I talked about the promising startup picoChip some time back here.

Let me summarize the progress they have made in recent times since then:

• picoChip has partnered with L&T Infotech in India to design WiMAX and advanced DSP systems.  This $5 billion venture has tremendous potential, given the depth of L&T’s technical expertise.
• Berkeley Design Technology (BDTI) recently published results of its study of various processing benchmarks, which give high marks to the picoChip PC102 multicore signal processing device.  Running at 160MHz, its 308 processors showed a 40-fold advantage in price-performance over traditional DSP processor solutions in key communication benchmarks.
• picoChip closed another round of funding, this time led by Samsung Ventures America (estd. 1999).  While the exact amount that Samsung has invested is not known, it represents the strong confidence of the Korean giant in the technology that picoChip is pushing.
• picoChip joined the Femtocell Forum.  Femtocells are cellular access points that connect to a mobile operator’s network using residential DSL or cable broadband connections.  A recent study from ABI Research forecasts that by 2011 there will be 102 million users of femtocell products on 32 million access points worldwide.

Of course, there are plenty of challenges facing picoChip.  Fundamental is the ease of adoption.  Traditional DSP companies have always offered multicore DSP which have managed some degree of success when it was easy to partition the applications to use the multicores.  The new multicore DSP startups face the same problem.  They will have to work hard to make the software development easy enough to partition the problem into parallel streams for using the potential of the multicores.  picoChip has realized this and hopes its picoTools will deliver in the long run.