Altera has been known but its so sad. They bought them less than a decade ago for $13B and did nothing with it. They really got themselves into a great position for Heterogeneous computing and then just did nothing.
It's depressing. This should be an age of new hardware, with near-memory computing and flexible architecture. Instead it's GPUs for the foreseeable future, Nvidia has a near monopoly, AMD seems uninterested (despite also owning Xilinx), and Intel's throwing in the towel.
I wanted to prototype a systolic array coupled with a DRAM slice to hold the weight matrix, but there's not even DRAM blocks distributed on any of the FPGAs I saw. I guess Xilinx has HBM but still not quite what I wanted.
They tried a lot of things with Altera and failed. The real problem is the price they paid (it was more than $13B). At the time they bought it, Altera had net revenues of $0.5B. They simply paid too much to ever recoup the expense.
