The hottest Signal Processing Substack posts right now

And their main takeaways
Category
Top Technology Topics

It's all a blur

lcamtuf’s thing • 11631 implied HN points • 06 Feb 26
🕹 Technology Signal Processing
  1. Averaging-based blurs are linear and often reversible, so knowing the filter and padding lets you set up simple equations to recover original pixels.
  2. A right-aligned moving average makes iterative reconstruction straightforward and can reveal fine detail even with large blur windows, though 8-bit quantization adds visible noise.
  3. Two-pass (X then Y) blurs can still be inverted if the filter biases the current pixel, and recovered images can survive normal lossy formats like JPEG unless compression is very heavy.

See it with your lying ears

lcamtuf’s thing • 6938 implied HN points • 10 Jan 26
🕹 Technology Signal Processing
  1. Images and audio are both sampled data so you can apply similar transforms to both, but ears and eyes perceive artifacts very differently so the same operation can look fine and sound awful.
  2. Pixelating or reducing bit depth in audio creates stair-step or high-frequency errors that produce metallic squeals or hiss, and those artifacts are typically removed with lowpass/rolling-average filtering or proper DAC anti-aliasing.
  3. Frequency-domain editing works well if you process short, overlapping windows with a Hann (sin^2) weighting and 50% overlap so the attenuations cancel out, avoiding clicks and enabling effects like pitch shifting and vocoding.

It's hard to build an oscillator

lcamtuf’s thing • 7958 implied HN points • 21 Nov 25
🕹 Technology Signal Processing
  1. Building a reliable oscillator is tough because it needs gain to work. Without gain, any oscillation will die out quickly.
  2. Using a Schmitt trigger can help create an oscillator with no stable midpoint. This means the circuit will keep switching back and forth, creating consistent oscillations.
  3. Different methods exist to build oscillators, like using op-amps or creating resonance with phase shifts. Each has its own way of generating oscillation, but they all need a careful balance of components.

Cursed circuits #4: PLL frequency multiplier

lcamtuf’s thing • 4081 implied HN points • 26 Dec 25
🕹 Technology Signal Processing
  1. Latches and clocked D flip-flops store single bits and let signals be sampled on clock edges, providing the basic timing building blocks for digital circuits.
  2. A digital phase detector uses flip-flops to see which clock edge arrives first and produces pulses that indicate whether a tested clock is running too fast or too slow.
  3. A PLL closes the loop by using that detector to steer a VCO, and by inserting a divider in the feedback the VCO will lock at an integer multiple of the reference frequency, turning a low-frequency clock into a higher-frequency, phase-aligned clock.

More bits is better, right?

filterwizard • 59 implied HN points • 01 Oct 24
🕹 Technology Signal Processing
  1. Increasing the bit width of an ADC can improve data accuracy, but it doesn't always work as expected.
  2. Quantization can cause significant errors, especially with low-level signals, leading to misleading results.
  3. Using dither helps improve the accuracy of the signal output from an ADC, making it better for capturing lower signal levels.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

Cursed circuits #3: true mathematics

lcamtuf’s thing • 3877 implied HN points • 22 Dec 25
🕹 Technology Signal Processing
  1. An op-amp simply amplifies the voltage difference between its inputs by a huge factor, and with feedback you force its inputs to be nearly equal so passive parts (resistors, diodes, caps) can be arranged to perform math instead of just gain.
  2. Addition and subtraction are straightforward: resistor networks can average or sum signals and a non‑inverting amplifier scales them to produce a true sum, while difference amplifiers give Vout ≈ VA − VB and can be biased to work on a single supply.
  3. Harder operations are possible too: multiplication/division can be done with log/antilog converters that use the diode’s exponential V–I curve plus a summing stage, and integration is implemented by charging a capacitor with a controlled current to produce precise ramps, though these analog tricks need careful biasing and have practical limits (rails, linearity, noise).

Cursed circuits #2: switched capacitor lowpass

lcamtuf’s thing • 3265 implied HN points • 04 Dec 25
🕹 Technology Signal Processing
  1. You can build a lowpass filter using just capacitors and a switch instead of resistors. This method is simpler and can lead to interesting circuit designs.
  2. The switch in this setup changes the connection of the capacitors, allowing them to charge and affect the signal based on their voltages. This simulates resistor-like behavior, even though no resistors are used.
  3. By adjusting the frequency of the switching, you can control how the filter responds to different input signals. This gives you flexibility in analog signal processing.

Optical Conversion of CMOS <> Superconducting Signals

Tapa’s Substack • 119 implied HN points • 10 Sep 24
🕹 Technology Signal Processing
  1. There's a new idea to use light to switch superconducting signals with CMOS circuits. This could help in areas like quantum computing and sensors.
  2. Using light for this switching can provide thermal isolation, which is a big advantage. It makes the connection between different technologies safer and more efficient.
  3. Two methods to switch are discussed: thermal and non-thermal. Non-thermal switching might be better because it's more efficient, using light directly instead of heating things up.

Analyze FIR filters using just high-school algebra

filterwizard • 39 implied HN points • 23 Sep 24
🕹 Technology Signal Processing
  1. FIR filters have a finite impulse response, meaning they only remember a limited amount of past input. This makes them predictable and stable, especially for applications needing fast settling times.
  2. You can think of FIR filter coefficients as a polynomial, which allows you to use algebra to analyze and create filters. This approach helps in understanding how changing coefficients affects the filter's behavior.
  3. By factoring the polynomial of an FIR filter, you can create smaller filters that combine to produce the same overall effect. This technique allows for a deeper exploration of filter design, giving you more control over the filter's characteristics.

4.4 Signals Along a Transmission Line

Fields & Energy • 319 implied HN points • 07 Aug 24
🕹 Technology Signal Processing
  1. Long telegraph cables can cause delays and signal blurring, which was a problem when laying the first transatlantic cable.
  2. Using too much voltage to fix signal issues can break the cable, leading to more problems rather than solutions.
  3. The first successful transatlantic cable started working in 1866, just after an important theory on electromagnetism was published.

Now synthesize your FIR filters using high-school algebra

filterwizard • 19 implied HN points • 27 Sep 24
🕹 Technology Signal Processing
  1. You can create FIR filters by breaking them down into smaller parts using simple math. This makes it easier to understand how each piece works together.
  2. The sharp notches or deep points in a filter's response happen because of certain factors in the polynomial. Each notch can be traced back to specific frequencies based on these factors.
  3. To improve a filter's performance, you can add more mathematical pieces to make the response smoother in certain areas. This way, you can customize how the filter behaves at different frequencies.

4.3.3 Cross Coupling

Fields & Energy • 259 implied HN points • 31 Jul 24
🕹 Technology Signal Processing
  1. Thaddeus Cahill invented an early electronic music system called the Telharmonium in 1897, aiming to broadcast music through telephone lines. However, his venture failed because the music interrupted phone calls, causing complaints from users.
  2. Cahill's difficulties were largely due to a problem called cross coupling, where signals from one line affect nearby lines. This was common back in the day when many phone lines ran close together.
  3. The situation shows that electrical signals can spread beyond their wires, not just following what we'd think of as direct paths. This understanding is important in telecommunications and electrical engineering.

Which filters are Noisier – Analog or Digital? Part 2

filterwizard • 19 implied HN points • 19 Sep 24
🕹 Technology Signal Processing
  1. When comparing analog and digital filters, analog filters tend to perform better in terms of noise, especially at low frequencies. Digital filters can introduce quantization noise that isn't present in analog filters.
  2. Digital filters, specifically the Direct Form filter, can have significant noise gain, which means they can amplify noise from quantization, making their performance worse in certain situations.
  3. To improve the noise performance of digital filters, increasing the bit depth of the processing can help, but there are also alternative filter topologies that can reduce noise without needing more bits.

When the mask slips - a cautionary (if ancient) tale of IC quality

filterwizard • 59 implied HN points • 01 Sep 24
🕹 Technology Signal Processing
  1. Don't assume that all ICs perform the same, even if they look similar. Small changes in production can lead to big differences in quality.
  2. Working with audio equipment requires attention to detail in filtering processes. It's essential to ensure that all components meet specific performance standards.
  3. When using older components, always check for changes in manufacturing. Even slight variations can drastically affect audio quality, as seen with the NE5532 op-amps.

Which Filters are Noisier – Analog or Digital? Part 1

filterwizard • 19 implied HN points • 18 Sep 24
🕹 Technology Signal Processing
  1. Analog filters can generate noise from several sources like opamps and passive components. Understanding where this noise comes from helps in designing better filters.
  2. Capacitors don’t create noise themselves, but they can hold noise sampled from resistors. This means their role in noise management in filters is important.
  3. The noise contribution of a filter stays consistent if you keep the capacitor values the same while changing resistors. This knowledge simplifies filter design.

Choosing an op-amp for your project

lcamtuf’s thing • 4081 implied HN points • 03 Jan 25
🕹 Technology Signal Processing
  1. When selecting op-amps for projects, avoid using older models like LM741 and LM324, as modern options perform much better and are easier to use.
  2. Look for op-amps with rail-to-rail input and output capabilities, which allow for better voltage range handling and simplify your circuit design.
  3. Focus on key parameters like bandwidth, output current, and noise specifications, but remember that many modern op-amps have decent performance that meets the needs of most hobby projects.

The marvelous disappearing capacitor

lcamtuf’s thing • 4285 implied HN points • 07 Dec 24
🕹 Technology Signal Processing
  1. Bootstrapping can significantly improve photodiode amplifier performance by minimizing the impact of parasitic capacitance. This helps in amplifying fast-changing signals better.
  2. A voltage follower in the circuit helps keep the photodiode at the same voltage, preventing internal charging and making it act like an ideal current source.
  3. While bootstrapping boosts performance, real-life limitations exist, like bandwidth and impedance, which need to be considered for accurate designs.

Domain Transformations: The Art of Finding Easier Spaces

Software Bits Newsletter • 103 implied HN points • 05 Jan 26
🕹 Technology Signal Processing
  1. Transform hard problems into easier ones by moving to a different domain, doing the simpler computation there, and (if needed) transforming the result back; this is worth it when the transform cost plus the easier computation is less than solving the original problem.
  2. Use well-known transforms to fix numerical and computational issues: log-space turns tiny-product underflow into stable sums (use the log-sum-exp trick to add probabilities safely), Fourier turns convolution into cheap pointwise multiplication, and embeddings or kernels lift data so linear methods work.
  3. Always check that a transform preserves what you need and that the round-trip cost is justified; the best algorithms exploit problem structure by finding the space where the computation becomes simple.

Why does the frequency response of my DAC droop?

filterwizard • 19 implied HN points • 31 Aug 24
🕹 Technology Signal Processing
  1. A DAC's output might not represent the input signal accurately because it holds samples longer than expected. This can result in a drooping frequency response instead of a flat line.
  2. The output is shaped by a sinc function, where certain frequencies lose energy and create unwanted noise, making the signal less clear.
  3. Modern DACs, like sigma-delta types, don't have this droop problem. They use faster processes and digital filtering to provide a smoother, more accurate sound.

Not so fast, Mr. Fourier!

lcamtuf’s thing • 119 HN points • 12 Mar 24
🔬 Science Signal Processing
  1. The discrete Fourier transform (DFT) is a crucial algorithm in modern computing, used for tasks like communication, image and audio processing, and data compression.
  2. DFT transforms time-domain waveforms into frequency domain readings, allowing for analysis and manipulation of signals like isolating instruments or applying effects like Auto-Tune in music.
  3. Fast Fourier Transform (FFT) optimizes DFT by reducing the number of necessary calculations, making it more efficient for large-scale applications in computing.

Bonus post - Noisy Monitors

Sunday Letters • 59 implied HN points • 07 Feb 22
🕹 Technology Signal Processing
  1. Noisy monitors can create problems for engineering teams. They get too many alerts that can drown out the important signals, making it hard to focus on real issues.
  2. Accumulating unresolved bugs and warnings can lead to confusion. Teams might ignore older and less critical issues, which can mask newer problems that need attention.
  3. It's important to maintain discipline and organization in monitoring systems. Just like a chef cleans his kitchen daily, teams should regularly tidy up their alerts and processes to stay effective.

Satellite receivers

Only Wonder Knows • 2 HN points • 12 May 23
🕹 Technology Signal Processing
  1. Satellite receivers use a device called Low Noise Block (LNB) to amplify signals from satellites.
  2. Satellite reception projects involve intricate components like amplifiers, mixers, and oscillators.
  3. The functionality of a satellite receiver is controlled by factors such as polarization selection and band choice.

Post-Transformers - Hyena Hierarchy

Why Now • 8 implied HN points • 04 Sep 23
🕹 Technology Signal Processing
  1. Hyena clans have a linear dominance hierarchy with one-to-one chain of command
  2. LLMs like Transformers face challenges with attention mechanisms due to scaling limitations
  3. Hyena proposes a sub-quadratic solution to attention via long-convolutions and data-controlled gating