Welcome to symthic forums! We would love if you'd register!
You don't have to be expert in bit baking, everyone is more than welcome to join our community.

You are not logged in.

Hey! If this is your first visit on symthic.com, also check out our weapon damage charts.
Currently we have charts for Battlefield 3, Call of Duty: Black Ops 2, Medal of Honor: Warfighter and Call of Duty: Modern Warfare 3

  • "3VerstsNorth" started this thread

Posts: 1,614

Date of registration
: Apr 12th 2013

Platform: PC

Location: Guilin Peaks, Finland

Battlelog:

Reputation modifier: 14

  • Send private message

71

Friday, November 15th 2013, 1:29am

Update!

I noob-failed a couple times in computing new tables. So not today but at some time tomorrow, there will new tables with confidence limits and the putatively correct movement speed compensation for moving ADS min spread.

Meanwhile, please go see post #3. To answer the earlier questions on T100 distributions, I show them and the percentiles there.

I also wrapped up some of the headshot efficiency analyses I have been working on. All feedback welcome and appreciated!

@bogeymanuk
np, all (well, most) discussion is useful :). I am a strong proponent of human experiments myself too.
"Less is more? How can that be? How could less be more, that's impossible. More is more." Yngwie Malmsten
"Many bullets help." WoopsyYaya
"most rhetorically legitimate ad hominem 2015" ToTheSun!

Posts: 199

Date of registration
: Nov 12th 2013

Platform: PC

Battlelog:

Reputation modifier: 3

  • Send private message

72

Friday, November 15th 2013, 3:22am

well, i guess that's that... (puts SCAR-H to head (of opponent, hopefully)). i was sort of hoping that the headshot effect wasn't quite so... reliable as that, but those medians and 90% intervals are damning evidence. in practice i'm sure it's not quite that bulletproof, and there does seem to be a regime where it does pay to start going to the body with the AEK (though that's out to 80m, admittedly), but for the SCAR it's headshot dominance all day every day.

still, this does give the guns some interesting niches for particular playstyles; i don't personally run around at 60m trying to shoot people from moving ADS, but if someone else does then they should by all means give the AUG a shot, i guess. the AK-12 and ACE-23 do well as torso rifles with good mid-range performance, which matches up with their apparent popularity well too.

while the i understand the revision of the tables due to errors is still ongoing, thinking ahead it would be interesting to see how the top configurations for standing ADS do at moving ADS, and vice versa in the next iteration. could be some of the balancing is in the one-trick pony factor with the attachments; e.g. heavy barrel/potato grip might be very suboptimal for CQB on the SCAR-H, but might not handicap the M416 as much up close. burst-fire effects might also be an interesting offset (particularly for something like the CZ that doesn't really have anything else to recommend it from the raw numbers).

Posts: 67

Date of registration
: Oct 31st 2013

Platform: PC

Location: United States

Battlelog:

Reputation modifier: 3

  • Send private message

73

Friday, November 15th 2013, 3:42am

Theorycrafting is useful for finding the optimal setup for certain restricted conditions but as someone has already pointed out you would be hard pressed to find someone who can double tap a SCAR-H to the head consistently from any range.


Sure, I agree. The aim here is not so much to find the optimal setup but rather map most of the relevant (ADS/HIP, Stand/Move, ...) conditions to (i) identify which weapons have the greatest potential in each condition and (ii) to quantify how much one looses in this 'condition-specific' potential per condition when deciding to use a given compromise setup.

Quoted

Along with the theory there could be some form of experimental data to put alongside and create a more balanced picture, even if merely anecdotal compared to thousands of theory parses.

There are enough matches played by us to form some form of limited dataset with regards to differing setups. Say I play 10 matches with the M416 with nothing but a sight on as a baseline, note the figures from the matches then play 10 with the ergo 10 with stubby 10 with angled etc, doing all barrel/grip permutations. Although not entirely scientific as the variations will be down to more factors than just the attachments there would presumably be at least a little correlation between players who did the same thing? At the very least it should make an interesting experiment.


Experiments are doable but incredibly expensive in terms of time. Here we are mapping 27.4 million combinations in some hours. With in-game experiments, you need many hours or tens of hours to test a single combination, and even then the result is not very accurate because of game-to-game, day-to-day, and so on variability in the player and in the overall game environment. See: The RDS Experiment.


I personally feel identifying which guns are "superior" is easy. The hard part in BF4 has been the attachments. My primary interest is in finding which attachments are optimal per gun for all-around, long range, and whatever other conditions.

We have certainly been inching closing but we need more practical consideration.
Hail to the king, baby!

  • "3VerstsNorth" started this thread

Posts: 1,614

Date of registration
: Apr 12th 2013

Platform: PC

Location: Guilin Peaks, Finland

Battlelog:

Reputation modifier: 14

  • Send private message

74

Friday, November 15th 2013, 7:29am

@Shinryu
@Cyphur
To estimate the best compromises, we need the condition-specific optimals first. And getting those for moving ADS was not possible earlier as there was some apparent controversy, or disagreement that turned out to be agreement or something, blabla, with respect to the issue movement speed vs. minimum spread.

The server is ready with these in an hour. I can then compute both the optimals (as earlier) but also tweak the code to identify the attachment set-ups that are optimal compromises for all ranges and for conditions.
"Less is more? How can that be? How could less be more, that's impossible. More is more." Yngwie Malmsten
"Many bullets help." WoopsyYaya
"most rhetorically legitimate ad hominem 2015" ToTheSun!

  • "3VerstsNorth" started this thread

Posts: 1,614

Date of registration
: Apr 12th 2013

Platform: PC

Location: Guilin Peaks, Finland

Battlelog:

Reputation modifier: 14

  • Send private message

75

Friday, November 15th 2013, 10:05am

... reliable as that, but those medians and 90% intervals are damning evidence. in practice i'm sure it's not quite that bulletproof,

Indeed. Miss your aim by half a head and you would have been much better off aiming at the torso. Establishing the statistical limits still pays off I think, because there are players that are just inhumanly good and really can push the guns to their limits in game.
still, this does give the guns some interesting niches for particular playstyles;

This. I think that even for us mortals, the theoretical analysis can be useful in revealing which guns are best suited for which playstyles or that what is the most effective way of using a gun-of-choice. Much more work needs to be done with the visualisation and compromise search though...
"Less is more? How can that be? How could less be more, that's impossible. More is more." Yngwie Malmsten
"Many bullets help." WoopsyYaya
"most rhetorically legitimate ad hominem 2015" ToTheSun!

Posts: 803

Date of registration
: Feb 18th 2013

Platform: 360

Battlelog:

Reputation modifier: 4

  • Send private message

76

Sunday, November 17th 2013, 6:47pm

DMRs need a buff, Shotguns need a Nerf and the Magnum and Rex need a huge buff or removal of the delayed firing action.

  • "3VerstsNorth" started this thread

Posts: 1,614

Date of registration
: Apr 12th 2013

Platform: PC

Location: Guilin Peaks, Finland

Battlelog:

Reputation modifier: 14

  • Send private message

77

Sunday, November 17th 2013, 10:42pm

Hi all, sorry for the delay. I have the simulations ready for standing/moving ADS/HIP etc. + identification of the best compromises across conditions ... but, bear with me, I need to do check a little thing or two WRT Min Spread that make a gigantic difference in the weapon performance estimation.

@lammeatt
wrong thread much? :D
"Less is more? How can that be? How could less be more, that's impossible. More is more." Yngwie Malmsten
"Many bullets help." WoopsyYaya
"most rhetorically legitimate ad hominem 2015" ToTheSun!

Posts: 30

Date of registration
: Nov 11th 2013

Platform: PC

Battlelog:

Reputation modifier: 3

  • Send private message

78

Tuesday, November 19th 2013, 4:02am

Just to clarify, before your charts were taken down I spent a lot of time reading them.

The suggested attachments for guns didn't take recoil into account, did it?

  • "3VerstsNorth" started this thread

Posts: 1,614

Date of registration
: Apr 12th 2013

Platform: PC

Location: Guilin Peaks, Finland

Battlelog:

Reputation modifier: 14

  • Send private message

79

Tuesday, November 19th 2013, 9:31am

Just to clarify, before your charts were taken down I spent a lot of time reading them.

The suggested attachments for guns didn't take recoil into account, did it?

The revised Optimal tables are now online. See the text below "2x2 Playstyle variables".

...this is not the announcement for v1.0 release as the Compromise tables remain WIP.
"Less is more? How can that be? How could less be more, that's impossible. More is more." Yngwie Malmsten
"Many bullets help." WoopsyYaya
"most rhetorically legitimate ad hominem 2015" ToTheSun!

Posts: 30

Date of registration
: Nov 11th 2013

Platform: PC

Battlelog:

Reputation modifier: 3

  • Send private message

80

Tuesday, November 19th 2013, 11:47am

So the Heavy Barrel does affect moving ADS?

Is it enough to be worth the recoil increase though?