Easton Man's Channel

22:57 · Jan 14, 2026 · Wed

Daniel Lemire's blog
How stagnant is CPU technology?

Sometimes, people tell me that there is no more progress in CPU performance.

Consider these three processors which had comparable prices at release time.

1. The AMD Ryzen 7 9800X3D (Zen 5, with up to 5.3 GHz boost) was released in 2024.
2. The AMD Ryzen 7 7800X3D (Zen 4, with up to 5.1 GHz boost) was released in 2023.
3. The AMD Ryzen 7 5800X3D (Zen 3, with 3.4 GHz base) was released in 2022.

Let us consider their results on on the PartialTweets open benchmark (JSON parsing). It is a single core benchmark.

In two years, on this benchmark, AMD more than doubled the performance for the same cost.

So what is happening is that processor performance is indeed going up, sometimes dramatically so, but not all of our software can benefit from the improvements. It is up to us to track the trends and adopt our software accordingly.

source

21:46 · Jan 9, 2026 · Fri

08:49 · Jan 8, 2026 · Thu

Daniel Lemire's blog
What I Got Wrong About “Hard Work” in My 20s

Telegraph | source

Telegraph

What I Got Wrong About “Hard Work” in My 20s

When I was younger, in my 20s, I assumed that everyone was working “hard,” meaning a solid 35 hours of work a week. Especially, say, university professors and professional engineers. I’d feel terribly guilty when I would be messing around, playing video games…

08:56 · Jan 7, 2026 · Wed

Daniel Lemire's blog
A bit of glass and freedom is all you need

Telegraph | source

Telegraph

A bit of glass and freedom is all you need

Generated by RSStT. The copyright belongs to the original author. Source

05:36 · Jan 7, 2026 · Wed

Chips and Cheese
CES 2026: Taking the Lids off AMD's Venice and MI400 SoCs
#ChipAndCheese

Telegraph | source
(author: George Cozma)

Telegraph

CES 2026: Taking the Lids off AMD's Venice and MI400 SoCs

Hello you fine Internet folks, Here at CES 2026, AMD showed off their upcoming Venice series of server CPUs and their upcoming MI400 series of datacenter accelerators. AMD has talked about the specifications of both Venice and the MI400 series at their Advancing…

ChipAndCheese

05:37 · Jan 5, 2026 · Mon

Chips and Cheese
Diving into Qualcomm's Upcoming Adreno X2 GPU with Eric Demers
#ChipAndCheese

Telegraph | source
(author: George Cozma)

Telegraph

Diving into Qualcomm's Upcoming Adreno X2 GPU with Eric Deme…

Hello you fine Internet folks, Today we are talking about Qualcomm's upcoming X2 GPU architecture with Eric Demers, Qualcomm's GPU Team lead. We talk about the changes from the prior X1 generation of GPUs along with why some of the changes were made. Hope…

ChipAndCheese

10:24 · Jan 4, 2026 · Sun

https://www.nite07.com/zh-cn/posts/git-includeif/

Nite

Git includeIf：优雅地管理多个 Git 身份

作为开发者，你是否遇到过这样的尴尬：在公司项目中不小心用个人邮箱提交了代码？或者在开源项目中暴露了工作邮箱？如果你同时维护着工作项目和个人项目，手动切换 Git 配置不仅繁琐，还容易出错。
幸运的是，Git 提供了一个强大而优雅的解决方案：includeIf。这个功能可以让 Git 根据不同的条件自动加载不同的配置文件，彻底解决多身份管理的难题。
为什么需要 includeIf？在日常开发中，我们常常需要在不同的身份之间切换：

22:05 · Jan 1, 2026 · Thu

Daniel Lemire's blog
Technology is culture

Telegraph | source

Telegraph

Technology is culture

We are experiencing one of the most significant technological breakthroughs of the last few decades. Call it what you will: AI, generative AI, large language models… But where does it come from? Academics will tell you that it stems from decades of mathematical…

02:57 · Jan 1, 2026 · Thu

Daniel Lemire's blog
The culture war that we won

Telegraph | source

Telegraph

The culture war that we won

Culture wars are real. They occur when a dominant culture faces a serious challenge. But unless you pay close attention, you might miss them entirely. As a kid, I was a “nerd.” I read a lot and spent hours on my computer. I devoured science and technology…

18:48 · Dec 31, 2025 · Wed

Chips and Cheese
Inside Nvidia GB10’s Memory Subsystem, from the CPU Side
#ChipAndCheese

Telegraph | source
(author: Chester Lam)

Telegraph

Inside Nvidia GB10’s Memory Subsystem, from the CPU Side

GB10 is a collaboration between Nvidia and Mediatek that brings Nvidia’s Blackwell architecture into an integrated GPU. GB10’s GPU has 48 Blackwell SMs, matching the RTX 5070 in core count. The CPU side has 10 Cortex X925 and 10 Cortex A725 cores and is therefore…

ChipAndCheese

15:06 · Dec 31, 2025 · Wed

Daniel Lemire's blog
By how much does your memory allocator overallocate?

Telegraph | source

Telegraph

By how much does your memory allocator overallocate?

How much virtual memory does the following C++ expression allocate on the heap? new char[4096] The answer is at least 4 kibibytes but surely more. Firstly, each heap memory allocation requires some memory to keep track of what has been allocated. You are…

05:26 · Dec 30, 2025 · Tue

Matt Keeter
Caves of Qud: A review

source
(author: Matt Keeter (matt.j.keeter@gmail.com))

23:02 · Dec 29, 2025 · Mon

Daniel Lemire's blog
Freedom from incompetence

Telegraph | source

Telegraph

Freedom from incompetence

Many people say that they crave more freedom.But what do we mean by “freedom”?Being free from constraints? Is that what we mean? Would you feel “freer” if you could walk outside in your underwear?It is almost surely not what you mean by “freedom.”I submit…

15:11 · Dec 28, 2025 · Sun

Daniel Lemire's blog
Don’t be so eager to rewrite your code

Telegraph | source

Telegraph

Don’t be so eager to rewrite your code

I used to always want to rewrite my code. Maybe even use another programming language. « If only I could rewrite my code, it would be so much better now. » If you maintain software projects, you see it all the time. Someone new comes along and they want to…

15:11 · Dec 28, 2025 · Sun

Daniel Lemire's blog
Parsing IP addresses quickly (portably, without SIMD magic)

source

Daniel Lemire's blog

Parsing IP addresses quickly (portably, without SIMD magic)

Most programmers are familiar with IP addresses. They take the form of four numbers between 0 and 255 separated by dots: 192.168.0.1. In some sense, it is a convoluted way to represent a 32-bit integer. The modern version of an IP address is IPv6 which is…

15:42 · Dec 25, 2025 · Thu

杰哥的{运维，编程，调板子}小笔记
2025 年我是怎么使用 AI 的

Telegraph | source

Telegraph

2025 年我是怎么使用 AI 的

2025 年我是怎么使用 AI 的¶ 前言¶ 经常看我博客的读者应该能看出来，我研究的主要是计算机系统结构方向，特别是处理器的微架构，几乎没有涉及到 AI 的内容，我也确实不喜欢 AI 研究，仅关注但不参与。但今年，因为各种 AI 技术尤其是 LLM 的发展，我确实成为了很多 AI 技术的用户，可以 2025 年是我正经大规模用 AI 的元年，所以在年末做一个简单的总结。我不想在这里给大模型厂商打广告，所以相关的名字我都会按照某 PDF 的方法进行打码，有需要的朋友可以自行查看实际的内容。 Vibe Coding¶…

16:23 · Dec 21, 2025 · Sun

Daniel Lemire's blog
Performance trick : optimistic vs pessimistic checks

source

Daniel Lemire's blog

Performance trick : optimistic vs pessimistic checks

Strings in programming are often represented as arrays of 8-bit words. The string is ASCII if and only if all 8-bit words have their most significant bit unset. In other words, the byte values must be no larger than 127 (or 0x7F in hexadecimal). A decent…

16:23 · Dec 21, 2025 · Sun

Daniel Lemire's blog
JSON-complete data formats and programming languages

Telegraph | source

Telegraph

JSON-complete data formats and programming languages

Much of the data on the Internet is shared using a simple format called JSON. JSON is made of two composite types (arrays and key-value maps) and a small number of primitive types (64-bit floating-point numbers, strings, null, Booleans). That JSON became…

03:05 · Dec 21, 2025 · Sun

Arch Linux: Recent news updates
NVIDIA 590 driver drops Pascal and lower support; main packages switch to Open Kernel Modules

With the update to driver version 590, the NVIDIA driver no longer supports Pascal (GTX 10xx) GPUs or older. We will replace the nvidia package with nvidia-open, nvidia-dkms with nvidia-open-dkms, and nvidia-lts with nvidia-lts-open.

Impact: Updating the NVIDIA packages on systems with Pascal, Maxwell, or older cards will fail to load the driver, which may result in a broken graphical environment.

Intervention required for Pascal/older users: Users with GTX 10xx series and older cards must switch to the legacy proprietary branch to maintain support:

● Uninstall the official nvidia, nvidia-lts, or nvidia-dkms packages.
● Install nvidia-580xx-dkms from the AUR

Users with Turing (20xx and GTX 1650 series) and newer GPUs will automatically transition to the open kernel modules on upgrade and require no manual intervention.

source
(author: Peter Jung)