[A13] Combat Realism v.1.6.5 (08.05.16) Final release

Started by NoImageAvailable, January 29, 2015, 12:27:41 PM

Previous topic - Next topic

NihilRex

Quote from: NoImageAvailable on April 10, 2015, 03:50:12 PM
Quote from: NihilRex on April 10, 2015, 02:29:45 PM
Is it at all possible to make Mortars more realistically accurate?

At no point should Circular Error Probability exceed burst radius, and with the sheer amount of aiming time spent, the aiming error should be rather small as well.

Max accuracy should be at 50% of maximum range, with worst accuracy trending towards the minimum range limit.  The accuracy degradation from midrange to max range should occur at about half the rate as the degredation that occurs between midrange and minimum range.

As a former 0341 Infantry Mortarman, the mortars in this game make me very sad.

Changing the way aiming works is pretty trivial, it mostly comes down to creating a mathematical algorithm to calculate the accuracy at any given range (and not kill your CPU in the process). Current calculations are based on a CEP of 50m and a scale of 1 game cell = 5m. Using the game's ForcedMissRadius (works by generating a random angle and a number between 0 and ForcedMissRadius and applying the resulting vector to the original target you get the cell the mortar will actually shoot at) I set it to 20 so half the rounds should land within 10 cells.

So going with your model and assuming the Rimworld mortar is 120mm with a max range of 6-7km (?) that would put maximum range at 1400 and maximum accuracy at 700 game cells. Now I never operated a mortar so I don't really know what kind of accuracy you can expect at that range and at minimum range (40 cells) but provided some accurate data on the start and end points it should be simple to make the accuracy scale linearly with target distance. Also, what exactly do you mean by burst radius?

On that note, depending on how accurate the game's ForcedMissRadius models CEP that can be adjusted too. I'd love to have some input from an actual expert on military matters when it comes to implementing these things ingame, rather than having to rely on stuff I read on the Internet.

Edit: I've been considering changing the scale of cells to meter from 1 cell = 5m down to 1 cell = 7.5m at least for the long range weapons (=everything except SMGs, pistols and shotguns) so range would be bit more balanced in relation to the game's map size (maximum is 400x400 cells). Anyone have thoughts on that?

Burst radius of the shell - the area covered by the 50% probability of causing lethal injuries.

Rimworld Mortars are probably 60mms or 81mms, not 120mms, as 120s cannot be single manned.

Infantry mortars can be aimed without an FDC, and 60mms can be operated by a single person.  The amount of time the pawns spend aiming the mortar makes me think that they are doing their own FDC math, not just aiming by eye.

Modern 60mm infantry mortars have an effective range of 3490meters with the common M888 HE round.  The lethal radius is approximately 28meters.  http://www.globalsecurity.org/military/library/policy/army/fm/7-90/figb-5.gif for reference, Vietnam\Cold War era.

EDIT: This would make the mortar bursts have a smaller radius than current, but be much more accurate.  If there is a 6tile lethal and a 12tile casualty radius, impact error should be 8-11 tiles.  Aiming error at 50% range should be no more than 2 tiles.  Obviously, this would need to be tested for playability, and the superscience used by the colonists, but the current system is nowhere near realistic.

dustysniper

Quote from: NihilRex on April 10, 2015, 11:17:16 PM
Quote from: NoImageAvailable on April 10, 2015, 03:50:12 PM
Quote from: NihilRex on April 10, 2015, 02:29:45 PM
Is it at all possible to make Mortars more realistically accurate?

At no point should Circular Error Probability exceed burst radius, and with the sheer amount of aiming time spent, the aiming error should be rather small as well.

Max accuracy should be at 50% of maximum range, with worst accuracy trending towards the minimum range limit.  The accuracy degradation from midrange to max range should occur at about half the rate as the degredation that occurs between midrange and minimum range.

As a former 0341 Infantry Mortarman, the mortars in this game make me very sad.

Changing the way aiming works is pretty trivial, it mostly comes down to creating a mathematical algorithm to calculate the accuracy at any given range (and not kill your CPU in the process). Current calculations are based on a CEP of 50m and a scale of 1 game cell = 5m. Using the game's ForcedMissRadius (works by generating a random angle and a number between 0 and ForcedMissRadius and applying the resulting vector to the original target you get the cell the mortar will actually shoot at) I set it to 20 so half the rounds should land within 10 cells.

So going with your model and assuming the Rimworld mortar is 120mm with a max range of 6-7km (?) that would put maximum range at 1400 and maximum accuracy at 700 game cells. Now I never operated a mortar so I don't really know what kind of accuracy you can expect at that range and at minimum range (40 cells) but provided some accurate data on the start and end points it should be simple to make the accuracy scale linearly with target distance. Also, what exactly do you mean by burst radius?

On that note, depending on how accurate the game's ForcedMissRadius models CEP that can be adjusted too. I'd love to have some input from an actual expert on military matters when it comes to implementing these things ingame, rather than having to rely on stuff I read on the Internet.

Edit: I've been considering changing the scale of cells to meter from 1 cell = 5m down to 1 cell = 7.5m at least for the long range weapons (=everything except SMGs, pistols and shotguns) so range would be bit more balanced in relation to the game's map size (maximum is 400x400 cells). Anyone have thoughts on that?

Burst radius of the shell - the area covered by the 50% probability of causing lethal injuries.

Rimworld Mortars are probably 60mms or 81mms, not 120mms, as 120s cannot be single manned.

Infantry mortars can be aimed without an FDC, and 60mms can be operated by a single person.  The amount of time the pawns spend aiming the mortar makes me think that they are doing their own FDC math, not just aiming by eye.

Modern 60mm infantry mortars have an effective range of 3490meters with the common M888 HE round.  The lethal radius is approximately 28meters.  http://www.globalsecurity.org/military/library/policy/army/fm/7-90/figb-5.gif for reference, Vietnam\Cold War era.

EDIT: This would make the mortar bursts have a smaller radius than current, but be much more accurate.  If there is a 6tile lethal and a 12tile casualty radius, impact error should be 8-11 tiles.  Aiming error at 50% range should be no more than 2 tiles.  Obviously, this would need to be tested for playability, and the superscience used by the colonists, but the current system is nowhere near realistic.

120mm mortars can be two manned. You can single man a 81mm mortar. It is most likely a 81mm mortar not a 60 (it is way to big), 60s can be held and carried by one man thats why infantry platoons carry a mortar section with 60s on patrol. And yes mortars can be aimed without FDCs through direct line but these mortars are lobbing over mountains and structures good luck hitting anything without line of sight or a FO. And they could possibly be using a plotting board since it takes them so long. As far as I have ever known the 60mm mortar has a kill radius of 25m 81mm 35m and 120 70-75m.

Also remember at 3490m a 60 is shooting charge 4.

NihilRex

Quote from: dustysniper on April 10, 2015, 11:41:29 PM
Quote from: NihilRex on April 10, 2015, 11:17:16 PM
Quote from: NoImageAvailable on April 10, 2015, 03:50:12 PM
Quote from: NihilRex on April 10, 2015, 02:29:45 PM
Is it at all possible to make Mortars more realistically accurate?

At no point should Circular Error Probability exceed burst radius, and with the sheer amount of aiming time spent, the aiming error should be rather small as well.

Max accuracy should be at 50% of maximum range, with worst accuracy trending towards the minimum range limit.  The accuracy degradation from midrange to max range should occur at about half the rate as the degredation that occurs between midrange and minimum range.

As a former 0341 Infantry Mortarman, the mortars in this game make me very sad.

Changing the way aiming works is pretty trivial, it mostly comes down to creating a mathematical algorithm to calculate the accuracy at any given range (and not kill your CPU in the process). Current calculations are based on a CEP of 50m and a scale of 1 game cell = 5m. Using the game's ForcedMissRadius (works by generating a random angle and a number between 0 and ForcedMissRadius and applying the resulting vector to the original target you get the cell the mortar will actually shoot at) I set it to 20 so half the rounds should land within 10 cells.

So going with your model and assuming the Rimworld mortar is 120mm with a max range of 6-7km (?) that would put maximum range at 1400 and maximum accuracy at 700 game cells. Now I never operated a mortar so I don't really know what kind of accuracy you can expect at that range and at minimum range (40 cells) but provided some accurate data on the start and end points it should be simple to make the accuracy scale linearly with target distance. Also, what exactly do you mean by burst radius?

On that note, depending on how accurate the game's ForcedMissRadius models CEP that can be adjusted too. I'd love to have some input from an actual expert on military matters when it comes to implementing these things ingame, rather than having to rely on stuff I read on the Internet.

Edit: I've been considering changing the scale of cells to meter from 1 cell = 5m down to 1 cell = 7.5m at least for the long range weapons (=everything except SMGs, pistols and shotguns) so range would be bit more balanced in relation to the game's map size (maximum is 400x400 cells). Anyone have thoughts on that?

Burst radius of the shell - the area covered by the 50% probability of causing lethal injuries.

Rimworld Mortars are probably 60mms or 81mms, not 120mms, as 120s cannot be single manned.

Infantry mortars can be aimed without an FDC, and 60mms can be operated by a single person.  The amount of time the pawns spend aiming the mortar makes me think that they are doing their own FDC math, not just aiming by eye.

Modern 60mm infantry mortars have an effective range of 3490meters with the common M888 HE round.  The lethal radius is approximately 28meters.  http://www.globalsecurity.org/military/library/policy/army/fm/7-90/figb-5.gif for reference, Vietnam\Cold War era.

EDIT: This would make the mortar bursts have a smaller radius than current, but be much more accurate.  If there is a 6tile lethal and a 12tile casualty radius, impact error should be 8-11 tiles.  Aiming error at 50% range should be no more than 2 tiles.  Obviously, this would need to be tested for playability, and the superscience used by the colonists, but the current system is nowhere near realistic.

120mm mortars can be two manned. You can single man a 81mm mortar. It is most likely a 81mm mortar not a 60 (it is way to big), 60s can be held and carried by one man thats why infantry platoons carry a mortar section with 60s on patrol. And yes mortars can be aimed without FDCs through direct line but these mortars are lobbing over mountains and structures good luck hitting anything without line of sight or a FO. And they could possibly be using a plotting board since it takes them so long. As far as I have ever known the 60mm mortar has a kill radius of 25m 81mm 35m and 120 70-75m.

Also remember at 3490m a 60 is shooting charge 4.

Charge 4 is no issue.

81s can be 1manned, but not accurately.  How are you gonna hold the adjustable leg down and safely drop the round?  You can sandbag the fixed leg, but the adjustable leg needs your palm, using your foot would either jostle it or let it move too much for accuracy.

60s on charge 1, handheld, max out at either 400m or 500m IIRC, and fire a LOT faster than these.  It is still possible for a 1st year enlistee to get steel on steel hits on a tank hulk.  It was either a Patton or an Abrams before it became a target.  (I know, because I've both done it and seen it done by others.) 

60mm mortars have a minimum range of 70-100meters depending on datasource, my memory claims that was 75meters was the short impact range we had aiming data for.

81mm mortars has a minimum range of 83m per http://www.militaryfactory.com/smallarms/detail.asp?smallarms_id=146

If the mod author prefers 81s, it is 6490m, with a lethal burst radius of 8tiles and a casualty fragmentation radius of 16, with CEP of 0-7 tiles, and aiming error of 0-2tiles at medium range.  (All tile numbers based on 5m tiles.

dustysniper

QuoteCharge 4 is no issue.

81s can be 1manned, but not accurately.  How are you gonna hold the adjustable leg down and safely drop the round?  You can sandbag the fixed leg, but the adjustable leg needs your palm, using your foot would either jostle it or let it move too much for accuracy.

60s on charge 1, handheld, max out at either 400m or 500m IIRC, and fire a LOT faster than these.  It is still possible for a 1st year enlistee to get steel on steel hits on a tank hulk.  It was either a Patton or an Abrams before it became a target.  (I know, because I've both done it and seen it done by others.) 

60mm mortars have a minimum range of 70-100meters depending on datasource, my memory claims that was 75meters was the short impact range we had aiming data for.

81mm mortars has a minimum range of 83m per http://www.militaryfactory.com/smallarms/detail.asp?smallarms_id=146

If the mod author prefers 81s, it is 6490m, with a lethal burst radius of 8tiles and a casualty fragmentation radius of 16, with CEP of 0-7 tiles, and aiming error of 0-2tiles at medium range.  (All tile numbers based on 5m tiles.

I dont know where these facts are coming from. As long as rounds are prepped a gunner can 1 man a 81mm you dont need to hold the adjustable leg, the collar locking mechanism is designed to do that. And after the first round it should be seated into the ground with the base plate. I have one manned 81mm mortars before  hung rounds and gunned during a fire for effect and still had all rounds complete before first impact. Charge 1 60mm mortars hit out to 1340 m, charge zero is 722m, Im looking at a wiz wheel right now. 60s are incredibly accurate in handheld if you know how to probably use elevation bubble. Ive seen MGs hit at 1200m and hit targets at 1000m. 75m is the min range of a 60 for FPF. The max effective range of a 81 is 5608m and a FPF of 83m. You can hit closer than a FPF but your likely to hit friendlies and yourself. This is coming from the mortar bible fm 3-22.90. The issue from charge 4 comes from tube wear you fire to many to fast your tube well glow even under a sustain ROF.  Could you realistically be your own FDC prep rounds and gun I doubt it. It would take way to long.

NoImageAvailable

Quote from: NihilRex on April 10, 2015, 11:58:44 PM
81s can be 1manned, but not accurately.  How are you gonna hold the adjustable leg down and safely drop the round?  You can sandbag the fixed leg, but the adjustable leg needs your palm, using your foot would either jostle it or let it move too much for accuracy.

Keep in mind that Rimworld doesn't use modern day mortars but some kind of homemade that is a) a fixed installation and b) uses a control panel to operate. Most of the mechanical tasks are probably done by machines and the operator only has to load and adjust angle and elevation on the panel. When comparing to modern day equipment it would be closer to something like a 2S9 than a man-portable mortar. That said I'll probably go with a 81, simply because the lower range is more meaningful on Rimworld scale.
"The power of friendship destroyed the jellyfish."

LustrousWolf

How do I defend against mechs? they just shoot at my colonists and 1 by one they all die, and my turrets, none of them stand a chance against the mechs .-.

NoImageAvailable

Quote from: LustrousWolf on April 11, 2015, 06:05:52 AM
How do I defend against mechs? they just shoot at my colonists and 1 by one they all die, and my turrets, none of them stand a chance against the mechs .-.

Scythers should go down fairly quickly to some concentrated fire. Centipedes you get a lot of colonists to focus fire on one. Position your colonists so that they can always break line of sight by retreating behind a wall or other obstruction, do this whenever a Centipede takes aim at a colonist, then move them back out after the mech is done firing.
"The power of friendship destroyed the jellyfish."

LustrousWolf

Quote from: NoImageAvailable on April 11, 2015, 06:46:36 AM
Quote from: LustrousWolf on April 11, 2015, 06:05:52 AM
How do I defend against mechs? they just shoot at my colonists and 1 by one they all die, and my turrets, none of them stand a chance against the mechs .-.

Scythers should go down fairly quickly to some concentrated fire. Centipedes you get a lot of colonists to focus fire on one. Position your colonists so that they can always break line of sight by retreating behind a wall or other obstruction, do this whenever a Centipede takes aim at a colonist, then move them back out after the mech is done firing.

Yes that is how you can fight them, but with CR its really difficult to do that :/ Like you tell a colonist to move, 5 seconds later they try to move and get mowed down by the rain of bullets from the minigun :C so i was wondering if you had to fight them differently with CR installed?

NoImageAvailable

#143
Quote from: LustrousWolf on April 11, 2015, 07:03:45 AM
Quote from: NoImageAvailable on April 11, 2015, 06:46:36 AM
Quote from: LustrousWolf on April 11, 2015, 06:05:52 AM
How do I defend against mechs? they just shoot at my colonists and 1 by one they all die, and my turrets, none of them stand a chance against the mechs .-.

Scythers should go down fairly quickly to some concentrated fire. Centipedes you get a lot of colonists to focus fire on one. Position your colonists so that they can always break line of sight by retreating behind a wall or other obstruction, do this whenever a Centipede takes aim at a colonist, then move them back out after the mech is done firing.

Yes that is how you can fight them, but with CR its really difficult to do that :/ Like you tell a colonist to move, 5 seconds later they try to move and get mowed down by the rain of bullets from the minigun :C so i was wondering if you had to fight them differently with CR installed?

Some weapons like the survival rifle have high cooldown times (it takes time to operate a bolt, tear down an MG etc.) so they can't move immediately after firing. MGs also have a long burst duration. You need to be especially careful with these weapons when employing them against Centipedes/pirates with MGs.

If people have trouble with mechs I'm considering making an instructional video to showcase some tactics I personally use against various types of raids.
"The power of friendship destroyed the jellyfish."

bartekkru100


NihilRex

Quote from: NoImageAvailable on April 11, 2015, 01:47:35 AM
Quote from: NihilRex on April 10, 2015, 11:58:44 PM
81s can be 1manned, but not accurately.  How are you gonna hold the adjustable leg down and safely drop the round?  You can sandbag the fixed leg, but the adjustable leg needs your palm, using your foot would either jostle it or let it move too much for accuracy.

Keep in mind that Rimworld doesn't use modern day mortars but some kind of homemade that is a) a fixed installation and b) uses a control panel to operate. Most of the mechanical tasks are probably done by machines and the operator only has to load and adjust angle and elevation on the panel. When comparing to modern day equipment it would be closer to something like a 2S9 than a man-portable mortar. That said I'll probably go with a 81, simply because the lower range is more meaningful on Rimworld scale.

I was not the Artillery, so towed\self-propelled systems are beyond my experience.  I would think they were more accurate than the man-portable and man-transportable options though.

Also, Id never heard of the 2S9, so I looked it up.  Neat toy, someone should drop one off in my driveway for my birthday. An amphibious 120mm mortar in an articulated turret?  The local duck population will never recover.

NoImageAvailable

Speaking of fun, I finished my work on accuracy: forcedMissRadius was reduced from 20 to 17.4 and is now multiplied by 1 - targetDistance / maxRange. So it would be 0 tiles at range 500, 10.44 at 200 tiles and ca. 16 at the minimum range of 32 tiles (any closer and the mortar could score a direct hit on itself and that's just silly) corresponding to a CEP of 0-8 tiles from max to min range. For reference, fragments go for up to 16 tiles.

I have now started conducting some ingame tests to balance lethality :D



My metric is that balance is right when a shell hitting the center knocks out half the pawns in the blue circle (~7 tiles).

In unrelated news I also made some tweaks to the AI and if I'm right pirates should actually engage from much larger ranges now instead of wading through MG fire for no reason. I'm currently doing some playtesting to confirm this and as soon as that's done the new version will go up.
"The power of friendship destroyed the jellyfish."

NihilRex

Quote from: NoImageAvailable on April 11, 2015, 12:53:39 PM
Speaking of fun, I finished my work on accuracy: forcedMissRadius was reduced from 20 to 17.4 and is now multiplied by 1 - targetDistance / maxRange. So it would be 0 tiles at range 500, 10.44 at 200 tiles and ca. 16 at the minimum range of 32 tiles (any closer and the mortar could score a direct hit on itself and that's just silly) corresponding to a CEP of 0-8 tiles from max to min range. For reference, fragments go for up to 16 tiles.

I have now started conducting some ingame tests to balance lethality :D



My metric is that balance is right when a shell hitting the center knocks out half the pawns in the blue circle (~7 tiles).

In unrelated news I also made some tweaks to the AI and if I'm right pirates should actually engage from much larger ranges now instead of wading through MG fire for no reason. I'm currently doing some playtesting to confirm this and as soon as that's done the new version will go up.

From what little I can see, that looks awesome.

NoImageAvailable

Version 1.1.2 is up

Rebalanced mortars and tweaked raider AI, for more details see changelog.

PS To anyone having trouble with Centipedes, in my recent test game I finally got around to using an EMP mortar and it is extremely useful. One shell will stun mechanoids in a large area and for a long duration while colonists are free to take potshots at them, so if you're having trouble try one.
"The power of friendship destroyed the jellyfish."

NihilRex

Quote from: NoImageAvailable on April 12, 2015, 03:45:18 PM
Version 1.1.2 is up

Rebalanced mortars and tweaked raider AI, for more details see changelog.

PS To anyone having trouble with Centipedes, in my recent test game I finally got around to using an EMP mortar and it is extremely useful. One shell will stun mechanoids in a large area and for a long duration while colonists are free to take potshots at them, so if you're having trouble try one.

Im still testing, but so far, Im not seeing the new more accurate behavior.  Will continue to test.