I know newsletters are all the rage, but with a bit of light organizing (or even without), your electronics content would make great web content. I'd love to be able to still find your writing on electronics in the future, and I feel like 8 years from now, substack content is going to be buried. If it were also on the web, it would be easier to find, bookmark, etc.
Fair enough. I guess the difference for me is mental. Would I go read a bunch of articles on a really good electronics site? Yes. Would I go dredge through someone's back catalog of substack newsletters? Not likely. Would I go read through back articles on what looked like a blog? Somewhere in the middle. [Even as I write this, clearly, I don't make sense!]
I suppose if I was linked to one of your electronics articles, I'd probably/hopefully realize it was unlike normal newsletters, and go exploring.
I guess I'm just prejudiced against newslettery things 😂
Good article, but I think this sentence is misleading:
> The beauty of coaxial cables is that virtually the entire electromagnetic field is contained within the structure of the cable, so the signal can’t take any shortcuts — even if the coax is coiled on the workshop floor
The reason it's misleading is about "signal can't take shortcuts", because coax cables radiate a lot, here 3 examples with H-field probe and +20 dB LNA measuring the leakage of cable shown on oscilloscope:
Even though this signal is "weak", lot of side channel attacks are based on EM emanations like these. For example devices like ChipWhisperer and ChipShouter.
Sure, I wasn't trying to imply that coax is secure - just that it is designed to contain the field very well, and that there is no way for the input signal to make it to the other side with any practically relevant amplitude through some unintended "side" coupling in a 100 ft spool of coax - not that you can't snoop on it with sufficiently sensitive equipment.
Nope, cool! It's nice that we're finally getting some options that don't cost an arm and a leg.
On the blog, I'm generally trying to stick to real-world demonstrations to explain the "why" in simple terms, but I'll check it out for other reasons =)
Thanks for explaining this so well, I've learned something from you again.
I know newsletters are all the rage, but with a bit of light organizing (or even without), your electronics content would make great web content. I'd love to be able to still find your writing on electronics in the future, and I feel like 8 years from now, substack content is going to be buried. If it were also on the web, it would be easier to find, bookmark, etc.
Substack is primarily just a blog-like website. You can access every article by a stable URL (e.g., https://lcamtuf.substack.com/p/signal-reflections-in-electronic), there is chronological archive (https://lcamtuf.substack.com/archive), etc. Maybe I'm missing the point, but what else would you like to see?
The subscription feature is there and some people use it, but it's a roughly 10:1 ratio between web and email viewers.
Fair enough. I guess the difference for me is mental. Would I go read a bunch of articles on a really good electronics site? Yes. Would I go dredge through someone's back catalog of substack newsletters? Not likely. Would I go read through back articles on what looked like a blog? Somewhere in the middle. [Even as I write this, clearly, I don't make sense!]
I suppose if I was linked to one of your electronics articles, I'd probably/hopefully realize it was unlike normal newsletters, and go exploring.
I guess I'm just prejudiced against newslettery things 😂
Good article, but I think this sentence is misleading:
> The beauty of coaxial cables is that virtually the entire electromagnetic field is contained within the structure of the cable, so the signal can’t take any shortcuts — even if the coax is coiled on the workshop floor
The reason it's misleading is about "signal can't take shortcuts", because coax cables radiate a lot, here 3 examples with H-field probe and +20 dB LNA measuring the leakage of cable shown on oscilloscope:
No coax cable in "empty space", no device nearby: https://matrix.envs.net/_matrix/media/v1/download/matrix.org/ZqFxmXgpMbsoHdNpEtXGgXpc
H-field probe near a coax carrying signal, without proper termination: https://matrix.envs.net/_matrix/media/v1/download/matrix.org/RmBZVRvOgNFkqYobdYvaAXSh
H-field probe near a coax carrying signal, with 50 Ω termination: https://matrix.envs.net/_matrix/media/v1/download/matrix.org/JjhBGYwbumQaLXIWFlYRuazG
Even though this signal is "weak", lot of side channel attacks are based on EM emanations like these. For example devices like ChipWhisperer and ChipShouter.
Sure, I wasn't trying to imply that coax is secure - just that it is designed to contain the field very well, and that there is no way for the input signal to make it to the other side with any practically relevant amplitude through some unintended "side" coupling in a 100 ft spool of coax - not that you can't snoop on it with sufficiently sensitive equipment.
I admit I could've used more precise language!
Have you seen the stuff at https://antmicro.com/blog/2023/11/open-source-signal-integrity-analysis/ ?
Nope, cool! It's nice that we're finally getting some options that don't cost an arm and a leg.
On the blog, I'm generally trying to stick to real-world demonstrations to explain the "why" in simple terms, but I'll check it out for other reasons =)
Ah ha! Thank you.
Honestly, still, many, many years later one of the absolute best technical authors we have. Amazing piece.