Embedded Forum - Recent Threads

RE: Find Wireless ARM MCU

cylee_acc — Mon, 02 Mar 2026 04:37:19 GMT

It’s an academic project on a miniaturized telemetry system using chiplet/2.5D integration. We’re considering bare die for advanced packaging research, since packaged parts add footprint and parasitics. This is for low-volume prototyping, and we’re exploring whether KGD or wafer-level options are available.

Find Wireless ARM MCU

cylee_acc — Wed, 11 Feb 2026 01:49:09 GMT

Hello everyone,

I’m looking for recommendations for a wireless ARM-based MCU for a research prototype.

Target specifications:

32-bit ARM MCU with operating frequency > 78 MHz
On-chip ≥ 512 KB Flash and ≥ 64 KB RAM
Integrated low-power wireless radio (Sub-GHz preferred)
Integrated power amplifier ¶ supporting up to 20 dBm TX output
Small package options preferred (QFN or WLCSP)

In addition, I’d like to know whether any suitable devices are available as bare die, known good die (KGD).

If possible, please also share links to datasheets, reference designs, and current availability.

Thanks in advance for your help.

RE: Find Wireless ARM MCU

balajivan1995 — Thu, 12 Feb 2026 07:23:01 GMT

I think this chip is closer to your requirement.

www.st.com/.../stm32wl-series.html

RE: Find Wireless ARM MCU

BigG — Wed, 11 Feb 2026 15:40:16 GMT

Mainstream - Microchip WLR089U0 and TI's CC111x, etc.

Niche - https://miromico.ch/en/products/modules

Otherwise can find Semtech modules bundled with MCU's.

RE: Find Wireless ARM MCU

michaelkellett — Wed, 11 Feb 2026 09:18:26 GMT

If you need information of that sort you will need to establish dialogue with the manufacturer or possibly their local agent.

Bare die only make sense in special circumstances (usually only in very large numbers) and are rarely stocked by distribution.

What is your project, and why do you want to use an un-packaged device ?

RE: Find Wireless ARM MCU

cylee_acc — Wed, 11 Feb 2026 05:30:14 GMT

Thank you for the suggestion.

We are currently prioritizing Sub-GHz solutions for range considerations. However, if Nordic offers devices with similar specs and especially bare die or KGD options, I would be interested in reviewing them.

RE: Find Wireless ARM MCU

cylee_acc — Wed, 11 Feb 2026 05:29:06 GMT

Thank you for the suggestion.

I have reviewed the STM32U series; however, I was not able to find any information indicating the availability of a bare die or known good die (KGD) option.

Could you please confirm whether a bare die or KGD version is available for any devices in the STM32U family?

Thank you for your assistance.

RE: Find Wireless ARM MCU

dougw — Wed, 11 Feb 2026 05:17:14 GMT

Nordic Semiconductor has some nice MCUs with wireless, but usually it is 2.4GHz.

RE: Find Wireless ARM MCU

embeddedguy — Wed, 11 Feb 2026 05:06:19 GMT

Have you seen STM32U series?

STM32U5 series of Ultra-low-power MCUs Enhanced Security for IoT and Embedded Applications - STMicroelectronics

RE: Is fancy code completion really worth it ?

metaprog — Mon, 05 Jan 2026 09:41:13 GMT

It's no different from human-written code though. And yes, you absolutely can check every path in the generated code, this is what formal verification is for.

Is fancy code completion really worth it ?

michaelkellett — Wed, 10 Dec 2025 10:35:26 GMT

Of credible AI horror stories I think this is about as bad as it gets !

https://www.tomshardware.com/tech-industry/cyber-security/researchers-uncover-critical-ai-ide-flaws-exposing-developers-to-data-theft-and-rce

I'm sticking with the "legacy" Keil IDE for C coding on ARM - it may be a little slower but at least I'm fairly sure that nothing gets into my code that I didn't put there !

RE: Is fancy code completion really worth it ?

shabaz — Sun, 04 Jan 2026 06:25:02 GMT

I use vi too : )

(Admittedly only for short pieces of code or edits).

I can't imagine explaining to a non-tech person how vi is like MS Word, but no buttons to click on, and no icons, i.e. not even Teletext level. A bit like trying to explain that the Internet was text-only for some people too, until Netscape Navigator came along... and pictures were often created in text characters, because it took too long to transfer images, and there was nowhere to store them for long, apart from floppy disks (or print on fanfold paper!). My e-mail account storage limit was 150 kbytes at uni.. (that was for all emails combined : )

Nowadays there's AI for vi (I've not tried it).

I don't know if you've ever watched Halt and Catch Fire, but perhaps you might like it. It's fictional, but many elements are close to reality. I really enjoyed it.

RE: Is fancy code completion really worth it ?

DAB — Sat, 03 Jan 2026 19:28:56 GMT

Unless you can check each path, you still never know what the AI is doing or what issues generate the answers.

Risk Analysis 101, if you cannot verify the results, you do not trust them. Ever!

RE: Is fancy code completion really worth it ?

metaprog — Fri, 02 Jan 2026 09:58:10 GMT

There is one way AI can earn way more trust than humans (your local ones or third party somebody) - AI is great at writing proofs. If you generate a certified code with AI, it outperforms humans already. Writing proofs is way too tedious for humans, but AI have infinite patience. I generated fully formally proven SystemVerilog modules with AI, fully certified code in SPARK and Coq. And I only ever use local models for this.

RE: Is fancy code completion really worth it ?

metaprog — Fri, 02 Jan 2026 09:52:42 GMT

AI is great. Just always use the local models, they are more than powerful enough. Do not forget to use tools - give LLMs an access to LSP (clangd is ok), to a sandbox with simulation and tests, etc.

RE: Is fancy code completion really worth it ?

dougw — Thu, 01 Jan 2026 01:10:02 GMT

Fake news seems to be a popular application for AI. Perps don't even care if the AI is hallucinating.

RE: Is fancy code completion really worth it ?

bradfordmiller — Thu, 01 Jan 2026 00:05:26 GMT

You forgot about the companies that will exploit AI to make their competitors look bad. "It's not enough to win, you have to make everyone else look foolish for trying..."

RE: Is fancy code completion really worth it ?

dougw — Wed, 31 Dec 2025 20:54:54 GMT

AI is still at the scary stage where it cannot be trusted, but you absolutely need to figure out how to make it work in order to stay competitive from a productivity point of view.

There will be many companies that forbid the use of AI because the risks are too high.

There will be companies that use AI and run into costly or disastrous situations.

There will be companies that successfully figure out how to leverage AI to improve productivity and corporate performance by at least an order of magnitude.

There is no question which companies will win in the end.

RE: How do you decide which CVEs actually matter in embedded products?

dougw — Wed, 31 Dec 2025 20:44:24 GMT

Almost by definition critical vulnerability means it needs to be addressed. It boils down to assessing whether it is critical or not. For us it means everything critical gets addressed and then we address as many non-critical issues as we have time and budget to tackle. In some cases critical issues have to be addressed with operational procedures as opposed to hardware or software fixes. The actual decision about what is critical is usually assessed using a risk vs cost matrix (if managers are involved). If it is just engineers, they have their own priorities, based on what they are willing to sign off on.

How do you decide which CVEs actually matter in embedded products?

micheal.embedded — Wed, 31 Dec 2025 12:53:36 GMT

In embedded products, patching every CVE often isn’t realistic due to certification, long test cycles, limited update windows, or system stability risks.

I’m curious how others handle this in practice. When a CVE affects a component in your system, what factors determine whether it gets patched immediately, deferred, or accepted as risk?

Do you prioritize based on exploitability, attack surface, physical access assumptions, update cost, or something else?

RE: Is fancy code completion really worth it ?

DAB — Wed, 31 Dec 2025 19:24:26 GMT

I find it is always best to use what you know.

If you try to learn the latest thing out, you spend all of your time as a just learning developer.

Give me someone who knows what they are doing and have years of experience in understanding the tools they use.

RE: How do you decide which CVEs actually matter in embedded products?

BigG — Wed, 31 Dec 2025 18:31:57 GMT

Great question. Everything needs planning but common sense should still prevail.

Wasn't it Warren Buffett who once said "It takes 20 years to build a reputation and five minutes to ruin it. If you think about that, you'll do things differently."

Hence be proactive. Trying to recover from the scenario where someone finds a critical bug that you've been ignoring, or worse, trying to hide and it's generally tickets.

The goal isn't to have zero CVEs (which is impossible); it's to try to have zero surprises.

RE: Is fancy code completion really worth it ?

colporteur — Wed, 31 Dec 2025 16:41:42 GMT

I still use Vi. I won't try and argue merits. I cut my teeth with Ed in the infancy of editors. Vim upgrade provided some options. Something to be said for confidence in your tools.

RE: How do you decide which CVEs actually matter in embedded products?

colporteur — Wed, 31 Dec 2025 16:35:45 GMT

My experience is five years old and may be stale dated.

Our dev environment conditioned what happened in production. It also made it difficult to know the condition of dev with all the testing.

Level of exposure was one deciding factor. Yeah sure it is possible but how possible in our environment. Having more than one mind to do the evaluation helped considerably. It help balance the focus of I might get fired if it happens here against we are to good for it to happen here.

RE: How do you decide which CVEs actually matter in embedded products?

obones — Wed, 31 Dec 2025 13:11:57 GMT

I'm trying to assess the risk, with a "score" given to all parameters.
For instance, if it affects a one off demonstration project that is seldom connected to the network, I'm 99% sure I'll simply ignore the CVE.

But if it impacts devices located at client's location and the CVE may given unwarranted access to their network, then I'm in favor of issuing an update (OTA, ideally). But then, it's the client's responsibility to apply it, and we've all seen what happens in real life.

The update cost is rarely taken into account, because the liability clauses in the contract mean that I would bear the cost of any wrongdoing arising from me not acting against a reported issue.

In the end, if it's a CVE that needs physical access, the default password not changed, and allows to simply change the appearance of the GUI, well, tough luck. I mean, if there is already physical access for the attacker, there are much more interesting things to compromise than defacing a product.

I've also had a Java CVEs that required remote access to give root access to the attacker. Very much an issue, quite high CVE index (8-9) but I never patched it. Why? Because the device does not have any network interface whatsoever, so the attack cannot happen to begin with. But that's a rare situation, now that everything is (has to be?) connected.