Jump to content

Pilot skill or probabilistic hit locations?


244 replies to this topic

Poll: Pilot skill or equipment? (357 member(s) have cast votes)

How should hit locations be determined?

  1. Pilot skill: To the steadiest hand go the spoils. (185 votes [51.82%] - View)

    Percentage of vote: 51.82%

  2. Probabilistic: Those gyro stabilizers aren't perfect you know. (160 votes [44.82%] - View)

    Percentage of vote: 44.82%

  3. Target Designation Only: Declare targets like in TT game, let the firing computer do the rest. (12 votes [3.36%] - View)

    Percentage of vote: 3.36%

Vote Guests cannot vote

#141 Riptor

    Member

  • PipPipPipPipPipPipPipPip
  • Veteran Founder
  • Veteran Founder
  • 1,043 posts

Posted 11 November 2011 - 11:08 AM

Heres an idea... each arm and each torso section has its own small cone of fire, but tall weapons installed in one location (ig. you shoot your right arm weapons) will hit the same location on the enemy?

This would make putting all your weapons in one location so they all hit the same spot when fired very accurate but also very risky since someone could just shoot your arm off or demolish your right torso and take all your weapons off by chance.

On the other hand you would cause massive damage if you hit your target since all the weapons you installed in that one location would hit the same section on the enemy.

The cones would be kept small so that you dont necesarly miss your target but big enough to keep you from headshooting everything that comes along with one salvo and end a match in 30 seconds.

#142 Creel

    Member

  • PipPipPipPipPip
  • 189 posts
  • Google+: Link
  • LocationFort Worth, TX

Posted 11 November 2011 - 11:09 AM

View PostCavadus, on 11 November 2011 - 10:50 AM, said:


Just give each section inside of a "meta-section" the same armor values. I mean, why divide it?

So if you have an Atlas that would have 72 points of armor on the CT and then break up the CT into six sections it would be silly to divide that 72 armor between them for a total of 12 armor per "CT-section".

Give all sections inside of the CT 72 armor.


I can see that working, but it would still require a large amount of testing for balance and perhaps adjustments of weapon values. Overall, I like it and didn't intend to suggest otherwise.

#143 TheRulesLawyer

    Member

  • PipPipPipPipPipPipPipPip
  • 1,415 posts
  • LocationChicagoland

Posted 11 November 2011 - 11:11 AM

View PostCreel, on 11 November 2011 - 07:47 AM, said:


As a pistols instructor, I'll give that a qualified yes.

I'm saying that, depending on the quality of the control system (human or mechanical), and the stability of the firing platform (offhand, or with a rest, or hardmounted), the projectile can be reasonably expected to contact the target within a range anywhere from 1" to 1 meter of where the shooter thinks they're aiming, at a distance of 25 feet. Granted, the 1m only applies to people with horrific technique, however it's not at all uncommon for beginners. That extreme will, of course, not be the case with computer controlled mechanical systems.

One thing to think about is Minute of Arc. (http://en.wikipedia....i/Minute_of_Arc) This is a mathematical term which measures angular difference, and is used to calculate course and trajectory.

a difference of 1/60 of 1 degree translates into shot deviation of 1" at ~30 meters. This means that if a weapon is misaligned by as little as 1 MOA. at 100m youre shot has deviated 3", and at a thousand the projectile hits 3 feet from your designated target spot. Even the best high powered rifles only guarantee accuracy up to about one MOA, which means that you can reasonable expect 6" groups at 100 yards, or 5ft groups at 1000 yards.

Miniscule differences affect trajectory in a massive way, and when you're dealing with weapon mounts that weigh tons which are bouncing up and down at 50kph. calibration of one sixtieth of one degree seems pretty unattainable. With only 1 degree of misalignment you are going to be hitting anywhere in a 60,000ft circle


You're a pistol instructor and you think 1moa= 6" at 100yrds? Yikes.

#144 Creel

    Member

  • PipPipPipPipPip
  • 189 posts
  • Google+: Link
  • LocationFort Worth, TX

Posted 11 November 2011 - 11:17 AM

View PostTheRulesLawyer, on 11 November 2011 - 11:11 AM, said:


You're a pistol instructor and you think 1moa= 6" at 100yrds? Yikes.


I had a bad math moment. I converted from metric very badly. for some reason I converted 30m to 100yds in my head, rather than to 100 feet.

#145 Creel

    Member

  • PipPipPipPipPip
  • 189 posts
  • Google+: Link
  • LocationFort Worth, TX

Posted 11 November 2011 - 11:18 AM

so 20,000ft for a 1 degree misalignment...I think the point still stands.

#146 TheRulesLawyer

    Member

  • PipPipPipPipPipPipPipPip
  • 1,415 posts
  • LocationChicagoland

Posted 11 November 2011 - 11:29 AM

View PostCreel, on 11 November 2011 - 11:09 AM, said:


I can see that working, but it would still require a large amount of testing for balance and perhaps adjustments of weapon values. Overall, I like it and didn't intend to suggest otherwise.


What might be interesting to do with subsection armor is to give weapons an area of effect. That is weapons could then damage multiple location simultaneously. Lasers inherrant accuracy could be balance by a very small area. Pretty much always only hitting one sub-section. An ac20 OTOH might damage 8-9 subsections (depending on how many there are of course)

#147 Creel

    Member

  • PipPipPipPipPip
  • 189 posts
  • Google+: Link
  • LocationFort Worth, TX

Posted 11 November 2011 - 11:31 AM

View PostTheRulesLawyer, on 11 November 2011 - 11:29 AM, said:


What might be interesting to do with subsection armor is to give weapons an area of effect. That is weapons could then damage multiple location simultaneously. Lasers inherrant accuracy could be balance by a very small area. Pretty much always only hitting one sub-section. An ac20 OTOH might damage 8-9 subsections (depending on how many there are of course)


careful, there...you're getting awfully close to the autocannon caliber debate...I think we can all agree to stay out of that snake pit.

#148 TheRulesLawyer

    Member

  • PipPipPipPipPipPipPipPip
  • 1,415 posts
  • LocationChicagoland

Posted 11 November 2011 - 11:52 AM

View PostCreel, on 11 November 2011 - 11:17 AM, said:


I had a bad math moment. I converted from metric very badly. for some reason I converted 30m to 100yds in my head, rather than to 100 feet.


Fair enough.

Just to go on the with the accuracy issue. A really good rifle with good ammo fired from a fixed stable position can expect about 1moa. A rack grade military issue rifle in good condition with military ammo typically is closer to 2moa.

A hex in battletech is 30m That's pretty dang close to 100ft.

So average military rifle is 2" at 1 hex. 4" at 2 hex, 6" at 3 hex.
So 3/6/9 is 6"/12"/18"

Doesn't sound like a ton a spread, right? Well, those are perfect conditions. Known distance so calibrated drop, no wind, no movement, no error. Figure thats a 0/0 pilot standing still firing at another mech standing still over clear terrain in battletech terms.

Its worthwhile to note that these are small group numbers. There are still outliers as well. The more shots fired, the more that MOA number increases. You're really talking about something like 1 standard deviation for most MOA numbers. (approximate)

Add in all those environmental factors, and firing what amounts to snap shots you you can easily see numbers that are an order of magnitude higher than best case numbers before even getting into skill. A meter or two spread at 270m is hardly unreasonable.

The other thing to remember is that you shouldn't get a completely random distribution about an aim circle. It'll be normal- that is center weighted.

Edited by TheRulesLawyer, 11 November 2011 - 12:08 PM.


#149 Creel

    Member

  • PipPipPipPipPip
  • 189 posts
  • Google+: Link
  • LocationFort Worth, TX

Posted 11 November 2011 - 12:00 PM

View PostTheRulesLawyer, on 11 November 2011 - 11:52 AM, said:


Fair enough.

Just to go on the with the accuracy issue. A really good rifle with good ammo fired from a fixed stable position can expect about 1moa. A rack grade military issue rifle in good condition with military ammo typically is closer to 2moa.

A hex in battletech is 30mm That's pretty dang close to 100ft.

So average military rifle is 2" at 1 hex. 4" at 2 hex, 6" at 3 hex.
So 3/6/9 is 6"/12"/18"

Doesn't sound like a ton a spread, right? Well, those are perfect conditions. Known distance so calibrated drop, no wind, no movement, no error. Figure thats a 0/0 pilot standing still firing at another mech standing still over clear terrain in battletech terms.

Its worthwhile to note that these are small group numbers. There are still outliers as well. The more shots fired, the more that MOA number increases. You're really talking about something like 1 standard deviation for most MOA numbers. (approximate)

Add in all those environmental factors, and firing what amounts to snap shots you you can easily see numbers that are an order of magnitude higher than best case numbers before even getting into skill. A meter or two spread at 270m is hardly unreasonable.

The other thing to remember is that you shouldn't get a completely random distribution about an aim circle. It'll be normal- that is center weighted.


and to further emphasize, 1 MOA is 1/60 of 1 degree. If the weapon is misaligned by a tenth of a degree, which would actually be some fairly impressive precision considering all of the factors involved, your 18" spread at 6 hexes becomes the diameter of an entire hex.

edit: this is misleading. miscalibration would not change the diameter of the accuracy "cone", but shift it. so rather than being the diameter of a hex, it would be the same diameter, but aiming one hex off.

Range of motion varying up to 1/10 of a degree is what would increase the diameter.

Edited by Creel, 11 November 2011 - 12:03 PM.


#150 UncleKulikov

    Member

  • PipPipPipPipPipPipPip
  • Bridesmaid
  • 752 posts

Posted 11 November 2011 - 12:20 PM

I think it's important to declare wants and intents that these mechanics choices are based on.

I want it to be hard to hit things.
I want it to be harder to hit things when doing advanced maneuvers, like jumping or running or firing full automatic.
I don't want coring to be a standard practice, but still achievable if you take the time and are skilled enough to land all your shots on the center torso.
I don't want a significant advantage given to macros.

Whatever mechanics achieve those goals, I support.

#151 Creel

    Member

  • PipPipPipPipPip
  • 189 posts
  • Google+: Link
  • LocationFort Worth, TX

Posted 11 November 2011 - 12:23 PM

View PostUncleKulikov, on 11 November 2011 - 12:20 PM, said:

I think it's important to declare wants and intents that these mechanics choices are based on.

I want it to be hard to hit things.
I want it to be harder to hit things when doing advanced maneuvers, like jumping or running or firing full automatic.
I don't want coring to be a standard practice, but still achievable if you take the time and are skilled enough to land all your shots on the center torso.
I don't want a significant advantage given to macros.

Whatever mechanics achieve those goals, I support.


on the subject of wants.

http://mwomercs.com/...__fromsearch__1

#152 Tierloc

    Member

  • PipPipPipPipPipPip
  • 231 posts
  • LocationWAR_Homeworld

Posted 11 November 2011 - 12:24 PM

Depending on the size of the subsections, and the angle, I'm not sure why you think an AC weapon type would spread. It's the difference between buckshot and slugs.

It seems to me you guys are bending several rules just to make up for point and shoot on my joystick. Specifically with a laser.

I didn't invent the laser, or "direct fire". I've read some neat stuff about the length of time required to do X damage instead of instant and I checked today that the pulse laser tracks the same target instead of strafing (something it should do imo) but without the manual element of aiming the weapon being present, IE it is computer guided by my direction (2x IE my joystick), having a magical floating recticle or circle of confluessence or focusable relay targeting delay turret amounts to nothing more than covering up for some pilots admitting they just plain suck at targeting and shooting things and need other people to be forced to have a random chance at not only weapon issues (overheat, misfire, jam, destroyed) but also correctly targeting an object and yet still missing.

This already happens. This is called "lag hax ***" in the current game, btw.


You can talk about flight patterns, but let's talk about scale.

Have you seen this image from another thread?

Posted Image

Left to Right (Scale 1:1)

5.7mm x 28 P90 (Five-seveN, P90) Ball SS190
5.56mm x 45 NATO (FN Minimi, Galil, L85A1, L86A1, M16A2) Ball M855
7.62mm x 51 NATO (FN FAL, FN MAG, G3, HK21, HK23, L1A1 (SLR), L7A1 (GPMG), M14A1, M60, MG3) Ball L2A2
.50" Browning (12.7mm x 99) (.50" M2, M2HB, Barrett) Armor-Piercing-Incendiary-Tracer M20
20mm x 102 Vulcan (M39, M50, M61A, M61A1, GAU-4, Mk. 22 Mod. 2) Target Practice M55A2
20mm x 139 HS 820 (HS820, KAD, Rh202, M139) Armor-Piercing-Incendiary
25mm x 137mm (KBA, GAU-12/U, M242, Rh205, Aden 25) Target Practice


Third one from the right is a 20mm vulcan cannon round. Reportedly the BT size of the machine gun rounds, arguably one of the smallest projectiles of the entire ballistic arsenal of the video game. Another user also posted some of the other sizes of weapons:

Quote

From books and reads,it gave weapons sizes. AC-2 was 25 to 30mm chain guns . AC-5 were like 40 to 75mm. but the Maruader AC-5 was a 120mm due to its low rate of fire . The Atlas AC-20 is a 8in gun that fires a 4 round burst, while the Victor AC-20 was a 100mm gun with a higher rate.


For the vulcan, at 1030 m/s, the standard payload has 6.3 mm RHA penetration at 0 degree impact angle and 1000 m range. The max range for these weapons in the game is 400 m. I would argue that based on bullet size alone, if you take the "max range" of the weapons as they are restricted, that most of the limitations in the game would be well below standards.

Actually you can look up some of the 75, 100, 120 etcs. It would be like shooting your 9mm pistol < 10 feet, in relation to the maximum distance. I do not think you can argue having a large spread at 10 feet.

#153 Creel

    Member

  • PipPipPipPipPip
  • 189 posts
  • Google+: Link
  • LocationFort Worth, TX

Posted 11 November 2011 - 12:34 PM

But it isn't 10 ft. It's 400 meters, and that gun's spread at full range is HUGE. Weapon accuracy is not dependent upon maximum range, but on actual range. the 9mm will still be accurate only within the cone defined by it's inherent accuracy whether at 10ft or 1000.

#154 diana

    Member

  • PipPip
  • 41 posts

Posted 11 November 2011 - 12:44 PM

AC shells might as well be guided to some degree with deployable fins, like some modern tanks, so they can be nearly pinpoint if devs will it.

#155 Tierloc

    Member

  • PipPipPipPipPipPip
  • 231 posts
  • LocationWAR_Homeworld

Posted 11 November 2011 - 12:45 PM

View PostHalfinax, on 11 November 2011 - 10:27 AM, said:


The inherent issue with this is that with the examples you provided the weapons in question are set to a static convergence. In previous MW games your weapons effectively shot out of the reticule there was no compensation for how near or far they were to you, or what the convergence range was set for. If the target was 10 meters away the weapons were just as precisely clustered on the reticule as when the target was 260 meters away.


I can agree with the static convergence. I guess instead of hard mounts their individual turrets in static locations with varying degrees of rotation. I will say though that your shots can be blocked by map objects or other mechs (in paricular teamates) if you are zoomed in or even shooting around a building, as far as MW4 is concerned. The origin of the weapon fire does not come from the cockpit, in other words.

Hitting the same target at 10m and 500m.. again with actual turrets and mechanical arms.. it would be quite feasible.

#156 Tierloc

    Member

  • PipPipPipPipPipPip
  • 231 posts
  • LocationWAR_Homeworld

Posted 11 November 2011 - 12:48 PM

View PostCreel, on 11 November 2011 - 12:34 PM, said:

But it isn't 10 ft. It's 400 meters, and that gun's spread at full range is HUGE. Weapon accuracy is not dependent upon maximum range, but on actual range. the 9mm will still be accurate only within the cone defined by it's inherent accuracy whether at 10ft or 1000.


Yeah the machine guns in the game at 400m spread pretty good too.

If you look at some of the RL ranges of the 100mm and 120mm cannons, they're like 8000m. The howitzers go like 75000. All I'm suggesting, is without knowing the payload to launch the projectile, if you got into one of the tanks with a 8000m effective range, and could only fire it MAX 350m, I'm pretty sure at the m/s you'd be pretty **** accurate.

#157 Creel

    Member

  • PipPipPipPipPip
  • 189 posts
  • Google+: Link
  • LocationFort Worth, TX

Posted 11 November 2011 - 12:54 PM

View PostTierloc, on 11 November 2011 - 12:48 PM, said:


Yeah the machine guns in the game at 400m spread pretty good too.

If you look at some of the RL ranges of the 100mm and 120mm cannons, they're like 8000m. The howitzers go like 75000. All I'm suggesting, is without knowing the payload to launch the projectile, if you got into one of the tanks with a 8000m effective range, and could only fire it MAX 350m, I'm pretty sure at the m/s you'd be pretty **** accurate.


Yes, those guns are very accurate, they are calibrated perfectly, they are immensely stable platforms.

Mechs aren't any of these things.

#158 Tierloc

    Member

  • PipPipPipPipPipPip
  • 231 posts
  • LocationWAR_Homeworld

Posted 11 November 2011 - 12:57 PM

View PostCreel, on 11 November 2011 - 12:54 PM, said:


Mechs aren't any of these things.


That's what computer guidance is for.

#159 Creel

    Member

  • PipPipPipPipPip
  • 189 posts
  • Google+: Link
  • LocationFort Worth, TX

Posted 11 November 2011 - 01:04 PM

View PostTierloc, on 11 November 2011 - 12:57 PM, said:

That's what computer guidance is for.


Computers in the btech universe are **** by our standards. That's an underlying assumption of the entire system.

#160 Tierloc

    Member

  • PipPipPipPipPipPip
  • 231 posts
  • LocationWAR_Homeworld

Posted 11 November 2011 - 01:17 PM

Oh I don't know.. if we had to go back to cobol so we could pilot 80 tons of *** kickery, would you not consider it? Could it do it? I think so. I didn't write a short story that turned into a franchise but I'm pretty sure it would be possible.

Edited by Tierloc, 11 November 2011 - 01:38 PM.






8 user(s) are reading this topic

0 members, 8 guests, 0 anonymous users