@xoofx Using LLMs to fix up grammar is fine, sure. For everything else, I suggest trying first and coming up with your own criteria; my perspective is simple. It's human to err, and we used to think improving your writing is part of the process. Your "awkward wording" is just fine actually.
zeux@mastodon.gamedev.place
Posts
-
I don’t fully understand the motivation to use LLMs to write blog posts. -
I don’t fully understand the motivation to use LLMs to write blog posts.@Doomed_Daniel At this point I would literally take photos of napkins with jotted notes over whatever the LLM produces when you feed the photos in.
-
I don’t fully understand the motivation to use LLMs to write blog posts.I don’t fully understand the motivation to use LLMs to write blog posts. The loss of personal voice is so striking that it immediately rings alarm bells in my head. You can fact check the output although even that is spotty but you can’t inject the soul back.
And if you stop writing yourself, what makes you think you still can? And what is it that makes you - you?
-
Someone on Lobsters wondered "how a modern compiler would fare against hand-optimized asm" in reference to Abrash's TransformVector (3x3 matrix-vector multiply) hand-written x87 routine in Quake.@rygorous @TomF @wolf480pl @pervognsen I think I posted stats for some random programs a little while back and it was amusing to see x64 average close to 4 bytes per instruction. Lots of 1-2 bytes instructions but also lots of 5-8+ so it all blends. Although I didn’t do instruction *counts* which might be a little smaller vs fixed length instruction sets.
-
Someone on Lobsters wondered "how a modern compiler would fare against hand-optimized asm" in reference to Abrash's TransformVector (3x3 matrix-vector multiply) hand-written x87 routine in Quake.@rygorous @wolf480pl @pervognsen ARM64 kinda looks like huffman coding sometimes, only without variable length output you’re reduced to truncating inputs arbitrarily.
-
Someone on Lobsters wondered "how a modern compiler would fare against hand-optimized asm" in reference to Abrash's TransformVector (3x3 matrix-vector multiply) hand-written x87 routine in Quake.@pervognsen A question I'm more curious about is what is the delta on a modern OOO CPU?