Artyom Bologov<p>So I have both float-><a href="https://merveilles.town/tags/minifloat" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>minifloat</span></a> and minifloat-><a href="https://merveilles.town/tags/float" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>float</span></a> conversion in my <a href="https://merveilles.town/tags/CommonLisp" class="mention hashtag" rel="nofollow noopener noreferrer" target="_blank">#<span>CommonLisp</span></a> minifloat library now. And a lot of constants—NaNs, infinities, most and least representable floats etc. It's basically ready, though severely untested. But hey, I didn't expect I can ever make a float decoding library, let alone in under one month!</p><p>I still have to find better algorithms. Some forgotten corners of the academic Internet have pearls of wisdom and I'll eventually find them. But, for now, my algorithms are alright too, running mostly in nanoseconds.</p><p>Find it at <a href="https://codeberg.org/aartaka/cl-minifloats" rel="nofollow noopener noreferrer" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">codeberg.org/aartaka/cl-minifl</span><span class="invisible">oats</span></a></p>