Chronograph & PF

[rhino]

I think truncating rather than rounding makes more sense.

1. You're trying to achieve a minimum value ... it doesn't matter how close you come to it, if you don't get it, you don't get it.

2. The power factor for major is so low now that if you don't make major, well ...

3. Worrying over propagation of errors from the scales, etc. will needlessly complicate something that need not be complicated.

So ... make a policy (I prefer truncate) and then stick with it. Then the shooters will know that what they need to do to accomodate any roundoff errors, etc.

It's like leaving early for work to accomodate bad traffic. It doesn't matter if you get there 99.9999999% on time ... you're late if you don't make it, and if you know what time you need to be there, you can add some extra time to trip to make sure.

[Scooter]

1st guy: PF= 181.2 * 938.4 / 1000 = 170.03808 => Major. 

2nd guy: PF= 181 * 938 / 1000 = 169.778 => Minor! 

How about we just use significant figures. I'm sure most of you have forgotten how to deal with significant figures.

1st guy: 181.2 & 938.4, that's 4 significant figures. So the pf is 170.0

2nd guy: 181 & 938, 3 significant figures. pf = 170.

Or 181.2 & 938, 3 sig. figs. pf =170.

But you really shouldn't be that close to the pf anyways. The chrono and scale will have a percentage of error, plus changes in atmospheric conditions.

[shred]

Ok, I hereby ban any more mention of "you shouldn't be that close anyway". We know. It's irrelevant to this discussion which is "where is the line and how do we know you're over it"

Focus on the points. Truncation will only make things worse. Would you prefer that your 124.9 grain bullet only weigh 124 grains for chrono purposes?

Significant figures is where it's at.

[TRNinTX]

Did someone go minor this weekend?

[Yoda]

Allright, I'll give it another go. I prefer doing the math(s) with the non-rounded/truncated values and then truncate the calculated PF to an integer value.

How about the following rules addendum:

IPSC 5.6.3.12 The values recorded for bulletweight and velocity must be with 1 decimal except when the equipment used does not allow this. The calculated PF will then be truncated to a whole number (no decimals)

This allows for Chrono's that don't display the decimal with velocities >999.9 fps, does justice to Shreks first question and the competitor.

How about it?

[ivanhu]

FACTS:

Bullet weight reading on a Lyman electronic scale: 181.2 grains.

Bullet average velocity reading on a Chrony chronograph: 938.4 fps.

PROBLEM: arising from lack of directions in the rulebook; two different chrono guys performing PF calculations:

1st guy: PF= 181.2 * 938.4 / 1000 = 170.03808 => Major. 

2nd guy: PF= 181 * 938 / 1000 = 169.778 => Minor! 

There is nothing in the rulebook preventing the second guy doing it, and, yes, before you ask, I've seen the above done in different places (not the minor scoring, but the truncation of readings). This practice could lead to the above error.

Now, solutions welcome.

First of all, we must identify the problem.

Someone said that the problem is, the same ammo will be different in Italy than in Germany. Well, IMHO it's a non-issue. All we have to ensure is that it must be consistent throughout one particular match only.

If we agree, then all the rule book has to deal with is to ensure that.

Therefore, the "Rool Buke" should say (among the description of thechrono procedure) something like this:

x.y.z. The measured bullet weights and speeds might be rounded. The MD must declare before the match starts that how many significant digits will be used after the decimal point. This decision must be posted.

x.y.z.1. In case of the bullet weight, no more than 2 (two) significant digits are allowed after the decimal point.

x.y.z.2. In case of the bullet velocity, maximum one significant digit is allowed after the decimal point if the speed is less than 1000 fps, and zero if it is greater.

x.y.z.3. When rounding is used, then the first digit after the last significant one determines the final number: if it is between 0 to 4, then the last significant digit be unchanged; if it is between 5 to 9, the last significant digit be increased by one.

It doesn't really matter whether the chosen method will screw someone or not. The only that matters is whether it is the same for all or not.

Just my 0.2

[shred]

Allright, I'll give it another go. I prefer doing the math(s) with the non-rounded/truncated values and then truncate the calculated PF to an integer value.

Agreed.. do the math, but then round the answer. It's the only way to be fair.

How about the following rules addendum:

IPSC 5.6.3.12 The values recorded for bulletweight and velocity must be with 1 decimal except when the equipment used does not allow this. The calculated PF will then be truncated to a whole number (no decimals)

This allows for Chrono's that don't display the decimal with velocities >999.9 fps, does justice to Shreks first question and the competitor.

How about it?

Ok except for the truncation part. Truncation bad. How many times do I have to say it?? If you're OK with truncating the PF, then you must be OK with truncating a 124.9 grain bullet to 124 grains.

And to answer some of Vince's question on translation, my IBM National Language Reference Manual Vol 2, says that with the sole exceptions of Argentina and 3-digit decimals, and Swiss federal currency transactions, every country in the world uses the following rounding system:

nnn.0 through nnn.4 round to nnn

nnn.5 through nnn.9 round to nn(n+1)

You people would rather hang a few innocent people than let one guilty one free. I prefer the other way around.

Round to one decimal place if you must-- you'll hang less innocents that way.

The rule as I see it would be simple: The competitors calculated PF is rounded to the nearest whole number. For example, a 164.7 PF would round to 165 and 164.4 PF rounds to 164. This I think is the fairest, but it does let some blatant scofflaws drop their load a whole 1 fps and still make it.

Should you prefer one decimal place, the rule goes: The competitors calculated PF is rounded to the nearest tenth. For example, a 164.95 PF would round to 165.0 and 164.94 PF rounds to 164.9.. You still hose a few people that really are shooting Major, but less pesky scofflaws get by, and people can still make fun of others for shooting 164.9.

(and I made a 167 PF at Area 4 this weekend, thank you very much)

[Vince Pinto]

The rule as I see it would be simple:  The competitors calculated PF is rounded to the nearest whole number.  For example, a 164.7 PF would round to 165 and 164.4 PF rounds to 164.

As a rule writer (not a mathematician), this concept looks clear and workable, however we must use "NNN.5" in the example, to further avoid ambiguity. It would also be better to use a reference integer for Minor, which is common to all divisions.

Here are two possible solutions (common proposed text in red, additional proposed text in green):

-------------------

Version A:

5.6.3.4 Power factor is calculated using the bullet weight and the average velocity of the 3 rounds fired, according to the following formula:

Power Factor = bullet weight (grains) x average velocity (feet per second) / 1000

The resultant power factor is rounded to the nearest whole number. For example, a result of 124.5 (or higher) rounds up to 125, and a result of 124.4 (or lower) rounds down to 124.

-------------------

Version B:

5.6.3.4 Power factor is calculated using the bullet weight and the average velocity of the 3 rounds fired, according to the following formula:

Power Factor = bullet weight (grains, rounded up to 1 decimal place) x average velocity (feet per second, rounded up to 1 decimal place) / 1000

The resultant power factor is rounded to the nearest whole number. For example, a result of 124.5 (or higher) rounds up to 125, and a result of 124.4 (or lower) rounds down to 124.

-------------------

Will one of these work? If so, which is better? If I can get consensus here, I will take it to the Rules Committee.

[mactiger]

Either one would work for me, but I prefer A, just because it doesn't have too many "gimmes" in it.

Troy

[Tman33_99]

A is simpler to understand, but I don't see a problem with the current rule. This would basically lower minor PF to 124.5 and major to 169.5 or 164.5.

[Vince Pinto]

Hi guys,

As an aside, I asked my old drinking mate Chuck Hardy for information about CED products, and his reply was:

1. Both the D-Terminator and CED Professional Scales as well as the CED Digital Pocket scale display in hundreds for grams and tenths for grains. Thus, you can measure 124.6 grain bullet.

2. Chronograph measures velocity in feet (1324), meters (404), or decimeters (404.4). When you input the bullet weight for power factor it will accept to the tenth (124.4) and will then give you the power factor to the tenth (164.7)

FYI.

[Skywalker]

Vince,

I already posted it before: proposed solution A doesn't help in the case of the numerical example I made before. A shooter actually shooting major can get scored minor due to lack of accuracy in calculations. The rules shall clearly state what's the accuracy to be used throughout the whole calculation, not only for the results.

Moreover, if you just round or truncate the result of the calculation, and nothing is specified about the accuracy of the data to be used as input for the calculations, there is an inconsistency with the stage and match results calculations, where it is clearly specified the accuracy of the data (i.e. time) to be recorded and used as input for calculations.

I will prefer Yoda's version over your proposed solution B, because it specifies what shall be done according to the available equipment.

I'm not sure about rounding or truncating the final result (i.e. calculated PF): Shred, I'm probably missing something from your posts, but I am not able to figure out how (if you use sufficient accuracy for the input data), truncating the result should be bad. Would you mind making a numerical example?

BTW, Vince, your mate just confirmed what I posted was my experience with several different brands of chronos and scales...

[kimel]

Displaying digits out to x decimal places means NOTHING if it isn't accurate and reproducible.

This is the whole idea behing the Significant Figures stuff Shred mentioned earlier.

The dirty little secret behind the chrono stage is that everything involved in that stage is temperature sensitive to some extent. Chronos will read slightly differently at 0C than at 25C. Powders act differently at different temps. Barrels produce different velocities at different temps. And digital scales wander all over the blooming map with temperature changes if they aren't recalibrated frequently...especially if running on battery power and the cells are weakening.

So shooter A that chronos at 0800 when it is 0C makes minor but shooter B shooting the same exact load in the same type of gun makes major when he chronos at 1500 when it is 25C.

Until we have created a method to appropriately calibrate chronographs out to 10ths of a FPS and we are reasonably certain that this calibration is held throughout the entire match then we cannot in all good conscience use the decimal places for anything more than rounding per the significant figures approach.

In fact, until we have a method to calibrate a chronograph at any place on earth to the SAME STANDARD then the chronograph stage is actually worthless because it is not consistent and fair.

Who is willing to guarantee that their chronograph is accurate to +/ 0.1 FPS? Oehler doesn't even warn you of a problem with their "proof channel" chronos until there is a 5 fps difference between the two channels (best conditions).

Yes, I agree that we should all be loading so that we have a safe margin above the cutoff. But if you are going to draw a hard line then you had damn well be able to justify that line and prove that it is the same right now as it was yesterday and will be tomorrow to the same precision as you are holding the shooters to.

5.6.5.1 says that the chronograph must be accurate to within 25 fps of the original reading with the popper calibration gun/ammo(paraphrased). Think about it...you are trying to hold a competitor's reading to 10ths of a fps but the chrono itself is allowed to wander around for 25 fps?

And if you think I am being silly about these temperature swings then you haven't shot in my neck of the woods. There is a reason we invented the "layered clothing approach" here in the Pacific Northwest.

[Vince Pinto]

I already posted it before: proposed solution A doesn't help in the case of the numerical example I made before. A shooter actually shooting major can get scored minor due to lack of accuracy in calculations. The rules shall clearly state what's the accuracy to be used throughout the whole calculation, not only for the results.

So what's wrong with Version B?

BTW, I generally like Yoda's proposal too, but Shred and others here strongly object to truncation, for the reasons they've given. Now what? If we change Yoda's proposal from "truncate" to "round up" (with slight polishing) would that cause an impromtu Mexican wave?

POSTSCRIPT: Here's a slightly reworded version of Yoda's proposal:

Draft text: The values used for bullet weight and velocity must be rounded up to 1 decimal place, except in respect of equipment used which only displays whole numbers, in which case face values will be used. The resultant power factor is then rounded up to a whole number (no decimals).

[Yoda]

-------------------

Version B:

5.6.3.4 Power factor is calculated using the bullet weight and the average velocity of the 3 rounds fired, according to the following formula:

Power Factor = bullet weight (grains, rounded up to 1 decimal place) x average velocity (feet per second, rounded up to 1 decimal place) / 1000

The resultant power factor is rounded to the nearest whole number. For example, a result of 124.5 (or higher) rounds up to 125, and a result of 124.4 (or lower) rounds down to 124.

-------------------

Will one of these work? If so, which is better? If I can get consensus here, I will take it to the Rules Committee.

Vince,

I agree with B except for the latter part. The competitor already get's the benefit by rounding UP to one decimal in the values for velocity and bullet weigth, IMO the resultant PF must be rounded DOWN (=truncated) to a whole number.

As I mentioned before in this thread - the real outcome is moot considering the other influences which are WAY more significant than the rounding of decimals. But ... we need to establish a procedure without ambiguity for the factors we can control.

Postscript

This can happen when we are both online in the same topic Second draft from Vince modified by Yoda:

POSTSCRIPT: Here's a slightly reworded version of Yoda's proposal:

Draft text: The values used for bullet weight and velocity must be rounded up to 1 decimal place, except in respect of equipment used which only displays whole numbers, in which case face values will be used. The resultant power factor is then rounded down to a whole number (no decimals).

[Neil Beverley]

POSTSCRIPT: Here's a slightly reworded version of Yoda's proposal:

Draft text: The values used for bullet weight and velocity must be rounded up to 1 decimal place, except in respect of equipment used which only displays whole numbers, in which case face values will be used. The resultant power factor is then rounded up to a whole number (no decimals).

My instinct is to truncate but I can live with rounding and would be happy enough to support Vince's revised wording of Yoda's proposal but changed so that the final result is rounded rather than rounded up. (Edited)

Kimel makes a good point about 5.6.5.1.

[Skywalker]

Kimel,

I understand your point, and at some extent Shred's one too: I agree wholeheartedly with your view of temperature being the principal cause of different readings. This is one of the reasons we (here in Italy) usually start chronoing people from 10:00/10:30 a.m., no earlier, to ensure everybody gets chronoed at (almost) the same temperature...

The point I was trying to make was that, even starting from the very same same readings, two different chrono guys could get different PF results due to lack of guidelines in the rulebook.

This is (IMHO) the biggest problem. At present, there is no guaranteed consistency in the calculations method throughtout the whole rulebook; while the same issue is addressed for stage and match results calculations.

Vince, as I said, I'm not sure I'd prefer rounding or truncating the final result, and I'm still trying to figure it out. The reasion I preferred Yoda's version over your Solution B was that it took into account the possible accuracy of the available equipment. Your reworded proposal looks fine to me, provided we all have the same meaning for rounding, that's the one Shred pointed out.

[shred]

Here's an example why truncation is bad. The same logic applies to the final PF value. Say a shooter's bullet weighs 124.999 grains.

If we use that number as-is, the shooter would need approximately 1000.01 fps to make Minor PF

Truncating that to 124 grains requires a velocity of 1008 fps to make Minor.

Rounding 124.999 to the nearest whole number in this case (NOT rounding up, that's different) is equivalent to giving the shooter the benefit of the doubt for all of 0.01 fps. Truncating is equivalent to stealing 8 fps from the shooter.

Which is better?

Note that no chrono we've got can even measure 0.01 fps

[Skywalker]

Shred,

my perplexity arises from this: if you apply rounding method to the input data (i.e. weight and velocity) to allow some tolerance on the measured values, is truncation of the final result (i.e. PF) still that bad?

I can't figure it out from your example.

[shred]

The problem is you can't do much rounding with the inputs. You can't round 1325 fps because you don't know what's after the decimal place since the chrono won't show it to you (in this example, the real velocity value could be anywhere from 1324.50 to 1325.49). If you rounded a bullet weight of 180.4 grains to 180, shooters would complain (and rightfully so)

Rounding goes both ways, up and down. We need to make sure that if it goes down, it only hurts the shooters that are already too low.

Do the math with whatever you measure and then round the final result. Problem solved.

(After some more Excel crunching, I may be leaning towards rounding to the nearest tenth. Thus 124.95 or better would be needed to make 125.0 PF. A few people could still get hosed by the chrono & scale roundoff, but it does seem a bit more stringent, and might make the 'death-of-IPSC-film-at-11 crowd quiet down a bit. It would not help with daily chrono fluctuations, which are far larger than 0.05 PF)

[Vince Pinto]

OK guys,

Now do you see why it takes 12 months, 7,000 emails and 4 days of face-to-face meetings to produce a decent rulebook?

Neil: No need to answer. You were there, every agonising inch of the way. How about we go early to Bali and get sloshed. How about right now?

Yoda: Please reconsider. I just find it hard to swallow having "round-up" and "truncate" in the same rule. Let's be consistent and stick with "round-up". Please? Pretty please? With a tulip on top

Shred: Hang in there, dude. I think Yoda is about to crack

Skywalker: Look towards the light. You are getting sleepy. Verrrrry sleepy. Repeat after me: "Rounding up is better than truncating". Good. When I snap my fingers, you will be assimilated. SNAP

Troy: Where the hell are you? I need my wing man

And, yes, I'll take my medication now .............

[Skywalker]

Skywalker: Look towards the light. You are getting sleepy. Verrrrry sleepy. Repeat after me: "Rounding up is better than truncating". Good. When I snap my fingers, you will be assimilated. SNAP 

I am sleepy...I am sleepy...I'm not seeing the light...my eyes are closing...

for the only reason my little earthquake decided at 3 a.m. to jump in my bed, thus in a few minutes I was (almost) sleeping on the edge of the mattress, my wife was in my place, kid was laying eagle spred on the whole bed, and cat was chasing my right foot out of the blanket...

I woke up at 6 a.m. not relaxed at all....

[Yoda]

Vince,

you know that I graciously yield to the majority. So far it looks like I'm outnumbered and certainly outweighed

I won't crack but go with the flow.

It is the consistent rounding UP which is against my instinct of what is right. As said before - the difference in rounding is minor when looking to the other variable influences. I'll think I go to the chrono between 2:00 and 4:00 PM, last week I chrono'ed my revolver load @ PF 190 so I need all the help I can get

[Jim Norman]

A couple points. The issue of rounding and significant numbers was addressed last night on the IPSC list.

So, WHy are we rounding numbers we can't measure? YOu have a weight expressed in 1/10th grain andd a speed in whole numbers. The math yeilds one decimal point

Chronos and scales as a rule both report 4 significant numbers. XXXX and XXX.X are the same in that respect. Therefore you can not have more than 4 significant figures in your answer. (XXX.X * XXXX)/1000=YYY.Y If YYY.Y >=165.0 you are major, if less you are minor, there is no rounding. no spreadsheet required.

Where is everyone getting these Scales and chorons that report back in 1/1000 grains and FPS?

165.0 tolerance -0.0 plus whatever you want makes major, you go to a match at 165.3 and show as minor, nect time you won't play so close to the edge. Temperature, inherent differences in equipment all make minor differences. As a rule the matches around here us the CED Chronos, two in series, you shoot over both at the same time, and the higher of the two readings is used. THe chrono does the match, you input the bullet weight in XXX.X and the speed is recorded as XXX.X or XXXX, the unit does the math and viola, you get a PF of YYY.Y

As to a decent rulebook, obviously a lot was moissed and now we are faced with "Fixing" a problem via a rules "Interpetation"

Nuf said on that

Jim

[Skywalker]

Jim,

what we are debating here is not an equipment issue, although it might become because not everybody uses the CED Millennium chrono.

What the (general) equipment is capable of has been already pointed out in terms of significant figures and decimals (I guess not to issue rules that will outdate the actual equipment available to everybody).

We are debating what has to be done with equipment readings.

The rulebook nowhere says what is the accuracy of the readings to be used for calculations, and this is what has to be fixed, because right now a chrono officer could (theoretically) decide to truncate chrono and scale readings to integer figures, because the rulebook doesn't specify the accuracy to be used.

OTOH, it clearly states what has to be done with time readings for calculating the stage results.

Theoretically speaking, you could use readings from a timer that records 1/1000 of second, but according to the rules you're not allowed to use readings from a timer that records 1/10 of second only.

This very same clearness on chrono and weight measurements is what is lacking, and what needs to be addressed (IMO).

Your proposal?