, I was an independent inventor and the CTO of another LCOS startup.  The largest part of my career was with Texas Instruments (TI) where I became the youngest TI Fellow in the history of TI (at age 33) for my work on Media Processing DSPs, Graphics Processors/Accelerators, Video interface , Synchronous DRAM, Video DRAM (VRAM), two 16-bit microprocessors, and the Video Display Processor (VDP, the first “Sprite Chip”).    So I have seen the electronics industry first hand from both a large company and a small startup company perspective in my 34 years in the industry.  I also have to date 142 U.S. patents so I am quite familiar with the inventing/patent process.  For more details on me see “About Karl Guttag.”

What I hope what will make this blog different is that I first hand technical experience in designing devices and understand the much of they underlying technology.  For example, there are a number of sites such a PicoProjector-Info (see their interviews with me on Jul 22, 2009, May 20, 2010, and Dec 5, 2010 that proved to be controversial) that cover the news on pico projectors, but with my experience I can interpret and explain what is going on and separate marketing hype from technical reality.    I will try and boil the technology down so that the average person will be able to read and understand it.  I like to say that a most of my career I had to translate between the engineers doing the “real work” and the management whom didn’t have time to understand the technical details.

Now for the elephant in the room; after 7 years as a founder, Director, and CTO of the startup Syndiant I have decided to leave (effective December 16th).   I want to make it clear that Syndiant had some great technical people and some great technology and one of the best team of technical people I have ever worked with.  I also firmly believe there will be a big volume market for pico projectors with LCOS.   Furthermore everything I have been able to gather, Syndiant is the technological leader in small high resolution display devices and as of right now it is the only company with a very small HD microdisplay that works (and produces some fine looking images).  So please don’t take my leaving Syndiant as any sign that pico projectors or LCOS is not going to happen, because I’m still a big believer in LCOS based pico projectors and particularly the future of laser illuminated LCOS pico projectors.

Read more