It is sort of painstaking to try and guess values for these figures from the graphs. Do you have the original equations for the curves somewhere in the archives? I would like to update the MS Excel spreadsheet I created to calculate these, because try as I might, I can't quite get my values to correspond with those on the curve.

Cheers,

quote:

---

Originally posted by yadernye:
Do you have the original equations for the curves somewhere in the archives

---

Actually, we have both a chart and a formula that does that. I'll post that up tomorrow or Wednesday. It is actually better to use the chart then try to figure it out yourself. Using the chart gives you a consistent set of values, as there really isn't a lot of variation between artillery peices of the same size.

Yes, please, post the equation(s). It would be great to see how they are handled.

"The following table gives estimated rates of fire (rds/hr) for non-automatic weapons when the caliber (mm) is known.

Caliber (mm) RF
20 295
30 260
40 225
60 175
75 152
81 144
90 134
105 113
120 99
130 90
155 60
175 47
203 30
225 25
250 22
275 18
300 16
325 15
350 13
375 12
400 10
450 8
500 7
550 5
600 3
650 2
700 2
800 2
900 2
1000 2

If additional interpolation is necessary, Figure 1 may prove useful."

If the Dupuy Institute considers these equations to be proprietary information, I will understand if you don't wish to share them.

Cheers,

As far as the formula's, they are not published in our documentation. I would have to go into the source code to extract them. I may do that at some point, but not tomorrow.

The complication is that all these graphs were originally hand drawn (we are talking about a model created in the 1970's), so they were never created by a formula. The TNDM, formula's were created to match the "hand drawn" curves, which usually meant that a fit required not only a formula, but also table of corrections.

I'll take a look at the curves to see what the problem is with the "trough" (not sure I understand it).

I will e-mail you the Excel chart I made (unless you know how I can post it here). If you compare that curve to the one in NPW, you will see they differ slightly. The curve in NPW continues downward from 75mm before flattening out around 250mm. When you graph the values you posted, you find that the curve begins to flatten out around 75mm, and then dip downward again around 155mm.

It could be that these discrpancies are meaningless, especially when you consider that the hand-drawn graphs in NPW were probably idealized.

Cheers,

"If the area of burst is not known, a value may be estimated from the table below based on caliber.

For RoF, I include:
a). Maximum RoF per Minute
b). Sustained RoF per Hour
c). Magazine Capacity (or equivalent Ammunition On-Hand)
d). Re-supply Factor (Weight-Size Based)

For Targets per Strike, I include:
a). Lethal Radius (from formula)
b). Penetration of HE/HEP-HESH/HEAT
c). Impact Measurement for KE Rounds (separate formula integrated with (a-b).

These two have gone well, although I would still like to see the formulae used when you have a chance, Chris.

The difficult one is the Range Factor - finding a good formula that will measure Range for all rounds. Does anyone have one?
As well, what formula was used on the weight to Calibre graph? Is it possible to share that - understanding of course that - as already mentioned - you may not be able to share this with us?

As this is a "scoring system", its purpose is to produce a final number that can them be summed with other weapons to give a rough approximation of combat power (not firepower!). As such, I would not put a lot of weight into each individual element of the formula's.

The QJM/TNDM is to be used for analysis of combat, not analysis of weapons.

quote:

---

Originally posted by Korman:
These two have gone well, although I would still like to see the formulae used when you have a chance, Chris.

---

Which formula? Most of the formula for calculating OLIs is in Numbers, Predictions and War.

quote:

---

The difficult one is the Range Factor - finding a good formula that will measure Range for all rounds. Does anyone have one?

---

Nothing specific besides what is in the OLI formula (which gives an OLI value, not a measurement of range).

quote:

---

As well, what formula was used on the weight to Calibre graph?

---

The graph was created in the 1970s and a formula to represent the graph was only created in the early 1990s (and is buried in the code of the program). It will not give you the answer I think you are looking for. It does not measure weapons effectiveness, but is simply used as part of a scoring system.

Now the US Army's attempt at measuring weapons effectiveness is handled by AMSA (sp?). They come up with the tables of munitions effectiveness and the SSPKs (Single Shot Probability of Kills) that a number of combat models use.

quote:

---

Originally posted by Chris Lawrence:
As this is a "scoring system", its purpose is to produce a final number that can them be summed with other weapons to give a rough approximation of combat power (not firepower!). As such, I would not put a lot of weight into each individual element of the formula's.

The QJM/TNDM is to be used for analysis of combat, not analysis of weapons.

---

Thank you for the clarification. This certainly makes sense within the overall QJMA philosophy.

I apologize if my requests for information have been perceived as being critical or untoward. My interest lies only in getting Col. Dupuy's methodology correct.

Cheers,

Shawn

In a series of e-mail swappings with James Dunnigan, he said very much the same thing - that the inter-relational values of weapons (and even of the value of individual weapons) was one of the least important variables in the total combat power of an army or unit.

Thanks again, but if you do get a chance at some point to dig out those specific formulae for RoF and Targets per Strike, I would like to see the process (if it's possible).

quote:

---

Originally posted by yadernye:
I apologize if my requests for information have been perceived as being critical or untoward.

---

Absolutely not. Basically combat models end up with only about two ways of calculating weapons effects. One is some form of scoring system (OLIs, WEI/WUVs, TASCFORM, etc.). The other is providing a Single-Shot Probability of Kill (SSPK) or some other direct measurement of weapons effects (like the JMEMs handbooks). Each has their problems.

For direct comparison of weapons, and doing weapon trade-off studies, obviously something based upon the actual weapon performance is of more use. A model built up from these measured effects (which invaribly come from a test-range environment, not a real-world environment) may allow one to actually determine what is the better gun to use for their tank. Beyond that, the use of this data to build up models has all kinds of problems and becomes very difficult to calibrate into something that measures the whole (to assemble the whole from its component parts may require more knowledge about the inter-relationships and nature of warfare than anyone currently has).

Scoring systems cannot be used for weapon trade-off analysis as they are simply an assigned value. They have fallen out of favor for a lot of models, although RAND's JICM/RSAS still uses them. At least with the QJM/TNDM system, they are used in an attempt to measure the "combat utility" of a weapon system, not its firepower. By the time you multiply a weapon system score by the square root of its range, by its speed, its ammunition capacity, account for its armor and sensors, etc., what you have is something very far removed from a "firepower" score. What you have a combat utility score. It is an approximation.

The real advantage of the QJM/TNDM system of scoring is two-fold over all other scoring systems (I'm not sure I want to enter the debate here of the advantages/disadvantages of scoring systems over other systems). First, because the OLIs are based upon a formula, they are consistent. Some other scoring systems relied upon "professional judgement" of military officers, and as such, each weapon score tended to vary depending on the branches of the latest group of officers assigned at that particular time. The OLI system gives you a structured, systematic set of scores.

The second issue is that they have been vaidated within a model. Unfortunately, there is no way to independently test or validate any scoring system outside of the model it is used it (this is also true of SSPKs, although I do not think many of the users of these systems appreciate this point). As such, the only way you can validate a scoring system, is that you have to validate it as used within the model. If the model vaidates (to real-world data), then the scoring system, overall (although not necesarily in particular), must have some validity. Obviously, having a scoring system that is consistent is useful for such testing.

To date, the QJM/TNDM is the only scoring system type models that has been validated. There have been at least two attempts that I am aware of to use a scoring system drawn from another source and mixing it with QJM or QJM-like charts and tables. Of course, you have now sewn together elements created for different purposes and from differing methods that is intellectually inconsistent. Neither of these other two modeling attempts have been validated.

quote:

---

Originally posted by Korman:
In a series of e-mail swappings with James Dunnigan, he said very much the same thing - that the inter-relational values of weapons (and even of the value of individual weapons) was one of the least important variables in the total combat power of an army or unit.

---

I would still like to - again, if possible - see any of the graph formulae if you can get them: RoF, Tgts@Strike, and Weight-to-Calibre.

The QJM/TNDM is to be used for analysis of combat, not analysis of weapons.

I understand this. Yet the numbers that come up for lethal area of bust confound me. IE 155mm: Most sources I read from the Army state about 50 meters.

http://nigelef.tripod.com/wt_of_fire.htm

I suspect this is what you are looking for. PM me if you want more info.

I cant seem to find a PM option, though.