Complex "calculated" question

Complex "calculated" question

by Marko Pinteric -
Number of replies: 55

I use calculated questions because I want each student to receive the problem with different data.

Now I have a different type of problems whose solutions can only be obtained by solving transcendental equations.

I would like to provide my own list of wildcard values as well as the answers to the questions (which are calculated externally by a Python program) and upload them to Moodle. Is this possible?

Average of ratings: -
In reply to Marko Pinteric

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
"External service" sounds like a candidate for STACK. But without specific example questions, one can never know. Depending on the "complexity" you might escape with the Formulas question type.
In reply to Visvanath Ratnaweera

Re: Complex "calculated" question

by Marko Pinteric -
The simplest possible question would be:

For a temperature {th}°C and relative humidity {phi}%, read the specific enthalpy h and mass ratio of water vapour and dry air x from a psychrometric diagram.

The formulas for h and x are too complex to be simply put in solution box of Moodle calculated question. So I could calculate a table

th1 phi1 h1 x1
th2 phi2 h2 x2
th3 phi3 h3 x3

myself and Moodle will only have to replace {th} and {phi} in the text of the problem and check {h} and {x} values with a certain relative error.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
This one is a clean candidate for a Formulas question. But one shouldn't go by the simplest possible. What if the next question can not be handled by Formulas?

But judging by the "calculated" in the subject, I think you can survive with Formulas. You don't expect symbolic mathematics, do you?
In reply to Visvanath Ratnaweera

Re: Complex "calculated" question

by Marko Pinteric -
Well, since I need to write a Python script to calculate h and x, I would probably do all the other calculations in that script as well. So, yes, a solution that doesn't provide symbolic maths is fine by me.

By the way, I do not currently have Formulas questions installed on my system. I need something that works for sure so I can get our system administrator to add it to Moodle.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
Well, I agree that solution boxes in Moodle don't have space for formulas: https://moodle.org/mod/forum/discuss.php?d=456089#p1832096. ;) But you said, you'll provide a table for h and x. And tables were a thing a half a century ago - those ugly thermodynamic tables! I expected them to be completely replaced by just-in-time calculations by those i9 13 Gen blasters on our desks, but surprised to find them in existence even today, albeit in colour and beautifully formatted: https://engineering.wayne.edu/mechanical/pdfs/thermodynamic-_tables-updated.pdf.

Yes, Formulas has the disadvantage that it is an additional plug-in. But then it has the advantage of not needing anything external to Moodle, like Maxima (STACK) or Jobe Server (CodeRunner) or VPL Jail Service (VPL). Even those solutions are easier to get going compared to interfacing Moodle with Python.

Aren't we are in engineering: Demanding problems need demanding solutions?

Anyway, to convince you and your IT that Formulas is the right thing you need a test Moodle, which AL Rachels offered. When it comes to the learning part I can offer you access to the material of a webinar its lead maintainer conducted for us. Otherwise visit the Formulas web site, https://dynamiccourseware.org/, containing tons of examples.
In reply to Marko Pinteric

Re: Complex "calculated" question

by AL Rachels -
Picture of Core developers Picture of Particularly helpful Moodlers Picture of Plugin developers Picture of Testers
Just wondering, have you tried this with the Variable Numeric set of question types? Back before I retired, I always felt that I could do so much more with them than I could with the calculated type of question.
Average of ratings: Useful (1)
In reply to AL Rachels

Re: Complex "calculated" question

by Marko Pinteric -
I do not currently have Variable numeric question installed on my system.

I need something that works for sure so I can get our system administrator to add it to Moodle. Can anyone confirm if this will solve my problem?
In reply to Marko Pinteric

Re: Complex "calculated" question

by AL Rachels -
Picture of Core developers Picture of Particularly helpful Moodlers Picture of Plugin developers Picture of Testers

If you send me an email, tomorrow (I am logging out for the day right now) I can give you access via IP, to a Moodle 4.1.9+ that has both formula and variable numeric question types installed, with a number of examples using each type. drachels@drachels.com

Average of ratings: Useful (2)
In reply to AL Rachels

Re: Complex "calculated" question

by AL Rachels -
Picture of Core developers Picture of Particularly helpful Moodlers Picture of Plugin developers Picture of Testers
Hi Marko,
I tried to email you this morning, and just now discovered the email was not delivered. I just sent the info in a private message and the email listed in your profile. I hope you get one of them.
In reply to AL Rachels

Re: Complex "calculated" question

by Marko Pinteric -

Al Rachels

Thank you for this opportunity.

I have tested Variable numeric set and it seems to do what I intend. I create precalculated sets of input and output numbers and enter them into Moodle. I wish it was possible to import data instead of typing it into the fields individually.

Best regards,

Marko

In reply to Marko Pinteric

Re: Complex "calculated" question

by AL Rachels -
Picture of Core developers Picture of Particularly helpful Moodlers Picture of Plugin developers Picture of Testers
Hi Marko,

Glad to read that you were able to do what you needed to. That site is the one I abuse by trying anything and adding any plugin that catches my eye. If you want to try any other plugin, just let know and I will add it.

Best regards,
In reply to AL Rachels

Re: Complex "calculated" question

by Marko Pinteric -
Do you know why the input fields for Formulas question on your website are so small? I can barely see three digits. Is this a website setting or a problem with the Formulas?
In reply to Marko Pinteric

Re: Complex "calculated" question

by AL Rachels -
Picture of Core developers Picture of Particularly helpful Moodlers Picture of Plugin developers Picture of Testers
Hmm, not sure...I can place 5 eights next to each other, in all the ones I've checked.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Hello Marco,

I'm solving the following transcendental equation directly in the Formulas question using Newton's method. The equation has to accept values of \(a\) and \(b\) ranging from 0 to infinity and it is not obvious to find a solution without a starting point. However, by breaking the function into a few intervals, I get a solution, even without a starting point, for the entire range of \(a\) and \(b\) values.

\(\dfrac{ab}{4}\left(\dfrac{\pi}{k}\right)^2+\dfrac{a+b}{2}\left(1-\dfrac{\dfrac{\pi}{k}}{tan \dfrac{\pi}{k}}\right)+ 2\dfrac{tan\dfrac{\pi}{2k}}{\dfrac{\pi}{k}} =1\)

Please give your equations with the range of values over which they apply, so I can have a look at them when I return to my office next wednesday.

Average of ratings: Useful (2)
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
Well, here is the set of psychrometric equations that connect θ, x, φ and h via p and psat
$$h = x \, h_\text{v} + (x \, c + c_\text{d}) \theta$$
$$\varphi = \frac{p}{p_\text{sat}}$$$$x = 0.622 \, \frac{p}{p_\text{atm} - p}$$
$$p_\text{sat} = 610.5 \exp \left(\frac{17.269 \, \theta}{237.3+\theta}\right)$$

So imagine you have h and φ and you have to calculate θ and x.  And this is only a start, the obtained values are in the second step used to make further calculations.  This is why psychrometric charts are used instead of calculations...

I think I will be able to write a Python script to calculate that, but I doubt it will be possible to do that directly in Moodle.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Marko Pinteric -
To make matters a bit worse, the actual formulas for psat are
$$p_\text{sat} = 610.5 \exp \left(\frac{17.269 \, \theta}{237.3+\theta}\right), \: \theta \ge 0$$
$$p_\text{sat} = 610.5 \exp \left(\frac{21.875 \, \theta}{265.5+\theta}\right), \: \theta < 0$$
I have already made a Python script that does the calculation using secant method, with -20 < θ < 50.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers
It doesn't look that bad. The functions behave smoothly. I will have a look at it next week. In the mean time could you provide a complete numerical example with explanations so it will be easier for me to check what I will propose to you.
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
Typical psychrometric problem would be something like that:
- a pair of values is provided; a student must read out the rest.
- a process with a constant parameter follows, until another parameter hits specified value; student must find all values in the end
- the determined values are then used for calculations

Example
- θ1, φ1, provided, determine x1 and h1
- keep x constant until φ hits value φ2, determine, θ2 and h2
- use h1 and h2 for further calculations.

I have already written a Python script that makes table of values
θ1, φ1, x1=x2, h1, θ2, φ2, h2, ...
One possibility is that Moodle gives student θ1, φ1 and φ2 and checks if student has determined θ2, h1 and h2 correctly, with a reasonable precision, of course.

In reply to Marko Pinteric

Re: Complex "calculated" question

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers
Marko,

I asked you for a numerical example (with numbers) with detailed explanations, i.e. step by step equations to be solved, transcendental and/or simultaneous, of course without reference to the psychrometric chart, and rigorous nomenclature of constants and variables, as well as data given to students and answers sought.

In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
Dominique

OK, let me start to replicate the exact formulas I use

$$h = x \, h_\text{v} + (x \, c + c_\text{d}) \theta$$ (1)

$$\varphi = \frac{p}{p_\text{sat}}$$ (2)

$$x = \frac{M_\text{v}}{M_\text{d}} \, \frac{p}{p_0 - p}$$ (3)

$$p_\text{sat} = 610.5 \exp \left(\frac{17.269 \, \theta}{237.3+\theta}\right), \: \theta \ge 0$$ (4a)

$$p_\text{sat} = 610.5 \exp \left(\frac{21.875 \, \theta}{265.5+\theta}\right), \: \theta < 0$$ (4b)

Constants: p0=1.01325e5, Md=0.028965, Mv=0.018015, hv=2501.1, cd=1.005, c=1.926
Average of ratings: Useful (1)
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -

Dominique

Let me now pose a moderately simple problem: "For gas h=30, φ=0.4 find the dew point. (Solution: θ=3.70)"

How do I solve this in Python?

In the first step, I have two functions p(θ) using formulas (2)+(4) (eliminating psat, taking given φ) and p(θ) using formulas (1)+(3) (eliminating x, taking given h). I find the intersection of this two functions using the secant method and get θ=17.43. Then I can calculate p and x=4.92 (3).

In the second step, I take the x obtained in the previous step and combine it with φ=1 (dew point). This step is much more straightforward. From x=4.92 to p (2), from p to psat (1), and from psat to θ=3.70 (4).
Average of ratings: Useful (1)
In reply to Marko Pinteric

Re: Complex "calculated" question

by Marko Pinteric -
Dominique

I have just tested the formulas. They work perfectly, except for two things I do not know how they do:
- compound function
- finding the intersection

For the intersection, the secant method should be used instead of the Newton method because the derivative of the compound function is not a smooth function.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Hello Marko,

For gas h=30 and φ=0.4, find the dew point... In the first step... obtain θ=17.43.

It took me two minutes to calculate θ = 17.42997 using Newton's method, starting with θ values of either -20, 0 or 50. It's quite simple.

I'm not sure what the second step of the calculation entails, but since you mentioned it is straightforward, you should be able to complete that part yourself.

You will find the XML file of the question attached.

Average of ratings: Useful (1)
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
Hello Dominique

As I understand it, you are simply doing Python programming in Global variables input form?
In reply to Marko Pinteric

Re: Complex "calculated" question

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers
Marko,

Yes. It's just a 'for' loop with the equations. Look at the code, it is quite straightforward.

You can change the width of the input boxes with a simple CSS. I did it with jQuery. Look at the code.

Let me know if the method converges to the correct answer for any value of θ in the range you specified, i.e. from -20 to 50.
Average of ratings: Useful (1)
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
Dominique,

I propose that by default input fields are larger.  Of course I do edit my questions in HTML, but most users don't.

I will look at the details of your solution later. But I am already convinced that the Formulas are the way to go. They are indispensable for more complex physical and technical problems.

My current problem is how to get our computer administration to integrate them into Moodle. They are extremely reluctant to add anything at the request of users. Usually they are concerned about (a) how additional applications will affect the stability of Moodle (i.e. will the system collapse because of a bug in the app) and (b) whether this application will be maintained in the future (what happens if this application is no longer available in a year's time). I need to address these issues to be successful, and I am fighting an extremely uphill battle. 
In reply to Marko Pinteric

Re: Complex "calculated" question

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Marko,

These are legitimate concerns from the administrators of your site. However, these concerns are unfounded. Indeed, the Formulas plugin has existed since 2010, and it has always been properly maintained, updated, and even improved over the years. It is installed in nearly 2000 sites in universities and educational institutions worldwide. It is widely agreed to be the best plugin for creating numerical questions in the fields of sciences, technology, engineering, and basic mathematics. Installing the Formulas plugin, which can be done easily and quickly, would provide your institution with a very useful tool for improving its teaching.

Here are some explanations about the code.

# t: theta, the root we are seeking.
# Depending on the problem, you may want to adjust this value
# in order to get a convergent solution.
# Initial value of theta:
t = 0;  

# Small increment used to calculate numerically the
# derivative of the function from eq.(1):
# del: delta
del = 0.0001;

# You can increase the number of iterations
# in order to get a more precise result:
for (i:[0:20]) {

# Eq.(4a):
  psatp = 610.5*exp(17.269*t/(237.3+t));

# Eq.(4b):
  psatn = 610.5*exp(21.875*t/(265.5+t));

# Select eq.(4a) or eq.(4b):
  psat = t >= 0 ? psatp : psatn ;

# Eq.(2):
  p = phi*psat;

# Eq.(3):
  x = Mv/Md*p/(p0-p);

# Eq.(1) evaluated at t:
# ft: f(t)
  ft = x*hv + (x*c + cd)*t - h;

# Eq.(1) evaluated at t+del:
# ftd: f(t+delta)
  td = t + del;
  ftd = x*hv + (x*c + cd)*td - h;

# Numerical approximation of the derivative of ft at t:
# fpt (f prime t): f'(t)
fpt = (ftd-ft)/del;

# Newton's method.
# New value of t:
  tn = t-(ft/fpt);
  t = tn;
}
Average of ratings: Useful (2)
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
Dominique,

you did not need to explain the code, I have already looked at it and understood it.

But I must thank you for your effort in writing the response to the university's computer administration's reservations, which I will include in my proposal. Let us keep our fingers crossed.
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
FYI: In the example above, the Newton method requires 70 steps and the secant method 26 steps to obtain the solution to 6 digits. Furthermore, the number of steps in secant method is the same regardless of the input.
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -

Correction: I used the bisection method. Because there is no "if" clause, this is the only method that provides predictable accuracy since it depends simply on the number of steps. I don't really know why it converges faster than Newton method; theoretically, it shouldn't.

If anyone is interested, my code is below. It could probably be written even more efficiently...


p0 = 1.01325E5;
Md = 0.028965;
Mv = 0.018015;
hv = 2501.1;
cd = 1.005;
cv = 1.926;

h=30;
phi=0.4;

t1 = -10;
t2 = 40;

for (i:[0:100])
  {
  t3=(t1+t2)/2;
  x1 = (h-cd*t1)/(hv+cv*t1);
  x3 = (h-cd*t3)/(hv+cv*t3);
  psat1 = t1 >= 0 ? 610.5*exp(17.269*t1/(237.3+t1)) : 610.5*exp(21.875*t1/(265.5+t1)) ;
  psat3 = t3 >= 0 ? 610.5*exp(17.269*t3/(237.3+t3)) : 610.5*exp(21.875*t3/(265.5+t1)) ;
  d1 = phi*psat1-x1*p0/(Mv/Md+x1);
  d3 = phi*psat3-x3*p0/(Mv/Md+x3);
  t1 = d1*d3 > 0 ? t3 : t1;
  t2 = d1*d3 > 0 ? t2 : t3;
  }


In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
Dominique,

I tried with secant method too, and it gives the same precision in only seven steps. However, when I apply it to Moodle Formulas, I get error "19: Some expressions cannot be evaluated numerically." I am clueless what is wrong.


p0 = 1.01325E5;
Md = 0.028965;
Mv = 0.018015;
hv = 2501.1;
cd = 1.005;
cv = 1.926;

h=30;
phi=0.4;

t1 = -10;
t2 = 40;

for (i:[0:100])
{
psat1 = t1 >= 0 ? 610.5*exp(17.269*t1/(237.3+t1)) : 610.5*exp(21.875*t1/(265.5+t1)) ;
psat2 = t2 >= 0 ? 610.5*exp(17.269*t2/(237.3+t2)) : 610.5*exp(21.875*t2/(265.5+t2)) ;
x1 = (h-cd*t1)/(hv+cv*t1);
x2 = (h-cd*t2)/(hv+cv*t2);
d1 = phi*psat1-x1*p0/(Mv/Md+x1);
d2 = phi*psat2-x2*p0/(Mv/Md+x2);

t3 = (t1*d2-t2*d1)/(d2-d1); # last three formulas seem to be the problem
t1 = t2;
t2 = t3;
}

th3=t3;
In reply to Marko Pinteric

Re: Complex "calculated" question

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers
Marko,

As you mentioned, the method converges rapidly. After nine steps, both d2 and d1 become so small that they are essentially considered zero. Consequently, d2 - d1 equals zero, triggering a division by zero error, even when you're in edit mode so the question won't crash when executed.

To avoid this problem, consider limiting the number of steps, perhaps to seven, i.e. 'for (i:[0:7]) {'.
Average of ratings: Useful (1)
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
Thanks. Apart from reducing number of steps, adding absurdly small number to the fraction

t3 = (t1*d2-t2*d1)/(d2-d1+1e-99);

solved the problem. I just want to make sure that there is no problems if it converges in fewer than specified number of steps.
In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
@all

I attach the final XML for the question with streamlined code:

p0 = 1.01325E5;
Md = 0.028965;
Mv = 0.018015;
hv = 2501.1;
cd = 1.005;
cv = 1.926;
t1 = -10;
t2 = 40;
psat1 = t1 >= 0 ? 610.5*exp(17.269*t1/(237.3+t1)) : 610.5*exp(21.875*t1/(265.5+t1));
x1 = (h-cd*t1)/(hv+cv*t1);
d1 = phi*psat1-x1*p0/(Mv/Md+x1);
for (i:[0:10])
  {
  x2 = (h-cd*t2)/(hv+cv*t2);
  psat2 = t2 >= 0 ? 610.5*exp(17.269*t2/(237.3+t2)) : 610.5*exp(21.875*t2/(265.5+t2));
  d2 = phi*psat2-x2*p0/(Mv/Md+x2);
  t3 = d2-d1 == 0 ? t2 : (t1*d2-t2*d1)/(d2-d1);
  t1 = t2;
  d1 = d2;
  t2 = t3;
  }

Thank you for your help. I consider this question answered, but (if you are interested) I can still report back on how my request to install the plugin goes.


Average of ratings: Useful (1)
In reply to Marko Pinteric

Re: Complex "calculated" question

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Marko,

The University of Maribor prides itself on being among the top 4% of universities worldwide, and the second largest and oldest university in Slovenia. It claims to be recognized as a successful scientific institution and comprises seventeen faculties covering the natural sciences, mathematics, engineering, medicine, biotechnology, social sciences, humanities, and other sciences.

And yet, it wouldn't install the Moodle Formulas plugin, 'the' plugin for the fields of science, technology, engineering, and mathematics? I find that hard to believe. That would be a loss for you, your colleagues, your students, and your institution.

Your IT department has no valid reason not to install the Formulas plugin. Until they make a decision, I offer to host your question(s) for free on the website DynanicCourseware.org, at the following address: https://dynamiccourseware.org/course/view.php?id=128&section=1➚.

In reply to Dominique Bauer

Re: Complex "calculated" question

by Marko Pinteric -
Dominique,

having tested it extensively, I can only agree that this is a really good and important plug-in that would greatly benefit lecturers in at least seven university faculties (Natural Sciences, Chemical Engineering, Electrical Engineering, Mechanical Engineering, Civil Engineering, Energy Engineering and Logistics). And fortunately there are other people at the university who share my view and are willing to help.

I can assure you that I have not given up the fight yet, although I have already received a preliminary negative response from the university's IT department. It's sad that I am spending my time lobbying and persuading for something that should be self-evident instead of preparing materials for my students. But hey, if that was the only nonsensical task I had to do in my job...

By the way, I plan to make my formula questions available as free supplements to my book and problem book on building physics so that lecturers using my books can also benefit from them.

Thank you for your offer. I do not know if I could use it for my students as I do not know how to transfer their data to your server, and I was generously offered access to AL Rachel's sandbox for personal testing. What exactly do you have in mind?
Average of ratings: Useful (1)
In reply to Marko Pinteric

Re: Complex "calculated" question

by Dominique Bauer -
Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Plugin developers

Something like this:

https://dynamiccourseware.org/course/view.php?id=136&section=1➚.

Individual students and teacher accounts should be possible.

In reply to Dominique Bauer

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
Dominique, checking whether I understood your example correctly: As it is any "guest" can practice solving that question on-line. But under a generic account. So the the "sending" site doesn't know what they've did or how they've did, if at all. So far clear.

When you said "Individual students and teacher accounts should be possible", do you mean you would and you site can cope with creating personal accounts for students and staff of a major university?

Intriguing idea, providing a service like LTI or SCORM, etc. to remote Moodle sites sending back records of attempts, grades, etc. I can't think of a technology which does that - to provide a Moodle question as a service to somebody else's Moodle site. Do you? Ideas from the others too are welcome!
In reply to Marko Pinteric

Re: Complex "calculated" question

by Marko Pinteric -
And now, dear friends, the final update.

In the absence of meaningful support from the university's Department of education, I have had to abandon my wish to use Moodle for quizzes for my students.

But do not worry, not all your and my time has been in vain. I will be developing a few dozen Moodle quizzes as supplementary material to my book so that professors around the globe who use my book can use them for their students. What an irony, the author himself will not be able to use them.

Formulas is a really cool addon and an absolute necessity if you want to use quizzes for more technical problems or problems with multiple sub-questions, which in practice accounts for more than 90% of engineering problems.

Anyway, thank you for your time and support and to everyone who has given me access to your own Moodles, and maybe I will see you again in the future.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
The link psychometric charts returns an error. Anyway, the subject is the Enthalpy-entropy chart. The English Wikipedia page is disappointingly short. The German Wikipedia is much better.
In reply to Visvanath Ratnaweera

Re: Complex "calculated" question

by Marko Pinteric -
Visvanath Ratnaweera

You can't access https://www.pinteric.com/books/mollier.pdf ? That worries me a lot as it is material for the readers of my book. Can you at least access https://www.pinteric.com/books ?

Yes, Wikipedia is not very helpful.  This is why I needed so much time to learn everything, including making my own diagrams, myself.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
Hi

No, both links time out. BTW, the mollier.pdf is a single page, a full-page h-x diagram, right? I get it when I 'wget' from the First World. It looks like your CDN cdn.jsdelivr.net discriminates the Third World. I'll PM you my current IP address, the one being blocked.

BTW, found your book! Congrats. And nicely surprised to see that analog calculations still exist. Can't remember consulting h-x diagrams, but the steam tables were a thing half a century ago! But they were ugly tables, https://pages.mtu.edu/~tbco/cm3230/steamtables.pdf for example. These h-x diagrams are art!
Average of ratings: Useful (1)
In reply to Visvanath Ratnaweera

Re: Complex "calculated" question

by Marko Pinteric -
Hello

I am very disturbed by your information. I host my personal website and also the website for my readers (including supplementary videos and documents) with a relatively cheap US hosting service. What advice would you give me?  To use a more expensive hosting service?

Why are you surprised by analogue calculations? Nowadays most things are automated, but students are still learning the principles. If nobody learns the principles, who is going to automate the procedures?

Also, psychrometric diagrams are a great resource for creating relatively simple, cool exam problems. Students do not have to do the calculations, they just use the chart, which is much easier. But if I want to use different input values, I have to calculate the solutions, which is much more difficult.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
Hi

About hosting the website for the readers of your book: I always liked the idea when I can contact the author of a book. And for the author it brings valuable feedback from the readers. May be others here can report the availability of https://www.pinteric.com/books/.

Analogue calculations: I am not from a field where people use charts. Just my ignorance.

Same with making questions, whether to learn or for assessing, no experience. Just the chart brought some nostalgie.
smile
In reply to Visvanath Ratnaweera

Re: Complex "calculated" question

by Marko Pinteric -
I not only provide the website with supplementary material, but have even included the website address in my book (with a hyperlink for the electronic version). It is therefore very important to me that the website is accessible.

Honestly, I was hoping for more feedback, and now I am worried that some of the feedback has been lost because of the problem you had. (Please check the private messages for more details.)

As for the analogue calculations, I was making a general comment. I work at Faculty of Civil Engineering and teach building physics. Other professors teach structural analysis. There are many many programs that make the complete structural analysis of a building. Nevertheless, the students learn how to calculate simpler cases manually so that they understand the principles. I think that's important.
In reply to Visvanath Ratnaweera

Re: Complex "calculated" question

by Ken Task -
Picture of Particularly helpful Moodlers
As per Mr. V's request to see if I could access from San Antonio, Tx ...
Yes I can:

curl -I https://www.pinteric.com/books/
HTTP/2 200
content-type: text/html
last-modified: Mon, 05 Feb 2024 16:11:55 GMT
accept-ranges: bytes
content-length: 9768
date: Tue, 05 Mar 2024 05:52:33 GMT
server: LiteSpeed
alt-svc: h3=":443"; ma=2592000, h3-29=":443"; ma=2592000, h3-Q050=":443"; ma=2592000, h3-Q046=":443"; ma=2592000, h3-Q043=":443"; ma=2592000, quic=":443"; ma=2592000; v="43,46"

wget https://www.pinteric.com/books/mollier.pdf
--2024-03-04 23:54:02--  https://www.pinteric.com/books/mollier.pdf
Resolving www.pinteric.com (www.pinteric.com)... 198.251.89.144
Connecting to www.pinteric.com (www.pinteric.com)|198.251.89.144|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 78808 (77K) [application/pdf]
Saving to: ‘mollier.pdf’

mollier.pdf         100%[===================>]  76.96K   242KB/s    in 0.3s    

2024-03-04 23:54:03 (242 KB/s) - ‘mollier.pdf’ saved [78808/78808]

M.PINTERC.PROBLEMS IN BILDING PHYSICS.2024

'SoS', Ken

Average of ratings: Useful (2)
In reply to Ken Task

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
Ken, thanks a lot!

@Marko, this is the situation:

- St. Antonions TX can access https://www.pinteric.com/books/mollier.pdf

- So do the "Zürcher"

- Out of five Sri Lankans that tried four couldn't. Under those who couldn't, there was one from SLTMobitel broadband, one in the Dialog mobile network, one in SLTMobitel mobile network and one in LEARN, the Internet provider of the universities. The only one who could was also SLTMobitel broadband customer but the geo-location databases in the web points her address to Luxumboug. Possibly the "country filter" gets the same wrong information.

I conclude that your hosting provider blocks IP addresses from "not so affluent" places. This not only limit your readers but also discrimination. You should complain to your hosting provider. I can send you the IP addresses by PM. Nothing can be done from the other side other than an intrusive network debug with your support. But that will counter saving you make by going to a cheap provider. If you want cheap and competent providers look in Germany. But that is OT, I'm signing off from this provider discussion. Your question topic is more interesting. I see Dominique closing in.
wink

In reply to Visvanath Ratnaweera

Re: Complex "calculated" question

by Marko Pinteric -
I am willing to pay more if it gives access to all my readers and I can afford some extra (personal) money for that purpose. I have already contacted my hosting provider again and if I do not get a satisfactory explanation, I am moving.

I am a little worried that they might open up access to Sri Lanka to appease me, but leave other "not so affluent places" blocked. I have no way of knowing.

But let us continue this discussion in private. I hope you can help me further in this regard.
In reply to Marko Pinteric

Re: Complex "calculated" question

by Marko Pinteric -
To @all:

I have just been informed that STACK has already been proposed for Moodle at the University of Maribor (14,000 students), but was rejected because "it inconveniently overloaded the entire system". No comment.

So please do not test the STACK option, because I will not be able to use it. Maybe Formulas or Variable numeric sets will be approved.

In reply to Marko Pinteric

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
STACK is immensly powerful for university mathematics, in handling symbolic mathematics. ("it inconveniently overloaded the entire system" is a declaration of incompetence of your IT.) I don't think the numerical calculations you are talking about would profit much from STACK. My initial guess was the Forumlas question type. Reading your other thread I think Variable numeric is cumbersome. So let's wait for Dominique to return.
Average of ratings: Useful (1)
In reply to Visvanath Ratnaweera

Re: Complex "calculated" question

by Marko Pinteric -
​I have just tested the formula question and am pleasantly surprised! It works very intuitively, each answer has its own input field and is evaluated separately.  Such solution would work with the majority of my problems. This should be included as standard in Moodle.

I still do not know how to implement the composite function, function intersections and why the input fields are sooooo small.

Average of ratings: Useful (1)
In reply to Marko Pinteric

Re: Complex "calculated" question

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
"It [the Formulas question type] works very intuitively". Not everybody agrees. I am slow in grasping it. I believe the reason is that I am making questions for others not teaching my own subjects any more. Hopefully it'll change soon. Now there is so mush need and activity around producing OER, I'm returning to Linux and Networking basics.