4 ms·
I feel people have underestimated how potentially impactful the AMD/XLNX merger can be. AMD now has a complete stack to compete in the DCG. They have bespoke a
by kd913 5y ago
I feel people have underestimated how potentially impactful the AMD/XLNX merger can be.
AMD now has a complete stack to compete in the DCG. They have bespoke accelerators, storage, FPGA, CPU (x86, ARM), (lowest latency) smartNICs and GPU.
They now are a player in the 5g communications market.
Automotive they are already competing with multiple deals with Tesla.
They can now compete stronger on the ML front with their integration with Xilinx.
You name your niche, AMD can build it for you.
The accelerators are a future bet to eventual physical limitations with Moore's law.
- TomVDB 5y agoHow did Intel leverage the acquisition of Altera? And why would it be different for AMD?
- kd913 5y agoThe market and tech is in a completely different position to what it was back then. Xilinx is well ahead of Altera. Xilinx already is extremely profitable in it's current markets already. Xilinx and AMD provide extreme supply overlap, and greater bargaining position with TSMC. AMD can leverage the IP much better within it's own products and vice versa. The times of easy performance wins are gone. Accelerators are the future now. Market is much better suited, software toolkits are better.
- pclmulqdq 5y agoXilinx got ahead because of the intel-altera merger. Stratix V was better than the Xilinx 7 series. Now they are giving Intel the opportunity to catch up. My guess is that AMD bought Xilinx either for fab capacity or for some underlying tech expertise that they needed.
- JonChesterfield 5y agoIntel appears to have squandered it. Strangled in bureaucracy at a guess. Not much sign of the fabled FPGA/x64 hybrid from Intel. AMD + ATI didn't look totally smooth from the outside but the graphics cards are still with us and I'm using an APU as my desktop.