"As many have said GPUs are so fast because they are so efficient for matrix multiplication and convolution, but nobody gave a real explanation for why this is so. The real reason for this is memory bandwidth and not necessarily parallelism."
Sorry but please don't take Eigen (https://eigen.tuxfamily.org) away from me. Can't speak for others, but the scientific code I work on would become unreadable like that.
I think the problem exists: the externalities of the approach to laptops you're describing are very bad. Framework is trying to promote the three R's to help combat this. Reduce, reuse, and recycle.
I think most folks here are advocating for MacBooks, which you absolutely could not build or buy an equivalent of in the 1000$ range. Desktops are a different class of hardware altogether.
I think we can all agree that maximizing profit is a cause, while also acknowledging that the situation is not ideal. There are many solutions to this self-imposed problem, like standardizing the battery interfaces.
I agree most company's have no incentive to do that. I think those incentives should be put in place though.
Intact Solutions applies state of the art research to simulation for engineering, design, and manufacturing applications. We create finite element analysis solutions that can operate on a wide variety geometry and material types, providing unique solutions for generative design and additive manufacturing.
Intact is a small R&D technology company. We are supported by federal research grants, software licensing and products, as well as industry partners. We value both the exploration of ideas as well as the construction of solid foundations for future endeavors.
We're looking for new software engineers! If you have an interest or past experience with
* mechanical engineering / finite element analysis
* computational physics and / or mathematics
* computational geometry, HPC, numerical issues
then please consider reaching out! We are always looking for candidates who are excited about pushing the state of art in modeling and simulation.
Our core software is written in C++. We are exploring the use of Rust for our core products. In addition, we support a Python interface and a rails based web application.
Intact Solutions applies state of the art research to simulation for engineering, design, and manufacturing applications. We create finite element analysis solutions that can operate on a wide variety geometry and material types, providing unique solutions for generative design and additive manufacturing.
Intact is a small R&D technology company. We are supported by federal research grants, software licensing and products, as well as industry partners. We value both the exploration of ideas as well as the construction of solid foundations for future endeavors.
We're looking for software engineers interested in geometry, materials, and simulation. If you have experience with mechanical engineering, computational physics and mathematics, FEA, HPC, computational geometry, etc, consider reaching out! We are always looking for candidates who are excited about pushing the state of art in modeling and simulation.
Our core software is written in C++. In addition, we support a Python interface and a rails based web application
I don't think looking at hacker news as a hive mind is a good idea tbh.
I'm on the side of those unionizing. Work conditions don't need to be unsafe for folks to unionize, and I'm fairly confident in saying that the vast majority of folks everywhere (in tech anyway) are paid an unfairly low amount for the value they generate.
I work on scientific software, so the biggest technical issue I face day-to-day is that OpenMP based threading seems almost fundamentally incompatible with M1.
The summary of the issue is that OpenMP threaded code typically assumes that a) processors are symmetric and b) there isn't a penalty for threads yielding.
On M1 / macOS, what happens is that during the first OpenMP for loop, the performance cores finish much faster, their threads yield, and then they are forever scheduled on the efficiency cores which is doubly bad since they're not as fast and now have too many threads trying to run on them. As far as I can tell (from the linked thread and similar) there is not an API for pinning threads to a certain core type.
Intact Solutions applies state of the art research to simulation for engineering, design, and manufacturing applications. We create finite element analysis solutions that can operate on a wide variety geometry and material types, providing unique solutions for generative design and additive manufacturing.
Intact is a small R&D technology company. We are supported by federal research grants, software licensing and products, as well as industry partners.
We value both the exploration of ideas as well as the construction of solid foundations for future endeavors.
We're looking for software engineers interested in geometry, materials, and simulation. If you have experience with mechanical engineering, computational physics and mathematics, FEA, HPC, computational geometry, etc, consider reaching out! We are always looking for candidates who are excited about pushing the state of art in modeling and simulation.
Our core software is written in C++. In addition, we support a Python interface and a rails based web application
https://intact.design
If you're interested in learning more, send your resume to
[email protected]
(We can only hire folks authorized to work in the U.S.)