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

Posts: 30

Date of registration
: Oct 1st 2013

Platform: PC

Battlelog:

Reputation modifier: 3

  • Send private message

11

Sunday, November 10th 2013, 11:39pm

I read over your posts and may have missed it, but what is the % chance of actually achieving the TTK outlined in the T100 tables? I've been working on some "Effective Range" tables myself and have values that differ slightly from yours.

Darktan13

Unhappy Camper

(4,678)

Posts: 1,413

Date of registration
: Feb 13th 2012

Platform: PC

Location: Europe

Battlelog:

Reputation modifier: 15

  • Send private message

12

Sunday, November 10th 2013, 11:39pm

Try repeating this using Hitrater - Hitrate simulator Not sure if thats the most up to date version, but you can always ask.

interested if the results differ at all.

Quoted from "J0hn-Stuart-Mill"


  • "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

13

Sunday, November 10th 2013, 11:45pm

Great work!

Well, at least i think it is. Far too complicated for me to be sure 8|

Why no AUG A3?


And why is the optimal burst size for some weapons so low at 5m?


Thanks. Good catch with AUG A3! There was a table-sizing problem. Corrected now - AUG is at the bottom with or without comp.

At 5m, almost always a double-headshot does the trick and hence many burst sizes and attachment combinations have identical T100 values, and the selection of the smallest is slightly spurious because of rounding errors.
"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

14

Sunday, November 10th 2013, 11:58pm

I read over your posts and may have missed it, but what is the % chance of actually achieving the TTK outlined in the T100 tables? I've been working on some "Effective Range" tables myself and have values that differ slightly from yours.

The T100 tables give the estimated expection value (...the average of all realized T100s). I haven't had time to look at the distribution of T100s at all, so dunno if it is skewed or not. If not very skewed, the % chance is close to 50 %. I'll need to look into this.

Try repeating this using Hitrater - Hitrate simulator Not sure if thats the most up to date version, but you can always ask.

interested if the results differ at all.


From the algorithm-side, Hitrater (Hitrate simulator) should be very close except that it is based on the HRecoil algorithm A and my code uses B as presented here:Algorithms for the Shooting Mechanics in BF4. There also could be differences in the relative sizes of the hitboxes. (We should get a Symthic-concencus on the hitboxes and then use the same in every code.)

The bigger difficulty in comparing these two is that the outputs are quite different. Hitrater gives the fraction of hits in a single burst with given parameters. My code shoots bursts, with inter-burst intervals considered too, until the target is dead and then reports the "time-to-kill" exactly at the shot that caused the kill.

@Sao
Awesome! :)

Edit: H Recoil info for Hitrater corrected. Thanks tehmoriz!
"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!

This post has been edited 2 times, last edit by "3VerstsNorth" (Nov 11th 2013, 8:50am)


Posts: 30

Date of registration
: Oct 1st 2013

Platform: PC

Battlelog:

Reputation modifier: 3

  • Send private message

15

Monday, November 11th 2013, 12:25am

Not sure if this is the right place to post this, but it's for discussion. I wasn't originally going to post this until I finished a total comprehensive analysis, but I figured that I may as well post what I have so far:

I've been trying to determine "effective ranges" for the weapons along with semi-realistic TTK values. Obviously you've done a much better job than I ever could, but I've approached it from a slightly different angle. I determined the TTK for naked weapons while trying to maintain 100% hitrate on target - this is why I wanted to know what your T100 hitrates were. My thought process was that if you aren't hitting your target, you're not using the weapon with 100% effectiveness (my goal was an effective range calculation, after all). This means that I had to make sure that spread recovered to an acceptable level such that the target could be hit with 100% chance at a given range.

REMOVED. Will be moved to separate topic.

I'm still working on other calculations, but you may have done the majority of what I was planning on doing, so props to you.

This post has been edited 2 times, last edit by "Vintar" (Nov 11th 2013, 1:55am)


Posts: 328

Date of registration
: Jun 12th 2013

Platform: PC

Location: US

Battlelog:

Reputation modifier: 5

  • Send private message

16

Monday, November 11th 2013, 12:44am

This is the best single target weapon analysis I have seen on symthic yet.

Some of the burst sizes seem to be huge, like the scar-h 10 burst at 20m. I bet someone could single shot 2-3 at the head faster and more accurate than someone 10 round bursting in full auto. With burst sizes that high another metric that includes mag size and reload speed will be important for evaluating multiple target performance. I personally don't care so much about how fast I can kill a single target, but efficiently I can keep engaging targets and still have the killing speed to take them down.

The best way I can think of for deciding optimal weapon combinations is to have accurate measurements of the BF4 maps with noted engagement hotspots. You should be able to define engagement "zones" and then decide which play styles are most viable in the zone. For example, in rush it might be better for the attacker going for the mcom to focus on moving stats, while the defenders to go for longer range stationary stats. I don't think one optimal compromise should be used, but have an optimization for each zone. Another thing that should be considered, is how much cover is available in each zone. This would force a change in the aiming point for a target behind cover and effectively change all the charts.

Recoil compensation difficulty is a bit trickier as people use different methods to compensate. Some people change their initial aimpoint and let the recoil move the weapon over the target. Others will use the same aimpoint across weapons and change the burst length/rate of fire to manage the recoil. Some will change the amount of movement they make with the mouse, and others will change the mouse sensitivity to get the same movement with different weapons. Ultimately what happens is that gamer will build up muscle memory and instinct, then when in a heated firefight will revert to it.

You would have to use some kind of standardized units that could be converted into mouse dpi + in game sensitivity settings (and console joystick angles).

Edit: Was thinking only about PC when I wrote this so another thing that should be consider is console auto aim performance. In some games people rapidly pop in and out of ADS to let the auto aim do most of the work.
The fact that someone has an opinion, doesn't make their opinion a fact. Making just arguments first requires an acknowledgement of intellectual humility, while valid arguments require you to not commit fallacies of logic and rhetoric.

This post has been edited 1 times, last edit by "untamedone" (Nov 11th 2013, 12:52am)


Posts: 328

Date of registration
: Jun 12th 2013

Platform: PC

Location: US

Battlelog:

Reputation modifier: 5

  • Send private message

17

Monday, November 11th 2013, 1:15am

This means that I had to make sure that spread recovered to an acceptable level such that the target could be hit with 100% chance at a given range.

Are you waiting for full spread recovery for each weapon or is it a different amount on each weapon/range?
The fact that someone has an opinion, doesn't make their opinion a fact. Making just arguments first requires an acknowledgement of intellectual humility, while valid arguments require you to not commit fallacies of logic and rhetoric.

  • "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

18

Monday, November 11th 2013, 1:19am

@Vintar

I think your analysis converges with mine in that beyond ~30 m, 2-shot microbursting is the way to go. I also think you should write this out into a separate thread with the assumptions and methods explained so that we could better understand where the numbers come from.

This is the best single target weapon analysis I have seen on symthic yet.
Some of the burst sizes seem to be huge, like the scar-h 10 burst at 20m. I bet someone could single shot 2-3 at the head faster and more accurate than someone 10 round bursting in full auto.

Thanks! Re burst sizes, I suspect that these short-range anomalies are caused by very rare incidents that bias the mean T100 by microscopic amounts. Since the global minimum is reported, these biases influence the outcome under circumstances where many parameter combinations have nearly identical values.

I am now recomputing all tables with median T100 instead of mean to get more robust estimates.

Quoted

With burst sizes that high another metric that includes mag size and reload speed will be important for evaluating multiple target performance. I personally don't care so much about how fast I can kill a single target, but efficiently I can keep engaging targets and still have the killing speed to take them down.

(The burst sizes will likely go down a lot with the new T100 estimator, but still...) The mag size considerations should definitely be included, maybe into another metric. I also will gladly accept a penalty of some tens of milliseconds in the effective TTK just to conserve ammo for another kill before reload.

Quoted

The best way I can think of for deciding optimal weapon combinations is to have accurate measurements of the BF4 maps with noted engagement hotspots. You should be able to define engagement "zones" and then decide which play styles are most viable in the zone. For example, in rush it might be better for the attacker going for the mcom to focus on moving stats, while the defenders to go for longer range stationary stats. I don't think one optimal compromise should be used, but have an optimization for each zone. Another thing that should be considered, is how much cover is available in each zone. This would force a change in the aiming point for a target behind cover and effectively change all the charts.

It would be great to get a community opinion on exactly what ranges and cover conditions are relevant in different scenarios. Could be tricky though with so much variability in the game.

Quoted

Recoil compensation difficulty is a bit trickier as people use different methods to compensate. Some people change their initial aimpoint and let the recoil move the weapon over the target. Others will use the same aimpoint across weapons and change the burst length/rate of fire to manage the recoil.

IMO, (unless the initial aimpoint was lower than intended) not compensating for the recoil is always a weak solution because the beneficial or harmful effect of V Recoil is very dependent on range, which is difficult to estimate accurately (see the "no-compensation" tables: the optimal target point varies between head to stomach in just ten or twenty meters).
"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
: Oct 1st 2013

Platform: PC

Battlelog:

Reputation modifier: 3

  • Send private message

19

Monday, November 11th 2013, 1:40am

This means that I had to make sure that spread recovered to an acceptable level such that the target could be hit with 100% chance at a given range.

Are you waiting for full spread recovery for each weapon or is it a different amount on each weapon/range?


It's uniform spread recovery between each burst such that the killing shot will retain 100% accuracy (just barely). This means that the spread recovery between each burst is a bit more than the minimum recovery needed for the next shot/burst to be 100% accurate (to have equal times between bursts). Thus the spread recovery time changes depending on the distance the target is at (which changes the allowable max spread), the damage of the gun (dictates BTK), and the accuracy/spread parameters of the gun itself (which restricts max burst length).

I'm gonna go ahead and redact the above post and move it into its own topic, since I don't want to derail this one.

Posts: 1,487

Date of registration
: Jul 20th 2012

Platform: PC

Battlelog:

Reputation modifier: 15

  • Send private message

20

Monday, November 11th 2013, 4:53am

The latest version of hitrater automatically compensates for horizontal drift. At least, I remember giggles mentioning it at one point.