[gsmecher]

> everything comes down to datapath.

Ain't that the truth.

However: Xilinx's DSP58 blocks (Versal devices), and older Intel DSPs, do integrate floating point into the DSP tiles - which does narrow the gap between the datasheet's DSP count and the number of floating-point operations achievable per clock.

[mathisfun123]

> Xilinx's DSP58 blocks (Versal devices)

this is correct (i am currently working on designs that target these parts) but those things are more expensive than comparably sized/equipped nvidia gpus so (in the context of this discussion) it's moot.

[rzzzt]

Ballas mentions in a neighboring reply that the eGPU from the paper is using an Agilex 7: https://ark.intel.com/content/www/us/en/ark/products/217645/...

[mathisfun123]

Yes but as you can see by reading, this comment chain starts with someone linking to a xilinx product