Startup Knupath offers world a new CPU architecture A company led by a former NASA boss wants to take tricky and compute-intensive algorithms off general-purpose silicon and has just popped out of stealth mode to show off its first efforts. The de-cloaking, which came after an uncharacteristically long period of ten years in “stealth mode”, revealed to the world a digital …
OK ... so now I can search for cat pictures in record time.
Like the race for thinner phones, there came a point where I said, screw the thinness, I want something that feels solid and fills my needs.
I'll admit, like many users I still haven't signed off on my actual needs yet... but I'm working on it.
Yeah... I'm going to get flamed to hell for this, aren't I.
Many problems in image analysis (and physics) can be cast into sparse matrix form, so I will certainly take a look, especially for the gigapixel and terapixel images we are getting. I also wonder whether many graph-based methods could benefit, especially if the average number of edges per vertex is low. Many of those algorithms may well be suitable for this sparse-matrix architecture.
I also wonder whether many graph-based methods could benefit, especially if the average number of edges per vertex is low.
Denser graph problems can often be recast into sparser ones within good constraints. See e.g. Fung et al., "A general framework for graph sparsification". So this chip could be broadly applicable to graph problems.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
