811
New
The latest blog posts to take off on Hacker News. 6,299 posts.
812
Google Cloud TPU Multislice Training
We used Multislice Training to run the world’s largest LLM distributed training job on a compute cluster of 50,944 Cloud TPU v5e chips.
813
Don't build AI products the way everyone else is doing it
More and more, people are building products as thin wrappers over other models, especially LLMs. This can cause major issues, and there is a better way.
814
Dungeon KeeperFX 1.0
All original Dungeon Keeper code has been rewritten, establishing KeeperFX as a true open-source standalone game. Ownership of the original game is still and will always be required for copyright reasons. When installing KeeperFX 1.0, perform a fresh installation without overwriting any previous versions. Saved games…
815
Gigantic heat caverns in Mustikkamaa have now been filled with water (2021)
The filling of underground heat caverns in Mustikkamaa with water has been completed this week, and the use of the heat caverns for heating homes in Helsinki is a step closer.
816
Interview with Jeff Geerling
Andrew Gregory of HackSpace magazine sat down with the one and only Jeff Geerling to chat about the wholesome corner of the internet he created.
817
Nordic is getting involved in RISC-V
What RISC-V will do is give our customers is an extra option when seeking to minimize power consumption in certain applications where the (many) tradeoffs of not using an Arm-based core are acceptable.
818
We're sorry we created the Torment Nexus
( This is the text of a talk I delivered at the Next Frontiers Applied Fiction Day in Stuttgart on Friday November 10th, 2023. Note: early draft, contains some typos, I'll fix them next week when I get home. )
819
820
We used to build steel mills near cheap power. Now we build datacenters
Why are people so concerned with hardware power consumption nowadays? Some common answers to this question are that power is critically important for phones, tablets, and laptops and that we can put more silicon on a modern chip than we can effectively use . In 2001 Patrick Gelsinger observed that if scaling continued…
821
Implementing Mario's Stack Blur 15 times in C++ (with tests and benchmarks)
Implementing Mario Klingemann's Stack Blur algorithm in C++ 15 ways, with tests and benchmarks, to build fast CPU drop shadows for JUCE.
822
Moving our Encrypted DNS servers to run in RAM
We recently announced the completion of our migration to remove all traces of disks in use on our VPN infrastructure.
823
Let's create a Python Debugger together
An article series where I show you how to write a fully functional Python debugger.
824
Lessons learned by a software guy venturing into hardware
Join me as I transition from soft to hardware with the SidecarTridge Multi-device. Insights and challenges of creating a ROM cartridge emulator for Atari ST
825
826
Albert Einstein in 1939 composed a message for the people of AD 6939
Preparing a time capsule in 1939, the Westinghouse Electric & Manufacturing Company asked Albert Einstein to compose a message for the people of AD 6939. He sent this: Our time is rich in inventive minds, the inventions of which could facilitate our lives considerably. We are crossing the seas by power and utilize…
827
I skipped to the ending
When I was 8 years old I started making websites with my friend. His grandmother had purchased him a domain name and web hosting. It was 2003 and the world was whispering of Web 2.0. Books I checked out of the library talked about DHTML (dynamic HTML, basically what any web app is today). There was no mobile app…
828
829
Link-Time Optimisation (LTO)
I recently started exploring link-time optimisation (LTO), which I used to think was just a single boolean choice in the compilation and linking workflow, and perhaps it was like that a while ago… I’ve learned that these days, there are many different dimensions of LTO across compilers and linkers today and more…
830
Cursorless is alien magic from the future
Xe Iaso's personal website.
831
832
Typing fast is about latency, not throughput
any speed above, say, maybe 80 wpm , isn’t going to hinder anyone when trying to put anything into the computer
833
On medieval cats
The other night I was mindlessly scrolling about on my phone through Instagram, as one does when they really ought to go to bed but are stuck being entertained instead of resting, when I came upon …
834
835
Reverse engineering the Intel 386 processor's register cell
The groundbreaking Intel 386 processor (1985) was the first 32-bit processor in the x86 line. It has numerous internal registers: general-pu...
836
What I learned getting acquired by Google
Our 10 person startup gets acquired by Google, we rebuild our product the Google way, and begin to understand that amazing things are possible at Google, if you play the Google game
837
Preview of CLion Nova, which uses the ReSharper C++/Rider C++ language engine
Update: CLion Nova is now available in CLion Classic TL;DR Today, we’re announcing a free early preview of CLion, which uses the ReSharper C++/Rider C++ language engine instead of the CLion "legacy"
838
Servo announces grant from the NLnet Foundation
Earlier this year in July, Servo received a NLnet grant to enhance several aspects of Servo.
839
840