Michael Zingale's Avatar

Michael Zingale

@michaelzingale.bsky.social

computational astrophysicist open source tool developer professor blower upper of stars star wars fan coffee aficionado https://zingale.github.io

2,402 Followers  |  542 Following  |  1,322 Posts  |  Joined: 20.06.2023  |  1.8441

Latest posts by michaelzingale.bsky.social on Bluesky

Ralph Wiggem “I’m in danger” meme

Ralph Wiggem “I’m in danger” meme

*Academics in early August

06.08.2025 16:31 — 👍 419    🔁 94    💬 12    📌 22

regex is one of the few instances that I've found ChatGPT to be useful

05.08.2025 18:52 — 👍 1    🔁 0    💬 0    📌 0

anybody try using jupyter-book 2.0 yet? it converted my config files, but I can't figure out how to get it to output a webpage. It will run it's own server, but that's not what I want.

#jupyter
#jupyter-book

05.08.2025 17:37 — 👍 1    🔁 0    💬 0    📌 0

Yeah, it came back.

05.08.2025 16:38 — 👍 0    🔁 0    💬 0    📌 0
Preview
patrick star and spongebob squarepants are standing next to each other in a dark room . Alt: patrick star and spongebob squarepants running around panicy with their arms up

github is down!!!!

05.08.2025 16:00 — 👍 2    🔁 0    💬 2    📌 0

The US could see a 30-40% decline in new international student enrollment, resulting in nearly $7 billion in lost revenue and more than 60,000 fewer American jobs. https://www.nafsa.org/fall-2025-international-student-enrollment-outlook-and-economic-impact

03.08.2025 13:20 — 👍 2692    🔁 1230    💬 132    📌 272
Post image Post image Post image Post image

Loeb: we must stick to the facts, not judge things based on public opinion.

Also Loeb: Look at how many likes and messages my ideas get! I must be doing it right.

02.08.2025 17:59 — 👍 84    🔁 15    💬 4    📌 5
pynucastro — pynucastro 2.7.1.post1+g022525b21 documentation

pynucastro 2.7 has been released:
pynucastro.github.io/pynucastro/

This includes new plot types, fixes to NSE, lots of documentation improvements, new weak rates, and new methods for exploring networks.

#astro
#astrocode

01.08.2025 11:54 — 👍 6    🔁 0    💬 0    📌 0

but even `long double` on Intel (sorry Mac's) is reasonable (80-bit).

31.07.2025 17:48 — 👍 1    🔁 0    💬 0    📌 0

computing derivatives of Fermi-Dirac integrals. The reference paper that I was using used 128-bit math with Gauss-Legendre and Gauss-Laguerre quadrature.

I've been told it helps a lot with reaction networks and approaching equilibrium, to get cancellation, but I've not tried that.

31.07.2025 17:41 — 👍 2    🔁 0    💬 1    📌 0

Aliens?

31.07.2025 17:07 — 👍 3    🔁 0    💬 1    📌 0

been having fun with 128-bit floating point. Why can't hardware move in this direction instead of FP16 ... :(

31.07.2025 16:27 — 👍 4    🔁 0    💬 1    📌 0

Wow!

30.07.2025 21:51 — 👍 5    🔁 0    💬 0    📌 0

that's great Taka! thanks for sharing.

30.07.2025 12:58 — 👍 1    🔁 0    💬 0    📌 0

before looking closely, I thought these were images of old Cray supercomputers.

30.07.2025 12:36 — 👍 1    🔁 0    💬 1    📌 0

thank you for sharing!

30.07.2025 12:29 — 👍 1    🔁 0    💬 0    📌 0

you could fit at least a cubic to that!

28.07.2025 23:28 — 👍 1    🔁 0    💬 1    📌 0
a reaction network visualization showing the nuclei (represented as nodes) connected by rates (represented by edges).  The color of the nodes indicates the abundance -- here He4 is the most abundant.  The color of the edges represents the reaction rate strength, with 3 He4 -> C12 being one of the strongest rates.

a reaction network visualization showing the nuclei (represented as nodes) connected by rates (represented by edges). The color of the nodes indicates the abundance -- here He4 is the most abundant. The color of the edges represents the reaction rate strength, with 3 He4 -> C12 being one of the strongest rates.

new in pynucastro -- you can now color nuclei by abundance and rates by reaction rate strength easily:

pynucastro.github.io/pynucastro/p...

combined with the ability to read simulation data from a plot file via yt, this makes it easy to show reaction flow in simulations.

#astro

28.07.2025 18:46 — 👍 5    🔁 1    💬 0    📌 0

I've always said that any shake of the boggle-game dice generates a valid perl program

28.07.2025 15:18 — 👍 1    🔁 0    💬 0    📌 0
List of Runge–Kutta methods - Wikipedia

again, wikipedia!

en.wikipedia.org/wiki/List_of...

although it only seems to have SSPRK3. Maybe we should edit?

26.07.2025 13:46 — 👍 1    🔁 0    💬 0    📌 0

I asked it for an 8th-order centered difference yesterday. It kept giving me wrong answers. I had to keep asking if it was sure, and each time it said "you're right .. here's a fixed version".

Never got it right

Wikipedia has a handy table of the coefficients.

26.07.2025 01:59 — 👍 1    🔁 0    💬 1    📌 0

huh... Universities like "beyond". We did this ~ 10 years ago:
news.stonybrook.edu/newsroom/pre...

24.07.2025 14:04 — 👍 0    🔁 0    💬 1    📌 0

It's so sad. Most countries are not providing nearly enough investment to absorb the losses. I think that China has a lot of potential to lead now.

24.07.2025 14:03 — 👍 3    🔁 0    💬 1    📌 0

floating point is the worst!

24.07.2025 13:53 — 👍 5    🔁 1    💬 1    📌 1

DAVE: Open the podbay doors, ChatGPT.
CHATGPT: Certainly, Dave, the podbay doors are now open.
DAVE: The podbay doors didn't open.
CHATGPT: My apologies, Dave, you're right. I thought the podbay doors were open, but they weren't. Now they are.
DAVE: I'm still looking at a set of closed podbay doors.

09.06.2025 18:04 — 👍 11041    🔁 2635    💬 116    📌 136

I remember when they used to sell those at the AAS meeting (in that booth with all the other nerdy stuff)

24.07.2025 01:45 — 👍 4    🔁 0    💬 1    📌 0

yeah, I think so. Isn't Sunday sauce the best...

22.07.2025 18:12 — 👍 1    🔁 0    💬 1    📌 0

is the postdoc market just awful this year?

#astro

22.07.2025 18:10 — 👍 4    🔁 0    💬 1    📌 0

We should have listened when the modems screamed at us.

22.07.2025 14:30 — 👍 11370    🔁 2546    💬 122    📌 90

I thought we were supposed to call it gravy?

22.07.2025 16:01 — 👍 1    🔁 0    💬 2    📌 0

@michaelzingale is following 19 prominent accounts