Soviet SMGs II

[c3k]

^^^
Perhaps concealment means less firing? Therefore, cover is not a factor in the comparison between the two target types?

[Sgt Joch]

This is interesting and tactically useful information.  

 

Does this 200 meter hard coded limit include both the PPSh and the PPS-42?  The Red Thunder game manual pages 68 & 69 have the effective range for the PPSh at 250 meters and the PPS-42 at 200 meters.  The RT manual, page 102, has the effective range of the Kar 98 at 500 meters.

 

(I have learned that changes are made in the game after the manuals are created so the manuals are not always 100% accurate.)

 

If the PPSh and PPS-42 are hard coded to only fire 200 meters and the German Kar 98 can shoot effectively out to 500 meters (a scoped 98k shoots further yet according to the manual), I think there is a tactical solution to the Soviet SMGs in many situations.

yes PPSH and PPS43.

[Sgt Joch]

Migo's tests piqued my interest so I have been running my own tests, same troops as described previously at 150 meters, but in various terrain.

I suspect the issue of more casualties with foxholes has more to do with the fact that troops in wheatfield break LOF more easily. I ran tests in a low grass map, with and without foxholes and foxholes do seem to result in less casualties.

I am not convinced SMGs are overpowered in game, I have had a fair number of tests where the Germans pretty much wipe out the Soviet SMG squad. You just need an early break where the Soviets get 2-3 casualties early and morale hit and suppression does the rest.

It would be nice if we could get more accurate info on the accuracy of the PPSH so we have an idea what to look for. All I have been able to find is that it had an effective range of 125-250 meters, depending on the source.

[Migo441]

Back to the data to see about the cover vs. concealment.

 

Foxhole vs. Foxhole at 180m

 

 

 

 

There were five cases where the Soviets suffered no casualties.  Those squads fired 3,401 rounds and caused 21 German casualties for an average of 161.95 rounds per casualty.

 

There were two cases where the Germans suffered no casualties.  Those squads fired 675 / 86 (7.92mm / 9mm) rounds and caused 12 Soviet casualties for an average of 56.25 / 7.17 rounds per casualty.

  

Grain Field vs. Grain Field at 180m 

 

 

 

There were seven cases where the Soviets suffered no casualties.  Those squads fired 544 rounds and caused 12 German casualties for an average of 45.33 rounds per casualty.

 

There were six cases where the Germans suffered no casualties.  Those squads fired 272 / 24 (7.92mm / 9mm) rounds and caused 12 Soviet casualties for an average of 22.67 / 2 rounds per casualty.

 

But here I want to be extra-cautious about drawing conclusions because our sample size is dropping.  In fact, in three of the six German no casualty cases for Grain Field vs. Grain Field at 180m, the Germans fired NO ammo.  And in one other, they fired a grand total of three 7.92mm rounds.  So shots per squad would be sketchy (point being sometimes the opposing forces didn't gain full spots and didn't engage), but the shots per casualty is sound (again, with the small sample size caveat).

 

@Sgt. Joch, absolutely, there were cases at each range grouping where the Grenadiers won the engagement but, on average, in every case I tested, the SMGs are favored.

[JasonC]

So we are seeing 2-3 times per shot accuracy edge for the rifles and MGs but 2-5 times shots fired from tne SMGs. In the worst cases for the SMGs they cancel, but the normal case is 3/2 to 2 times the effective firepower.

At 75 meters I'd buy that. At 180 meters, I don't remotely buy it.

[c3k]

I, too, don't like the full-auto accuracy we're seeing with smg's out past 100m. However, as useful as this test is, it does not compare a German rifle squad to a Soviet smg squad. It shows what the two can do when going toe to toe.

 

The next test would/should replace the Soviet smg squad with a German rifle squad (engaging the same target the Soviet smg squad has).

 

As mentioned by me, and tested by Sgt. Joch, the apparent ineffectiveness of the foxholes when compared to wheat was just chaff. It was the lack of LOS to the targets which stopped the firing. (Another test condition would be to Area Target an occupied zone and see whether the smg's edge out bolt actions. Shrug.)

 

Not liking results is not a valid reasoning which will lead to change. What should it be? How much should the muzzle climb, etc? How many of the test casualties were caused by the first round fired in a burst? What about effects which lead to increasing effectiveness? E.g., once the target is pinned, they don't shoot back, the firing unit gains superiority and racks up greater kills. The suppression is heavily weighted (IME) to incoming firepower: of course smg units will pin a target faster than a bolt-action unit. This does not mean the smg unit has "greater" accuracy. They have greater williingness to shoot. Etc.

 

This is a great first look and shows what needs to be examined.

 

Ken

[Migo441]

Ken,

 

Absolutely.  Everything at this stage is very preliminary; I'm not crying "The sky is falling!" and making any dramatic claims about what is "broken" or what needs to be changed.  

 

My tests are not focused narrowly on the accuracy of SMGs or Rifles(+LMG) but rather, as you point out, about what tends to happen when they face on another.  This was inspired by JasonC's comments earlier in the thread about "real world" conditions.  Definitely more variables,  definitely more challenging to draw clear conclusions, etc...  

 

So I think we're all in agreement about approaching these "results" with caution.

 

I never intended to claim (or indeed, I hope, claimed) that units were worse off in foxholes all else equal.  But I did think it was interesting that the concealment benefits of the grain outweighed the cover benefits of the foxholes at 180m.  

 

Your last paragraph there says it well.  Stuff like this a big can 'o worms or big ball 'o wax or big <something> 'o <something>.  And it's precisely BECAUSE of this that we love the game.  It's deep man!  What fun would it be if we had perfect information and everything was crystal clear?  We'd be playing checkers.

 

But hopefully we can start diving into this a little deeper and these preliminary tests suggest further tests, etc...  That's the great thing huh?  There's always more to test than anyone has the time and inclination for.  It's the gift that keeps on giving.  Oftentimes giving frustration, but giving nevertheless!
 

[JasonC]

Foxholes aren't just better cover than wheat, they can be used to gain immediate full concealment when needed by going heads down, they provide safety at will to rally better, they give a stable firing platform, etc.  Yes we all "get" that the way concealment is helping here is by reducing firing occasions not just making resolved fire occasions less effective, but if the game is showing that are more powerful in a wheatfield than it is with foxholes, the game is wrong about that, too.  No real commander would consider his forces better protected by wheat than by foxholes, and they are just right in that assessment.

 

As for the proper 180 meter accuracy with SMGs, even the relatively flat-shooting PPsH, at 100 yards that round is down to 1300 feet per second and have dropped about 4 inches from its point of aim.  The total flight time to that point is about 0.2 seconds.  The flight time fully doubles from that to the 180 meter distance, thanks to declining velocity.  The bullet drop is more like 2 feet at that range. Meanwhile, single shot practical accuracy with a PPsH puts all the rounds in about a 6 inch diameter circle at 50 yards.  At 180 that will have grown to 22 inches - and we are not talking about single shots here.  We are talking about a circle 2 feet around held over the apparent point of aim by 2 feet of height, for the first trigger press, with everything the recoil on full auto does to the gun dispersing the remainder of the burst more widely than that.

 

These are not small differences.  4 inches of drop and a 6 inch circle is a practically accurate weapon for typical exposure of a man in combat trying to use cover.  2 feet of each is maybe effective against an upright stationary man, but you aren't going to have that as your target in combat.

 

For comparison, a K98 is going to put rounds within 3 or 4 inches of the aim point at 180 yards, as long as the shot is fired from a prone position or a supporting rest.  The total bullet drop over that distance is 5 inches.  In other words, the K98 is going to be significantly more accurate at 180 than the PPsH is at 100 yards.  The MG34 or MG42 is firing the same bullet with the same ballistics, from a heavier and more stable platform, full auto.  It will still "wander" more than the rifles will, due to recoil and the dispersion full auto induces, but much less than the PPsH will.  And it is starting from a round that "sees" 180 yards distance as "nearer" than the 100 yard distance "looks" to the PPsH...

[Bulletpoint]

Maybe a bit off topic, but since the test was based on foxholes:

 

Soldiers stick out too far from foxholes, I think. They seem to have most of the torso exposed.

Edited August 31, 2015 by Bulletpoint

[Sgt Joch]

On the bullet drop, these are all valid points, but given enough practice a good shooter can compensate.

 

For example, video of Sterling MK 4 (L2A3) SMG being fired at a target 200 yds away:

 

https://www.youtube.com/watch?v=XYi9-gY6ymM

 

 

granted it is single shots, but the Sterling is not that different in design from the PPSH. The shooter in the video seems to be getting 80-90% hits.

 

In tests by Migo and me, the accuracy of the PPSH fired in 1-2 second bursts at 150-180 yds is in the 0.5-2% range, i.e. 1 shot out of 50 to 1 shot out of 200. That does not seem out of the ballpark.

[c3k]

Foxholes and wheat: this is off topic. Wheat and foxholes cannot be compared (yet, that is happening). What SHOULD happen is units in wheat and units in foxholes IN wheat should be compared. 

 

More, later...

[JasonC]

Sgt Joch - and Jerry Miculek can hit a 6 inch balloon 400 yards away with a 9mm revolver, double action - with his pinky finger, held upside down - twice.  

He has already done it right side up at 1000 yards.  But he is the best shooter in the history of the world, so no it is not exactly what you can expect every private to do on a battlefield.

 

Meanwhile the average trained private can hit a pop up target at that distance with a K98 at least 70% of the time - or he doesn't pass basic marksmenship - and 100% of the time if he can take his time about it.  But that's all on firing ranges.

 

Range stuff isn't going to tell us anything.  The differential though is telling us something.  The full rifle caliber bullets should be pulling ahead at 180 yards because those rounds are at least as accurate at that distance as the SMG is at 100, and the SMG should be seeing a degradation in accuracy of 4 to 6 times at that range.

 

Quantity can be more important at 50 and 100 yards when the practical accuracy of the SMG per bullet is high, like 4 to 6 inch circles.

But not when it is hitting a 2 foot diameter circle held 2 feet above the target.  Then per shot quality is going to matter more than quantity.

[poesel]

So we have something to test.

 

"How does the Kar98 compare to a Soviet SMG at 180m?"

 

Setup: flat earth, no cover. Targets are German 50% fanatic scouts with CA dispersed in an arc 180m from shooter. Scenarios here:

https://dl.dropboxusercontent.com/u/8811801/SMG%20test.zip

 

Shooter A: regular, fanatic driver with Kar98

Shooter B: regular, fanatic 2 man scout team with PPSh & PPS-43

 

Test ran until subjects didn't shoot any more.

 

 

Result:

 

Shooter A: 51 shots - 3 hits - 17 shots/hit

Shooter B: 490 shots - 7 hits - 70 shots/hit

 

The Kar98 in this test is about 3x as accurate as a Soviet SMG.

 

The number of hits is too low for a good statistical result but the direction is as expected.

 

 

(*) side note: can someone remember enough statistics to calculate how 'good' this result is and how often I would have to run the test to get 90%+, 99%+ accuracy?

[JasonC]

Unfortunately, what it is actually saying is that a 3.75-4 rounds burst is as accurate as a rifle shot at 180 yards, which isn't even close to correct...

[c3k]

Against a moving target, which is better? Smg's stream of lead (with impacts giving adjustment feedback to the shooter), or one shot rifles? If a rifle misses, it misses. An smg can get the stream of lead to walk into the target.

 

Just an example...

[poesel]

c3k - the targets were all stationary.

 

So we finally have found something we think the game does wrong wrt SMGs?

 

Could someone please take a look at the scenarios to see if my setup is somehow flawed? I could run it a few more times to get better numbers.

[JasonC]

c3k - sure, one of the reasons I wanted realistic examples is that SMGs might be favored by massed targets, lack of cover, and movement, as all things that can be benefited by spray and pray and hits on targets *other* than the intended one, especially at close range.  Whereas a longer range shot at a stationary and small / single unit target in good cover should bring actual accuracy to the forefront.

 

When 3-4 round bursts from an SMG are outperforming aimed fire even at small stationary targets in cover at 180 meters, it is not because of those legitimate edges to more lead in the air.  It is because the benefit of actual MV and MOA accuracy isn't being modeled as important, and thus any automatic is systematically doing better, even without thus favorable conditions.

[Migo441]

180m - Back to the Range with the Kar98k

 

In the search for MOAR data (to reinforce Poesel's latest tests), I rounded up some Sicherungs troops.  After the assault-split, I had a team of 5 shooters universally equipped with the Karabiner 98k.

 

They deployed in open ground and oriented downrange towards 10-man squads from Soviet '44 SMG Companies (the same guys from the earlier tests) also in open ground.  However, the targets were now upgraded to Fanatic and given limited target arcs so they would not shoot back.

 

We're removing any suppression/casualty effects on the shooters and focusing purely on the effectiveness of the Kar98k according to the standard CM "combat conditions" modifier.  

 

Four lanes and five iterations generated 20 total iterations and a potential 500 shooter-minutes (five shooters per lane for five minutes).

 

Total Rounds Fired: 1,381

Shots per man: 13.81  (100 total shooters)

Shots per man per minute: 2.76 (5 minutes)

 

Casualties*:  50

Rounds per Casualty: 27.62

 

There was a single iteration where the shooters didn't fire a single shot in spite of having a full contact.  Believe me, the Feldwebel managing the range (me) was unimpressed.  When I next saw a team opt not to fire on a full contact, I issued a fire order.  But that was only a single instance out of 100 section-minutes, otherwise it was TacAI all the way.

 

Observed shots per man (team based) ranged from 5.40 to 26.20 over 5 minutes or 1.08 to 5.24 on a per minute basis.

 

Observed rounds per casualty ranged from 11.33 to 90.00.

 

There were two iterations with no casualties caused: the aforementioned section that was dressed down by the Feldwebel and another team that missed with each of 34 shots and was assigned remedial marksmanship training.

 

* Here, I continued to use the more restrictive definition and did NOT count light wounds.  Just important to be consistent when running further tests for comparison's sake.

[JasonC]

22 seconds per shot and 4% chance to hit, a standing target in the open at 180 meters.  

 

OK, real combat accuracy shouldn't be firing range accuracy.  

But that is a really low ROF, and a really poor accuracy when nothing is against the shot (reply fire / suppression, target movement or cover etc).

 

I get that the baseline has to be set low to reflect realistic achieved per shot accuracy under more realistically degraded combat conditions.

If you substitute a sniper team, does the per shot hit rate rise by a factor of 5 to 10?  

Because I haven't see unrealistically low hit chances with those, where it is easier to see each shot etc.

[JasonC]

From the 2:35 mark to the 3:15 mark, 10 rounds from stripper clips, aimed fire, standing unsupported.

 

 

That's 4 seconds per shot including reload time.  Maybe 5.  Not 22.

Granted, they would not always be firing at a max deliberate fire rate, because they need to see a target, might be suppressed, etc.

Longer range, into cover, with LOS obstructions, firing unit suppression, could all reduce ROF to something like the 3 shots a minute the game is showing.

But without those, 3 shots a minute is a solid performance for a man firing a smoothbore musket, not a magazine rifle.

Edited September 1, 2015 by JasonC

[Migo441]

Yep, I think what's reflected here is that even "open ground" isn't open ground.  And the targets aren't necessarily standing, some are kneeling and some are prone.  At a closer range, the spotting would be more comprehensive but out at 180m it seems like only a subset of the shooters sees a subset of the targets at any given time so that reduces the ROF accordingly.

 

So next thing might be to elevate the firing position slightly to give more preferential LOS and place the hapless targets on dirt, pavement, etc...

 

Again, in the interest of trying to dial in on <insert shooting weapon here> accuracy as closely as we're able.

[Sgt Joch]

yes, 1 in 27 does seem on the low side, even in a combat situation.

 

I would also be interested to see the results with sniper teams.

 

If you want to take the terrain completely out of the equation, use a paved road.

[Sgt Joch]

nice video on the Mauser. I have a 60's Parker-Hale .30-06 rifle in my collection with a Mauser Bolt, same type as shown in the video, also with a 5 round internal magazine. Basically the same design as a Kar 98.

 

The Mauser Bolt is heavy, precise and utterly reliable, but it does not like to be rushed. Personally, I would not go as quick as the shooter, I would be too worried about jamming a round or damaging the rifle. It is also not a natural movement to use the bolt when the rifle is at shoulder level, normal practice is to lower the butt to chest level to make the movement easier. When you bring it back to shoulder level, you then have to realign the target. In a pinch, yes, you can probably squeeze off 5 shots in 20 seconds like in the video, but it will not be carefully aimed shots.

Edited September 1, 2015 by Sgt Joch

[c3k]

c3k - sure, one of the reasons I wanted realistic examples is that SMGs might be favored by massed targets, lack of cover, and movement, as all things that can be benefited by spray and pray and hits on targets *other* than the intended one, especially at close range.  Whereas a longer range shot at a stationary and small / single unit target in good cover should bring actual accuracy to the forefront.

 

When 3-4 round bursts from an SMG are outperforming aimed fire even at small stationary targets in cover at 180 meters, it is not because of those legitimate edges to more lead in the air.  It is because the benefit of actual MV and MOA accuracy isn't being modeled as important, and thus any automatic is systematically doing better, even without thus favorable conditions.

 

My bold.

 

I think that -may- be part of it. As well, I think recoil is not being weighted enough, if at all. Call it "muzzle rise" if you prefer. (Something to note: Tigers were phenomenally stable firing platforms. The gunner sight barely moved when the cannon fired. Compare that to tanks which "bounce" when they shoot. Shot sensing, repeatable shots, and consistent aimpoints are possible with one, but not the other. All IRL, not in-game. In-game, there does not seem to be a benefit to having a stable firing platform.)  Regardless of what mechanism we think is modeled or not, the ONLY arguments which will hold water are those which are, forgive the pun, water-tight. 

 

What is needed is real-world effectiveness data and then in-game data under the same conditions. Then, it must be convincingly shown that IT MATTERS. That's the big one, by the way. If it doesn't matter...most of the time...in the game, then nothing that is shown will mean anything.

 

Ken

[JasonC]

Sgt Joch - firing a bolt gun without taking it off the shoulder is not hard, it is just a matter of a little training.  I do it routinely, with a Remington 700 and a Tikka T3 not a K98, but not a big difference.  4-5 seconds per shot does include time to aim, though if one takes 6 to 8 seconds the accuracy might be a bit higher, at longer range, with a scope, etc.  There is no way it takes 22 seconds, even with reloading time, to fire a second aimed shot after the first.  Basically,(short range) rapid fire is about 5 times the rate modeled and deliberate fire is easily 3 times as fast.  The world record with a Lee Enfield - 10 rather than 5 round mag - is 38 aimed shots in 1 minute - at 300 yards. 

Edited September 1, 2015 by JasonC