Guy makes a semiconductor lab in a shed and makes a RAM.
Okay I thought this was one of those funny haha gimmick videos at first, but the more I watched the more impressive it became. Hell I haven't heard this many new terms in a year. This guy (or a group of them) built a lab for semiconductors. Let's pretend we want to replicate this process and thus we need to extract as much info as possible. Not only he needed the knowledge for electronics, but also dyi building, airflow engineering, prob chemistry, engineering the actual equipment...
First, build the environment.
The shed is separated into the gowning area and the cleanroom area. It is imperative that the cleanroom stays as clean as possible, as single speck of dust can ruin a chip.
1) The shed is insulated (there's a lot of information on cleanroom insulation online),
2) some framing is added so we can set a cleanroom grade HEPA filter with an air intake filter just under the roof. I cannot tell which filters he got from the video.
3) A dedicated electrical circuit is installed. No information in the video on the specifics.
4) Heating and cooling with a mini split. (Would be nice to specify which role heating and cooling plays)
5) The room is divided and now every space has to be sealed. (?) Instead of using expensive plastic, he uses flame-resistant drywall with a water-based epoxy to create a smooth particle-free surface. The attic space above is also sealed and controlled for airflow.
Now, the airflow part is very important (1st pic) the cleanroom is also kept at a positive pressure (how?) to keep the contamination from coming in. Massive HEPA filter scrubs the air and recycles it a few hundreds per hour. (because the room volume is small)
After all's said and done, a particle counter is brought to the cleanroom for testing. It shows 40 particles/40ft^3 in the middle, and sub-100 at the edge of the room. This makes it a class 100 cleanroom, which OP claims is on the level of Samsung, Intel, TSMC.
Let's move onto equipment.
The gowning room has (1) power management. Grid power. Solar power. Local breakers. Isolated and controlled before reaching the lab (how?) (2) Clean room suits and supplies. Battery backup. Chemical storage. One must obviously use clean room suits and gloves before work, so you keep skin flakes and other particles out of the air.
Cleanroom has (1) Plasma etching, which removes nano scale films. (Sputter coater Model: Technics Hummer V) (2) High temperature processing (appears to be 1200C Mini Tube Furnace) (3) Eye level photography and sub micron powdering created manually from a microscope with custom software (?!?!?!?!?! whaaat) (4) Fume hood for chemical processing, spin coating (the spin coater is 3d printed) (5) Automated sample cleaning with a robot arm (6) Thin film deposition system for building atomic levels
Just who the hell is this guy?? I wanted to describe the ram creation process too but this feels already long enough
There's a few articles written on this guy, so far I found 2 (one is Japanese)
https://news.livedoor.com/article/detail/31080664/
https://www.tomshardware.com/tech-industry/semiconductors/ambitious-semiconductor-enthusiast-builds-diy-class-100-cleanroom-in-his-garden-shed-contains-a-plasma-etcher-vacuum-furnace-and-even-custom-software-driven-lithography-machine
They have their own interpretations of the video (especially the Japanese one)
That's pretty interesting, I hadn't heard about KVM Qemu.It seems like a linux only thing though, so it wouldn't fit as Windows malware analysis probably?
I see quite a few businesses use proxmox, good to know.
It'd be even better if we could trick malware into ignoring stuff like wireshark and debuggers I think. There's probably more advanced ways to detect analyzers than just process name, so it'd be interesting to look into. I have a few channels on videogame cheat development and they talk about bypassing anticheats who use all kinds of heuristics I'm curious about.