Daniel Strano

Formally educated in physics, works as a full stack developer for Unitary Fund, interested in video game development, GPGPU, and quantum computation

[Daniel Strano @Dan_Strano]

It's amazing how QC simulator developers can report "quantum supremacy" experiment fidelity, and then the hardware providers argue its invalid because the noise profile is nonphysical, and we're all like, "Yeah, that makes sense: the wrong part of the answer is wrong!"

[Daniel Strano @Dan_Strano]

@Grownded @WKCosmo Made sense when I learned it via the physics department. Some of us would rather not invest the energy to learn (complicated, formal) subjects that list as selling point that have no application to human life and physical reality.

[Daniel Strano @Dan_Strano]

@WKCosmo In my first high school calculus class, I solved (I believe successfully) the questions on the first test with pure geometry; I saw no motivation for calculus. I was advised to drop the class.

It was when I took physics, that I suddenly saw the point; that made it make sense.

[Daniel Strano @Dan_Strano]

We quote Noether as source, but it seems like it's to support the opposite of the point she proved.

[Will Kinney @WKCosmo]

Two things I find graduate students are almost universally unwilling to accept are:

— Energy isn't conserved in cosmology

— Black holes are vacuum solutions

Not sure what that says about how we teach physics.

[Daniel Strano @Dan_Strano]

Get started with (and master) #Qrack in about 3 seconds! (You might need to install OpenCL, first.)

https://t.co/PueV27Ky5g

[Daniel Strano @Dan_Strano]

#PyQrack passed 1/2 a million recorded downloads, today. (#Qrack)

[nick farina @nick_farina]

this is INSANE; the city is red and you can smell smoke. it looks and feels like the city itself is burning to the ground.

[Claudia Gonzaga-Jauregui @cgonzagaj]

"Bullying is a means for mediocre scientists to rise to the top. Some star academics reached their position because they are bullies, not in spite of it." - Excellent piece about bullying in academia (and really any other professional environment). https://t.co/JTqGBaAuuf

[Daniel Strano @Dan_Strano]

By the way, Schwarzschild metric as a canonical solution to the Einstein field equations lacks accounting for thermodynamic evaporation... which seems wrong, by today, if regarded as a complete model in itself.

[Daniel Strano @Dan_Strano]

Q&A:

"Dumb" Q: Why aren't Schwarzschild coordinates valid on the interior?

"Smart" A: They just make no physical sense.

My "dumb" 2019 A: They actually do, if we accept the sign flip on "g_tt" and "g_rr" and effectively on the metric tensor signature.

https://t.co/WOzSISgtf1

[Daniel Strano @Dan_Strano]

Was this confused hobbyist work? That's what I was told, and maybe.

It "looks like" FLRW, at least superficially, when this code runs, by the way. (And it's de Sitter.)

[Daniel Strano @Dan_Strano]

By the way, avail yourself of the common misconception that CUDA is inherently faster; in single-GPU DFT benchmarks to 2^27 amplitudes in width, both execute in almost exactly the same time. (We produced QEngineCUDA by an almost line-for-line API translation from QEngineOCL.)

[Daniel Strano @Dan_Strano]

#Qrack has always supported #OpenCL acceleration, but now you get optional #CUDA, as well! As a result, we should be able to leverage NVIDIA's data transfer hardware gimmicks like NVLink and SLI, in Qrack's default optimal layer stack (with "QUnit")!

https://t.co/S3FswQ5Ev9

[Daniel Strano @Dan_Strano]

NAND prices are cratering at the moment, and that's half the system build!

We also gave a total debugging and parallelization overhaul on the (CPU-based) quantum binary decision tree ("QBdt") layer... 🤔

(Search "Robert Wille" for a line on the research that inspired our QBdt.)

[Daniel Strano @Dan_Strano]

By the way, we realized around Christmas that 8 NVMes in RAID 0 configuration (with at least 4 PCIe lanes apiece) as Linux swap disk can just barely or almost keep full utilization pressure on a recent 64 hyper thread AMD running Qrack, (but ask @twobombs).

Better than DIMMs! 🤣

[Chad Orzel @orzelc]

I've been a professor for 22 years now, and written five pop-science books, but when I need to type "Schrödinger" I still copy-and-paste it from a Google search as the fastest way to get the dots over the o in the right place.

[Daniel Strano @Dan_Strano]

I mean, for example, Sanderson clearly had a better theoretical physics model of atomic electronegativity, as far as theoretical physics goes, but Pauling wasn't grossly wrong, and Pauling's scale is still seemingly in more common usage.

[Daniel Strano @Dan_Strano]

Believe it or not, a hypothesis doesn't have to "come from anywhere" in order to make a falsifiable prediction that can ultimately be proven true or accurate, empirically. (You can absolutely just literally guess.)

[Daniel Strano @Dan_Strano]

(Depending on how bad of a guesser you are, you might empirically select many null hypotheses, this way.)