By: hobold (hobold.delete@this.vectorizer.org), October 28, 2020 12:44 pm
Room: Moderated Discussions
who (who.else.delete@this.nobody.com) on October 28, 2020 1:07 pm wrote:
> anon (an.delete@this.n.net) on October 28, 2020 11:31 am wrote:
> > After AMD bought ATI for US$ 5 billion they spent years hyping "AMD Fusion", APUs,
> > HSA, Streams, OpenCL, etc, but it was too much hype for too little results.
> > "APU" means "Accelerated Processing Unit" but most of what it "accelerates" are just games,
> > and they are still slower than dedicated videocards despite all the hype to the contrary,
> > they also have worse drivers compared to Nvidia and lack something like CUDA.
> >
> > So, after looking at what happened to ATI after its acquisition by
> > AMD, I'm not automatically optimist about the Xilinx acquisition.
>
> Well, here we are 14 years after said acquisition, and after many false starts, it finally
> seems "AMD red" is not just firing on all cylinders, but has a competitive hunger again.
> So, not entirely doom and gloom for Xilinx, if one has a longer-term outlook.
Do FPGA prototypes still have a part in modern CPU/GPU design? If so, then having your very own FPGAs might give you an advantage over the competition, perhaps?
> anon (an.delete@this.n.net) on October 28, 2020 11:31 am wrote:
> > After AMD bought ATI for US$ 5 billion they spent years hyping "AMD Fusion", APUs,
> > HSA, Streams, OpenCL, etc, but it was too much hype for too little results.
> > "APU" means "Accelerated Processing Unit" but most of what it "accelerates" are just games,
> > and they are still slower than dedicated videocards despite all the hype to the contrary,
> > they also have worse drivers compared to Nvidia and lack something like CUDA.
> >
> > So, after looking at what happened to ATI after its acquisition by
> > AMD, I'm not automatically optimist about the Xilinx acquisition.
>
> Well, here we are 14 years after said acquisition, and after many false starts, it finally
> seems "AMD red" is not just firing on all cylinders, but has a competitive hunger again.
> So, not entirely doom and gloom for Xilinx, if one has a longer-term outlook.
Do FPGA prototypes still have a part in modern CPU/GPU design? If so, then having your very own FPGAs might give you an advantage over the competition, perhaps?