Because of differences in opinion on bullet drop in this thread I decided to have a close look in controlled conditions to see if I could get a firmer idea on which rifle has more or less; along the way I found out some very important things I wasn't looking for.
I used the big three from each point and fired about 100 rounds in total. I'd appreciate it if at least one pair could get onto Teamspeak and repeat these tests. Also, someone with a better graphics card will have to test any longer ranges, I can't see player models beyond about 330m. I don't want to colour people's judgement ahead of time so I won't report all the results yet; I want to see if others get a replication of some of my results (especially the weird stuff).
Sharqi
Shooter standing on the top floor of the south building, in line with the flag (north-south line) aiming down the road due west.
Target standing just beside the telephone pole (right under the streetlight) on the north verge of the road; distance to Construction Site flag is 173m.
Wake
Shooter in a boat near the southern limb of the island; 327m from South Base flag.
Target standing by the north-west corner of the building to the north of the flag.
I'm lucky in that I can get access to two computers with BF2 installed in the same room, connected to the Internet through the same router, so I had the advantage of being able to listen to what the target heard directly for each shot, which lead to some of the very interesting results that I wasn't expecting.
The servers I found all reported pings of less than 60 and they were empty of course, so needless to say there wasn't any load to account for anomalies in hit detection; I kept a close watch on ping and it didn't appear to alter during the tests.
Okay to get this out of the way, at the ranges I tested there appears to be nothing between the big three: no certain differences in the amount of bullet drop. The reasons I'm not certain are because the changes in aiming point are very subtle, mere pixels, but also because of the unexpected findings (basically the amount of misses). I'm also not sure if we should be judging from the imaginary centre of the scope for the M95 as the aiming point, or use the underside of the horizontal bar of the reticle.
On Sharqi to ensure I was being as accurate as I could be with regard to vertical alignment I was aligning vertically off to one side, then tapping left to line up the crosshair where I wanted to aim. For consistency the target in each case was a sniper, scoped in, looking directly at the shooter.
M95
I started about three or four pixels (checking in Photoshop, three exactly) above the head, aligned for the shot. No headshot. Tried again. No headshot.
Lowered the aim point (two pixels). No headshot.
Lowered the aim point again until the bar was touching the head. No headshot.
One pixel above the head. No headshot. WTF? Surely if two pixels above isn't a headshot and touching the head isn't either then shouldn't one pixel be the sweet spot?
Touching the head. No headshot.
And so on. There's a lot more of these and not in order because it's hard to judge differences of a single pixel and I adjusted a little left and right to try to position for the location of the hit box in relation to the head model.
M24
Two pixels above the head. No headshot.
Touching. No headshot.
Touching, adjusted to the right. No headshot.
One pixel above. No headshot.
Two pixels above, exactly the same relative position left to right as previous shot. Headshot.
L96A1
Looked like the same aiming point as the last shot on the M24 but checking in Photoshop it's about one pixel to the right (the player model isn't exactly the same shape). No headshot.
One pixel above the top of the head. Headshot. Checking you can see it's one pixel to the left too... hang on a second though, aren't these two supposed to aim the same?!
Touching the head, same relative position left to right as above. Headshot.
One pixel lower, one pixel further to the right. Headshot.
...
I went back to the M24 and M95 after this and one successful aiming point with the M24 was with the bar touching the head but two pixels more to the left than the previous time.
I stopped taking screenshots after this but you get the basic picture.
Now if someone else could try some tests with the same positions. I'm especially interested if you get any of the same anomalies as I got. Bear in mind I have two computers in the same room I could hear what the target is supposed to hear (the whiz of the bullet closely followed by the crack of the shot) but that's not always the case... also please have a look at the distances from the centre aim the player names show up, see if I'm going crazy.
I used the big three from each point and fired about 100 rounds in total. I'd appreciate it if at least one pair could get onto Teamspeak and repeat these tests. Also, someone with a better graphics card will have to test any longer ranges, I can't see player models beyond about 330m. I don't want to colour people's judgement ahead of time so I won't report all the results yet; I want to see if others get a replication of some of my results (especially the weird stuff).
Sharqi
Shooter standing on the top floor of the south building, in line with the flag (north-south line) aiming down the road due west.
Target standing just beside the telephone pole (right under the streetlight) on the north verge of the road; distance to Construction Site flag is 173m.
Wake
Shooter in a boat near the southern limb of the island; 327m from South Base flag.
Target standing by the north-west corner of the building to the north of the flag.
I'm lucky in that I can get access to two computers with BF2 installed in the same room, connected to the Internet through the same router, so I had the advantage of being able to listen to what the target heard directly for each shot, which lead to some of the very interesting results that I wasn't expecting.
The servers I found all reported pings of less than 60 and they were empty of course, so needless to say there wasn't any load to account for anomalies in hit detection; I kept a close watch on ping and it didn't appear to alter during the tests.
Okay to get this out of the way, at the ranges I tested there appears to be nothing between the big three: no certain differences in the amount of bullet drop. The reasons I'm not certain are because the changes in aiming point are very subtle, mere pixels, but also because of the unexpected findings (basically the amount of misses). I'm also not sure if we should be judging from the imaginary centre of the scope for the M95 as the aiming point, or use the underside of the horizontal bar of the reticle.
On Sharqi to ensure I was being as accurate as I could be with regard to vertical alignment I was aligning vertically off to one side, then tapping left to line up the crosshair where I wanted to aim. For consistency the target in each case was a sniper, scoped in, looking directly at the shooter.
M95
I started about three or four pixels (checking in Photoshop, three exactly) above the head, aligned for the shot. No headshot. Tried again. No headshot.
Lowered the aim point (two pixels). No headshot.
Lowered the aim point again until the bar was touching the head. No headshot.
One pixel above the head. No headshot. WTF? Surely if two pixels above isn't a headshot and touching the head isn't either then shouldn't one pixel be the sweet spot?
Touching the head. No headshot.
And so on. There's a lot more of these and not in order because it's hard to judge differences of a single pixel and I adjusted a little left and right to try to position for the location of the hit box in relation to the head model.
M24
Two pixels above the head. No headshot.
Touching. No headshot.
Touching, adjusted to the right. No headshot.
One pixel above. No headshot.
Two pixels above, exactly the same relative position left to right as previous shot. Headshot.
L96A1
Looked like the same aiming point as the last shot on the M24 but checking in Photoshop it's about one pixel to the right (the player model isn't exactly the same shape). No headshot.
One pixel above the top of the head. Headshot. Checking you can see it's one pixel to the left too... hang on a second though, aren't these two supposed to aim the same?!
Touching the head, same relative position left to right as above. Headshot.
One pixel lower, one pixel further to the right. Headshot.
...
I went back to the M24 and M95 after this and one successful aiming point with the M24 was with the bar touching the head but two pixels more to the left than the previous time.
I stopped taking screenshots after this but you get the basic picture.
Now if someone else could try some tests with the same positions. I'm especially interested if you get any of the same anomalies as I got. Bear in mind I have two computers in the same room I could hear what the target is supposed to hear (the whiz of the bullet closely followed by the crack of the shot) but that's not always the case... also please have a look at the distances from the centre aim the player names show up, see if I'm going crazy.
Last edited by Todd_Angelo (2006-06-28 05:39:15)
