|
Post by timeitself on Jan 23, 2012 17:49:13 GMT -5
edit: updated spreadsheet:https://docs.google.com/spreadsheet/ccc?key=0ApQI-G2bQMZPdC1rcGxSM0NseWhLdFVzN3pRTDJ1RXc#gid=26 I wanted to get a better idea for time to kill with shotguns and akimbo machine pistols. Since these weapons spray everywhere the only reasonable way seems to be to compare probability of getting a kill at a given distance and time. This obviously means coming up with some approximation of the probability the bullet or pellet will hit the target given distance and spread. I took the quickest, and laziest approach I could come up with. I know it's not perfect, but neither is anyone's aim, as I wanted to see where it would take me. I assumed the rounds/pellets to be distributed independently and were shot at a rectangular target that was inside the spread. At least its a start. From there I used the rate of fire and pellets per shot info to come up with a quick chart. Here it is for 300 u (about 7.5m) I didn't want to go onto a comprehensive collection of ranges and proficiencies without getting some kind of feedback. First, the 1887 RoF at 240 looks to be off. Anyone have a better number? I used a very quick and dirty method to estimate probabilities, I'm guessing I'm going to need to go back and take some screen shots and known rages and count pixels. Even as close as 300u, still max damage for all but the AA12, it takes 3 shots with most shotguns to be fairly sure of a kill.
|
|
|
Post by RageHulkSmash on Jan 23, 2012 18:54:11 GMT -5
I'm confused on what the x and y-axis represent.
|
|
|
Post by llednik on Jan 23, 2012 19:47:04 GMT -5
Fairly certain that the model 1887 fire rate is exactly 60 rpm. Took exactly 30 frames to fire a second shot in a test that I did at 30 fps.
Question, I'm assuming you are not using steady aim and all of the shotguns are hip fired, is this correct? Also, You mentioned the model firing rate as being off but if I am reading this right Both the spas and the ksg are being shown as firing a second shot in less than half a second but they should both be closer to a full second for the second shot. One last thing, I'm also assuming that you were not using damage, range, or extended mags, is this correct?
Scythemage, I believe the x axis is total time(in milliseconds). As in From the time the very first bullet is fired till the kill is certain. The Y axis represent the probability of a kill at a given time. 1=a guaranteed kill and .5 is a 50/50 chance of a kill.
|
|
|
Post by brightknight141 on Jan 23, 2012 19:47:39 GMT -5
Can you link it to a larger picture? It's not clear-also label the X and Y Axises. (Axi, Axisis, Axises?)
|
|
|
Post by timeitself on Jan 23, 2012 20:24:56 GMT -5
larger image img141.imageshack.us/img141/8121/shotgunprobabilitiesv23.pngx axis is milliseconds (ms) after first shot is fired y axis is probability that fatal damage has been done I used spreads without steady aim from XboxAhoy, 5-8 degree cone and used the highest value for the machine pistols for standing and moving/firing. At this range my, again very crude, estimation of the probability was 0.25 to 0.5 for the range of 5-8 degrees. That I am sure will be important to refine at least to the point that aim becomes a question (especially for the machine pistols. I used the rate of fire that is on the widely circulated weapon stats chart. It has the ksg at 163 and spas at 127 rpm.
|
|
asasa
True Bro
fuck
Posts: 4,255
|
Post by asasa on Jan 23, 2012 20:46:16 GMT -5
Wow, and thats not even with akimbo
facepalm at how OP striker and mps are
|
|
|
Post by llednik on Jan 23, 2012 20:49:39 GMT -5
Ah that would explain it, the fire rates of the spas and ksg are much much lower than that. At 127 the spas would be firing faster than this and that clearly isn't the case. If I'm not mistaken 65 rpm is generally the accepted spas fire rate. At 163 the ksg would be faster than this which is also clearly incorrect. I believe the estimated rpm for the ksg is around 60 rpm but I plan on doing some more frame by frame tests just like with the model and I will get back to you with some precise fire rates.
|
|
|
Post by timeitself on Jan 23, 2012 21:57:00 GMT -5
KSG and SPAS rates of fire updated and corrections to my probability calculations. Been a while since I took that class I guess. Thanks for the info.
|
|
|
Post by randomguy987 on Jan 23, 2012 22:59:53 GMT -5
Nice graph. It "tells a story" very well.
If you have the time/inclination, it would be interesting to see the same comparison w/ Damage + Emags on the shotguns.
Or is Damage/Emags already included? (It kind of looks that way since the SPAS soundly beats the 1887 on that graph . . . unless the colors associated with those two shotguns got mixed-up in the key maybe?)
|
|
banana
True Banana
Zoro > Law
Posts: 1,577
|
Post by banana on Mar 28, 2012 9:55:08 GMT -5
does this include damage
|
|
|
Post by timeitself on Mar 28, 2012 12:24:06 GMT -5
this is pre-shotgun patch without steady aim, range or damage
|
|
banana
True Banana
Zoro > Law
Posts: 1,577
|
Post by banana on Mar 28, 2012 18:08:04 GMT -5
it'll be dope if u made one post patch with damage or is it too much work ?
|
|
|
Post by timeitself on Mar 28, 2012 18:46:45 GMT -5
This was a quick a dirty run through most of the necessary steps. It isn't terribly accurate, mostly just a fair comparison between the guns. Going back to do everything right is more time consuming and I don't have all the skills especially for plotting results that I want and so have to teach myself as I go. It's not that going back to this spreadsheet and tweaking the values is hard. It's that the results aren't to the standard most people would expect. About all I'm going to say is that the AA12 sucks. But, just to satisfy some curiosity here is post patch with steady aim at 300 u (about 7.5 m). Probabilities for individual hits ranged from 0.75 to 0.46 Once I can refine that I can report some more reliable numbers.
|
|
banana
True Banana
Zoro > Law
Posts: 1,577
|
Post by banana on Mar 29, 2012 21:25:57 GMT -5
can u make one for black ops ? ;p
|
|
banana
True Banana
Zoro > Law
Posts: 1,577
|
Post by banana on Apr 4, 2012 18:21:27 GMT -5
Now that i think about it 300u seems too close because most of the guns hit 85+% in 200 ms. try 400u and are the machine pistols akimbo'd if not then i think you should make them.. just some helpful criticism
|
|
|
Post by timeitself on Apr 21, 2012 23:21:08 GMT -5
I finally got around to doing a better job of this. I used a screen shot and took data points for angle and length around a center point. Then integrated using midpoint to get the area of the player inside of the spread. gets me to a curve fit for radius of the cone of spread on the target I'd wanted to do some nice graphs but it was taking too long so I went back to the simple ones like I'd already posted here. There are so many that I'm only going to post a few, otherwise refer to this video or the slideshow. Only includes base and steady aim for now. 5 m 7 m 5 m with steady aim - everything but the AA12 is essentially at 1 7m with steady aim 10m with steady aim
|
|
|
Post by timeitself on Apr 28, 2012 8:56:55 GMT -5
Maybe this way makes more sense. Think of it as a topo map for probabilities. Black = Default Green = Steady Aim Red = Damage + Steady Aim Blue = Range + Steady Aim (note that a lot of this overlaps with just steady aim) Solid ~ 0.97 Dash ~ 0.86 Dash-Dot ~ 0.63 ++++ ~ 0.5 Would probably help to put the range where the number of hits to kill changes up there too.
|
|
banana
True Banana
Zoro > Law
Posts: 1,577
|
Post by banana on Jun 17, 2012 19:10:28 GMT -5
Any news on your sa+damage and sa+range charts
|
|
|
Post by timeitself on Jun 30, 2012 14:58:07 GMT -5
After some in game testing my estimation of the size of the player hit box (head on) seems to have been fairly undersized. With some testing I found the area to be 1.23 m^2 and repeated my calculations. Video has a bit more detail but here are the graphs. for the MW3 SPAS-12
|
|
|
Post by timeitself on Jul 5, 2012 8:38:20 GMT -5
USAS added. Interesting result. The thing puts out so many pellets that I'm going to have to recommend range over damage. See for yourself.
|
|
banana
True Banana
Zoro > Law
Posts: 1,577
|
Post by banana on Jul 5, 2012 15:50:08 GMT -5
Very interesting stuff
|
|
|
Post by cashmoves on Jul 6, 2012 7:44:50 GMT -5
youre telling me the usas is better with range than damage? i just got it gold a couple days ago, and i think that damage really increases its cqc one shot kill ratio... and with shotties, the most important thing is knowing when you are in your wheelhouse range, you WILL WIN 100% of the time, or as close as possible. and, if im reading your chart correctly, your chart shows that damage excels over range within the shotty's divine range ~9 meters.
ill admit though, i leveled the gun up the ENTIRE way without using steady aim... so maybe ill have to give steady aim and range a go when i get home. this is very interesting though.
|
|
|
Post by timeitself on Jul 6, 2012 8:27:13 GMT -5
Without steady aim damage would be the better choice. Result is somewhere between the stock weapon and steady aim. The trouble with steady aim of course being that you have to be more accurate to make use of it.
If you look at the number of hits it takes to kill (chart is in the video somewhere) you'll see the distance where ranges becomes just as effective as damage (~10.4 m), and the distance where range is better than damage (>12.7m). With steady aim the USAS is putting out enough ammo to have a good chance of getting a kill at those ranges so range competes with damage even on the first shot. Damage certainly makes the weapon more consistent but range isn't without it's selling points. Especially with the less lethargic fire rate of the USAS.
|
|
|
Post by cashmoves on Jul 6, 2012 9:22:15 GMT -5
Without steady aim damage would be the better choice. Result is somewhere between the stock weapon and steady aim. The trouble with steady aim of course being that you have to be more accurate to make use of it. If you look at the number of hits it takes to kill (chart is in the video somewhere) you'll see the distance where ranges becomes just as effective as damage (~10.4 m), and the distance where range is better than damage (>12.7m). With steady aim the USAS is putting out enough ammo to have a good chance of getting a kill at those ranges so range competes with damage even on the first shot. Damage certainly makes the weapon more consistent but range isn't without it's selling points. Especially with the less lethargic fire rate of the USAS. this makes sense. considering i use the USAS most with overkill, i like it with damage and a wider spread. this make it more consistent in the CQC range as well as widening the 'kill zone' to potentially take out/damage multiple targets with one burst. every now and then ill get a double kill with one shot fired. as your sole primary, paired most likely with a launcher/mp9, i can see how range and steady aim will make it a more versatile weapon. still, i think you'll have to keep a cool head. those very close quarters (commando knife range) hitmarkers can be INFURIATING. damage in those instances insures more kills and a cooler head, at least for me. just so i totally understand you the dominance of all potential setups (excluding specialist bonus) is like this: 0-10m damage w/o SA 10-12m damage + SA 12M+ range + SA
|
|
|
Post by timeitself on Jul 6, 2012 10:10:00 GMT -5
You're getting into the area of player accuracy and playstyle. Think of this as a little more akin to time to kill numbers, except with the whole probability thing because of the shotgun spread. You can add rapid fire to an SMG and the TTK won't suffer at all. But we all know that it will be harder to use at longer ranges because of the increased recoil.
SA w/o damage looks better than damage w/o SA at all of the constant probability lines I have. Again, provided you can be on target. The plot for damage w/o SA is at 5:25 in the video. The whole fewer pellets to kill vs lower probability of getting a hit is interesting but in this case it looks like SA is the winner if accuracy is good.
for the USAS
0 - whatever range you're comfortable using SA at damage
your SA range -10.7m damage + SA
10.7m - 12.7m damage/range + SA tie
12.7m - ... whatever the max range on the USAS was.. ~20m range + SA
My personal preference would be to have SA for anything past 5m even if it gives me trouble with the close range swipe shots. The USAS is fairly forgiving but it looks like that 83% line is around 9m. If I got into a gunfight at 10m using damage w/o SA I'd be pulling the trigger twice without any questions.
|
|
|
Post by cashmoves on Jul 6, 2012 10:20:36 GMT -5
SA w/o damage looks better than damage w/o SA at all of the constant probability lines I have. Again, provided you can be on target. My personal preference would be to have SA for anything past 5m even if it gives me trouble with the close range swipe shots. The USAS is fairly forgiving but it looks like that 83% line is around 9m. If I got into a gunfight at 10m using damage w/o SA I'd be pulling the trigger twice without any questions. this actually helps me a lot. im going to change around my search classes to reflect this research. currently, my usas overkill class utilized damage WITHOUT steady aim and an assault streak. im going to change that to damage/range WITH steady aim and specialist so that i can have steady aim as well as sit rep (cause i whore that shit yo). thanks for the work, i finally got a chance to watch the videos (im at work) and they were very well done.
|
|
|
Post by cashmoves on Jul 6, 2012 12:51:18 GMT -5
i think i remember hearing that damage is not provided with the specialist bonus. is that true even if you have damage unlocked? starting the usas with range would have that minor knock against it as well. if you start with damage, at least you can get range with the specialist bonus...
|
|
cmck
True Bro
Hit him again!
Posts: 1,752
|
Post by cmck on Jul 6, 2012 13:36:38 GMT -5
Damage is better hands down for specialist if you're looking to get the most out of your shotgun because range is given, but damage is not. If you're not using specialist I found damage and SA to be very reliable in close. I didn't really like range too much for the bs stabs in the face that were happening to me after I shot them once. I think I'll try it again though.
|
|
|
Post by cashmoves on Jul 6, 2012 14:02:23 GMT -5
Damage is better hands down for specialist if you're looking to get the most out of your shotgun because range is given, but damage is not. If you're not using specialist I found damage and SA to be very reliable in close. I didn't really like range too much for the bs stabs in the face that were happening to me after I shot them once. I think I'll try it again though. i went home for lunch and played a couple rounds of FFA with the usas + range and SA. compared to how ive been playing (damage without SA) i felt like i got fewer one shot kills but more consistently had two shot kills... with the flinch this bad boy puts out, you have a pretty good percentage of killing even when it takes 2, as the chart shows above. on rare occasions ill have 3-4 shot kills, and in those instances, i probably shouldnt have engaged them with my shotty anyway (thats why theres the mp9 ;-) ) im gonna go back, play a couple more rounds with range + SA then give damage + SA a go and compare, at least on on a subjective level.
|
|
|
Post by cashmoves on Jul 8, 2012 12:17:28 GMT -5
i think you're right, the usas with steady aim and range is pretty dominant. honestly though, ive been messing around with silencer + range and steady aim on search and its fucking awesome haha. no one knows what the hell happened.
i run usas + range and silencer smaw (for long range shots, clear out buildings, or stack my streak with a uav or care package heli) throwing knife
extreme conditioning (which stacks hella long) or sleight of hand bc youve had to give up extended mags hardline steady aim
specialist: sit rep, dead silence, whatever
ive got oma bonus several times playing with this setup. its extremely fun and NO ONE uses a silenced USAS haha. its great fun.
|
|