neverlast74
True Bro
Clan:DarkBadEvil Fightaz of DeathCommando - Squad: Elite Ninjas
Posts: 236
|
Post by neverlast74 on Dec 1, 2011 15:05:46 GMT -5
|
|
|
Post by jackson on Dec 1, 2011 15:54:39 GMT -5
Thank you!!! Perfect. Definitely adds some credibility to my sim. Let's see if Imgur is a little easier to view. Some improvements on presentation made too. imgur.com/a/4GyQGAlso fixed the right/left recoil directions. Symthic had them reversed before.
|
|
|
Post by rudybojangles on Dec 1, 2011 16:07:46 GMT -5
Thank you!!! Perfect. Definitely adds some credibility to my sim. Let's see if Imgur is a little easier to view. Some improvements on presentation made too. imgur.com/a/4GyQGAlso fixed the right/left recoil directions. Symthic had them reversed before. Looking HAWT. *see what I did there?* The RPK is a motherfudging monster. Straight vertical recoil, little spread, just hot lead-based death. Me gusta. AS VAL and PDW are also nice looking. Any other bro conclusions?
|
|
Lexapro
True Bro
PSN: Lexa_pro
Posts: 1,066
|
Post by Lexapro on Dec 1, 2011 16:27:49 GMT -5
Oh that's coool
|
|
neverlast74
True Bro
Clan:DarkBadEvil Fightaz of DeathCommando - Squad: Elite Ninjas
Posts: 236
|
Post by neverlast74 on Dec 1, 2011 16:32:38 GMT -5
Any other bro conclusions? Agree RPK is still very strong in the support area. BTW: The last version of the symthic grid shows that PKP now is the only 34 dmg gun - I only recall reading the nerf the scar and g3a3 but not 249,240b, and m60.... (although the effect is only important on the first 10 meters - which is not really a criteria per se. how ever even more a reason for PKP and against the 249,240b,and m60 dmg statswise. not to forget the low PKP 1.0 max recoil in all positions ...)
|
|
|
Post by jackson on Dec 1, 2011 16:44:19 GMT -5
I'm already doubting these plots a little. I think my method of randomly picking where the shot goes in the cone of spread is fundamentally flawed. Updating my code now.
|
|
|
Post by kirbyderby on Dec 1, 2011 17:11:54 GMT -5
The PKP is pretty much a mid-range, mobile, area denial spray beast. I got 8 combat efficiency ribbons in a round just pa-trolling the attacker spawns on Damavand Peak Rush.
|
|
|
Post by cthuzael on Dec 1, 2011 17:33:34 GMT -5
jackson, if you need further sceenshots to keep up your good work, simply state your wish which you need.
|
|
sleep
True Bro
Posts: 10,189
|
Post by sleep on Dec 1, 2011 19:24:02 GMT -5
not to forget the low PKP 1.0 max recoil in all positions ...) just a note, 1.0 max spread (when ADS) in all positions. the recoil is still pretty wild as you would expect from an LMG if you're not using a bipod.
|
|
|
Post by yrrnn on Dec 1, 2011 22:44:36 GMT -5
Hmm, based on this and the other stats that have been uncovered, with a foregrip, the KH2002 is looking like it might be the best choice for Assault...
|
|
sleep
True Bro
Posts: 10,189
|
Post by sleep on Dec 1, 2011 23:11:10 GMT -5
Hmm, based on this and the other stats that have been uncovered, with a foregrip, the KH2002 is looking like it might be the best choice for Assault... yeah, my bro conclusion from these charts was that the KH2002 is underrated at the very least. it does have pretty monstrous reload times, but then so do the AEK-971 and F2000.
|
|
neverlast74
True Bro
Clan:DarkBadEvil Fightaz of DeathCommando - Squad: Elite Ninjas
Posts: 236
|
Post by neverlast74 on Dec 2, 2011 3:27:22 GMT -5
The PKP is pretty much a mid-range, mobile, area denial spray beast. I got 8 combat efficiency ribbons in a round just pa-trolling the attacker spawns on Damavand Peak Rush. I agree and that makes perfect sense balance wise. I also conclude from this that support has the supression/spaying/ammo role but not so much "the fast kills 1-1 lonewolf role" as they lack accuracy (compared to assault weapons) So at least 2 assault guys in every squad becomes a must given the weapon accuracy and the alternatives. 1 support for ammo and suppression and the 4th place is depending on the map - engineer/recon. would be a setup that might work ok on some maps.
|
|
neverlast74
True Bro
Clan:DarkBadEvil Fightaz of DeathCommando - Squad: Elite Ninjas
Posts: 236
|
Post by neverlast74 on Dec 2, 2011 3:59:36 GMT -5
I'm already doubting these plots a little. I think my method of randomly picking where the shot goes in the cone of spread is fundamentally flawed. Updating my code now. besides I do not like the idea that they are flawed -- what makes you believe that they are flawed? My limited understanding is: the stats give you the place - and the tool renders the area where randomized shots are being shown. I think you model mimicries very well the game code.
|
|
|
Post by kirbyderby on Dec 2, 2011 4:04:25 GMT -5
The PKP is pretty much a mid-range, mobile, area denial spray beast. I got 8 combat efficiency ribbons in a round just pa-trolling the attacker spawns on Damavand Peak Rush. I agree and that makes perfect sense balance wise. I also conclude from this that support has the supression/spaying/ammo role but not so much "the fast kills 1-1 lonewolf role" as they lack accuracy (compared to assault weapons) So at least 2 assault guys in every squad becomes a must given the weapon accuracy and the alternatives. 1 support for ammo and suppression and the 4th place is depending on the map - engineer/recon. would be a setup that might work ok on some maps. I would argue that, under the right conditions and with a certain playstyle, the Support would actually be an excellent lone wolf setup. One support guy using a bipod can hold a chokepoint from pretty much any range. The bipod also makes the LMGs pretty deadly weapons at long range, provided you have a well-covered position so you have time to deploy your bipod and find the enemies without getting sniped. The mortar rounds out the kit, giving it a decent chance against emplaced ground vehicles. This kind of setup would have good synergy with a MAV for easier mortaring/flank protection. Alternatively, the PKP makes for a great aggressive Support lone-wolf, allowing him to effectively hose any targets at up to medium distance with the low max spread. C4 is great for making more open/direct routes by blowing up walls, in addition to making traps on enemy vehicles if the opportunity presents itself. The main problem with this as a lone wolf class is the lack of a medkit, but there should be plenty of dead Assault guys to get them from.
|
|
|
Post by bel on Dec 2, 2011 4:04:59 GMT -5
jackson, if you need further sceenshots to keep up your good work, simply state your wish which you need. I'm not jackson, but I have a request. I would really like to see screenshots of a soldier standing front-on and side-on (and crouch/prone if you're feeling awesome), at explicit ranges (say 10m, 30m and 50m), at the highest possible resolution, viewed by a player standing unzoomed with a PP2000 (or prone with an unzoomed M4 if you don't have the PP2000 unlocked). That would be really helpful for gauging exactly how much surface area enemies take up at different ranges. Easiest way to get the range is to have a buddy stand on a cap point. I'd do it myself, but I'm pretty light on for buddies on servers that won't kick me for high ping.
|
|
neverlast74
True Bro
Clan:DarkBadEvil Fightaz of DeathCommando - Squad: Elite Ninjas
Posts: 236
|
Post by neverlast74 on Dec 2, 2011 8:11:07 GMT -5
I would argue that, under the right conditions and with a certain playstyle, the Support would actually be an excellent lone wolf setup. One support guy using a bipod can hold a chokepoint from pretty much any range. Alternatively, the PKP makes for a great aggressive Support lone-wolf, allowing him to effectively hose any targets at up to medium distance with the low max spread. hmm interesting - I would like to know what you mean saying - "under the right conditions and with a certain playstyle"? Because I have a hard time running alone on large maps (snipers will tag you and from that moment your fodder - on small maps there are less opportunities to position so once tagged the M320 will hit you. Picking running enemies from a 10-30 meter distance yes that works fine - a "tactical loitering" surprise attack. (will also work with an A M416 that also come with a bipod). But only once. Even a 2 man squad (with one assault) is a problem for me being alone with a bipod. Getting flanked with a bipod means I am dead. The horizontal bipod angle is maybe 70° So once they know that you are there on big maps they will come from the side. In 1-1 situations assault rifles once you are identified on the map are more accurate and as long as I am suppressing - the guy throws a grenade at me and I have to move - and having no grip on my PKP means I need a lot of bullets to hit the enemy ... so I do see a lot of weaknesses without even any special tactics from my enemy once I am tagged. pls. enlighten me the only thing is that I see might work - position yourself kill "say" 2 guys - and re - position wait until the tag wears off (does it??) and then position somewhere else ... that might work - but in conquest/rush maps I do not actively work on the prime goal as I concentrate on runners that are in areas between spawns bases and targets.
|
|
|
Post by jackson on Dec 2, 2011 8:47:00 GMT -5
I'm already doubting these plots a little. I think my method of randomly picking where the shot goes in the cone of spread is fundamentally flawed. Updating my code now. besides I do not like the idea that they are flawed -- what makes you believe that they are flawed? My limited understanding is: the stats give you the place - and the tool renders the area where randomized shots are being shown. I think you model mimicries very well the game code. My method of randomly generating shots within the cone of spread puts too many shots closer to the center. It is randomly distributed along the radial direction but because there is less area the closer you get to center, the shot density is higher in the center. It's an easy fix and I'm doing it now actually.
|
|
|
Post by cthuzael on Dec 2, 2011 9:57:13 GMT -5
jackson, if you need further sceenshots to keep up your good work, simply state your wish which you need. I'm not jackson, but I have a request. I would really like to see screenshots of a soldier standing front-on and side-on (and crouch/prone if you're feeling awesome), at explicit ranges (say 10m, 30m and 50m), at the highest possible resolution, viewed by a player standing unzoomed with a PP2000 (or prone with an unzoomed M4 if you don't have the PP2000 unlocked). That would be really helpful for gauging exactly how much surface area enemies take up at different ranges. Easiest way to get the range is to have a buddy stand on a cap point. I'd do it myself, but I'm pretty light on for buddies on servers that won't kick me for high ping. I will deliver as soon as I am back online and find a mate to take some beauty shots.
|
|
Lexapro
True Bro
PSN: Lexa_pro
Posts: 1,066
|
Post by Lexapro on Dec 2, 2011 9:59:53 GMT -5
What algorithm would you use to pick a point within a circle where every point has equal chance of being selected? random (x,y) between +/- radius (so a random point within the square inscribing the circle) then testing for X^2 + y^2 < radius? That's probably a terribly inefficient algorithm (rejection sampling), but I would like to know what you're using. EDIT: Alternatively, you could use this: dzindzinovic.blogspot.com/2010/05/xna-random-point-in-circle.html
|
|
|
Post by kirbyderby on Dec 2, 2011 12:06:05 GMT -5
I would argue that, under the right conditions and with a certain playstyle, the Support would actually be an excellent lone wolf setup. One support guy using a bipod can hold a chokepoint from pretty much any range. Alternatively, the PKP makes for a great aggressive Support lone-wolf, allowing him to effectively hose any targets at up to medium distance with the low max spread. hmm interesting - I would like to know what you mean saying - "under the right conditions and with a certain playstyle"? Because I have a hard time running alone on large maps (snipers will tag you and from that moment your fodder - on small maps there are less opportunities to position so once tagged the M320 will hit you. Picking running enemies from a 10-30 meter distance yes that works fine - a "tactical loitering" surprise attack. (will also work with an A M416 that also come with a bipod). But only once. Even a 2 man squad (with one assault) is a problem for me being alone with a bipod. Getting flanked with a bipod means I am dead. The horizontal bipod angle is maybe 70° So once they know that you are there on big maps they will come from the side. In 1-1 situations assault rifles once you are identified on the map are more accurate and as long as I am suppressing - the guy throws a grenade at me and I have to move - and having no grip on my PKP means I need a lot of bullets to hit the enemy ... so I do see a lot of weaknesses without even any special tactics from my enemy once I am tagged. pls. enlighten me the only thing is that I see might work - position yourself kill "say" 2 guys - and re - position wait until the tag wears off (does it??) and then position somewhere else ... that might work - but in conquest/rush maps I do not actively work on the prime goal as I concentrate on runners that are in areas between spawns bases and targets. To be fair, the whole "long distance bipod area denial" strategy I wrote up is largely situational. I haven't used it much since I've only recently unlocked the M240, and I like the PKP in more of an aggressive role. I've found playing defensively with a bipod is mostly effective in infantry-heavy maps like Metro. There aren't a lot of flank routes; most of the action is funneled into a few areas. Throw some claymores around and deploy around a couple of hallways and you've got yourself a slaughterhouse. More often, I'll play an aggressive Support guy. The PKP is excellent for mid-range shootouts due to the low max spread. You can spray at anything accurately, and they'll either be dead or too suppressed to do significant damage. C4 for sneaky traps, or mortar for long-range targets.
|
|
sleep
True Bro
Posts: 10,189
|
Post by sleep on Dec 2, 2011 13:53:54 GMT -5
"long distance bipod area denial" sounds like what i do on grand bazaar conquest large over all the major firing lanes: E->C road, E->C back alley, RU spawn -> C back alley if we lost C and are getting r@ped, C->B center alleyway etc etc. PKP with its working silencer is especially good for this since there's always a mortar dude or two on the other team looking waiting for juicy stationary targets to pop up on the map.
|
|
|
Post by jabberwockxeno on Dec 4, 2011 19:39:40 GMT -5
I've updated my code a bit to reflect recent findings on how recoil and spread behave. Changes: -First shot recoil modifier -Vertical recoil is no longer random -Reduced recoil recovery while firing to reflect the 1-2 degrees/second approximation -No spread recovery while firing -Colored points based on shot number 1. Blue 2. Green 3. Red 4. Cyan 5. Magenta 6. Yellow 7. Black (Then repeat order) Without the exact code, this can only be taken as a very VERY loose guide. I don't really know if any conclusions can be drawn from this yet. There are still too many assumption for me to put any confidence in the plots. The only thing I would put some faith in is the first shot accuracy since recoil and recovery haven't come into effect yet. There is an album of plots of all automatic primary weapons. No attachments. Standing position. Aiming down the sights. Stationary. Firing a 4 shot (or 3 shot for the burst-only guns) burst. 200 trials. Axes are unitless. Multiply by the range to target to get the x-y deviation of the shot in whatever units of range you want to use (meters, yards, feet, AU, lightyears, arshins, whatever). s80.photobucket.com/albums/j166/PvtDBJackson/BF3%20AccuracyI'm looking at the images, and the ak74m seems to use a different scale or something, there are only green, yellow, and red dots. I don't care about range and stuff, I just need to know rough shot pattern of full auto vs. burst (manual on auto move) vs. in game burst auto mode, each with and without ads. TL:DR : Ignoring all the colors and such, do these basically say that if X weapon's chart has a larger dot radius than Y's, then Y is more accurate than X?
|
|
|
Post by jackson on Dec 5, 2011 11:47:30 GMT -5
I've updated my code a bit to reflect recent findings on how recoil and spread behave. Changes: -First shot recoil modifier -Vertical recoil is no longer random -Reduced recoil recovery while firing to reflect the 1-2 degrees/second approximation -No spread recovery while firing -Colored points based on shot number 1. Blue 2. Green 3. Red 4. Cyan 5. Magenta 6. Yellow 7. Black (Then repeat order) Without the exact code, this can only be taken as a very VERY loose guide. I don't really know if any conclusions can be drawn from this yet. There are still too many assumption for me to put any confidence in the plots. The only thing I would put some faith in is the first shot accuracy since recoil and recovery haven't come into effect yet. There is an album of plots of all automatic primary weapons. No attachments. Standing position. Aiming down the sights. Stationary. Firing a 4 shot (or 3 shot for the burst-only guns) burst. 200 trials. Axes are unitless. Multiply by the range to target to get the x-y deviation of the shot in whatever units of range you want to use (meters, yards, feet, AU, lightyears, arshins, whatever). s80.photobucket.com/albums/j166/PvtDBJackson/BF3%20AccuracyI'm looking at the images, and the ak74m seems to use a different scale or something, there are only green, yellow, and red dots. I don't care about range and stuff, I just need to know rough shot pattern of full auto vs. burst (manual on auto move) vs. in game burst auto mode, each with and without ads. TL:DR : Ignoring all the colors and such, do these basically say that if X weapon's chart has a larger dot radius than Y's, then Y is more accurate than X? Sorry made some new plots and photobucket is a pain. Working with Symthic to get my plots on his site. But yes, if the dots are spaced closer, then the gun is more accurate.
|
|