And it makes me now unsure of the most proper way to Export/DumpSave and then Import/Get lists of numeric data if not with an .mx file. What is the most proper way, considering that the lists may or may not be ragged depending on the method used & information included? Can anyone advise on this?

@CATrevillian If you only need to work inside the Mathematica ecosystem, I think you'll be hard-pressed to do better than Import/Export with MX

@b3m2a1 I suppose that is true, it just seems like I’m being bogged down when using Import to load “many” files (~80x10, which does not seem like it should be considered as so large an amount to cause such a slowdown) due to the need to make definitions every time (according to the linked QA)

@CATrevillian I highly doubt that it's the definitions that make it slow. If you trace an Import call, you'll see that hell breaks loose behind the curtains.

Try the following:

Table[
  Export[
   StringTemplate["~/tmp/export/``.mx"][IntegerString[i, 10, 3]],
   RandomReal[{0, 1}, {500, 500}]
   ], {i, 500}];

This creates 500 files with matrices in ~/tmp/export

Now

files = FileNames["*.mx", {"~/tmp/export"}];
AbsoluteTiming[Import /@ files;]

Is done in 3.2 seconds here

However

AbsoluteTiming[
 Block[{System`Private`ConvertersPrivateDumpSymbol},
     Get[#];
     ReleaseHold[System`Private`ConvertersPrivateDumpSymbol]
     ] & /@ files;
 ]

Only needs 0.3 seconds. So an order of 10 faster. Not sure it's worth it for your case.

@halirutan if it was just a matter of 3 seconds versus 0.3 seconds, I would not be concerned! But it just took me ~6000 seconds to Import the .mx files I am working with. Considering that it constitutes only ~162 Mb, I am thinking there is a better way to go about this?

@CATrevillian The best way would be if you could make an artificial example that you can post as a question so that one could try it.

Hi all,

Is it possible to get all the steps in mathematica of differential equations.

And any algorithm which used by software.

But this is too small, right? You said something about 160MB. That's not even a single MB of memory.

Agh, sorry, having trouble with the formatting today. But that would represent 1 of the .mx files I am to Import, which takes something like ~5 seconds for me (still, this seems too long, or maybe I am asking for too much). Then I have maybe ~80x10 of these, which makes up the 160MB of memory.

@CATrevillian Do you store it on tape or a floppy disk?

@halirutan I store it on an external ssd that is connected via USB-C (so, presumably ~5-10 Gbps transfer rate?)

Look, for one these files of yours I get the following import timing:

In[30]:= AbsoluteTiming[Import["~/tmp/test.mx"];]

Out[30]= {0.021189, Null}

If you have an internal SSD, can you try to save a test-file there an import it? So that we can exclude the USB connection.

If I create 800 of these mx files, I can import back all of them in 0.2 seconds

I think you does not entertain/answer new guy.

@123 Not if "new guy" asks a question that is impossible to answer without further details. In general, the answer is no. You cannot get the steps from solving a differential equation automatically. However, depending on the kind of equation you have (ode, pde?) there might be ways to let Mathematica help you with the steps.

Why don't you ask a question on main where you give details about the differential equation and show how you solved it? It's a perfectly fine question and many more people would see it on the main site.

@user21 Good morning!

I asked this question because you guys are experts. I thought may be you found some solutions.

@halirutan :D means you are saying good bye me from this room in different words. :P

@123 Absolutely not!

@halirutan oh hej, just dropping by.

@halirutan I had the impression that in the last weeks the conversion here died down a bit - but that might just have been a wrong impression

@halirutan oof. It's completely the USB connection. Yowza. WOW....welp. I will consider this many hours gained, and not consider my many hours lost! Haha! Yeeesh. Yeah, one Import costs me 0.0367 seconds now...thanks for your help :D:D:D

@123 The thing is that many of our differential equation experts are not in this chat room, but they read the questions on the main site regularly. So your chances of getting an answer there are much higher. Here in chat, we just joke around and solve one or two USB mysteries..

@CATrevillian Well, well, well..

Problem is that if we don't get the steps. How we believe answer. Because it should have approximation it must error. How we compute the error percentage

@halirutan, now that 12.2 is almost done I hope I can participate a bit more again; 12.2 was a bit of a mad dash for me.... I think I addeed 35 new ref pages - never again, the amount of work to do that, keep them in sync make them useful etc is just crazy.

@halirutan how are things at you end.

@user21 If I step in, the silence is usually gone :)

@ ;-)

@user21 Oh.. I feel you. I guess you enjoyed the good old doc tools.

@123 regarding this in specificity, the software & associated algorithms are properties of WRI, so that is usually not something that users here have knowledge of. Specific questions about methods can likely be answered by a number of individuals here, but asking about specific algorithms they use is usually not a solvable task.

@halirutan Ookay.. It's good to joke for health and day become wonderful.

@halirutan often I just copy cells from other places and edit the raw cell...

@123 error percentage is a different story entirely!

@user21 you're awesome!! I'm excited to see the benefits of this hard work!!

@user21 Otherwise things are good at my end. You know, I live kind of remotely with lots of nature around me. And I can work from home for the past 8 months. I would lie if I'd say I don't enjoy this. I'll certainly miss my Fussball training once a week though.

Thanks @CATrevillian I need to show steps to students. Can we get the automatic steps of simple differentiation and integration (expilict and one/two variable)

@CATrevillian, thanks for the kind words. As always I am anxious that I screwed it up.... Not as many bug fixes as I would have liked to make

@123 Have you seen Rubi?

Once 12.2 is out, I am wondering if people here would like to have (virtual) FEM meetup and discuss current and future development? Is that something that would be of interest?

I have heard of rubi language not learned the coding. I know Javascript and python.

@123 Klick the link. It's a Mathematica package: Rule-based integration

@halirutan thanks.

@user21 Thank God you're anxious as well. I always try to triple-check everything when pushing a new version.

@halirutan, I think I need vacation ;-)

@user21 That's going to be a problem right now if you don't plan to build up a tent inside your home.

@halirutan, oh with vacation I just mean - not work -;) I am fine with staying at home, we are renovating a house, so there is plenty to do and keep my mind off of work things.

@halirutan, do you feel like a chat?

@user21 I heard from undisclosed sources that having 4 kids helps tremendously. Then, you just wish you could work for 5min straight :) Last night, I was playing Minecraft with them on 3 computers over network. That was nice though.

@user21 Sure. I have coffee meeting in a bit but I'd be ready at 10 our time.

OK, I'll send you a link

via email

Yep.

@halirutan You did an astounding job for the community with the highlight.js PR. I absolutely would not have been able to stick with something for so long. Well done and thank you.

@CarlLange Believe me, it's hard for me as well. Especially if it's only one of many side projects. No matter how endless the discussions there seem to be, Josh (the maintainer) was very reasonable and helpful. We had a nice long Zoom chat and he really tried hard to understand the nuances.

And along the way, we worked out some underlying inconsistencies and problems in Highlight.js itself. And it seems he wouldn't be opposed to having me in the team a bit longer :)

github.com/highlightjs/highlight.js/issues/…

But hell, how I hate JavaScript.

@halirutan I did a double take there, since in US that's what we call "table soccer" aka foosball <https://en.wikipedia.org/wiki/Fussball>

@ChrisK Point taken. But I knew user21 speaks my language :)

@ChrisK I've heard it called Kicker in Germany, which I find much more of an entertaining name than foozball :D

@halirutan Yeah, it was a really impressive open source conversation I think! Really well done.

@CarlLange Can't beat the French "Baby foot"

our daughter went to a french summer camp and "Baby foot géant" was on the social agenda. We were like WTF is "baby foot giant". Turns out it's foosball played with humans!? Strangely circular

@ChrisK I'm definitely calling it that from now on. And table tennis? Baby hand.