Call of Duty Weapon and Attachment Stats

[kylet357]

Feb 29, 2016 23:35:15 GMT -5 darkrangeresp said:

A quick question for Marvel. This is my first post on this forum, and I do appreciate all the work that you do with gathering the weapon data. I was looking at your chart, and I noticed that the fire rates on the Brecci and Drakon both seem to be reduced from the previous stat chart I was looking at. I am an Xbox one player, I went in game and manually timed them out. I was able to get the Drakon up 268 without rapid fire and 281 with it manually. Likewise, I clocked the Brecci at 208 and 221 with rapid fire respectively. I repeated both of these tests several times and my results were consistent give or take 5 RPM across firing 25 magazines with each weapon.

I saw that these weapons got a stealth nerf in another thread on this board reducing their fire rates. I was wondering if your weapon stat chart is PC specific or if you also take code from the console versions?

He gets the code from the PC version. For pretty much 99% of the time, Marv's charts are accurate across all platforms. There are very few weapons that are different across the platforms (the only one I can even think of is the MP40 in World at War). I'm not exactly sure what's going on with the console versions of the game, since I don't even have an X1/PS4 (I don't even have BO3).

[banana]

Feb 29, 2016 23:35:15 GMT -5 darkrangeresp said:

A quick question for Marvel. This is my first post on this forum, and I do appreciate all the work that you do with gathering the weapon data. I was looking at your chart, and I noticed that the fire rates on the Brecci and Drakon both seem to be reduced from the previous stat chart I was looking at. I am an Xbox one player, I went in game and manually timed them out. I was able to get the Drakon up 268 without rapid fire and 281 with it manually. Likewise, I clocked the Brecci at 208 and 221 with rapid fire respectively. I repeated both of these tests several times and my results were consistent give or take 5 RPM across firing 25 magazines with each weapon.

I saw that these weapons got a stealth nerf in another thread on this board reducing their fire rates. I was wondering if your weapon stat chart is PC specific or if you also take code from the console versions?

pc specific,

marvel you should add a note on the console rpms imo

[darkrangeresp]

I can see why they would do a PC specific patch for the gun. I have played a lot of pc shooters in my day and mouse aiming is very powerful with some of the weapons like this. It makes sense also that his chart is PC specific as it would be hard to access the code on console. I guess this is one of the first times Treyarc has made a game that has different weapon stats on different platforms.

[kylet357]

Mar 1, 2016 18:00:26 GMT -5 darkrangeresp said:

I guess this is one of the first times Treyarc has made a game that has different weapon stats on different platforms.

Nope. As I said before, MP40 on World at War (PC MP40 actually got nerfed, console one didn't). And if I remember correctly, Marv once said that getting into the console copies of the game would be just as easy/hard as it would on PC.

[djangoraw]

For the grip attachment, what does ADS Recoil Reduction Rate =1 and ADS Recoil Reduction Limit =.075 mean?

[probaddie]

Mar 2, 2016 18:18:49 GMT -5 djangoraw said:

For the grip attachment, what does ADS Recoil Reduction Rate =1 and ADS Recoil Reduction Limit =.075 mean?

Unfortunately, there's no hard evidence for how this works exactly. However, we (Marvel4 and I) have a good theory based on the values of these variables in Black Ops II and the apparent fact that the Grip actually works (for automatic weapons; more on this later) in Black Ops III.

So there are three variables that govern this mechanic: reductionRate, reductionLimit and returnRate. (I omit the "adsRecoil" prefix for simplicity.) In Black Ops II, every weapon had these values set to 0, 0 and 1, respectively - with one exception: the HAMR, which has values of 0.03, 0.15 and 1, respectively. Now, if you played Black Ops II, you'll remember that the HAMR gradually "settled" over prolonged fire. (This is also partly on account of the decreased fire rate after the sixth shot.) In Black Ops III, as you noted, equipping a Grip sets the reductionLimit and reductionRate to 0.075 and 1.

So this is the theory that is consistent with all of these facts (and the names of the variables). For each shot fired, the velocities drawn for the subsequent shot are scaled by reductionRate more than the previous value. So, for example, with the HAMR, the first shot is not reduced at all the second shot is reduced by 3% (corresponding to the 0.03 value for reductionRate), the third is reduced by 6%, etc. This stops when either the player releases the trigger or the reductionLimit is reached. So, with the HAMR, after the sixth shot is fired, the 15% reduction is reached and all shots fired thereafter have their kick reduced by 15%. When the trigger is released, the reduction is reduced at a rate of returnRate per fireTime of the weapon. That's just a fancy way of saying, "if it's time for you to fire another shot but you're not gonna, roll back the reduction by returnRate". With a returnRate of 1, you expect to be back at a 0% reduction the moment you release the trigger, since you lose 100% of the reduction.

All this implies two things for the Grip in Black Ops III. First, the Grip provides a 7.5% reduction to viewkick - its simply impossible to tell whether this affects gunkick at all - to every shot except the first. Second, the Grip is broken for semi-automatic weapons, since releasing the trigger is required to fire each subsequent shot. Now, that second consequence is why I have to stress that none of this is confirmed information. It's hard to tell whether the second point really is true, as it's very hard to test its validity directly; the Sheiva has very little recoil and the MX Garand, because of its very random recoil, wouldn't easily signal in a test that the Grip does not work as expected. If the Grip does work for semi-auto weapons, then there's something else going on that we don't understand.

[haoz]

I'm a little confused because I think you put some values out of order with respect to their labels? Can you clarify that ...

reductionRate - rate at which viewkick magnitudes are scaled as gun continues to fire
reductionLimit - maximum viewkick reduction
returnRate - rate at which viewkick magnitudes return to nominal while gun is not firing

What does it mean that grip sets reductionLimit to 1? You suggested that the HAMR reduction limit of 0.15 means that viewkick can only be reduced by 15%. Doesn't 1 imply a ceiling of 100% viewkick reduction then?

Is it necessary that the trigger be held for returnRate not to apply? What if return rate is inactive so long as the gun is being fired near its maximum rate?

[probaddie]

Mar 2, 2016 21:35:07 GMT -5 haoz said:

I'm a little confused because I think you put some values out of order with respect to their labels? Can you clarify that ...

reductionRate - rate at which viewkick magnitudes are scaled as gun continues to fire
reductionLimit - maximum viewkick reduction
returnRate - rate at which viewkick magnitudes return to nominal while gun is not firing

What does it mean that grip sets reductionLimit to 1? You suggested that the HAMR reduction limit of 0.15 means that viewkick can only be reduced by 15%. Doesn't 1 imply a ceiling of 100% viewkick reduction then?

Is it necessary that the trigger be held for returnRate not to apply? What if return rate is inactive so long as the gun is being fired near its maximum rate?

Yeah, I completely bungled the order of the numbers, which are corrected now in my previous post. ("Respectively" my ass.) You have exactly the right idea regarding the proposed theory.

Your idea for how returnRate works is possibly correct but raises a few questions. What would it mean for the gun to fire "near" its maximum rate? At what rate is the reduction being lost between each shot? Per second is too fast for nearly every gun, as even a weapon that fires at only 600 RPM would lose 10% of its reduction before the next shot. Granted it's possible that this is what happens, but from a coding perspective it's much harder than a simple query of "bro, u can fir nao. u fire? no? k u lose x reduction lol."

[probaddie]

These questions regarding the adsRecoil mechanic prompted Marvel4 to do some testing on my behalf, testing that almost perfectly confirms what I stated before. Here are the results that Marvel4 has confirmed:

• The adsRecoil mechanic does not affect gunkick.
• The returnRate is correct as described, but it is not applied continuously when the trigger is released. Rather, the game is simply querying if the user is taking a shot within one fireTime of the last. When a shot is being fired, the reduction is increased by reductionRate; if not, the reduction is decremented by returnRate. Thus, a returnRate of 1 necessarily implies that the reduction is completely reset after one fireTime if a shot is not fired.
• The reduction does not apply to the first shot. Consequently, this means that the first shot of each burst from a burst-fire weapon is also not affected, since a time of fireTime + burstFireDelay must pass before the next volley may be fired.
• The Grip works on semi-automatic weapons only when they are fired at the maximum rate. This prevents returnRate from being invoked.

Again, all of this assumes that the core theory is correct, but there's very little now that could be wrong given the evidence.

[Lexapro]

With the grip, is viewkick minimum magnitude factored in before or after the 7.5% viewkick reduction?

[imrlybord7]

probaddie Truly a dog of science if there ever was one.

[kylet357]

Mar 5, 2016 17:09:51 GMT -5 imrlybord7 said:

probaddie Truly a dog of science if there ever was one.

Not as much as these two

[probaddie]

Mar 4, 2016 17:31:57 GMT -5 Lexapro said:

With the grip, is viewkick minimum magnitude factored in before or after the 7.5% viewkick reduction?

After Marvel4 and I basically spent the whole night testing, we figured out that it doesn't matter, because the grip actually reduces minimum magnitude to zero.

[Lexapro]

Mar 8, 2016 2:48:41 GMT -5 probaddie said:

Mar 4, 2016 17:31:57 GMT -5 Lexapro said:

With the grip, is viewkick minimum magnitude factored in before or after the 7.5% viewkick reduction?

After Marvel4 and I basically spent the whole night testing, we figured out that it doesn't matter, because the grip actually reduces minimum magnitude to zero.

HOLY ****

[probaddie]

Mar 8, 2016 9:09:04 GMT -5 Lexapro said:

Mar 8, 2016 2:48:41 GMT -5 probaddie said:

After Marvel4 and I basically spent the whole night testing, we figured out that it doesn't matter, because the grip actually reduces minimum magnitude to zero.

HOLY ****

Yep. Apparently, adsRecoilReductionRate also reduces adsViewKickMinMagnitude. Hence, you get a 100% reduction to minimum magnitude. Real f***ing neato, huh?

[imrlybord7]

So does this mean Grip is good?

[banana]

Mar 8, 2016 13:28:33 GMT -5 imrlybord7 said:

So does this mean Grip is good?

just wait for the plots lol

[Gamma]

Why do the pros use long barrel on the VMP? Am I missing something? I thought it useless on the VMP and SMGs in general

[Lexapro]

On the VMP, it pushed the 5HK range from 1000m to 1250m. It's not a huge difference but it actually does help at medium range. The VMP and the Weevil are the SMGs that benefit from long barrel the most. Long barrel is USELESS on the Kuda though and I have no idea why some pros still use it.

That being said, a lot of pros are switching over to Quickdraw, Grip, Stock on VMP and Laser Sight on the Kuda.

[banana]

Mar 15, 2016 11:30:19 GMT -5 Lexapro said:

On the VMP, it pushed the 5HK range from 1000m to 1250m. It's not a huge difference but it actually does help at medium range. The VMP and the Weevil are the SMGs that benefit from long barrel the most. Long barrel is USELESS on the Kuda though and I have no idea why some pros still use it.

That being said, a lot of pros are switching over to Quickdraw, Grip, Stock on VMP and Laser Sight on the Kuda.

>Kuda w/ long barrel

They use it because they go off of feel and not stats. I don't think any will be able to tell you what exactly long barrel does. Fallible human feel is also why they use the m8 and not the xr2

Anyway their feel tends to be better than the average cod pubstomper which is cool I guess. My feel caused me to use the Vesper and MoW during the beta which is dope.

[haoz]

Mar 15, 2016 21:24:20 GMT -5 banana said:

Mar 15, 2016 11:30:19 GMT -5 Lexapro said:

On the VMP, it pushed the 5HK range from 1000m to 1250m. It's not a huge difference but it actually does help at medium range. The VMP and the Weevil are the SMGs that benefit from long barrel the most. Long barrel is USELESS on the Kuda though and I have no idea why some pros still use it.

That being said, a lot of pros are switching over to Quickdraw, Grip, Stock on VMP and Laser Sight on the Kuda.

>Kuda w/ long barrel

They use it because they go off of feel and not stats. I don't think any will be able to tell you what exactly long barrel does. Fallible human feel is also why they use the m8 and not the xr2

Anyway their feel tends to be better than the average cod pubstomper which is cool I guess. My feel caused me to use the Vesper and MoW during the beta which is dope.

Yes but going purely by stats is a sort of feeling as well. Or you're missing some sort of quantification of the following XR-2 deficiencies: long burst delay, side wobble in a burst, slower fire rate. The M8 is more accurate and more forgiving.

Wasting two class slots for long barrel on a Kuda is such a shame though.



I think the finding re. viewkick min magnitude to 0 explains why grip is so good on the VMP. It can sometimes shoot straight for ~10 bullets with grip.

[Lexapro]

I think some players have started switching to the XR-2. Results have been...mixed.

[JustABitAgroed]

Players are having a lot more success with it over in EU but the only NA players I've seen pull it out have been Sender and BLFire. Both with poor results. But if the M8 keeps getting nerfed/banned then I think players will start switching over. Or start going to the Shieva a lot more. It really depends. I don't think the low usage and poor results come from the weapon's performance though. Most players just found the M8 and stuck with it. Only now are they trying out the XR-2 a little bit.

Regarding the XR-2 itself, I don't think the wobble is as big of a deal as people make it out to be. Especially with Grip. (We'll have to wait for the recoil plots to see if that actually holds true or not though.) Also, the longer delay isn't all that much longer. 2 frames isn't an eternity, especially when you consider that most pros aren't going to be perfectly timing their bursts. A lot of them just oversample (coming from what I saw at UMG SC), and even if they do practice proper form, I think it's fair to say that they're not always (if not rarely) going to be able to consistently time their bursts in the heat of a match. Looked at within this context, I think the auto-burst feature is very undervalued. Also, 143 less RPM isn't an enormous disparity when considering that you need one less shot to kill. The only thing that the faster RPM really helps with is hitting multiple shots in a row, something that is obviously important with a burst weapon, but keeping the one burst TTK in mind, you both have to hit fewer shots in order to get the one burst kill, but you also have to stay on target for far less time (disparity of 40ms) which should make one bursting far easier if it weren't for the recoil pattern (which we'll have to wait to see just how bad it really is. Once again, I think people are overestimating how much recoil it has). Where I think the XR-2 fails is in its power to accuracy ratio. Specifically when having to shoot multiple bursts. If you don't one burst someone with the XR-2 then the M8 beats it out in regards to theoretical TTK when 2,3, and 4 bursting (hopefully pros aren't going to have to use more than 4 bursts). Even when factoring in the auto-burst X-factor, the weapon still doesn't have a significant enough TTK advantage over the M8 to justify its recoil pattern. So I think that's the last hurdle that the weapon has to get over. Either the accuracy needs to be improved. Or god forbid we bump the TTK a little bit, say to .066 firetime.

Overall though, I can't disagree with banana. At this point, the pros should be using the XR-2 on most maps, and maybe running a Shieva on Infection or Fringe. But I don't think the M8 has much of a place anymore.

And in my honest opinion, having pros run things like Long Barrel on the Kuda and Pharo is an embarrassment.

[banana]

Mar 16, 2016 9:48:36 GMT -5 JustABitAgroed said:

Specifically when having to shoot multiple bursts. If you don't one burst someone with the XR-2 then the M8 beats it out in regards to theoretical TTK when 2,3, and 4 bursting (hopefully pros aren't going to have to use more than 4 bursts).

Last time I calculated it, the xr2 still beats it in ttk even you have to use the next burst

[JustABitAgroed]

.07 * 3 + .2 = .41 // .06 * 4 + .166 = .406 // Firetime still counts on the last bullet of the burst, correct? Or is that already factored into burst delay?

[haoz]

I truly think the XR-2 is inferior to the M8. It's not accurate enough for long range. It's too slow for close/mid range.

Since it's not possible to oversample burst and semi-auto guns in this game (pulling the trigger before the end of burst delay will still make the gun fire after burst delay passes), auto-burst is a bit redundant for firing at max rate.

[JustABitAgroed]

Mar 17, 2016 21:47:25 GMT -5 haoz said:

(pulling the trigger before the end of burst delay will still make the gun fire after burst delay passes)

Now that's interesting. I noticed it did this on the Shieva but I didn't think it did for the M8 or RK5. Are you sure there's no penalty?

[Gamma]

What are your 3 preferred attachments on the XR-2?
Mine were QD, long barrel, rapid fire. But after the RF nerf, it doesnt seem all that useful.

High caliber isn't worth it for me, fewer than 1 in 10 of my kills are headshots

[haoz]

Red dot, QD, stock, rapid fire/grip.

[banana]

Red dot
QD
Rapid fire
Stock / Grip

or

Red dot
QD
Rapid fire
Stock
Gri[