Now Playing
Maia 300's Efficiency Bet, RTX 5090 Halt & the $1 Trillion Chip Milestone
Audio is available on Spreaker — see link below.
Microsoft's Maia 300 claims a 40% tokens-per-megawatt gain as Nvidia halts RTX 5090 production to feed data centres — and the semiconductor industry just crossed $1 trillion four years early. Everything investors and engineers need to know about the week's biggest AI hardware moves.
Audio is available on Spreaker — see link below.
Microsoft just reframed how the inference competition gets measured. The company announced Maia 300, its next-generation AI accelerator, built on TSMC's N2P process with eight stacks of HBM4 memory delivering two hundred and eighty-eight gigabytes of capacity.
Here's the important distinction. A chip win means you built something technically impressive.
Elsewhere, NVIDIA confirmed it has halted RTX 5090 production. The GB202 die is being redirected to data center and professional cards where margins are higher.
The semiconductor industry crossed one trillion dollars in annual revenue, four years ahead of the twenty thirty forecast that analysts were using as a reference point. TSMC posted a record forty-six point seven billion dollars in Q3 revenue, up fifty percent year on year, its sixth consecutive quarterly record.
On the IP side, a jury found that Arm violated its architecture license agreement with Qualcomm, failing commitments and deliberately disrupting Qualcomm's business. The ruling reinforces Qualcomm's design independence and shifts the balance between IP licensors and custom chip makers.
The two things worth tracking closely from here are both tied to the same constraint. TSMC's full earnings call on Wednesday October fifteenth will give clearer signals on N2P capacity allocation and whether AI chip demand is genuinely outstripping supply or being managed through pricing.
Chapter summary auto-generated from the verified script. Listen to the full episode for the complete content.