The Oracle of Death - Page 5

Omnimon · 2013-08-09, 02:25 PM

Originally Posted by Chewy

I can work with this, thank you for the information. Give me a moment to get a some good music playing, grab a drink, and put another mouth full of ranch sunflower seeds in a cheek. And yes Im going to derail a bit just to get a clear view on MAX AI weapons. Sorry about that.

First let me order them by faction. It makes things readable and much faster to find.

Code:

NC
Scatter-
24,731 Kills
5,036 Unique Killers
4.9 kpu

Mattock-
5,728 Kills
346 Unique Killers
16.56 kpu

Grinder-
9,610 Kills
792 Unique Killers
12.13 kpu

Hacksaw-
19,281 Kills
1,224 Unique Killers
15.75 kpu
===========

TR
Heavy C-
22,592 Kills
5,275 Unique Killers
4.28 kpu

Mercy-
20,399 Kills
1,587 Unique Killers
12.85 kpu

Mutilator-
14,659 Kills
1,311 Unique Killers
11.18 kpu

Onslaught-
5,361 Kills
426 Unique Killers
12.58 kpu
===========

VS
Quasar-
31,350 Kills
6,125 Unique Killers
5.12 kpu

Blueshift-
24,630 Kills
1,420 Unique Killers
17.35 kpu

Cosmos-
16,335 Kills
1,256 Unique Killers
13.00 kpu

Nebula- 
11,423 Kills
683 Unique Killers
16.72 kpu

With them in order by faction I can give better opinions. First thing I see is the kpu, thanks to your ordering, and that no TR weapon is placed till 6th out of 12. That is odd, and good proof that TR MAXes don't last long enough or have the least support compared to NC and VS MAXes.

===

Next is total kills. Damn are these numbers another story from kpu. Mattocks and Onslaughts are just shit with only 5k kills when everything else is at least 2 times that number but for grinders. Then you have Quasars on top with 31k kills. 7k over 2nd place of 24k, just under a 1/3 better, and Quasars have over 5 times the kills of last place!

===

Finally we have unique killers. This one is a bit harder to read for me without pulling them out and putting them next to each other. With pure totals you should see around 2 times the number of players that killed with MAX AI weapons (2 weapons for each player).

So for just totals you have
NC-7,398, TR-8,599, VS-9,484

That's telling me that VS is by FAR using the most MAXes and NC is pulling the least for AI work. NC is behind by over a thousand to TR and TR is behind by almost 900 to VS. You have to remember that this is for both arms and you need to cut the totals in half for AI MAXes getting kills but even then the story is the same with NC having the least MAXes, TR in the middle, and VS with the most by a large number.

Im not even going to get into how I make for 2 of the 346 of the Mattock users. It is just sad to see such a great weapon being so under used.

===

Using all 3 data points. I can't say much on on everything as a whole now and need more time thinking it over. VS still looks to have the best AI MAX but without having time to think things over and letting it sink in I can't give a good idea yet on how good they are. Id need to see a lot more of these things to get a full picture of MAXes as a class.

Can't wait to get my hands on this gold mine myself.

Don't forget to count me in as well . I bet if you took my stats out those #'s will look at least a few hundred(s) if not thousand(s) kills worse for mattocks .

KesTro · 2013-08-09, 02:40 PM

I suppose those roflpod stats shouldn't be too surprising given how much the TR loves their airzerg. Interesting none the less. As you've already stated it'll be great to see these if and when you've had more time to collect data.

maradine · 2013-08-09, 03:02 PM

Originally Posted by KesTro

I suppose those roflpod stats shouldn't be too surprising given how much the TR loves their airzerg. Interesting none the less. As you've already stated it'll be great to see these if and when you've had more time to collect data.

I read that as mostly the result of the current Striker situation, and not necessarily platform superiority. It will be difficult to make much of any air-related statistics until AA ubiquity is even across empires - assuming that's what they're tuning towards.

maradine · 2013-08-09, 04:13 PM

I present: the magic of LMGs. A bit of a mess as a text extract so I dumped it into excel for your sorting pleasure. My take away: damn, I am so using the wrong guns.

typhaon · 2013-08-09, 04:44 PM

And.... scene.

* I think on the LMG data you're missing the LA1 Anchor for NC and have some NS gun, instead.

maradine · 2013-08-09, 06:16 PM

Looks like I just flat out didn't record it right when I transferred to excel. Fixing.

Fixed. Also slathered some commas around in there.

fierce deity · 2013-08-09, 06:56 PM

Originally Posted by maradine

If you're looking for something to start a thread over, here's a little bit more of a statistical anomaly:

Code:

Harasser AI Weapons - by kpu, NS weapons aggregated

id 6122 P525 Marauder-H
31,626 Kills
2,190 Unique Killers
14.4 kpu

id 6115, 6116, 6117 M40 Fury (Combined)
20,476 Kills
1,964 Unique Killers
10.46 kpu

id 6125 Proton II PPA-H
3,990 Kills
666 Unique Killers
5.99 kpu

id 6100, 6101, 6102 M12 Kobalt (Combined)
1,399 Kills
346 Unique Killers
4.04 kpu

id 6124 Enforcer C85 Modified-H
317 Kills
92 Unique Killers
3.44 kpu

id 6112, 6113, 6114 M60-G Bulldog (Combined)
2,602 Kills
780 Unique Killers
3.33 kpu

id 6103, 6104, 6105 M20 Basilisk (Combined)
7,470 Kills
4,358 Unique Killers
1.71 kpu

Interesting info here. I'm surprised that anyone actualy uses the Enforcer C-85 Modified-H and gets any kills with it. I've never even seen one outside of VR. Then I can't believe how much the Marauder is outperforming everything else, even all 3 factions usage of the Fury. Even if you exclude vehicle kills since most of the other AI weapons are awful at AV, it would probably still be outperforming everything.

Originally Posted by maradine

New Empire-specific Pistols

id 7389 LA3 Desperado
3,804 Kills
682 Unique Killers
5.58 kpu

id 7413 Cerberus
1,306 Kills
253 Unique Killers
5.16 kpu

id 7401 TS2 Inquisitor
1,665 Kills
352 Unique Killers
4.73 kpu

Seems like the Desparado is by far the best, and the Cerberus and the Inquisitor to be roughly even. Sort of what I expected, but I thought the Inquisitor would have higher numbers than this.

Originally Posted by maradine

Code:

Really Big Sniper Rifles by kpu

id 26004 Parallax VX3
51,527 Kills
2,929 Unique Killers
17.59 kpu

id 25004 RAMS .50M
59,260 Kills
3,536 Unique Killers
16.76 kpu

id 24004 EM4 Longshot
47,515 Kills
2,867 Unique Killers
16.57 kpu

This just feels right. I get sniped by the RAMS much more than the Longshot. Maybe its because the NC & VS have the more long range options than the TR, so the TR are more inclined to use snipers at range than the NC & VS. Just wild speculation.

Selerox · 2013-08-09, 07:02 PM

I'm still amazed by the lack of people who use the SVA-88. It's just better than anything else VS has.

It's virtually the same up close as the Orion, but with a 50% bigger mag and far better attachment options. It's also got pretty much the same reload as well (0.02 seconds slower for a short reload, which is nothing).

I genuinely do not understand by anyone would use anything else. I can understand using the Orion if you're a CQC specialist, or the Ursa for long range work. But the fact is that the SVA-88 is almost as good at both of those in their specialist areas and is massively better in all other areas.

maradine · 2013-08-09, 07:41 PM

Originally Posted by fierce deity

Seems like the Desparado is by far the best, and the Cerberus and the Inquisitor to be roughly even. Sort of what I expected, but I thought the Inquisitor would have higher numbers than this.

Raw kills are interesting, but don't read too much into them - they tend to reflect a bunch of secondary factors, chief of which is popularity, which itself is influenced by the universe of other weapon choices in the slot. There are twice as many Desperado users as either of the others.

I read this as a minor performance advantage. I haven't decided where I draw the line on kpu parity for purposes of this exercise, but I'm personally only raising eyebrows when the top and bottom are more than 20% from each other.

Falcon_br · 2013-08-09, 08:08 PM

Originally Posted by maradine

Raw kills are interesting, but don't read too much into them - they tend to reflect a bunch of secondary factors, chief of which is popularity, which itself is influenced by the universe of other weapon choices in the slot. There are twice as many Desperado users as either of the others.

I read this as a minor performance advantage. I haven't decided where I draw the line on kpu parity for purposes of this exercise, but I'm personally only raising eyebrows when the top and bottom are more than 20% from each other.

All reviews on the Internet says the same:
Desperado it the new tx-1 the perfect pistol for close range! Maybe the nc will stop crying about the tx-1.
That's why the desperado is selling good, is the only new pistol that the reviewers are saying it is good to get it.
The other two are bad in comparison with the pistols the TR and VS already have. No point in using then!

This is the power the reviews have on this game.

fierce deity · 2013-08-09, 10:27 PM

Originally Posted by maradine

Raw kills are interesting, but don't read too much into them - they tend to reflect a bunch of secondary factors, chief of which is popularity, which itself is influenced by the universe of other weapon choices in the slot. There are twice as many Desperado users as either of the others.

I read this as a minor performance advantage. I haven't decided where I draw the line on kpu parity for purposes of this exercise, but I'm personally only raising eyebrows when the top and bottom are more than 20% from each other.

I understand that, Its completely reasonable that it got double the kills since it has double the users. I guess I worded my statement too strongly. By kpu they are balanced reasonably well, the desparado is better on that metric than the cerbeus by 16%, and the guns are spaced out in kpu by 8%. So the the desperado is above average and the cerberus is below average. That is perfectly fine by me, and I think it shows that the Cerberus and the Inquisitor did not deserve as much bad press as they recieved.

For snipers it really is a dead heat, with the TR ever so slightly leading the pack. Me getting sniped by TR a lot more than NC is just my personal experience.

Goldoche · 2013-08-09, 11:29 PM

Is there a way to only look up stats for BRs 50+ ? Players still learning the game tend to have less score on balanced yet harder to use weapons. Also, easy to use weapons that are balanced tend to have a disproportionate score because noobs can use them effectively.

maradine · 2013-08-10, 12:52 AM

Yes, but at a significant time and labor cost. Each kill would have have to be cross-referenced with an API pull to SOE to get the BR. Now that they provide joins, this might be possible with less calls, but no guarantees. I'll put it on the list of further research projects. What would be nice is direct query access to their SQL cluster, or the ability to run map-reduce jobs against their hive. I actually don't know what their back end looks like. It also starts looking suspiciously like work at that point.

Fulmir · 2013-08-10, 05:27 AM

Any chance this whole thing could come with a disclaimer about abuse of stats?

The Vortex/Raven/Fracture thing is a pretty good example, if you look at the PS2-stats data on those weapons it's pretty clear that the Vortex isn't used very much. If you look at the vehicle kills per use-time though it beats out both the Fracture and the Raven.

This would suggest that the issue is less with the Vortex and more with people's opinion of it.

EvilNinjadude · 2013-08-10, 05:46 AM

I say I say, this IS an interesting thread. Most of the good stuff has been covered so far.

It's not gonna be for balance purposes but for fun: O Great Oracle of Death, show me which faction stabs the most people on an average day.
(Give separate data MAX slaps for xtra kudos)