Uni-Pixel: A Picture Is Worth A Thousand Words

| About: Uni-Pixel, Inc. (UNXL)

About two weeks ago, we wrote an article questioning why Kodak never filed an 8-K covering their manufacturing agreement with Uni-Pixel (NASDAQ:UNXL). At the time, we deduced that Kodak probably didn't think much of Uni-Pixel's technology. In this article, we will explore why we believe that is and we will focus on what we believe to be the single over-riding problem with UniBoss: product quality.

One of the most interesting things about UniBoss is that very few people have seen the product outside of the limited presentation environment the company controls. Most investors have just seen the demos on the company's website and a few have visited the company. Almost nobody has actually gotten to play with the demo system and the company has, for about a year, refused to give investors samples. We're among the few, fortunate investors to have handled the product and viewed it under a microscope. We also took close-up photographs of the product when the management team wasn't looking. We can say unequivocally that the product does not work. The lines are too visible. No consumer electronics company would approve this in a display for their product. If it works, it'd be relegated to a niche, low-end product where optical quality doesn't matter.

Here's a close up of it so you can see for yourself. The UniBoss sample is placed over a white sheet of paper. We suggest you click on each photo to enlarge it in order to best see the quality.

UniBoss White Background

You thought it'd be clear and invisible, right? The metal grid pattern is clearly visible.

UniBoss Close Up With Finger

Here's a photo we took with a finger touching the UniBoss so you could get a sense of the scale. The grid pattern is clearly visible.

Here's a photo of it covering an iPhone so you can see it with a backlight. The UniBoss sheet covers the entire display, and we pulled up a white background on the phone.

UniBoss Over Phone

Again, visible metal grid lines. Want to see a "working" demo? Here's a photo of Uni-Pixel's newest demo unit using UniBoss. This photo was taken only a few weeks ago and uses the same unit as the company uses in its presentations and YouTube videos on its website.

UniBoss Broken Demo

There is a major problem with the demo. The sample has wide broken channels, which you can see create large gaps in the trace. These are caused by dead spots from defects in the UniBoss grid. If Uni-Pixel can't get a demo to work correctly, how can investors possibly think they will be able to produce millions of units every month?

We believe that these photos alone cast enough doubt on the potential for UniBoss - both due to its low quality and the production problems with the demo unit - that Uni-Pixel is a compelling short. Other than with our eyes, what makes us think it doesn't work? We have done extensive research on every part of Uni-Pixel and UniBoss. All the evidence indicates to us that Uni-Pixel is a fraud and that UniBoss will fail.

Extensive due diligence proves UniBoss does not work

In addition to testing UniBoss ourselves, we have spoken with engineers at touch screen companies who have tested UniBoss. They cite fundamental problems related to its electrical properties and optical quality. They don't think that Uni-Pixel can produce UniBoss. Even if Uni-Pixel can make UniBoss, our sources told us that it won't be successful in the marketplace due to the visibility of the metal lines.

Our confidence is increased by conversations with former employees and industry consultants. Key employees at Uni-Pixel have told us the company has never been able to produce UniBoss and they are not confident that Uni-Pixel can ever successfully produce it.

All the while, management keeps spinning a story that investors are eating up. However, research into the background of the management team reveals Uni-Pixel is led by charlatans with a history of misleading and defrauding investors out of money.

We think the end game is approaching and Uni-Pixel will again fail to deliver meaningful revenues from UniBoss. The message is clear: short Uni-Pixel. Uni-Pixel is the most attractive short in the market today. With a current price of $19, we have a fair value target of below $3/share.

Engineers who have tested UniBoss say it is unacceptable

All the extensive research we've done on the UniBoss indicates it doesn't work. We spoke with four touch engineers who have tested and evaluated UniBoss. Their findings are consistent. They found either that the samples were unacceptable and didn't hold up under testing, or that UniBoss did not pass visual quality tests.

We consulted with an engineer at Touch International, a leading manufacturer of specialty touch screens, who tested the latest 6-micron version of UniBoss. He obtained and tested a sample in 1Q 2013. The engineer found numerous problems with UniBoss and believed the product needed substantial improvements.

First, he said the lines were visible and that consumers are unlikely to accept it. In his judgment, the lines needed to be significantly thinner. Then he said the lines created a Moiré effect that caused visual problems. Lastly, he said that the electrical properties of the grid changed under high temperature and high humidity testing. This is due to the underlying chemical properties of copper, which are difficult to overcome.

Basically, he said it failed his tests on multiple accounts, didn't work, and couldn't work. This isn't an older version of the product. This is the latest version of the product that Uni-Pixel is marketing today.

The engineer at Touch International is not alone in his findings. Another industry consultant, who has a Ph.D. in Chemistry and has been in the touch industry for over 20 years, cited numerous technical problems. The most problematic feature he found was a significant variation in resistance across one of the planes due to printing geometry or variations in metal deposition. Essentially, the shape and thickness of the lines were inconsistent. This resulted in huge variation in trace resistance and problems with non-linearity.

We asked him to explain it to us in English. He said that means when you type "A" on your screen, it comes out as "S." Not only that, but one part of your screen will come out as an "S" while another part will come out as a "Z."

In essence, the touch wasn't accurate, and the inaccuracy had wide variability.

To Uni-Pixel's credit, his tests did find something positive. The consultant said the Moiré effect was minimized in his sample, and that the product handled the Moiré effect well. However, the consultant said the lines were too visible over certain background colors. We asked which colors. He responded, "Black, grey, blue, green, red, and white." That's almost every color in the rainbow.

Furthermore, he said while there were techniques Uni-Pixel and touch screen manufacturers could utilize to minimize the visibility of the lines, the lines would always be visible at the end.

We cut to the chase and asked him if what he found in the tests posed a problem. The consultant said it would not be a problem as long as people didn't care that the lines were clearly visible on the screen under most colors and didn't mind that the touch functionality was inconsistent across the touch plane.

Good luck with that, Uni-Pixel.

Another engineer at Synaptics recently tested a 6 micron sample of UniBoss and also noted that the lines were clearly visible. He has no axe to grind since his company is agnostic as to who they work with. Synaptics makes controllers so they want to enable all technologies that enable touch. In his judgment, while the visibility of the lines wasn't a deal breaker, he thought UniBoss would be relegated to the lower end of the market due to its optical quality. The Synaptics engineer also mentioned that UniBoss fell into a group of competing technologies and that Uni-Pixel hadn't differentiated themselves in his mind.

Lastly, we had a conversation with an engineer at Texas Instruments who works in their touch controller department. We spoke with him this past January. Unlike the others, he had not tested the 6 micron sample but had tested older versions such as a 20 micron and an 8 micron sample of UniBoss. The engineer told us Uni-Pixel had promised a newer 6 micron sample but had been unable to provide one. This surprised us since Uni-Pixel announced a partnership with Texas Instruments back in February 2012 and touts this partnership to investors. Nevertheless, he shared with us his evaluation of the samples. He said the lines are much too thick and too visible for use in a consumer device, that the samples had issues with reflection, and that the grid patterns were more visible under certain colors. Furthermore, the engineer wasn't even sure how Uni-Pixel made the samples since they did not have the full process in place, and he was very skeptical they could produce it in scale.

We asked him when he thought he was going to get a 6 micron sample of UniBoss. He told us he did not know and said it would take at least a few months. That shocked us given the strides Uni-Pixel claims they've made and the development partnership that Uni-Pixel has with Texas Instruments. Like the Synaptics engineer, he added that Uni-Pixel's technology is just one of the many ITO-alternatives Texas Instruments is testing and that Uni-Pixel isn't in the lead.

Perhaps the most shocking thing we learned happened when we called the engineer at Texas Instruments back earlier this month. He told us that Texas Instruments has decided to discontinue making touch controllers and that the partnership with Uni-Pixel was therefore effectively over. That made us wonder why Uni-Pixel never told investors about this.

That's four industry engineers who have directly tested the latest UniBoss product in the last four months. The message from all of them is consistent; UniBoss does not work and will not be accepted by the market for a variety of reasons.

Even former employees say Uni-Pixel cannot produce UniBoss

Along with engineers who have tested UniBoss, we spoke with two former Uni-Pixel employees. Both had been part of Uni-Pixel's sales efforts to push UniBoss adoption in the touch screen industry. They told us that Uni-Pixel had yet to prove the UniBoss technology by the time they had left the company and expressed significant skepticism about Uni-Pixel's ability to successfully produce UniBoss.

One former employee even said the demo sample Uni-Pixel trotted around to investors and customers was actually manufactured using the subtractive, photolithographic. He said Uni-Pixel was unable to produce UniBoss using the additive, direct printing process. Nevertheless, Uni-Pixel used the photolithographic sample to show that the metal mesh concept could work, but did not inform people that Uni-Pixel had not yet found a way to manufacture it. It was all smoke and mirrors.

Another former employee also told us Uni-Pixel was unable to properly produce UniBoss. Uni-Pixel enticed potential customers by producing small batches of samples in a lab and by promising significant improvements in short order. However, Uni-Pixel was unable to progress from lab to greater scale production, and he lost contracts with customers as a result. He cited two examples of failures with the 10 micron version of UniBoss. In each instance, Uni-Pixel was able to win initial contracts due to cherry-picked samples and big promises of advancements, but Uni-Pixel would ultimately fail when it came time to produce UniBoss in material volumes and in a real-life setting outside of the lab. He said such instances were very embarrassing and led to his decision to leave the company.

We asked him if there was a chance the company had fixed its consistency problems and made progress with UniBoss since he left in mid-2012. He said he didn't think the company had made much progress since he left. We asked him if he thought Uni-Pixel could ever manufacture UniBoss. He cautioned that Uni-Pixel lacked manufacturing DNA, pointing out that, if you look up and down the employees, it's all sales people and early stage scientists with no manufacturing experience and record of successful product commercialization.

But Uni-Pixel has all these presentations with data and pictures, right? These presentations have figures on UniBoss performance, show the progression they've made over time, and have close ups of the product. This shows UniBoss works, right?

One industry expert said that he had significant misgivings on trusting anything on Uni-Pixel's presentations. He has a Ph.D. in Chemistry and specializes in thin film and coatings and started his career at Kodak. He examined their investor presentation and noted the pictures they show on the slides are not consistent. The shape of the lines and the angles of the grid are different in each slide. That signifies that the company cannot consistently manufacture the products they present to investors. In order to show their product, he said it is clear to him that they had to pick the best samples, and even then, the company is having difficulty consistently manufacturing the samples. He said as a scientist, he would be uncomfortable presenting Uni-Pixel's data and specifications. To him, it was scientifically misleading, akin to a biotech company data-mining to show the best treatment results of a small sub-group of patients that cannot be replicated in a larger trial.

All of our research leads us to believe that Uni-Pixel cannot produce UniBoss. We find Uni-Pixel's disclosure of delivery and production delays to support our view. Just in case this isn't convincing enough, we'll be back in a few weeks with an expose on the shady characters all around the company - in the management team, the board, and among the owners.

Disclosure: I am short UNXL. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.

About this article:

Tagged: , , , Diversified Electronics
Want to share your opinion on this article? Add a comment.
Disagree with this article? .
To report a factual error in this article, click here