ThunderScope: The Open-Source Oscilloscope That Moves the Display and Processing to a PC
language
eng
date
Aug 12, 2026
slug
thunderscope-open-source-pc-streaming-oscilloscope
author
status
Public
tags
Hardware
Open Source
오실로스코프
Product
summary
ThunderScope is an open-source oscilloscope that streams four channels of 1 GS/s data to a PC in real time over Thunderbolt, USB4, or PCI Express. This article explores its PC-based display and processing architecture, the ngscopeclient and TS.NET software stack, multi-unit synchronization, and the potential for custom gateware.
type
Post
thumbnail
category
🤖 Computer Science
updatedAt
Aug 13, 2026 08:15 PM
Recently, I have been measuring and validating more high-speed signals at 25 Gbps and above, often saving waveforms and signal information as images and data. Most modern oscilloscopes let you connect external storage or save to internal storage, then export screenshots and waveform data over USB. But plugging in a USB drive and copying files every time became tedious, and even operating the oscilloscope by hand just to inspect a waveform started to feel cumbersome. That was when I came across an oscilloscope with a rather unusual concept: ThunderScope. Its bandwidth is not sufficient for directly validating the eye diagram, jitter, or rise time of a 25 Gbps signal. Although the inconvenience I experienced in 25 Gbps measurement work led me to this product, ThunderScope's real strengths lie closer to PC-based long-duration acquisition, protocol decoding, data post-processing, and measurement automation.
ThunderScope is an oscilloscope with a high-performance analog front end, but it does not take the form of a conventional standalone instrument. It comes in two versions: a portable ThunderScope connected via Thunderbolt or USB4, and ThunderScope PCIe for desktop systems. It samples at up to 1 GS/s in 8-bit mode and 500 MS/s in 12-bit mode. According to the official specifications, its analog bandwidth depends on the number of active channels: 350 MHz with one channel, 200 MHz with two channels, and 100 MHz with four channels. RF experiments can use up to 500 MHz of analog bandwidth with the anti-aliasing filter disabled. The Crowd Supply campaign completed funding in November 2024. According to the official July 2026 update, the Developer Edition and probe kit were nearing shipment, while assembly and calibration of the PCIe model and enclosure assembly of the USB4 model were underway. It is better described as being in early production and fulfillment than in beta testing.
ThunderScope is more than an oscilloscope that can simply be connected to a PC. As its name suggests, it connects over Thunderbolt, USB4, or PCI Express and delivers a 1 GS/s sample stream to the PC at more than 1 GB per second, leaving computation, display, and storage to the host system. Many standalone oscilloscopes confine users to their built-in compute resources and storage, as well as the user interface and software features chosen by the manufacturer. ThunderScope moves the small screen and physical controls outside the instrument and onto a PC, with the ambitious goal of using the PC's CPU, GPU, and storage to “unlock” those constraints.
Even more remarkable is that open source and the open-source community sit at the heart of the project.
Contributors from a variety of backgrounds work on different parts of the project. Source code is available for many components, including the hardware design, FPGA gateware, drivers, TS.NET, and ngscopeclient, so those with sufficient expertise can build and improve them directly. As the architecture transitions to LiteX, some repositories—such as the earlier XDMA implementation—have been archived. This is better understood as a change in implementation generation than as the project becoming closed source. The principal repositories continue to receive updates.
Official demonstrations include capturing and decoding USB MIDI transactions, as well as capturing a radio signal while simultaneously probing and decoding an SPI bus. I would like to see a broader range of real-world applications, but perhaps because the product is still in early production and fulfillment, examples posted by ordinary users remain difficult to find on YouTube.
Future Expandability
The product page emphasizes that ThunderScope can be expanded and customized even after purchase. An external 10 MHz reference clock can synchronize its sampling reference with other equipment. A pairing feature that links the clocks and synchronization signals of multiple ThunderScope units to increase the channel count is also under development, though it is too early to regard it as a finished production feature. Because the FPGA gateware and build process are public and JTAG access is available, advanced users can build and load their own bitstreams. This is not a simple plug-in feature, however; it is advanced customization for users who understand FPGA development environments, the hardware architecture, and calibration. Unlike unofficially hacking the FPGA in a commercial oscilloscope, ThunderScope is designed from the outset to let users inspect and improve both its hardware and software.
A PC Software Stack Built Around ngscopeclient and TS.NET
Processing data with PC hardware offers a different range of possibilities and scalability from a conventional standalone oscilloscope. ngscopeclient is an independent open-source instrument client used as ThunderScope's primary user interface. It provides waveform visualization, filter graphs, protocol decoders, and GPU-accelerated post-processing, and it can connect to other instruments as well. TS.NET is the PC host engine that receives ADC sample streams from ThunderScope, performs triggering and basic signal processing, applies calibration data, and exposes SCPI control and data sockets to external applications. Actual processing performance will depend on the PC's CPU, GPU, storage, and interface environment, but the ability to upgrade those host resources is itself an important advantage of ThunderScope.
Another promising area is measurement and test automation. TS.NET's SCPI and data sockets make it possible to control the hardware and acquire the required signals in code. Because ngscopeclient can work with several types of instruments, it should also be possible to automate and integrate measurements involving a function generator or other test equipment. Rather than describing this as a single .NET SDK, it is more accurate to say that ThunderScope provides an open-source host engine together with standardized control and data interfaces.
Crowdfunded purchases are sometimes viewed somewhat negatively in Korea, but this is the first piece of hardware in a long time that has made me want to participate in a campaign myself.
There are still risks: the product remains in early production, while its software and drivers continue to evolve. Its bandwidth also means that it cannot replace equipment intended for high-speed signal-integrity analysis. Even so, the ability to inspect the hardware and software, automate the measurement environment in code, and extend functionality with PC resources offers an appeal that is difficult to find in conventional oscilloscopes. If ThunderScope is viewed not as a finished instrument but as an open-source measurement platform that its users can help develop over time, its direction looks quite compelling.
Related posts
My First PCB Prototype with AI: The Connector Was Harder Than the Design
Aug 31, 2026
I designed and ordered my first small auxiliary PCB for work. AI accelerated circuit design and production preparation, but connector sourcing and validating the board in the real world still required careful manual work.
Cerebras WSE-3: Why Turn an Entire Wafer into a Single AI Chip?
Aug 12, 2026
Cerebras challenges the memory-bandwidth bottleneck in AI inference with WSE-3, a wafer-scale processor integrating 4 trillion transistors and 44 GB of on-chip SRAM. This post looks at the benefits of wafer-scale design, its defect-tolerant architecture, partnerships with OpenAI and AWS, and the competitive landscape with NVIDIA and Groq.
What does the CPU think? A great introductory article on building a 16-bit CPU directly in C.
Mar 15, 2026
This is a text that helps you understand registers, memory, ALU, flags, branching, stack, and assembler while directly implementing a 16-bit CPU in C code. It is especially recommended for those who want to connect computer architecture not through abstract explanations but through working code.