WWII German Ballistic Penetration Criteria

[jwduquette1]

Could be one or more different sub-species was left off the lists. I found one site that says there was a pzgr. 39/1 type round for this weapon.

 

39-1 is listed on the datenblatte for the Flak18/36.

[Harry]

It seem´s like a big missunderstanding.

Hope I can help a little.

The 9,5kg PzGr. (from Jentz book - battle in North-Africa) was only fired from the 8,8cm Flak 18, 36 und 37 never from the KwK 36 of the Tiger I. I think Jentz believed the 8,8cm Flak in his book about North Africa because in 1941 there was no Tiger in NA - and he wrote about the beginning battles in NA.

 

The problem with the PzGr. 39 and 773m/s is:

Many authors have there data about the KwK36 from the Datenblätter für Heereswaffen und Geräte. And here is the mistake.

The data for the PzGr. 39 and the SprGr. are confound.

In the Datenblatt for the KwK 36 the PzGr. 39 are declared with 9kg and 773m/s - that is wrong. The right weight is 10,2kg with 810/820m/s after H.Dv.481/60.

 

In the Datenblatt for the KwK 36 the SpGr. are declared with 10,2kg and 800/810m/s- that is wrong. The right weight is 9kg with 773/780m/s after H.Dv. 481/60.

 

There are light differences in the PzGr.39 versions, but only in the drive bands. The weight is always the same - 10,2 kg.

 

Harry

[jwduquette1]

It seem´s like a big missunderstanding.

Hope I can help a little.

The 9,5kg PzGr. (from Jentz book - battle in North-Africa) was only fired from the 8,8cm Flak 18, 36 und 37 never from the KwK 36 of the Tiger I. I think Jentz believed the 8,8cm Flak in his book about North Africa because in 1941 there was no Tiger in NA - and he wrote about the beginning battles in NA.

 

The problem with the PzGr. 39 and 773m/s is:

Many authors have there data about the KwK36 from the Datenblätter für Heereswaffen und Geräte. And here is the mistake.

The data for the PzGr. 39 and the SprGr. are confound.

In the Datenblatt for the KwK 36 the PzGr. 39 are declared with 9kg and 773m/s - that is wrong. The right weight is 10,2kg with 810/820m/s after H.Dv.481/60.

 

In the Datenblatt for the KwK 36 the SpGr. are declared with 10,2kg and 800/810m/s- that is wrong. The right weight is 9kg with 773/780m/s after H.Dv. 481/60.

 

There are light differences in the PzGr.39 versions, but only in the drive bands. The weight is always the same - 10,2 kg.

 

Harry

 

Hi Harry:

 

Thanks. Do you by chance have a scan of the Kwk 36 Datenblatt that you can post?

 

As to the rest I'm not exactly sure, but I think what Steve is trying to ask regards the FLAK 18/36 pzgr projectile. The US Army prepared a Tech Manual for the Flak 18/36 in 1943. Steve is hypothesizing that the pzgr. projectile described in the US Army Tech Manual is a significantly different projectile than the FLAK 18/36 pzgr. round being described by Jentz in "Tank Battles of North Africa" and "Dreaded Threat". These differences would be above and beyond simple driving band changes between KPS Fuhrung and FES Fuhrung. Moreover I think the dots he is attempting to contact would entail a 30% jump in armor penetration capability between his two FLAK 18/36 pzgr. projectiles.

 

NOTE: This is specific to the FLAK 18/36 pzgr projectile -- not pzgr39 or 39-1

 

Regards

Jeff

Edited January 5, 2007 by jwduquette1

[Mobius]

As to the rest I'm not exactly sure, but I think what Steve is trying to ask regards the FLAK 18/36 pzgr projectile. The US Army prepared a Tech Manual for the Flak 18/36 in 1943. Steve is hypothesizing that the pzgr. projectile described in the US Army Tech Manual is a significantly different projectile than the FLAK 18/36 pzgr. round being described by Jentz in "Tank Battles of North Africa" and "Dreaded Threat". These differences would be above and beyond simple driving band changes between KPS Fuhrung and FES Fuhrung. Moreover I think the dots he is attempting to contact would entail a 30% jump in armor penetration capability between his two FLAK 18/36 pzgr. projectiles.

Thanks Harry and Jeff.

Significatly yes.

From Jeffs datenblatt image:

http://www.panzer-war.com/Images/88mmflak1...nblattglm3a.jpg

#1 This might be the m. Bd. Z the Allies evaluated.

#2 Or not.

#3 What does "Ltgr." stand for?

#4 Here are two curious entries. The "Ltgr." round below, but what is the one above it?

Edited January 5, 2007 by Mobius

[Harry]

Jeff,

 

I will send you the Datenblätter tomorrow.

 

Harry

[Harry]

Thanks Harry and Jeff.

Significatly yes.

From Jeffs datenblatt image:

http://www.panzer-war.com/Images/88mmflak1...nblattglm3a.jpg

#1 This might be the m. Bd. Z the Allies evaluated.

#2 Or not.

#3 What does "Ltgr." stand for?

#4 Here are two curious entries. The "Ltgr." round below, but what is the one above it?

 

 

1. that is the normal base fuze for the Pz.Gr.

2. That is the Pz.Gr. as described in German tanks of WW2

3. Ltgr. means Leuchtgranate = illuminating

4. Siggr = Signalgranate

 

Harry

[Alvaro]

French post-WWII military publication ("Notice sur la DCA dans l'aemee allemande) from British data:

 

Paz. Gr. m. Bd. Z. penetration data:

 

meters...velocity..90º...60º

0...........793.......141...118

457........748.......130...110

914........703.......119...108

1371......658........109...94

1828......617........99.....87

2285......577........90.....80

 

Weight: 9.5 kg

MV: 810 m/s

Charge: 2.420

 

For comparison, Pzgr39 was listes as:

Charge: 2.950

Weight: 10 kg

MV: 800 m/s

[jwduquette1]

French post-WWII military publication ("Notice sur la DCA dans l'aemee allemande) from British data:

 

Paz. Gr. m. Bd. Z. penetration data:

 

meters...velocity..90º...60º

0...........793.......141...118

457........748.......130...110

914........703.......119...108

1371......658........109...94

1828......617........99.....87

2285......577........90.....80

 

Weight: 9.5 kg

MV: 810 m/s

Charge: 2.420

 

For comparison, Pzgr39 was listes as:

Charge: 2.950

Weight: 10 kg

MV: 800 m/s

 

 

Hi Alvaro:

 

I assume the above is from the report Frederic dug up from Saumur. Great images from the projectile testing against the Sherman and M10.

 

I wasn't able to find the note in the Saumur report indicating the penetration data for 88mm pzgr. was English based. Although it does look more like WO 185/178 data than the American TM data. Was the English origin something you and Fredric discovered? The 793m/s indicated in the Saumur report is also pretty much right out of the British WO 185/178 report -- moreover the British call Vo = 2600-fps (or about 792.5m/s). However they also indicate the following description:

 

"German 88mm (Anti-Aircraft mounted as Tank Gun) APCBC/HE MV 2600f.s. Weight 21 lb."

 

Specific projectile type -- beyond the above description -- is not indicated by the British

 

WO 185/178 Also Indicates:

Perforation @ 30-deg

500yrd = 114mm

1000-yrds = 105.5mm

1500-yrds = 96.5mm

 

It seems fairly evident to me whats going on. But I suppose folks should just look at the data and make up their own minds.

 

Jeff

[jwduquette1]

one more bit just for giggles -- Agar Hamilton wrote an excellent book on the Crusader Operations called "The Sidi Rezeg Battles, 1941". In it, Hamilton references a South African Army report dated 6 July 1942 which details penetration capability for a number of German guns. The penetration data was supposedly derived from captured German Documents. It indicates the Flak 88's "AP" projectile could penetrate the following levels of armor at 30-deg:

 

71mm @ 500m

67mm @ 1000m

65mm @ 1500m

63mm @ 2000m

 

He goes on to say that British Sources on the gun give a markedly better penetration for the 88mm.

[jwduquette1]

It seem´s like a big missunderstanding.

Hope I can help a little.

The 9,5kg PzGr. (from Jentz book - battle in North-Africa) was only fired from the 8,8cm Flak 18, 36 und 37 never from the KwK 36 of the Tiger I. I think Jentz believed the 8,8cm Flak in his book about North Africa because in 1941 there was no Tiger in NA - and he wrote about the beginning battles in NA.

 

The problem with the PzGr. 39 and 773m/s is:

Many authors have there data about the KwK36 from the Datenblätter für Heereswaffen und Geräte. And here is the mistake.

The data for the PzGr. 39 and the SprGr. are confound.

In the Datenblatt for the KwK 36 the PzGr. 39 are declared with 9kg and 773m/s - that is wrong. The right weight is 10,2kg with 810/820m/s after H.Dv.481/60.

 

In the Datenblatt for the KwK 36 the SpGr. are declared with 10,2kg and 800/810m/s- that is wrong. The right weight is 9kg with 773/780m/s after H.Dv. 481/60.

 

There are light differences in the PzGr.39 versions, but only in the drive bands. The weight is always the same - 10,2 kg.

 

Harry

 

Now this is rather facinating -- the schusstafeln for the KwK 36(L/56) pzgr39 FES and pzgr.39-1 and pzgr.39 Al indicates Vo = 780m/s. hmmm...course there is always the new gun -- average gun -- and old gun Vo's and associated drop in Vo as a result of decreasing obturation. Some folks calibrate their firing tables to new gun specs -- some folks like average gun specs.

[Mobius]

Some folks calibrate their firing tables to new gun specs -- some folks like average gun specs.

Hopefully not the same folks.

[Alvaro]

Hi Alvaro:

 

I assume the above is from the report Frederic dug up from Saumur. Great images from the projectile testing against the Sherman and M10.

 

I wasn't able to find the note in the Saumur report indicating the penetration data for 88mm pzgr. was English based. Although it does look more like WO 185/178 data than the American TM data. Was the English origin something you and Fredric discovered? The 793m/s indicated in the Saumur report is also pretty much right out of the British WO 185/178 report -- moreover the British call Vo = 2600-fps (or about 792.5m/s). However they also indicate the following description:

Specific projectile type -- beyond the above description -- is not indicated by the British

 

WO 185/178 Also Indicates:

Perforation @ 30-deg

500yrd = 114mm

1000-yrds = 105.5mm

1500-yrds = 96.5mm

 

It seems fairly evident to me whats going on. But I suppose folks should just look at the data and make up their own minds.

 

Jeff

 

Hi Jeff:

 

Although I think it is the same report you know, author is Colonel Michelet, but it is not from Saumur and it was not recovered by Frederic, but by another French man. It seems he (I just cannot remember now his name) got it directly from Col. Michelet!. It seems that the source of this and other French penetration data was British, at least this is what a friend, retired French Armor colonel, said to me when we joined and commented this report last year. It has sense, because ovbiously French could not make firing tests during the war. Of course we may be wrong. At the time this report was done only Pzgr39 was available.

 

Alvaro

Edited January 6, 2007 by Alvaro

[jwduquette1]

Hi Jeff:

 

Although I think it is the same report you know, author is Colonel Michelet, but it is not from Saumur and it was not recovered by Frederic, but by another French man. It seems he (I just cannot remember now his name) got it directly from Col. Michelet!. It seems that the source of this and other French penetration data was British, at least this is what a friend, retired French Armor colonel, said to me when we joined and commented this report last year. It has sense, because ovbiously French could not make firing tests during the war. Of course we may be wrong. At the time this report was done only Pzgr39 was available.

 

Alvaro

 

Thanks Alvaro. I don't disagree that the origins seem like they are British. I just thought I had missed a footnote or the like that indicated the original source of the data. Did any of these folks ever track down the sister report the French did post-war on the Panther?

 

I think you have ADM 213/951. Have a look at pages 20-25. It's a practical example of how the Germans were determining limit velocity for projectiles and plates. Pretty much a direct answer to the original intent -- as I see it -- for this thread. Everything else here is simply another excercise in clutter and meander.

 

Regards

Jeff

[Mobius]

914........703.......119...108

Alvaro, could you double check this figure?

Thanks.

[jwduquette1]

I was reviewing BIOS Final Report No 1343, “German Steel Armour Piercing Projectiles & Theory of Penetration”. One can find a copy of this report at the PRO archives. Similar to the BIOS report on German Armor Steel – UK sent a bunch of fellows over to Germany after the capitulation and interviewed key German scientistis involved with Armour Piercing projectile design and manufacture, as well as translated a great deal of German documentation on the subject. Anyway to cut to the chase, there is an example of how the Germans determined limit velocity for armor piercing projectiles on pages 21 – 25 of the report.

 

The Germans defined two critical limiting velocities – GS & GD. Pretty much analogous (in fact almost identical) to The United States methods of determing Army Ballistic Limit BL(A) and Navy Ballistic BL(N) limits.

 

GS is a plate specific limit in which the plate provides complete protection against a specific projectile. The break between partial penetration and complete penetration is defined as no cracking that allows light to be seen through the opposite side of the plate. “The Human Eye must be safe behind the plate”. Those familiar with how BL(A) is defined will note the almost identical definition between the German’s GS and BL(A).

 

GD on the other hand is projectile specific, and defines the limit velocity at which the projectile will just penetrate the plate with no remaining velocity. Again almost identical to the USA and their BL(N) limit velocity. Both GD and BL(N) are focused upon projectile passage and condition of the bursting charge cavity following passage.

 

There is an example of how both GS and GD are calculated on page 23 of the report. Below are the testing velocities, impact results, as detailed in the BIOS report example:

 

German Test Data; Determination of GD Limit Velocity

 

PP: Is a Partial Penetration, i.e. the projectile was not capable of passing through the plate at this velocity.

 

CP: Complete Penetration, i.e. the projectile passes through the plate at this specific velocity.

 

The GD limit velocity for this set of trials is indicated as 777.8m/s. The Germans would than round the number (+) upward to the nearest whole ten – so they would call it 780m/s.

 

The USA method for determination of BL(N) would be to employ an “up and down” firing method until 3 complete penetrations (CP) and 3 partial penetrations (PP) are obtained within a velocity spread no greater than 100-fps (~30m/s). Or; If the zone of mixed PP and CP results is greater than 100-fps, “fire additional rounds to obtain a minimum of 3 CP and 3PP in the mixed zone. Or; If no mixed zone is obtained, average the velocities of the lowest CP and the highest PP. Or; If the mixed zone is less than 100-fps, average the velocities of the 3 lowest CPs and three highest PPs. Or; If there are an uneven number of PPs and CPs within the mixed zone, than one of two weighted averages is employed to determine the limit velocity. The first weight average applies if the number of PPs in the mixed zone exceeds the number of CPs in the mixed zone:

 

EQUATION-1

BL = VA + [((Np – Nc)/(Np + Nc)) x (Vhp – VA)

 

VA = Average of all velocities in the mixed zone

Np = the number of PPs in the mixed zone

Nc = the number of CPs in the mixed zone

Vhp = the velocity of the highest PP

 

The second weighted average applies if the number of CPs in the mixed zone exceeds the number of PPs in the mixed zone:

 

EQUATION-2

BL = VA - [((Nc – Np)/(Nc + Np)) x (VA – Vlc)

 

VA = Average of all velocities in the mixed zone

Np = the number of PPs in the mixed zone

Nc = the number of CPs in the mixed zone

Vlc = the velocity of the lowest CP

 

Using the above methodology to determine BL(N), and employing the same velocity data detailed in BIOS for the German testing, the best mixed zone that also falls within the 100-fps (30m/s) test bracket requirement is shown by the data I have shaded in blue. Two PP at respectively 756.3-m/s and 768-m/s. Rounding out the six total shot requirement you end up with four CPs at 772.2, 773.1, 774.5 and 776.6-m/s. This leaves us with a mixed zone with a velocity bracket of less than 100-fps (30m/s). The mixed zone in this instance contains a greater number of CPs than PPs so we employ EQUATION-2 to determine the V50 BL(N). If you work through the math you should get the following:

 

VA of the mixed zone = 770.1m/s

Nc = 4

Np = 2

Vlc = 772.2m/s

BL(N) = 770.8m/s

 

The difference between the USA methodology for determining V50 BL(N) -- i.e. 770.8m/s -- and the German method of determining GD – i.e. -- 777.8m/s is about 1%. Rounding the German figure up to 780m/s leaves a percentage difference of about 1.3%.

 

What does this mean in practical terms? Here is an example...Consider the firing tables provided in TM E9-369A for FLAK 18/36 firing pzgr. Lets assume the above example applies to our 88mm pzgr round, and the limit velocity of our projectile for a specific plate is 770.8m/s (coincidently the same as the USA methods for determining BL(N) listed above). Looking at the firing tables we see that this velocity falls at a little less than 400-meters range – about 392meters. Now consider the German determined limit velocity of 777.8m/s. That’s is going to be a bit over 300meters range – about meters 321meters. So the contrast in range between a limit velocity of 770.8m/s and 777.8m/s works out to be a difference of 71meters. In terms of 780m/s the range difference is 92meters.

 

In terms of actual armor penetration vs. range, the differences are even less significant. Consider the Jentz penetration figures (German datenblatte) pzgr. fired by 88mm Flak 18/36. Penetration at 392meters works out to be about 94.1mm. The penetration at 321meters works out to be about 94.8mm, and the penetration at 300meters works out to be 95mm. The difference is peanuts. This level of error isn’t worth sweating over when one considers the sort of statistical slop entailed in both the test methodolgies. Variations in battlefield muzzle velocities from standard conditions, or from propellant temperature, or gun age, will introdce the same level of error or greater. Moreover if I have a new Tiger barrel shooting pzgr39 at Vo = 780m/s and an average barrel shooting at Vo = 773m/s should I worry myself over that extra 1/4mm of armor penetration the new Tiger barrel will do? What if the new Tigers interior space heater isn’t working, but the old Tigers heater is working. The cartridge propellant temperature in the new Tigers ready racks might be 30 or 40-degrees colder than the old Tiger’s ready racks (it’s a cold winter day on the Russian Steppes). Now my new Tiger shoots at only Vo = 770m/s and my old Tiger still shoots at Vo = 773m/s. Do I worry about that extra 0.157mm of armor that the old Tigers pzgr39 can punch through? If your answer is yes – well – good luck to yah -- but don’t loose the forest cause of all them darned trees.

 

If you understand the sorts of error sources involved with determining ballistic limits and plate penetration vs. the possible variables on the battlefield you should realize that normalizing German penetration data to an American or British standard is a waste of time -- particulalrly when a flawed slope effects model may be introducing a error of some 20% or 30% into the mess. To me it makes more sense to focus ones efforts on error sources that make a difference.

 

But hay, that’s just me and my two cents worth. Everyone needs to find their own path to the truth.

 

Regards

Jeff

[jwduquette1]

Alvaro, could you double check this figure?

Thanks.

 

double checked --

it's 108mm at 914m

[Alvaro]

Alvaro, could you double check this figure?

Thanks.

 

 

Mobius, yes, this is the exact transcription. But it seems more logical 103 instead, as an 3 can be mistaken more easy with 8.

 

Regards

Alvaro

[Mobius]

Thanks Jeff and Alvaro,

Jeff your calculations seem fine. For this one case. Though I think the data is biased. There should be something in the methodology where the velocity cannot be less than 2 m/s with another test.

 

But are you sure these are equivalent?

GS is a plate specific limit in which the plate provides complete protection against a specific projectile. The break between partial penetration and complete penetration is defined as no cracking that allows light to be seen through the opposite side of the plate. “The Human Eye must be safe behind the plate”.

 

PP: Is a Partial Penetration, i.e. the projectile was not capable of passing through the plate at this velocity.

 

The problem I have is reconciling the small differences per calculations between v50 and v80 you have from the seemingly large differences between v20 and v80

1 The Initial Penetration (IP) means the 20% probability of armor penetration. The Certified Penetration (CP) means the 80% probability of armor penetration;

found on the data on the Russian Battlefield site here: Russian Battlefield

The average differences found between IP and CP don't look like 2% me. It seems like the velocity would be much greater than 30m/s to produce such large differences between patial penetrations and full penetrations.

 

Plus they sometimes use a 'tabulated type' figure:

The tabular data were theoretical and were calculated by method of ARTKOM (Artillery Committee). This method was accepted in 1939 and its final result depended of two values:

- The Limit of the Through Penetration (LTP) when a whole projectile penetrate the armor and was found behind the armor plate.

- The Breaking Point of the Back Surface (BPBS) when a projectile didn't penetrate the armor but the back surface of the armor plate happened to be damaged.

 

The armor counted to be penetrated if at least 75% of a projectile's fragments happened to be found behind the armor plate. Most of the Soviet armor penetration tables based on this value.

 

I'm a litte confused on their 'theoretical' being from tests. But that's another matter.

Edited January 6, 2007 by Mobius

[Alvaro]

Plus they sometimes use a 'tabulated type' figure:

I'm a litte confused on their 'theoretical' being from tests. But that's another matter.

 

 

It is possible to make short tests of a new projectile at a given strive velocity to obtain the projectile/plate constant required to fill some ballistic equations, DeMarre, ArtKom, etc. and predict penetration at other velocities and sometimes… even at other angles of attack .

 

The accuracy of these mixed tested/theoretical curves is good provided the penetration mechanics remain the same, which is not always the case. For example if ArtKom curves where made for AP(BC) projectiles against very hard test plates, surely those projectiles would break in some extent at impact. (I never seen a test of uncapped projectile to remain intact against very hard plates). But if these same projectiles where tested again softer armor, perhaps they remain whole and then ArtKom equation become pessimistic. If this was the case, Russian penetration figures would be worst case ones with room for improvement when projectile damage was minor.

 

Best Regards

Alvaro

Edited January 8, 2007 by Alvaro

[Mobius]

I made few measurements from the illustrations of the two large burster types of AP shells described in the US Ordinance TM

These were based on the diameters being 88mm.

Pzgr.

Burster Cavity diameter: 53.26mm

Burster Cavity height:151mm

 

Pzgr. Patr. m. Bd. Z.

Burster Cavity diameter: 46.9mm

Burster Cavity height:124mm

 

Maybe this is why there is a penetration difference.

Edited January 8, 2007 by Mobius

[Harry]

From a other forum:

 

http://213.147.167.60/blaze/viewforum.php?f=7

 

 

"Looks like the only evidence for the 3 projectiles comes from the american TM pages (OP 1666 ?).

American references of those days are really poor, especially OP 1666.

For this you only have to look at page 447, where a 8,8 cm Pzgr 39/43 is portrayed as a 8,8 cm Pzgr 39/1.

On other pages, the information given is downright wrong; figure 361 and 369 for instance look nothing like the originals.

 

The only difference i can come up with between the 2 8,8 cm Pzgr. shown is the difference in drivingbands, but that's quit normal on german shells and cannot be interpreted as a new type of shell.

 

Original german documents are good sources, the rest should be read with caution."

 

Harry

[Mobius]

"Looks like the only evidence for the 3 projectiles comes from the american TM pages (OP 1666 ?).

American references of those days are really poor, especially OP 1666.

For this you only have to look at page 447, where a 8,8 cm Pzgr 39/43 is portrayed as a 8,8 cm Pzgr 39/1.

I can't tell the difference between the two and I'm not a member of that forum. But didn't the 39/43 have a big 'A1' near its nose? And 1/3 pound is closer to 150g burster the 39/1 had than (.3 pound) 137g buster of the 39/43, which is on page 442.

Its TM 9-1985-3/TO 39 B-1 A-10

Edited January 9, 2007 by Mobius

[Harry]

Compare the 88mm Pzgr. with 9.5kg and the Pzgr.39 with 10.2kg from the KwK36 the penetration at 30° is for the first around 97mm for the Pzgr.39 120mm. If you got to the 2000m range the figures are 73mm for Pzgr. and 84mm for the Pzgr.39.

 

So the difference in penetration going down from a plus at 100m of 23-24% to a plus at 2000m of only 15% for the Pzgr.39.

 

What is the reason for the difference going less?

 

Has the 9.5kg Pzgr. better longrange ballistics, or is the reason the hitting angle at this longrange?

 

Harry

[Mobius]

What is the reason for the difference going less?

Has the 9.5kg Pzgr. better longrange ballistics, or is the reason the hitting angle at this longrange?

Probably just velocity. The impact angle is 1.125°.

 

Maybe the large explosive chamber in the 9.2kg led it to fracture at the higher velocity.

Edited February 22, 2007 by Mobius

[wilsonam]

If you understand the sorts of error sources involved with determining ballistic limits and plate penetration vs. the possible variables on the battlefield you should realize that normalizing German penetration data to an American or British standard is a waste of time -- particulalrly when a flawed slope effects model may be introducing a error of some 20% or 30% into the mess. To me it makes more sense to focus ones efforts on error sources that make a difference.

Jeff,

 

Belatedly joining the postings here - but I think I'll wrap that quote around a large bat and use it to drum the issues into people's heads. I go slowly mental trying to explain the issues of modelling the effects based on shaky, guesswork measurements, anecdotal evidence and poorly-understood mechanics to the computer gaming world. I swear it has made me take up drinking. No, come to think of it, I already liked whisky