What I've done wrong?

@AjayMishra You want E not e.

Yep

The system doesn't provide any definitions for symbols starting with lowercase (ASCII) letter. So, it's E, Pi etc.

@b3m2a1 Thanks. But that's weird.

@AjayMishra If you want the lower-case looking e use esc ee esc

As in press escape then type ee then escape

Is there anyway, by which I can visualize fluid flow in 3D in mathematica. Just like one can do 2D fluid flow simulation at anvaka.github.io/fieldplay ?

@AjayMishra Weird but consistent

@ChrisK Calling Mma perfectly consistent would be a great overstatement, but when compared with competition... well, using any other system would induce considerably increased hair loss. (Or maybe it's just that after almost 25 years of use, Mma begins to feel natural.)

@kirma It is consistent in the sense that all built-in symbols start with an uppercase letter.

I'm using Python a lot these days, and not even the parts of the ecosystem that are supposed to work together, like Numpy and Pandas, are nearly as consistent (taken as a whole) as Mathematica.

how do you submit a question to those cool webinars wolfram u has?

I keep jamming enter, and nada haha

nevermind, my machine is a trickster

hi ramesh

Hi kglr

I have been trying to be in touch with you for the last six months.

how so?

Where are you from? Are you in the US?

i am in cupertino, ca.

Ok, then I am Humboldt, CA.

that's where teach?

I am trying to earn respect for my words: here is about me

business.humboldt.edu/people/ramesh-adhikari

very nice to know.

Can we chat over the phone? I can give you my number

sure...

You can call me now: 7072963883

Our offerings for solid-state chemistry are minimal right now.

I've not used CRYSTAL before, I will definitely look at it.

Would like to know what kind of things users would want to do with this type of a data structure, that would help in designing it.

Sorry I didn't answer all the questions there - was my first time doing a webinar and I was left to the end

Hah no worries @JasonB. it was my first time making it on time to one, hah!

We used crystal to model material interactions, give it a unitcell and propagate it, then test material parameters like sheer or what have you, older system, could be a neat thing to integrate/mimic/operate with, within mathematica and the wolfram language

I would like to note, also @JasonB, that we used it more on a quantum chemistry level, more like a physical chemistry simulator? I worked in a solid state physics group, they would do things like bringing strands of DNA closer to large Au nanodot, large like 70-90 atoms, not physically large, haha

Looking at the documentation for CRYSTAL, it does a lot of heavy lifting - wave-function and DFT methods for crystals just to start with

@JasonB. @CATrevillian yeah CRYSTAL is definitely not something I think of as a local-PC type package...

Right now I want to make WL the place to visualize and analyze results from electronic structure calculations

@JasonB. if you have the results from electronic structure I can give you anharmonically corrected vibrations, too.

Or any number of other nuclear-wavefunction-derived properties

When the paclet repo finally goes live I can contribute a bunch of that stuff

I also have code to pull potentials and force constants and all the rest of it out of Gaussian .log files and things that I could hook up to your Molecule class when I get the time.

@b3m2a1 you would do that based on the wavefunction, just the basis set and expansion coefficients?

or you would need the results from a harmonic vibration job?

@JasonB. honestly if you give me a potential surface I don't even need the harmonic vibration stuff

But if you give me the force constants already I can do it faster

If you give me force constants and 3rd and 4th derivatives I can do it even faster

but if you just had the orbitals - if you had the basis set and the c matrix, you have WL code to get the potential terms from that?

and also you knew you were at the minimum geometry

If you want to use some more empirical force field, too, there is nominally nothing preventing me from working with that.

@JasonB. Ehhhh... no not really. I just let electronic structure be its own little black box.

I mean nominally there's nothing stopping me from looking it up and implementing it, but I don't have anything prebaked like I do for the rest of it.

there are so many black boxes, each with their own output formats, would be nice if there were a common interchangeable format

I know CML was put forward as a standard, and has room for most everything these programs do

I agree wholeheartedly. I've written three parsers for three different ES output languages at this point and they're all annoying.

but I don't see any adoption of CML output from electronic structure

The people who write these packages don't really seem to care about portability. Like the fact that people don't just stop at what ES spits out doesn't seem to be a priority.

Even psi4 which is otherwise great in terms of it's ability to write hooks into doesn't play nice.

@JasonB @b3m2a1 Ah! I'm pretty inept at judging machine requirements like that, my usual MO is to just throw the biggest book of qualifications at a problem and shake it really really hard

/me nods sagely as though I understand a word

One thing, then, and it can probably happen, but it would be useful to at least accept results from that platform? Or, vice versa, make visualization and send it off to Crystal? I just like putting two things together is all, and I was laughing to myself that this finally hit V12, at least the visualization part--running the intensive calculations we used a computer cluster, but even still, only were enabled to use something like 2 or 4 nodes tops, seems like something a multicore PC could

handle these days, but then again, that's probably way too much added-functionality for it to be worthwhile. Add onto that I'm not sure where the software is today, having used a version now 10 years old.

@CarlLange hah! See, my mistake (and in-progress sin) is that I start talking with those words too--Here's an example, haha!

Who's got a good way to add an arrowhead? Good, or simple? Or pretty, and complicated? I'll poke around the site when I get home, but I have a 3D graphics primitive wherein I slice a segment of data, scale it, and then put it back into the 3D graphics primitive, so I now realize I have easy access to an endpoint, I should just be able to pop it on there hey? After I submit the abstract, I'll be inclined to share ;) I got a nice display going for 3D spirals so far!

essentially, I am showing direction of rotation :D

And I have no idea how to use tube, apparently, yay! learning! 3d arrowheads, buuuuuuuut but but but I have a line wrapping my data set, it looks so nice! Now I am determined AND curious!