Project Reality Forums

If this is the case then this needs to be fixed RIGHT AWAY. Not to put pressure on the DEVs but we cant have a rifle that 99%(gotta leave room for that x factor ) of the time shoots the the left of the intended target.

ironcomatose wrote:If this is the case then this needs to be fixed RIGHT AWAY. Not to put pressure on the DEVs but we cant have a rifle that 99%(gotta leave room for that x factor ) of the time shoots the the left of the intended target.

Now, now. Let us not jump to any conclusions about this stuff. There is still much more testing that needs to be done. And it needs to be repeatable, before it can be demonstrated to be an issue. Just because myself and VipersGhost witnessed this does NOT make it the Gospel truth. We could go back tonight and come up with a different result. More testing and cross checking is needed before anyone can say FIX or DON"T FIX anything.

As far as the DEVs goes, I don't think that ANY of them should be concerned about this or other issues right now. They have worked very, very hard to bring this latest installment of PR to us, and I'm sure they would like to have a short break and a chance to enjoy the fruits of there own labor. After all they like to play PR as well, so why not let them.

All work and no play makes jack a dull boy.

R

OK I ran the same test on the M16, G3, L85 and QBZ on a LOCAL SERVER. I used the Soflam to check the holes, I had to run up to 70m's and then take a screenshot real quick. Both times I was aiming with a 1-2 count at the center of the bigger red dot where the bars come together. Shots were from a prone, settled position. Everything is at 100 yrds. These guns seem to be fine...although the G3 you cannot aim as quicky I think and it is for sure less accurate. All 5.56 guns felt similar for the most part...the QBZ pattern here is nice but I think I had a couple of good slow shots. It does seem to help to wait 2 full seconds before firing for a very nice shot. Now the L85 works FINE on my local server???!??!? This is really weird. Rico and I BOTH tested it last night on the DVB server if I remember correctly and it was off for both of us MULTIPLE times. So we'll have to test in-game with the other guns and see if we can reproduce this on a live target maybe to make sure the decals aren't screwy.

M16-
[/IMG]


G3-
[/IMG]

QBZ
[/IMG]

L85
[/IMG]

I wonder if there is a built-in weapon sway, and depending on where you stop in the sway motion depends on where the shot grouping will occur.

For example, as you walk, does your weapon aim lightly to the left as your right foot moves forward, and slightly to the right as your left foot moves forward? Of course, the in-game animations will not match spot-on, but if such a mechanic does exist, this might explain why shot groupings are appearing to the left and right of the intended aim point seemingly at random. It might also explain why shot groupings are never above or below the aim point?

Just thinking out loud..

another thing I noticed (0.5 observation, but it may be still present) - sights (2D ones) may look more or less sharp on different video settings. It was textures on low that prevented me from accurate fireing from scoped QBZ-95.

Wolfe wrote:I wonder if there is a built-in weapon sway, and depending on where you stop in the sway motion depends on where the shot grouping will occur.

For example, as you walk, does your weapon aim lightly to the left as your right foot moves forward, and slightly to the right as your left foot moves forward? Of course, the in-game animations will not match spot-on, but if such a mechanic does exist, this might explain why shot groupings are appearing to the left and right of the intended aim point seemingly at random. It might also explain why shot groupings are never above or below the aim point?

Just thinking out loud..

Yeah but we tried all kinds of weird stuff with movements and everything was the same...now that I've tested on a local server I'm suspect there could be something else going on, like a server side code issue or maybe just the decals showing up improperly...need more testing. I do however like the shot groupings but I think they could be tightened up a bit...the old vets tell me you should be able to hit the center mass of a standing target at 300m's pretty easily if you have time to make a clean shot.

Mind you my shots aren't PERFECT every time and I don't have a competition target there either

going back to the topic - nedlands1's measures. I downloaded his spreadsheet and something just hit me: GL deviations. Lets take M16A2 Iron GL for example: MOA in stances:
Standing 7,92
Crouched 4,32
Prone 2,16
Isn't it harder to aim UGL in prone position when you pull the trigger with left hand finger and stabilizing weapon on left elwbow is less efective? Not to metion this that you aren't stabilizing weapon on elbow because of rising barrel must!

Shoudn't UGLs be much less accurate when lying on the ground? just a suggestion of mine.

the old vets tell me you should be able to hit the center mass of a standing target at 300m's pretty easily if you have time to make a clean shot.

I'm not a vet, but I am a soldier. Hitting human-sized target at 300m isn't SO easy... I hit my target with my AKMS with the first bullet, but most of my fiens didn't.*

EDIT:
* we were firing in prone supported position.

Not sure but I'm not sure those values are correct. The formula uses the "min" deviation modifier and my G3 pattern was quite smaller than the min...it should be quite bigger. Now things may not be to scale and other factors in BF2 nonsense but still I'm not sure. I know nedlands is the man at a lot of this stuff so who knows.

ViperGhost: when you say 100 yards did you move ~91.44m away from the target? If so the maximum spread should be in a circle with a diameter of ~0.48m (2 x 91.44 x tan 9/60) with the G3A3. Does the furthest round seem about 24 cm from the point of aim to you? I can't tell since I don't know how large the object is. Have you tried my test with two people on a door with the M16A4? Maybe you should.

EDIT: no offense taken.

Here is a little bit better reference image. This was from the exact spot I took the shots with the G3 and M16. Both guns fired a pretty decent pattern. Maybe when you are really steady and take your time...then the human deviation (not the moa) becomes null and isn't a factor. That would explain why the numbers on your chart seem higher than what I was seeing in-game. So I guess the question I have is what is the MOA value for the G3 and M16?


[/IMG]

Uncle Blues wrote:
I'm not a vet, but I am a soldier. Hitting human-sized target at 300m isn't SO easy... I hit my target with my AKMS with the first bullet, but most of my fiens didn't.*

EDIT:
* we were firing in prone supported position.

AKMS has a worse MOA than a standard M16, thats definitely a big factor when taking on 300m target. Eitherway I was just passing on what a vet was telling me. They said they went 38-40 consistently at 300yrds with standard issue M16 iirc.

Edit: He/She is a pretty good shot I recon though. When prone and on a rest it's not to hard to line up a steady shot...after that it's all the gun.

I'd say each small rectangular panel is about half a metre in width. Assuming that my spreadsheet values are correct and you took these shots in the right positions and at the right range (prone, stationary and at 91.44m) then these are the calculated maximum deviations from the aiming points for the different weapons:

G3A3
0.2394m from point of aim
(91.44 x tan 9/60)

M16A4
0.1596m from point of aim
(91.44 x tan 6/60)

Bear in mind these are the maximum values. To me, they look pretty right comparing the hits to the data.

nedlands1 wrote: The formula I have used is D=C*B*S*Z, where:

• D=Deviation in MOA (Diameter not radius of group)
• C=Constant
  =No. of MOA in a degree x 2(x 2 converts from radius to diameter)
  =60 x 2
  =120
• B=ObjectTemplate.deviation.minDev
  =Baseline deviation
• S=ObjectTemplate.deviation.devModStand, ObjectTemplate.deviation.devModCrouch or ObjectTemplate.deviation.devModLie
  =Stance deviation multiplier
• Z=ObjectTemplate.deviation.devModZoom
  =Zoom mode multiplier

I was looking at this Min property and thinking that the minimum deviation was being displayed for the entire result. This definitely explains a lot then. Yeah those values are fine then. Anyway could we see the minimum values? Normally the max is fine but with all of the extra factors it'd be nice to be able to see an average of what we should expect in-game.

VipersGhost wrote:I was looking at this Min property and thinking that the minimum deviation was being displayed for the entire result. This definitely explains a lot then. Yeah those values are fine then. Anyway could we see the minimum values? Normally the max is fine but with all of the extra factors it'd be nice to be able to see an average of what we should expect in-game.

Yeah the values are the minimum possible maxima . The minimum possible value for each weapon is zero. That "ObjectTemplate.deviation.minDev" isn't actually the minimum deviation. It's the starting point (or baseline as I've called it) for BF2's deviation calculations.

So to give you an example, the scoped G3A3, when standing and with the spreadsheet specified conditions, will fire rounds in 21.6 MOA circle. The furthest distance they will be from the mark is 10.8 MOA (ie the radius). As I said, that does not exclude rounds from landing closer to the place you aimed.

Well said and good job. No one can say it wasn't tested Those values are definitely right-on in my opinion. Good info to know for sure, glad I got my head around it...thanks a ton!