Talk:Nomogram

Not all of the examples here seem to be nomograms. Tephigrams, for example are not nomograms, but rather non-linear graphs. The defining characteristic of a nomogram is, I believe, that calculations are performed by drawing a straight line between labelled points on two scaled curves and seeing where the line intersects a third scaled curve. Comments? --macrakis (talk) 20:36, 22 January 2009 (UTC)Reply

I have updated the article to say that Tephigrams and other graphical calculators are not strictly nomograms, although they are often described as such. I have also replaced the top figure of a Smith Chart with a true parallel-scale nomogram. Rrrddd (talk) 01:08, 22 January 2012 (UTC)Reply

Actually, I believe this definition of nomograms is too restrictive, and applies to a particular subset of nomograms (alignment nomograms). The term has certainly been used far more broadly. Glenbarnett (talk) 03:50, 10 December 2012 (UTC)Reply

In Nomogram#Chi-squared_test_computation_nomogram, what does the ABCDE stand for? Mikael Häggström (talk) 10:31, 14 June 2010 (UTC)Reply

I have added a paragraph that describes the function of the five lettered scales. Rrrddd (talk) 01:08, 22 January 2012 (UTC)Reply

Wow, nomograms are AWESOME. Especially the Parallel-resistance/thin-lens nomogram. Can't believe I've never come across that visualization before. It is soooo insightful.

The article says "A nomogram consists of a set of n scales. Knowing the values of n-1 variables, the value of the unknown variable can be found ... by laying a straightedge across the known values on the scales" which suggests that nomograms can model functions of any number of variables.

But how can this be done for any value of n other than 3? Assuming the nomogram is printed on a flat surface, the straightedge has only 2 degrees of freedom. Therefore 2 known variables define the line (and therefore the result) completely, and no further known variables can influence the result.

The only way I can think of getting around this is to use more straight edges with "intermediate result" scales on the monogram. This would effectively mean combining multiple sequential monograms on the same surface, using, for example, a bivariate function nested within a bivariate function to model a trivariate function. But the article mentions nothing about multiple straightedges, so I'm wondering why it says "n-1 knowns" rather than "2 knowns".

--Tennenrishin (talk) 11:12, 4 October 2013 (UTC)Reply

Have a look at http://myreckonings.com/wordpress/2008/01/09/the-art-of-nomography-i-geometric-design/, search for "4-Variable Charts"

--alex (talk) 08:28, 28 September 2014 (UTC)Reply

Not sure if I understand this correctly, but we use what I believe is a type of nomogram for backup/degraded mode firing solutions in U.S. Field Artillery, called a Graphic Firing Table (GFT), derived from a Tabular Firing Table (TFT: talk about a redundant term). When we derive a firing solution from corrected data, we apply that solution to the nomogram using thin lead pencil or superfine alcohol pens. That correction can be a simple "single-plot GFT setting", two-plot, or multi-plot GFT setting. With the two-plot and multi-plot GFT settings, the user defines the graphic index line as lines between corrected trajectory data points from data corrections derived from meteorological data and weapon corrections (muzzle velocity, etc.). The more and closer together the data points, the more curve-like the index line.Caisson 06 (talk) 20:11, 11 May 2015 (UTC)Reply

A book of radio/electronic nomograms I once used, they were called ABACs. Doug butler (talk) 08:28, 11 May 2021 (UTC)Reply

1) I explain why a nomogram is not a graphical analog computation device.

2) Propose that a nomogram is a kind of slide rule specialized to compute a more complex function.

3) Other uses of monograms in data analysis

4) Propose a more general article comprising monograms among other graphs mentioned in the article.

(I forgot the wiki markup, feel free to reformat this, all the automatic "corrections" of the IA and jumps of the cursor drive me crazy, just make me loose the point I want to explain, sorry for that)

The text says:

"Like a slide rule, a nomogram is a graphical analog computation device."

It is wrong because a nomogram is a precalculated function to some useful precision. The fact that they were not computed with electronic digital calculators, don't make them analog devices.

An analog device is, evidently, an "analog computer", a device built with electronic components (capacitors, resistors, transformers,etc), to assembly a circuit analogous to a function to approximate it. For example the integral of a particular function use condensers in its circuit version to transform the input into an output voltage corresponding that function's integral interpreting the voltage levels. So, as its name says, an actual analog calculation device. That is very far from how a nomogram is used. (What if analog computations of some sort were made to build a nomogram? Who knows, it does not matter, in that sense it is a black box, which can be used with a know precision.)

Nomograms were accurate reproduced, I wonder how, to be used by those who require to eval the function in their everyday work.

A nomogram does not help to make a calculation like an abacus does, so is not a computation device.

A slide rule, may be some similar, in the sense that they have functions like logarithms, sine, cosine,.

It has logarithm scales because they are used to compute products by adding quantities, based in the property. But are also used along with a linear scale to read the logarithm or antilogarithm of a number. Some models also have trigonometric functions.

That scales were used in slide rules to perform elementary operations used in more complex functions.

On the other side the nomogram is a graphical representation of those more complex functions, which with the aid of a ruler one can find the value for a point n the domain of the particular function.

In this sense is an specialized slide rule, I was tempted to say a kind of slide rule in steroids, but each nomogram is specialized in a particular function not a general pattern. However the technique to develop monograms is slide rule making on steroids.

I propose this better description:

"Is a graphical representation of a relevant precalculated function used as an aid in the calculations needed in some knowledge domain. They are an alternative to precalculated tables like those often seen in book appendices."

The article mention other kind of graphs for the same purpose, but different in the way they are built.

All those instruments are invaluable for the execution of many engineering enterprises before the electronic computation era. (That does not make them "analog".)

Previous to, and during the 1970s, different kinds of graphs where taught to life sciences students not just as an aid in the calculations to perform some action, they were also used to analyze experimental data to visualize what kind of relations exist. The more simple and introductory examples was to use logarithmic and semilogarithmic paper to graph data, where exponential relations are shown as straight lines.

That was very important even when scientific pocket calculators appeared. Texas Instruments and Hewlett Packard calculators replaced slide rules to do the calculations, we used them to compute mean squares quickly, but the graphs allowed us to understand how the measured data were related.

That makes me think if a series of articles for calculating aid devices in the pre-electronic computation era, be first introduced in an article stating the general context and a brief description of the different solutions. ~2026-59143-97 (talk) 09:10, 8 November 2026 (UTC)Reply

A nomogram is most certainly a "computation device". It is also analog (as compared to, say, an abacus). You are mis-applying insufficiently general definitions of these terms to come to a poor conclusion. Your proposed alternate description is also more or less accurate, but is worse than the one that was already there. –jacobolus (t) 15:17, 8 November 2026 (UTC)Reply

But with that said: if you want to work on our existing articles about "calculating aid devices in the pre-electronic computation era", or add new ones about encyclopedic topics that we don't yet have articles for, please feel free. Make sure you base your work on reliable sources, and feel free to ask for help if you have questions or run into difficulties. –jacobolus (t) 17:40, 8 November 2026 (UTC)Reply

I disagree that the existing lede is inaccurate, and prefer the current wording. I disagree with your assertion that "analog device" only refers to an electronic analog computer. I don't know where you got that idea. ApLundell (talk) 17:15, 8 November 2026 (UTC)Reply