Why is disruptive technology so hard?
I mean: you've invented something that can change the world, so why aren't people throwing cash at you?
I thought I would write about not only my successful experience, but try to help others get through the tough times.
open.substack.com/pub/codeandr...
The instant loading from SD card would have been the most exciting thing for us back then!
I haven't played it before. I assumed it was hover bovver, but it didn't look right, so I've loaded it up on my handmade modern spectrum clone
Thanks for all the updates for those of us who can't be there. So much amazing work being done with SYCL and OpenCL. There's no reason for anyone to lock their software to processors from one vendor or one type of processor
OpenCL is still very much in active use for real Hugh performance workloads
That looks like an A600 doesn't it? Which means the date is wrong
Looks better like that. If it was dry it would just be a valley. The awful weather is what makes it special
After to crapshoot that was the launch of PVC/Alchemist, things are quite good these days on Intel GPUs.
SYCL looks pretty good.
And when Intel does decide to re-enter the GPGPU market, I expect the software to work.
This is a great list of misconceptions about compilers and explanations of what's really going on and why. These misconceptions are very common: I have had to explain probably all of them more than once. There's one main one missing: compilers are not magic, they follow well-specified engineering
I think for it to work you need tight coupling between different processors located in different parts of the system or you'll just get overloaded by the costs of data movement which is exactly the opposite of what you hope to achieve. That requires particularly heterogeneous friendly software
It is an interesting new field. Have you seen the work we've been doing with Samsung and SYCL? oneapi.io/event-sessio...
