Viewing profile — stephencanon

That’s exactly the same rule with more words (postfix = right, prefix = left).

This comes up every so often, and while it's sort of almost true and attractive, it isn't actually correct. The correct rule is "follow the C grammar". An easier to remember and al…

It's very cool. Thank you!

"Centrist" in practice means "more or less content with the status quo." That is fundamentally a conservative position orthogonal to any notion of "facts".

The work discussed in this post shipped in the OS last year (fall 2025), so nothing here is dependent on very recent changes.

Field hockey is almost exclusively a girls sport in the US, while boys have (American) football in the fall. Both draw from the potential pool of soccer players in US middle and hi…

“in a real race we have no actual chance of winning” is an absolutely wild thing to say in response to a link to a real race in which the human has won the last few years in a row.…

What sort of engineering standards are these Cybertrucks built to? Oh, very rigorous engineering standards. The wheels aren't supposed to fall off for a start.

We have a ground-source heat pump for our ADU. We did it because we were curious about just how efficient we could make the house, but I don't expect that it will ever break even f…

Prior to the current generation Intel designs, Apple’s branch predictor tables were a good deal larger than Intel’s IIRC, so depending on benchmarking details it’s plausible that A…

Must be the water.

The CORE-MATH project authors, most of whom are French academics (including the author of the linked paper). I don’t know of any interesting work in this space that came out of Red…

"I'd like to thank my mother, Ayn Rand, and God" is the usual example. Yes, you can reorder the list to remove the ambiguity, but sometimes the order of the list matters. The seria…

That would fall under "more constrained", due to process limits.

Enlarging a branch predictor requires area and timing tradeoffs. CPU designers have to balance branch predictor improvements against other improvements they could make with the sam…

Your students should be able to figure out if a computation is exact or not, because they should understand binary representation of numbers.

You can often eke something out for order-four, depending on uArch details. But basically yeah.

For throughput-dominated contexts, evaluation via Horner's rule does very well because it minimizes register pressure and the number of operations required. But the latency can be …

When Intel specced the rsqrt[ps]s and rcp[ps]s instructions ~30 years ago, they didn't fully specify their behavior. They just said their relative error is "smaller than 1.5 * 2⁻¹²…

For the asinf libcall on macOS/x86, my former colleague Eric Postpischil invented the novel (at least at the time, I believe) technique of using a Remez-optimized refinement polyno…

Ideally either one is just a library call to generate the coefficients. Remez can get into trouble near the endpoints of the interval for asin and require a little bit of manual in…

Newer rsqrt approximations (ARM NEON and SVE, and the AVX512F approximations on x86) make the behavior architectural so this is somewhat less of a problem (it still varies between …

These sorts of approximations (and more sophisticated methods) are fairly widely used in systems programming, as seen by the fact that Apple's asin is only a couple percent slower …

Yann is definitely more well-known outside of academia. Inside academia, it's going to depend a lot on your specific background and how old you are.

Mechanically, you're probably right, but the screen-centric controls of the newer generation are _awful_ by comparison to the F generation's physical buttons and dials (this isn't …