3 ms·
It is interesting that they would make this into an ASIC, provided how notoriously high the development costs for ASICs are. Are those costs coming down? If so
by hristov 10y ago
It is interesting that they would make this into an ASIC, provided how notoriously high the development costs for ASICs are. Are those costs coming down? If so life will get very hard for the FPGA makers of the world soon.
It would be interesting to see what the economics of this project are. I.e., what are the development costs and costs per chip. Of course it is very doubtful I will ever get to see the economics of this project, it would be interesting.
- zhemao 10y agoIt's mainly a question of volume. ASICs have a big economy of scale, so the cost-per-chip goes down considerably once you go over a certain number of chips. Plus, there's all the NRE costs of an ASIC design over an FPGA design. Google probably figured they could use enough chips to make the cost of manufacturing ASICs worthwhile. I don't think FPGAs are going to be beat out by ASICs for low volume applications anytime soon.
- indolering 10y agoBut even a custom run of silicon (on yesterday's technology) will only set you back $30K. That's one of the reasons there is so much interest in RISC-V.
- visarga 10y agoI think the initial investment was cheaper than the projected GPU energy cost and if it is actually faster, then there are advantages for faster iteration.
- TorKlingberg 10y agoMaking a custom ASIC is expensive for a startup, but not that expensive. Many companies are making their own high-end chips for things like signal processing.