Hacker News Reader

Comment by buildbot | Hacker News Reader

Very cool. Unrelated, but the technology and engineering behind ATLAS is immense. ~98 million data channels streaming out at 2Ghz (at least that is what I recall). Cut down to a manageable level by hundreds of FPGAs in the L1 trigger.

https://cds.cern.ch/record/2745767/files/ATL-INDET-PROC-2020...

View on HN

Yes, the FPGAs! Atlas and other experiments at CERN require such complex and hard-read time requirements, making them great applications of FPGAs. Right now, I am working on a project to try and deploy graph neural networks for track reconstruction and filtering (on the faster L1 trigger), as well as reconstructing other quantifiable measurements using graph neural networks (I believe this is not on the L1 trigger but requires some element of real-time). What's cool is they have copies of the FPGA boards that are at CERN at Fermilab, which makes remote development easier. Also, a fun fact: the detectors that surround the core are not all perfect concentric circles and are a big mix of different geometries and types of detectors, making the problem more interesting.

Oh that is really cool! Way back in my undergrad I worked on a tiny tiny part of the insertable B layer, writing some very bad FPGA firmware to control the T3MAPS test chip (I think this was for Phase 2 upgrades?). FPGAs everywhere in this domain!

And yeah the detectors are arranged in a really complicated way, I can imagine that is a hide headache.

I recall watching Marco Reps' video[1] where he built the (relatively) high-speed high-precision ADC they designed and published as an open hardware project[2], to be used to accurately monitor current in the magnets.

A small but important piece of a giant machine puzzle.

[1]: https://www.youtube.com/watch?v=D28uSzCs7-k

[2]: https://ohwr.org/project/opt-adc-10k-32b-1cha/wikis/home

I love how nearly everything hardware at CERN is published under the open hardware license. I think even the pixel chip designs maybe, but I could be wrong.

Timepix is not public but licensed to commercial vendors as far as I know.

But anyway I appreciate they make such (a readout chip for) detectors available so we could benefit in one of our previous projects (not about particle physics).

The Timepix is a TSMC chip, I do not see how you could make that opensource without breaking NDAs with the fab. Something like the Skywater PDK would be ideal for open science (provided it could perform as well under radiations of course).

Reply on news.ycombinator.com