Fatlimey retweeted
People often ask if they should use tau instead of pi, but the answer is that usually you don't need either. Most uses of those constants are because of functions that take radian parameters, which is almost always a bad idea:
computerenhance.com/p/turns-…
Fatlimey retweeted
Here's the first pop of Spectrum Chuckie Egg reverse engineered back to a fully assembler-able version. The code is pretty well documented; I just need to add some more stuff to the readme about how some quirky bits work. I'll update the github soon!
github.com/Paulie68000/ZXSpe…
Fatlimey retweeted
Trump wins the White House. Again.
The Democrats blew it. Again.
A depressing yet predictable result.
Fatlimey retweeted
how i unblurred the code from @CDisillusion
Fatlimey retweeted
I have respect only for modern x86 ASICs. It's an art and the highest level of architectural skill to implement and emulate an outdated ISA with such impressive techniques. de.wikipedia.org/wiki/Intel-…
Fatlimey retweeted
Books every physics student should have:
(1) The Feynman Lectures.
Nobel-winning theoretical physicist Richard Feynman asked to teach the undergraduate introductory physics course at Caltech. It was an inspirational but breakneck experience for the students.
Fatlimey retweeted
New blog post: "Zero or sign extend" fgiesen.wordpress.com/2024/1…
Fatlimey retweeted
Was Marvin Minsky single handedly responsible for slowing down the growth of neural networks by perhaps 20 years? Even though Minsky actually studied and wrote on neural nets he ultimately was one of the major evangelists for symbolic approaches, which mired the field of AI in a dead end for at least 20 years. Imagine if instead we had continued pushing on the connectionist and neural network approaches from 1968 onwards? Would we have gotten specialized hardware earlier that might have helped it (just like the symbolic AI crowd got the Symbolics machines in the 1980s for their beloved LISP?)
Alternate history!
This quoted post is unavailable.
Fatlimey retweeted
differential cable (capstan) drives aren't new.
if you think 1x Neo is impressive, check out the WAM arm from 15 years ago.
highly backdrivable, 0 backlash, super efficient (only rolling joint contacts), and capable of insanely low bandwidth direct joint-torque control.
it's also worth noting that this wasn't a research project, the WAM was in production. you could buy one for around $120,000.
so why didn't WAM dominate? In short, cables are a maintenance nightmare.
1x seem to know this and feel confident they've solved it.
in his S3 interview, the 1x CEO mentions that "like for us... [people say] 'oh you can't use tendons to drive a robot. They don't have reliability, durability and all these things. That's been tried many times before'... and then you try again, and try again for years, and you finally make it work".
btw Elon has said the latest gen Optimus hand is "tendon" driven too.
if you want to learn more about cables and their use in robotics, follow along below.
I'll go into technical details about cable drives, WAM arm and how 1x / Tesla could approach it differently:
Fatlimey retweeted
I've written a long blog post about poor foundations in geometric algebra:
terathon.com/blog/poor-found…
(This is the toned-down version, believe it or not.)
Fatlimey retweeted
Squigonometry: link.springer.com/book/10.10…
Fatlimey retweeted
I believe this is how triangles are composed out of vertices somewhere between the vertex shader and the fragment shader.
#CGFun
Fatlimey retweeted
When I first started teaching statistics, I was called in by my superiors to explain why my class's mean marks were higher than other professors'. My response: "Why are you comparing means and not medians?"
This was met with complete silence. I will never forget this.
Fatlimey retweeted
1/N Happy #FPGAFriday! I just wanted to give a quick description of why this is different than previous designs. None of my previous GPUs were designed to be practically useable. I hardcoded vertices/shaders/texels in a BRAM to test rasterization/texcache/and shaders. They were…
Fatlimey retweeted
Why are the oldest stars completely different to younger ones? What role does dark matter have in the way galaxies form? Join physicist Dr Emma Chapman on a journey to the earliest moments of the universe: #BritishScienceWeek