Thursday, December 13, 2012

Optical Interconnects

Some years ago, Gene Amdahl was mounting a second startup. His first, Amdahl Computer was reasonably successful and subsequently sold to Fujitsu. The second startup was called Trilogy, it was an effort to build macro-scale processors that occupied an entire silicon wafer rather than just a chip. Though Trilogy came and went, it was a big deal at the time. The reason why the company is now just a historical footnote is that they ran into an insurmountable technology issue. They had no way to package the wafer. Specifically there was no way to get signal from one side of the wafer to the other. At the time, optical communications technology was still in its infancy. Trilogy approached Corning about developing an optical package but, of course, it was well beyond the capability of the optical technology as well.

Over a decade later, the issue came up again. As processor chips became denser, not only clock skew but interconnect densities started to become a problem, there was just not enough linear space at the edge to keep up with larger and more complex chips. Again, the obvious solution was to move to optical I/O. After another decade plus period of development, IBM has announced that the technology is ready to go. What this means is a few things. First, Moore’s Law gets another one of those needed breakthrough’s to keep it on-track. Second, with chips communicating optically, it is likely that you will now star seeing optical cabling inside computing devices rather than just a T3 fiber line leading to the building where they are housed. Third, rather than fiber reaching into the home (FTTH) or premises, you might see optical pathways directly from the device reaching out to the network.

In the optical networking world, there is a concept called “transparency” keeping the signal optical for a long as possible. Before the development of optical amplifiers, optical signals had to be periodically converted to electronics and regenerated as an optical signal. Given the processing and rise time of the electronics, it was much like flying a plane across the country but have to land every few miles and change planes. Transparency meant that the plane no longer had to land until it was near its destination. Optical I/O, optical output from computing devices may mean that the optical signal will eventually no longer have to drive to the local airport to catch its plane but can take off from right inside your PC. The results will be data rates that make today’s connections look like dial up. This scenario is probably another decade or so away, but an important threshold has been crossed.

This may also have some tendency to roll back the current "mobility boom" in favor of wired connections. Although wired connections tend to have inherently more bandwith than wireless, the wireless industry has been quite clever regarding signal processing and compressing more and more data into the spectrum available. Transparency right up to the processor may be more than can be matched by increased data compression presuming there is some application that could make use of it.

Tuesday, December 11, 2012

More is Better II (more of the same)

An old joke…. A man sees another man standing under a streetlight staring forlornly at the ground. The first man asks, “What’s wrong”? The second responds, “I dropped my car keys and can’t find them.” The first man looks at the pavement, clean and uncluttered, no sign of the keys. He asks the second man, “Are you sure you dropped them here?” The second man responds, “No I dropped them over there”, pointing off in the darkness. The first man, confused, asks, “Then why are you looking here?” The second man responds, “The light is better here.” (Image from Monk Wisdom.)

The joke is an illumination of the sometimes absurdity of looking for answers in a place of convenience rather than where you are most likely to find them. I once worked on a product, an optical film that had severe manufacturing issues. Unfortunately the work group was arranged so that all of the chemists were on the east coast where the factory was located and all of the optics expertise was on the west coast. The factory continued to look for a chemical solution to what proved to be an optical issue until the business was closed, a possible optical solution waiting, untried.

The LCD and TV industries do not have fatal issues keeping the product from the market, but the industries do suffer from a lack of innovation. Tim Cook’s recent pronouncement, “When I go into my living room and turn on the TV, I feel like I have gone backwards in time by 20 to 30 years” refers to a product that has some considerable staleness. If you follow the industry, there has been no lack of innovation, but much of it has been as effective as looking under a random streetlight instead of looking at the issues. Indeed, Tim Cook’s own company is guilty of this. In a previous posting, “More is Better”, I refer to the continued growth in pixel count in mobile devices when they obviously have severe sunlight viewability problems. Although a shortage of pixels was a real impediment to running complex apps (Graff Spee) for mobile devices, that seems to be a solved problem as pixel densities exceed human eye resolution.

In TV there have been even more innovations: ever improved image processing algorithms, “Connected TV,” “Smart TV”, 3D, and as if the TV also suffered from a pixel deficit 4K. Raymond Soneira deals with the efficacy of the 4K set quite succinctly in “Your existing HDTV is already a true 'Retina Display' ” . 3D seems to have arrived ahead of its time. “Connected TV” and “Smart TV” seem to be getting some yawns from Apple that, if Mr. Cook is to be taken seriously, has a mind to upend those developments. As to the algorithms, every digital manipulation of the image leaves its own set of digital artifacts. Particularly, as now when people are watching both digital and analog content, the digital manipulation of faces can leave a flat and cartoonish image with the wrong content source. Further the delay caused by the image manipulation can leave the picture out of synch with the sound if the sound source is not the TV itself. Interestingly, on higher end sets, they frequently have a gaming mode where you can turn off the image manipulation to keep the delay from disrupting your game. In many cases, video content looks better in gaming mode as well.

If you look historically at what innovations in LCD and TV have and have not made a difference, there is something of a pattern. In CRT real improvements were made with improved design: Trinitron and Black matrix, improved materials: invar shadow masks, black (36% transmission glass), new sources: cable and the vcr, and an improved usage model: home theater. There were also, of course more minor tweaks in design and materials as well as such things as Picture in Picture (PIP) the came and went. In LCD there was improved design: IPS and overdrive, improved materials: BEF, LED backlights and the optical rubbing that enabled multi domain, improved usage models: touch and going to wide screen, and any number of other innovations that have yet to leave a substantial permanent mark. Although there are occasionally electronic manipulations that prove to be indispensable such as overdrive, the big changes are frequently the introduction of new materials that materially change the optics of the device.

The TV may be in for a round of innovation as Apple threatens to fundamentally re-think the TV usage model. If Apple follows true to form, they will certainly push the technologies but will not hinge their new product on fundamentally new technology or more pedestrian approaches such as improved software. In Digital Signage, there is substantial room for innovation as well. Though it is still common for new implementers of Digital Signage to use TVs, the design requirements are different. Indeed, even for purpose built Digital signage screens, many of the design trade-offs embodied in current LCD production are optimize around the TV application but may not fit well with the varying environments and applications of Digital Signage. A re-thinking of digital signage design may be in order and some of that re-thinking may find its way into the general TV market as well. Much of this may be connected with the optics and materials used in the display rather than just the electronics.

Of the more prominent hiccups with the iPhone 5, two of the three (the choice of the sapphire lens cover and the anodized aluminum case) involved some new materials experimentation. The original choice of Gorilla Glass, is a materials experiment that went well; it was helped by the glass expertise of those familiar with that materials system. The challenges of mobile devices are different from the challenges digital signage but the idea of looking for a solution in the materials may be common. This does not mean waiting for IGZO or large size OLED technology to come down in price.

Monday, December 10, 2012

The Antique in the Livingroom

Tim Cook, “When I go into my living room and turn on the TV, I feel like I have gone backwards in time by 20 to 30 years."

Isn’t it interesting that while a phone is principally an audio device, it is virtually impossible to find a cell phone without a built in camera; whilst a TV is a video device but does not have a built in camera? As an audio device, the phone has become a primary tool for selling music. As a mobile internet connection, it has other uses but I think most of the revenue is in music sales. Although there are other media, the TV is the primary tool not only for selling video but everything else. Much of the rumor about why the rumored Apple TV set has yet to make an appearance surrounds resistance from the video content owners from doing a deal on Apple’s terms.

When Apple’s CEO, Mr. Cook, refereed to stepping back in time 20-30 years when he watches TV, he probably did not mean the screen (there is much speculation that although an Apple TV set will be state of the art, it will state of the art for current screen technology). Mr. Cook could have been referring to the lack of interactivity. But he may have been referring to the entire usage model including the content. Not just the videos, but the way the TV is used as a marketing platform. Much of apple’s business in connection with the iPhone revolves around purchased content. The ad-supported content market is much bigger. Certainly, the counter to a new Apple TV set oriented around purchased content would be one that could more generally serve as an advertising platform as the current TV set does but with a more net centric marketing approach. Although Apple may launch a curve ball or two in the hardware of an apple branded TV set such as a new aspect ratio, the big surprise may not be in the hardware or the software for that matter but in the content generation, a new way for marketers to pitch their product interactively.

Thursday, December 6, 2012


The headline is, “Mitsubishi Drops DLP Displays: Goodbye RPTVs Forever”. Like 3D, projection TV got off to a bad start. The first widely available projection TV was the Kloss Videobeam (Henry Kloss was the K in KLH). It was based on a single color CRT (driven very hard) and projected on a specialty curved screen. The product had a number of difficulties. Being front projection, it had to be viewed in the dark. Even so it was dim. The curved screen had multiple problems as well. It could not be cleaned with conventional cleaners without leaving a lasting residual in the image on the screen… and it was added expense.

One by one, these problems all were fixed; but not by Kloss as they were immediately displaced from the market by the Japanese with a 3 tube projector design. Due to the loss of energy and light caused by the shadow mask in a color tube, a monochrome tube can be 5 times as bright as a color tube. Having three monochrome tubes (a red, green and blue) in a projection set made for an image that was 15 times brighter. Improved optics did away with the curved screen and switching from a front projection to a rear projection format meant that it could be viewed in a lit room; RPTV was born.

These improvements eliminated the initial problems with projection TV but created new ones as well. Early RPTVs were difficult to align and the sets in consumers’ homes were frequently not aligned. Also the lenticular front screens did a great job of spreading out the light from left to right but had little effect distributing light vertically. Many projection sets were used as public displays in bars and placed up high where the primary light from the display went over the heads of the viewers. The result was in your neighbor’s house and in public, when you saw an RPTV, it usually looked terrible. One other aspect remained as well, driving a CRT very hard meant that it necessarily had a short life span, significantly shorter than consumers were used to.

By the mid 1990’s, after 20 years of fine tuning, projection TVs had gotten to the point where consumers did have a hard time telling them from direct view CRTs. In consumer surveys taken at the time, many consumers reported buying 45” and 55” direct view sets even though direct view was never sold in these sizes; they were obviously projection sets. In addition to improving the screen and making alignment much easier, RPTV makers had incorporated clever folded optical paths, decreasing the RPT set thicknesses from feet to inches. Further, as HDTV with the transition to 16:9 was in the offing, RPTV looked forward to a bright future as CRTs were fundamentally ill-suited for the wider format.

Also beginning in the 90’s, new RPTV image engines were being developed. Most famously, TI developed the DLP, a micro-mirror technology using an external light source. Also Liquid Crystal on Silicon (LCOS) showed promise. The first DLP sets were expensive, and in-part to hold down costs, it was designed as a field sequential color set with a color wheel, as were the first color TV set designs from CBS. Of course, introducing a mechanical component into what had been a purely solid state device meant that these sets had maintenance issues. As with the original Kloss Videobeam, these single imager designs were rapidly dropped for 3 imagers. However even the improved design had maintenance issues as well as the high performance light source tended to have a life of only 2-3 years. This meant that the set still had to be actively maintained. Though makers took to packaging an extra bulb in the sets that they shipped, Consumers were frequently not aware of this and had to spend $200 to replace their bulbs the first time they burned out. Even so, replacing the bulb could be tricky as oils from your hand could cause early failure of the bulbs and even when the first bulb was used and installed correctly, the owner was still faced with making significant purchase every few years to maintain their set.

In spite of all of these issues, RPTV sales grew as they were actually a great value compared with large sized Plasma at the time. However, as Plasma got cheaper and LCDs grew in size and capability, RPTVs value story diminished. The loss of value was exacerbated by, in my opinion, a failure to tell a good marketing story to consumers. TI made some direct efforts to promote DLP but their commercials were ineffective and lacked a clear message to the consumer about why RPTV sets were better for them. Even today, with the HDTV transition having well passed, the average size LCD set being sold is only about 38”, much smaller than what is needed to fully take advantage of HDTV resolution. There have been significant pushes by the set industry as a whole on each new feature, frequently this in opposition to getting the consumer the proper sized set. Though both LCD and Plasma offer very thin sets, thinner than what can be made in RPTV, most American homes are not at a loss for space. American consumers really need to have larger sets and telling them this should have been the mission of the RPTV industry.

Mitsubishi positioned itself as “the large screen TV Company” and was the last maker of RPTV. However, lacking the marketshare and marketing dollars to make an impact, it was only a matter of time. The RPTV story has several lessons for other technologies as well. Being introduced with obvious visual problems, it took a couple of decades for the technology to shake off its initial bad impression. Plasma still suffers from a perception of image burn-in. Some of the initial 3D sets had bad flicker issues.

Most RPTV sales were in the US; in the beginning, virtually all RPTV sales were in the US. Though I sometimes criticize pyrrhic marketshare battles, there is significant value to a technology having a global appeal. Though the US is unique in being both wealthy and having large homes, the public information display/digital signage market is global. Significant numbers of RPTVs were bought here as public information displays, maybe as much as half in its early years. The set makers, in general never recognized this as a significant market until recently but the market has always been there. If RPTV had built some momentum in the early years it could have been doing fine in digital signage as well. But of course, in the US with the ADA requirement of nothing sticking out from the wall by more than 4” any RPTV installation has to be either high up or floor standing both of which are not particularly well suited to a retail environment.

Wednesday, December 5, 2012

How RCA Lost the LCD

From the IEEE Spectrum article, "Today, companies in Japan, South Korea, and Taiwan dominate the LCD industry. Meanwhile, the corporation that started it all has faded from memory, purchased by General Electric in 1986. Nevertheless, RCA’s technological legacy can be seen in every LCD wristwatch, calculator, laptop, and television. All of these screens trace their origins to that firm’s laboratories and factories. As much as they are portals to the digital future, liquid crystal displays are also reminders of a past filled with possibilities for the once-dominant American electronics industry. And in their story are lessons for any technology company willing to learn them."

Masahiro Kadomatsu, the former chairman of Asahi Glass used to remark about how US scientists could be so inventive while US management could be so awful, particularly being short sighted. At one point in time, Kodak was the world's sole source for liquid crystal, Corning's sunglases plant in Harrodsburg, KY was the sole source of LCD substrate glass. RCA developed the liquid crystal chemistry as well as switching mechanisms such as IPS and Westinghouse, invented the active matrix driving technique. Of these 4, only Corning retains a place in the LCD manufacturing supply chain. For the others, particularly for RCA, bad results from management decisions overwhelmed great technology coming out of their labs. There was a period of at least 50 years when every new major consumer technology went through a period where it only existed at RCA's Sarnoff Labs, now Sarnoff is no more. For a more complete history of LCD development see "Liquid Gold: The story of Liquid Crystal Displays and the Development of an Industry", there is an Amazon link to the book in the Recommended section on the sidebar. The book was written by Joe Castellano, a former RCA researcher and the inventor of IPS.