Reimagine the Future.

Introducing a flexible, power-efficient microprocessor architecture that makes system design and implementation painless. This architecture is flexible, highly scalable and ready to transform products in any industry. Whether in IoT, Robotics, Mobile or Data Center, replace FPGAs and GPUs with a solution that is easier to implement.


robtics IoT-GPU-DataCenters-Processor-data-flow

Transformative Computing from Robotics to Data Centers

Improve machine learning: optimizing speed, performance in multi-axis applications. On-the-fly control loops adapt to control structural stiffness and accuracy. Cycle-by-cycle motor control can extend motor torque, allowing smaller motors to be used, lowering effective inertia and raising performance.

Contact Us to Learn More
image
image
image
image
image
image
image
image

PROBLEM: INEFFICIENT BATCH PROCESSORS

GPU, FPGA, CPU and ASIC Processors all do a couple great things but can’t replace each other so they are hybrided together.



WHY?
Because I/O performance is bottlenecked by oversized centralized cores that try to batch process everything together. This mismatches the processing architecture for most algorithms and slows everyting down.

"As an example, for the machine learning training applications that Bolsens has seen in the field, somewhere around about 50 percent of the time is spent on moving the data around for your applications across clustered compute, and 50 percent or less of the time is spent on actually doing the compute itself."



SOLUTION: DECENTRALIZED DISTRIBUTED INTELLIGENCE

What's needed is a processor that can dynamically reconfigure the architecture in one clock cycle and have an individual core for every input to end the batch processing nightmare.

Real-Time Control like a racecar upshiting on the straightaway for speed, then downshifting into the turn to dynamically reconfigure the power to the track.

 


"We are currently stuck with architecture developed some 40 years ago that are not going to satisfy humanity's insatiable thirst for a lot more processing, much faster processing and much more energy-efficient processing.

Frankly, the only thing left to improve is the compute side of the equation."