cpu

56 stories and discussions about cpu, aggregated from every source we track.

1.
260 points•signa11•about 2 months ago•193 comments•
2.

CPU, memory, OS and Tetris, all running on the gates. Zoom in and watch the signals flow.

152 points•rossant•about 2 months ago•90 comments•
3.
116 points•jiegec•about 2 months ago•6 comments•
4.
63 points•calvin•about 2 months ago•1 comment
5.

Unlocking _everything_ on the CPU with DRAM scrambling - xoreaxeaxeax/skitter-creek-bath-salts

55 points•radex•3 months ago•17 comments
7.

This story begins, as they so often do, when I noticed that my machine was behaving poorly. My Windows 10 work machine has 24 cores (48 hyper-threads) and they were 50% idle. It has 64 GB of RAM an…

45 points•friendlysock•over 9 years ago•9 comments
8.

VSCode Version: 1.10.2 (8076a19) OS Version: macOS Sierra 10.12.3 VS Code uses 13% CPU when focused and idle, draining battery. This is likely due to the blinking cursor rendering. I think CPU usag...

30 points•av•over 9 years ago•16 comments
9.

Hackers can now reverse-engineer updates or write their own custom firmware.

18 points•r31r06•about 6 years ago•17 comments
10.
17 points•xutaxkamay•2 months ago•3 comments•
12.

Serving Gemma 4 E2B q4_0 through llama.cpp on one laptop, twice: CPU-only and on a 2021-era 4 GB GTX 1650 Ti. Same GGUF, same binary, same prompts, one flag apart. The card takes decode by 4.3x, and needs only 1598 MiB to do it.

15 points•xbill•2 months ago•6 comments
13.
13 points•xbmcuser•about 2 months ago•1 comment•
14.

Gemma 4 E2B q4_0 served by llama.cpp on one laptop, CPU-only and on a GTX 1650 Ti, rebuilt on CUDA 13.4 and re-measured in CPU, GPU, GPU, CPU order with a temperature gate. The card takes decode by 4.14x, and run order moves the answer by about 2%.

10 points•xbill•about 2 months ago•2 comments
15.

VUPEN Security provides vulnerability management and intelligence solutions (advisories, analysis, exploits).

10 points•jcs•about 14 years ago•0 comments
16.

It's a drop in replacement, it's faster, and it's cheap. What's not to like?

9 points•jnord•2 months ago•5 comments•
17.

In the post I will cover: Configuring an ASRock H370M-ITX/ac to allow DCI DbC debugging Using Intel System Studio and System Debugger to single-step a Coffee Lake-S i7-8700 CPU Debugging an example exploitable UEFI…

9 points•fcbsd•over 7 years ago•1 comment
18.

Local decision model with calibrated probabilities: send a state and yes/no, choice or score questions, get a probability for every option. 0.8B GGUF on CPU, Jev-style API. - kouhxp/gutsy

8 points•mrkn1•about 2 months ago•2 comments•
19.

Inside every modern CPU since the Intel Pentium fdiv bug, assembly instructions aren’t a one-to-one mapping to what the CPU actually does. Inside the CPU, there is a decoder that turns assemb…

8 points•bitfield•almost 2 years ago•0 comments
20.

You write System.out.println("Hello World"); and hit Run. Something happens. Text appears on your...

6 points•smtahosin•about 2 months ago•0 comments
21.

grep for meaning: filter, score and classify lines with a plain-English question. Local CPU decision model (Laya, ONNX), single Rust binary. - sfmqrb/gutcheck

5 points•sfmqrb•about 2 months ago•0 comments•
22.

Why is x86 little-endian? Why are registers called AX, BX, CX, and DX, and why DOS uses addresses in the format A000:0000? The answers are in the story of how each chip was born.

4 points•ibobev•about 2 months ago•0 comments•
23.

Light the led unless both switches are on.

4 points•charlesfrisbee•about 2 months ago•3 comments•
25.

Local decision model with calibrated probabilities: send a state and yes/no, choice or score questions, get a probability for every option. 0.8B GGUF on CPU, Jev-style API. - kouhxp/gutsy

3 points•mrkn1•about 2 months ago•1 comment•
26.

Why a sorted array processes 6x faster than unsorted data: CPU pipelines, 2-bit counters, TAGE predictors, pipeline bubbles, and writing branchless code.

3 points•syed_anzar•about 2 months ago•4 comments
27.

We Built a Game IDE in the Browser and Lived to Brag About It Hey, we're Evolved Tech —...

3 points•evolvedtech_27a34e6a5cf9•2 months ago•0 comments
28.

The last few days I spend some time on porting the beloved game DOOM to the ZPU CPU. For you who don't know what I am talking about, DOOM wa...

3 points•zhemao•over 12 years ago•0 comments
29.

Local decision model with calibrated probabilities: send a state and yes/no, choice or score questions, get a probability for every option. 0.8B GGUF on CPU, Jev-style API. - kouhxp/gutsy

2 points•mrkn1•about 2 months ago•1 comment•
30.

The 12-core Zen 3 gaming and productivity monster that never was

2 points•rbanffy•about 2 months ago•0 comments•
31.

What Matters Today is CPU and RAM Utilization (and infra costs) tl;dr: As agents take over writing code, the cost of validating generated code becomes a primary determinant of engineering throughput. Optimizing your…

2 points•roflc0ptic•about 2 months ago•0 comments•
32.

Anti-cheat systems for online multiplayer games can flag individual PC parts, which can be a problem if you're trying to find a deal on secondhand hardware.

2 points•josephcsible•about 2 months ago•0 comments•
33.

CPU-only LLM inference engine in pure Rust — 4-bit quantized models, hand-written AVX2 kernels, speculative decoding, and a local HTTP/MCP server. No GPU, no Python runtime. - Classevelabs/rai

2 points•simonpure•about 2 months ago•0 comments•
34.

Someone buys a used Ryzen 7 5800X3D. “Valorant” and other Riot games won’t start with it because the previous owner was a cheater.

2 points•rzk•about 2 months ago•2 comments•
35.

A reading path for CPU performance engineering, from one instruction to production inference. Primary sources only, with a runnable benchmark for every section. - usamahz/cpu-performance-engineering

2 points•bogdiyan•about 2 months ago•1 comment•
36.

The return of AMD's ambidextrous strategy?

2 points•peter_d_sherman•about 2 months ago•2 comments•
37.

Snap any image, screenshot, or webpage into plaintext. No GPU. No cloud. One command. - kouhxp/textsnap

2 points•mrkn1•about 2 months ago•0 comments•
38.

A new article series will examine the CPU caches that aren't called caches, but work similarly to transparently accelerate programs.

2 points•matt_d•2 months ago•0 comments•
39.

Sampling CPU and HEAP profiler for Java featuring AsyncGetCallTrace + perf_events - async-profiler/async-profiler

2 points•mooreds•about 7 years ago•0 comments
40.

Making long-running CPU intensive processes not to suck every bit of CPU time from other processes in Linux.

1 points•bilegeek•about 2 months ago•0 comments•
41.

An open speech recognition model that runs on the same CPU engine as Needle. It transcribes seven languages, reaches the first token in 11 ms, and loads beside Needle so one binary turns a clip straight into tool calls.

1 points•yarapavan•about 2 months ago•0 comments•
42.

Processor pipeline visualization tools are routine inside industry CPU teams, but few of them are described or released publicly. As a result, students, researchers, and other practitioners rarely see the tooling that…

1 points•matt_d•about 2 months ago•0 comments•
43.

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

1 points•mrkn1•about 2 months ago•0 comments•
44.

The Hutter Prize requires no validation data, so its scores cannot be inflated by data leaks. Three entries, each running on a single CPU core, just improved its record by nearly 10%.

1 points•jabowery•about 2 months ago•1 comment•
45.

No sound card driver. No audio library. Just the CPU and 44100 numbers a second. - whispem/asm.fm

1 points•ta988•about 2 months ago•0 comments•
46.

Snap any image, screenshot, or webpage into plaintext. No GPU. No cloud. One command. - kouhxp/textsnap

1 points•mrkn1•about 2 months ago•0 comments•
47.

Snap any image, screenshot, or webpage into plaintext. No GPU. No cloud. One command. - kouhxp/textsnap

1 points•mrkn1•about 2 months ago•0 comments•
48.

Demonstrating offlineisbetter models. Contribute to offlineisbetter/offlinedemo development by creating an account on GitHub.

1 points•chronicallyoffl•about 2 months ago•1 comment•
49.

The LEDs are connected directly to the die, which is embedded in a glass substrate.

1 points•rbanffy•about 2 months ago•0 comments•
50.

With over 2 million cores, 1% efficiency is worth millions. See how Uber tuned Go's runtime stack to boost CPU utilization for a service by 10%.

1 points•birdculture•about 2 months ago•0 comments•
51.

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

1 points•antonios_makro•about 2 months ago•0 comments•
52.

The LEDs are connected directly to the die, which is embedded in a glass substrate.

1 points•MC995•about 2 months ago•0 comments•
53.

A clean-sheet R4000-class CPU and SGI Indigo IP20 system running unmodified IRIX 5.3 on an FPGA.

1 points•karlgkk•about 2 months ago•1 comment•
54.

A new X meme called "Jevops" (Jevon's Law meets DevOps) took off after someone asked "Can Jev emulate a CPU?" — prompting the punchline: "the end game…

1 points•Bluestein•about 2 months ago•0 comments•
55.
1 points•dadawei•2 months ago•0 comments•

Related topics