ChipCrunch.com

In general, publicity is good thing for a company – so it is nice to see XMOS Semiconductor finally emerging from stealth-mode and providing some information to the public on their company and what they claim to be a revolutionary new type of programmable semiconductor technology called Software-Defined Silicon (SDS). Now don’t get too excited, the web site is still relatively sparse on details, but a quick email to XMOS resulted in a nice presentation style response that provided more insight. Founded in 2005 and backed by Amadeus Capital Partners and Esprit Capital Partners, XMOS currently consist of a team of 25 people out of Bristol, England. The company is headed by James Foster, who has previously held a CEO position at Oxford Semiconductor and before that held various engineering positions at Lucent and Lattice Semiconductor. The CTO for the company is David May, who is well known for his invention of the Transputer and the OCCAM programming language

According to XMOS, SDS will provide designers with the advantages that are usually associated with System on a Chip (SOC) solutions while at the same time providing the flexibility that FPGAs offer. SDS is implemented as an array of XCore event driven multi-threaded processor elements which are linked via an XLink inter-core communication link, both of which can be easily programmed and configured through XC - an XMOS C-style proprietary language. The advantage of this approach is that the whole system can be specified using a high-level language and no more RTL coding is necessary. Further, the whole system can be reconfigured in about the same time that it takes to recompile the actual code. The chart below summarizes the advantages of SDS versus competing technologies:

Designing ASICs is very risky unless you have really high volumes. ASSPs are less costly, but since they are sold to many companies, differentiation becomes difficult. FPGAs have almost everything going for them - except that they are very silicon inefficient, which makes them expensive and utterly unusable for high-volume consumer products that have very low margins.  The fashion criterion displayed in the chart is utterly beyond me, but as you can see SDS is fashionable as well. On the performance front, the XCore is expected to yield a performance in the 500 MIPS range. Also, developers will be able to dynamically partition resources between control processing (MIPS), DSP processing (MSPS), and I/O processing (Mbps). The key to SDS success will be whether or not XMOS will be able to deliver a robust and intuitive programming environment; without this ability most developers will likely shy away since there is nothing more frustrating than dealing with buggy tools. More technical details of the architecture and tools are scheduled to be released in Q4 of 2007 and the launch of the first device family with the complete tools set and IP library is to follow in Q1 of 2008.

Startups are exciting, whether you are simply interested in the new technology that they might be developing or are looking to join one early and hope to cash out big through an initial public offering or through being acquired. Maybe you are just looking for ideas for your own startup and browsing through what other startups might be doing might just give you that needed spark that will put you on the right path. In any case, EETimes.com just updated their list of the top 60 semiconductor startups for 2007. The list is not exactly detailed about what each company does, but at least it provides a quick snapshot on each company and a web address. Not that the web address is of much use in some cases; for example, XMOS Semiconductor and Unity Semiconductor have a little more than welcome pages and that’s about it. However, this seems to be an exception as most startups are eager to tell you a lot about their upcoming products and it also seems that most of them are actively hiring. While many of the startups on this year’s list can be found on previous lists, there are several new ones, including: Ambric, Artimi, Athena Design, Blaze DFM, Boston Circuits, Codeplay, Mirics Semiconductor, Nanoradio, Nemerix, Nemoptic, Prime Sense, Takumi Technology, Tilera, Unity Semiconductor, XMOS Semiconductor, and Xoomsys. You can find the complete list of all the selected startups here.

Printing high-performance semiconductors on a wide range of substrates – just imagine the possibilities. Turns out, startup Semprius Inc. , which was spun out from the University of Illinois research labs but is now located in Durham, North Caroline, has taken major steps in making this a reality. As a matter of fact, the company’s progress has been so encouraging that quite some money has been flowing their way as of late. At the end of April, Semprius received $4.1 million in Series A funding from Arch Venture Partners and Intersouth Partners - both firms have quite a bit of experience in funding semiconductor startups. Additionally, just a few days ago the National Science Foundation awarded the company a grant to pursue research in processes and materials for the display industry. Key to Semprius’ success so far is the elegant method that the company is proposing for transferring transistors from a traditional substrate onto other surfaces such as glass or plastic. In short, the transistors are manufactured on a traditional and well understood substrate, leveraging conventional manufacturing techniques, thus eliminating any problems that would be introduced using non-traditional substrates. The innovation stems from the company’s ability to then peel off the finished transistor from the traditional substrate and place it onto another one. A more detailed illustration of the process can be found here. One sample application for this technology would be placing transistors directly onto an LCD back panel, which would lead to significantly faster pixel response times for the display. Other applications include: flexible displays, large area sensors, RF devices, and OLEDs. With all these interesting business opportunities looming on the horizon for the company, it will be interesting to watch whether or not Semprius will be successful at commercializing their technology.

Companies, such as Sun and HP, seem to have tried many times before to get rid of office PCs by replacing them with thin clients whose sole purpose was to connect to the datacenter that hosted all the necessary applications, and server them to employees. Well, most of the initial experiments in this area don’t seem to have worked out very well, given the fact that the majority of office cubicles still host a PC or its equivalent; however, a startup called Taradici out of Vancouver, Canada hopes to change this in the near future.  Other than the fact that the company name sounds more like an Italian restaurant than anything else, the technology that Teradici has developed seems rather interesting. The idea behind the company is that through the use of specialized image-processing algorithms, in combination with a custom silicon implementation, the above mentioned concept will finally be realized. The image below, taken from their website, gives a nice little illustration about how this is going to work:

Teradici claims several major advantages, including: patented display compression and propagation over networks, hardware bridging for peripherals and audio, wire-speed encryption of all data traffic, and complete operating system independence. Additionally, due to the proprietary image data handling, the theoretical distance between the portal and the host can be extended by several orders of magnitude. All the above is implemented through a custom chipset, implemented in 130nm,  that comprises the TERA1100 PC-over-IP Portal Processor and the TERA1200 PC-over-IP Host Processor. The TERA portal as shown in the image above, consumes only 15W of power, which is significantly less than a standard desktop. But what about companies that give their employees laptops instead of desktops? Laptops already consume significantly less power, and using VNC or similar software, a datacenter hosting applications can be reached without a problem. Will Teradici be able to convince employers that their TERA portal is a better solution? This leads to an interesting questions: could an even lower power version of the TERA1100 processor be developed to be integrated into notebooks? This would enable employees to work better remotely because of the reduced application lag that is commonly associated with VNC over VPN connections. Further, security could also be improved because of the integrated encryption, and all this while leaving the workforce nice and mobile at the same time.