Explore >> Select a destination
|
You are here 
|
marc-b-reynolds.github.io | ||
| | | | |
nhigham.com
|
|
| | | | | The 2008 revision of the IEEE Standard for Floating-Point Arithmetic introduced a half precision 16-bit floating point format, known as fp16, as a storage format. Various manufacturers have adopted fp16 for computation, using the obvious extension of the rules for the fp32 (single precision) and fp64 (double precision) formats. For example, fp16 is supported by... | |
| | | | |
pavpanchekha.com
|
|
| | | | | [AI summary] Pavel Panchekha explains the inherent floating-point inaccuracies when evaluating polynomials near their roots and discusses why the automated tool Herbie struggles to find more accurate implementations in such ill-conditioned cases. | |
| | | | |
codingnest.com
|
|
| | | | | There is a lot of confusion about floating-point numbers and a lot of bad advice going around. IEEE-754 floating-point numbers are a complex beast, and comparing them is not always easy, but in this post, we will take a look at different approaches and their tradeoffs. | |
| | | | |
andrewkelley.me
|
|
| | | [AI summary] The blog post discusses the challenges and solutions for handling asynchronous I/O in Zig, focusing on cancellation, resource management, and concurrency. It highlights how cancellation is a key primitive for managing resources and avoiding leaks, while concurrency is essential for avoiding deadlocks in producer-consumer scenarios. The post also touches on the importance of using appropriate APIs like `concurrent` instead of `async` when true concurrency is needed, and the ongoing design work for stackless coroutines and I/O interfaces. | ||
