They want their *own* CUDA situation. if you're doing FPGA stuff, the Xilinx hardware is good and Vivaldi is really the only way to use it. AFAIK the open-source fpga situation is pretty far behind it (please correct me if I'm wrong, I work in a place that uses Xilinx FPGAs)
What better way to do that than decrease availability, I ask, both rhetorically and sarcastically! CUDA-proper did well, at least partially, because it was put in front of everyone. This is taking an exclusivity angle that doesn't make much sense for ecosystem development, IMO.
I suspect this goes to show how much influence/priority B2B carries, my point is it's a mistake they've made before. The maintenance/support costs I believe they're trying to remove have paid dividends for CUDA and countless others. Anyway, a play on the tagline before I go: "Together we advance... licensing and removal of [positive] feedback loops over product development"
I seriously think the CUDA situation was set up intentionally.
Jensen Huang (Nvidia CEO) is related to Lisa Su (AMD CEO).
A decade ago, I saw a demo of some AMD skunkworks GPU datacenter tech, that could execute CUDA natively on AMD/ATI graphics cards. Initially was half speed, but having the flexibility was crazy amazing. Created a big buzz in the big iron and educational markets.
Where'd it go? Buried. You cant even find articles about it. Its a few comments on edtech datacenters.
Now look at AMD's graphics line. Where's their ROCm LLM tooling? It's a fucking joke. Its like they're intentionally sinking it for her Nvidia uncle Huang. And Su takes the cheaper CPU market and offers better features than Intel.
I'm not trying to dismiss what you're saying as a possibility (AMD's behavior in many regards over the last 15 years or so is baffling to the point that a family conspiracy feels surprisingly plausible) but Huang isn't Su's uncle.
They are "first cousins once removed" meaning that Su is the child of one of Huang's cousins. Or put another way, one of Huang's grandparents is one of Su's great-grandparents.
Honestly, my mind could never wrap around genealogy and family relations well at all. I thought they were indirect cousin/uncle , or as you say, cousins once removed.
Also my understanding of many Asian cultures is they tend to have a much more tight-knit large family structure. And doubling that is the fact they're 2 heads of world-level hardware tech companies.
And, well, there's no such thing as coincidences. Having all of this line up, and for "some reason" AMD keeps missing when they could have owned a big chunk of the market has a certain family oligopoly smell to me.
They want their *own* CUDA situation. if you're doing FPGA stuff, the Xilinx hardware is good and Vivaldi is really the only way to use it. AFAIK the open-source fpga situation is pretty far behind it (please correct me if I'm wrong, I work in a place that uses Xilinx FPGAs)
> They want their *own* CUDA situation
What better way to do that than decrease availability, I ask, both rhetorically and sarcastically! CUDA-proper did well, at least partially, because it was put in front of everyone. This is taking an exclusivity angle that doesn't make much sense for ecosystem development, IMO.
I suspect this goes to show how much influence/priority B2B carries, my point is it's a mistake they've made before. The maintenance/support costs I believe they're trying to remove have paid dividends for CUDA and countless others. Anyway, a play on the tagline before I go: "Together we advance... licensing and removal of [positive] feedback loops over product development"
I seriously think the CUDA situation was set up intentionally.
Jensen Huang (Nvidia CEO) is related to Lisa Su (AMD CEO).
A decade ago, I saw a demo of some AMD skunkworks GPU datacenter tech, that could execute CUDA natively on AMD/ATI graphics cards. Initially was half speed, but having the flexibility was crazy amazing. Created a big buzz in the big iron and educational markets.
Where'd it go? Buried. You cant even find articles about it. Its a few comments on edtech datacenters.
Now look at AMD's graphics line. Where's their ROCm LLM tooling? It's a fucking joke. Its like they're intentionally sinking it for her Nvidia uncle Huang. And Su takes the cheaper CPU market and offers better features than Intel.
I'm not trying to dismiss what you're saying as a possibility (AMD's behavior in many regards over the last 15 years or so is baffling to the point that a family conspiracy feels surprisingly plausible) but Huang isn't Su's uncle.
They are "first cousins once removed" meaning that Su is the child of one of Huang's cousins. Or put another way, one of Huang's grandparents is one of Su's great-grandparents.
Honestly, my mind could never wrap around genealogy and family relations well at all. I thought they were indirect cousin/uncle , or as you say, cousins once removed.
Also my understanding of many Asian cultures is they tend to have a much more tight-knit large family structure. And doubling that is the fact they're 2 heads of world-level hardware tech companies.
And, well, there's no such thing as coincidences. Having all of this line up, and for "some reason" AMD keeps missing when they could have owned a big chunk of the market has a certain family oligopoly smell to me.
1 reply →