What Cartilage Is and Is Not
April 27, 2022

Cartilage concentrates on one consequential goal: replace the Turing-von Neumann programming bottleneck with a scalable machine for reactive, distributed computation. Ordinary silicon, local reconfiguration, and proven SerDes links supply the physical foundation.

The architecture moves operating-system responsibilities into hardware, removes call stacks and threads from the application model, and lets programmers reason directly about objects, state, ports, and dataflow.

The Final Chip Unifies General Computation

A network-on-chip links locally reconfigurable regions into one expandable fabric. The Cartilage runtime manages placement, routes, ownership, and object creation so application structure remains readable as hardware grows.

Specialized GPU, TPU, neuromorphic, and signal-processing blocks can attach where they add value. Cartilage carries the broad body of software that otherwise sprawls across processors, operating systems, and orchestration layers.

Build the Post-Processor Platform

The same machine opens a path toward programmable matter, local silicon production, and new physical computing systems. Its immediate value starts with a new chip architecture and the language that makes it usable.

Engineers, investors, researchers, and product partners can join the work before the platform reaches mass deployment. Bring a hard scaling problem, help build the runtime, or fund the path to silicon.

Run the Cartilage Definition as Runtime Architecture

Cartilage Runtime-Instantiation Architecture carries this system into nested modules that enter and connect inside a live spatial machine.

Cartilage Core · public source repository · Cartilage Runtime-Instantiation Architecture · Cartilage Visual Language · Why This Instrument Exists · https://youtube.com/watch?v=H8eP99neOVs&feature=share

Originally posted on LinkedIn

Brian Greenforest · (2022-04-27 11:19:00 UTC)

Open the original LinkedIn post · LinkedIn activity 6925040601075785728

LinkedIn status when archived: Visible to Brian’s connections.

#focus and say "no". At what I'm working on, there's no place yet to quantum computers, exciting photonics, and even deep reinforcement learning. Although I study these a lot, I do not consider them as major part of what this project is all about. Cartilage is made of ordinary chips. Ordinary old-school cheap and available silicon. Cartilage uses battle-proven SerDeses that are good enough to implement a massively scalable NoC, but that's it. There's no additional bleeding-edge innovation behind what we have. We have a working massively scalable BREAKTHROUGH ambitious new chip architecture, that does just that: it innovates upon the Turing/von Neumann processor design, and solves ALL problems that design has introduced at the first place. We get rid of the call stack and get rid of multithreading. And we run the entire OS on the hardware level. So you reason about your reactive code the first time in the history of computer science in a non-messy way! It's definitely coming. No matter us, or someone will replicate these ideas, or even steal them. Because these ideas are not anymore abstract, and we're seeking when this freaking CHIPS SHORTAGE ends that we can start giving value to our customers. We call it the Final Chip, like in Final Fantasy. Because with such chip, you won't need to buy MANY OTHER chips. There will be no such thing as "chips shortage" anymore. You can integrate any special-purpose built GPU, TPU, neuromorphic, and whatnot--to crank numbers in an accelerated manner. But for massive volume of code mess that you are tired paying SALARY TO YOUR ENGINEERS, our chips will feel like Messiah. Because these chips will naturally be. And I'm not saying it as a selfish self-promoting marketing BS to convince you to pay attention. I want you to NOTICE this amazing stuff BEFORE everyone else will accept it as a new "Thermonuclear Bomb Working." Do you want to join the XXI century's "Manhattan Project" and participate EARLY in making THE BOMB, or do you want to see us "blow up" the whole world later? Because sooner or later, us, or somebody else, definitely WILL. Because I have proven that the new model of computation is REAL. Come work with me. We have only real working stuff. With future plans and ideals of gradual focus on one thing at a time. Including democratizing silicon and PCB manufacturing. Including programmable matter. But new silicon chips running new model of computation with a new programming language, how simple is that? Because next step will be to monetize it. To PUSH IT through "we're fine". Everyone is fine. We all are busy. Hard to notice game changers. Too much inertia. Global warming. Innovation. Paradigm shifts.... I wish you happiness, prosperity, luck, FUN! Come work with me and let's make it funded--together. I won't offer this again when SALES START.

Original LinkedIn media

“Focusing is about saying no” — Steve Jobs video thumbnail
Link preview thumbnail shown with the original LinkedIn post. Open the saved full-resolution image.

Media source note: Current thumbnail from the video URL recorded in Brian Greenforest’s LinkedIn export.

Try Cartilage: Run the open one-slot multiplier.

To edit, load, save, share, and follow the browser, Raspberry Pi, and FPGA streaming work, request the private demo username and password on LinkedIn.