| Homepage | https://dockerr.com |
| Github | https://github.com/el-dockerr |
| Writers alter ego | https://swenkalini.com |
I sat down and did the math. In my new video, I show how I drastically reduced AI workloads and why big tech completely ignores it.
Let's bring efficient coding back: https://youtu.be/UkgKGQi6szE
#AI #MachineLearning #Coding #Tech #ElDockerr #SoftwareEngineering
Remember the 90s scam "SoftRAM" that promised to double your memory? It was totally fake—so I decided to actually build it. 💻✨
Watch the video: https://youtu.be/vz1zg8J9hoI?si=WSR5ysNLjaaLWYW0
One assembler to rule them all.
I'm building UA. A language that reads like high-level code but talks directly to the registers. The standalone asm compiler do PE, ELF, or raw binaries without any ext. linkers.
Full manifesto coming soon in the "BOOM".
New Code & Math Experiment! 🌍📐
Are we living on a globe or a disk? I take a deep dive into the Flat Earth phenomenon.
To settle the debate once and for all, I built a working simulation in the Godot Game Engine to see if a sunset mathematically functions on a flat plane. ☀️📉
Watch the full breakdown and simulation here: https://youtu.be/ImqlZZH3P_E
#GodotEngine #GameDev #Science #Physics #Math #Programming #EduTooter #Debunk #ElDockerr
I did some nice experiments, created a reversed raytracer and got some strange results.
https://youtu.be/0NMEjUv8Eq4
#Raytracing #CPlusPlus #GraphicsProgramming #Math #Coding #GameDev #PhysicsSimulation
Running out of DSP slices on your FPGA? I’ve been experimenting with Low-Rank Approximations for 3x3 Convolutions to solve exactly that.
My latest project replaces standard matrix multiplications with learned, hardware-friendly bit-shifts.
The result:
• 33% reduction in DSP usage (2 Muls instead of 3)
• <1% error (SSIM > 0.99)
• Ideal for SWaP-C constrained edge perception.
https://www.dockerr.blog/blog/lowrank-hardware-approximation
#FPGA #VHDL #EmbeddedSystems #ComputerVision #EdgeAI #OpenScience #Engineering
Released V1.1.O of GMlib. IO Swapping has now encryption. Plus improvements of memory compression on RAM.
https://github.com/el-dockerr/ghostmem
They said "just download more RAM" was a meme. I took it personally. 👻💾
Introducing GhostMem: A user-space C++ library that transparently intercepts virtual memory accesses. It traps page faults and compresses cold data on-the-fly using LZ4.
No kernel modules. No admin rights. Just more RAM.
https://github.com/el-dockerr/ghostmem
#cpp #lowlevel #performance #systemsengineering #lz4 #linux #windowsdev
GhostMem is a smart memory management system that extends available RAM through **in-memory compression** rather than traditional disk-based swapping. - el-dockerr/ghostmem