Tensormesh turns GPU memory into software
AMD partnership cuts latency, doubles density.
AMD partnership cuts latency, doubles density.
320 billion transistors, 72-GPU rack scale—Scale up, scale out.
This is what a turnaround looks like.
Chips, memory, consoles all cost more
The emphasis turns to memory, power, and racks as interconnected parts of the puzzle
New modules target mainstream edge AI
SmartRoto, an AI-powered plug-in for Nuke, is 4× faster than using just Nuke.
Live rendering reshapes architectural client meetings.
If so, we'd like to hear from you.
Watch CG legend Don Greenberg’s livestreamed presentation on the big screen at Siggraph or off-site for non-attendees.
Gradual rollout expected across product lines.
Siggraph programs highlight connection among CG, physics, and AI.