UART explained: check the wires, the ground, the speed and the volts

UART explained: not a voltage standard

UART explained: a data byte can say stop

UART explained: an error flag says what happened, not why

UART explained: the receive buffer, and why bytes are lost when it fills

UART explained: why a printer prints rubbish when the settings disagree

UART explained: why the receiver samples in the middle of each bit

UART explained: baud, and why 115200 baud is not 115200 bytes a second

UART explained: parity misses two flips (stop bit, parity and 8N1)

UART explained: the low bit goes first

UART explained: the line idles high, and a start bit is low

UART explained: two wires and no clock

ESP32-S3 upload failed? Manual vs automatic download mode

TCP and UDP explained: connected is not delivered

TCP and UDP explained: fast is not the question

TCP and UDP explained: FIN is polite, RST is not

TCP and UDP explained: slow down yourself

TCP and UDP explained: a slow reader slows a fast sender

TCP and UDP explained: one gap makes everything wait

TCP and UDP explained: TCP is a stream, not messages

TCP and UDP explained: three slips, and still no promise

TCP and UDP explained: the network can lose, delay, repeat and swap your cards

TCP and UDP explained: UDP is not a broken TCP

TCP and UDP explained: port 5000 on TCP is not port 5000 on UDP

TCP and UDP explained: the network makes no promises

SPI explained: when it goes wrong

SPI explained: Octal and the letters (D, Q, O)

SPI explained: Quad SPI, four data wires

SPI explained: Dual SPI, two bits per pulse

SPI explained: more than one device

SPI explained: how fast the clock can go

SPI explained: one transaction, from CS low to CS high

SPI explained: four modes, and why both sides must match

SPI chip select explained: one wire per device

SPI explained: sending and receiving at once

SPI explained: one byte, eight bits

SPI explained: the clock

SPI explained: the four wires

SPI explained: why another bus when I2C exists?

Choosing a display: start with your MCU, not your dream screen

eDP explained: video comes after the conversation

LVDS explained: why RGB wires cannot just become a differential pair

MIPI DSI command mode vs video mode: who keeps the picture?

MIPI DSI explained: a few wires can carry a lot of pixels

Refresh, tearing and double buffering: why a faster refresh does not fix it

RGB565, RGB666, RGB888: colour depth is not bus width

RGB timing: pixel clock, HSYNC, VSYNC, Data Enable and the porches

Frame buffer, PSRAM and DMA: the frame fits, but does it arrive in time?

Where does one frame actually live? A display with memory vs a stream

8080 parallel displays: no clock line, so what sets the beat? (WR, D/C, bus width)

QSPI displays: do four data lines make updates four times faster?

SPI TFT explained: open a window, then pour pixels (CASET, RASET, RAMWR)

I2C OLED explained: why half the screen still shows the old picture

A screen is not one thing: backlight, panel, controller and display memory

How much data is a screen? Count the pixels first

I2C 5V to 3.3V explained: why you need a level converter

I2C Too Many Pull-ups explained: why the sixth module breaks the bus

I2C Pull-up Value explained: 4.7k or 2.2k, and why

I2C Pull-up Resistors explained: why nothing answers without them

I2C Read explained: START to STOP, and two bytes into degrees