Have a cracking weekend boys and girls!

a side project given by my prof today had to store a 60second 60 fps video efficiently just by using C (no libraries allowed)

neural nets. My god

Cerebras is 10x faster than GPUs 10x = 3 Moore's Law doublings = 10 years It's basically like getting a GPU from 2034. When has this ever happened in tech?

No one understand the pain of Full Stack Developers 🤣🤣🤣

That "****video" looks like a pointer to a triple star rated video.👄👄 May be stop allocating at random locations. Either use strided memory access or Arrays of Structures of Structure of Arrays to keep related data in the vicinity of each other.

The Spending Law: Any growth beyond Moore's Law is achieved only by spending more From a presentation by Rio Yokota #AI #HPC via Hatem Ltaief

Are you ready for the VT220 demo? I've prepared scrolling stars Amiga-style for you, but with my Telnet session to the KianV RISC-V Linux FPGA SoC. Enjoy!

I'll never trust myself again 😎 github.com/PabloOyarzo/po…

AMD published its first official MLPerf results for #MI300X Roughly in line with Nvidia's #H100 in the Llama 2 70B model for #GenerativeAI, but it falls well behind the revamped #H200 tomshardware.com/tech-industry/… #AI #HPC #GPU via Tom's Hardware

For those who know the sound of the link negotiation from memory, and for those who have never heard heard it - this is the best write up of that soundscape i've ever found. windytan.com/2012/11/the-so…

Comparison between 20X / 40X microscope objectives. We can see on the right how much powerful the 40X is.

Love the cover! I found the Google Books preview: books.google.com/books?id=eD4ta…

After 27 years of providing in-depth coverage of the amazing world of PC and mobile hardware, AnandTech is saying farewell. We want to thank everyone from the AnandTech community for their support and passion for what we’ve done over the years anandtech.com/show/21542/end…

=> Micron: Patent Applications (P: Aug 2021) "Wafer-on-Wafer Formed Memory and Logic" patents.google.com/patent/US20230… Logic devices can be AI accelerators can provide a greater bandwidth Testing patents.google.com/patent/US20230… Signal Routing, Input/Output Connections, etc. for Cerebras WSE-x?

That makes a lot of sense... I always try to use "1D" Arrays when possible...

Holy shit, let's play Asteroids on the KianV RISC-V Linux SoC over Telnet. Let's play. One world, one Unix, one CPU RISC-V... aleeeeeee!

Goodnight all

End of Friday….