Captain
Join Date: Nov 2012
Posts: 3,672
# 21
01-24-2013, 11:04 PM
I don't think Cryptic has a parser. I would rather hear from Mancom and see what he thinks. Cause there being an issue with the log or the way the parser is working with it does seem like one possible reason we are seeing odd numbers with FAW.

The STO combat log is far from perfect... perhaps something is up withe the way the parser is sorting the data from it.
When the messenger comes to appropriate your profits ... kill the messenger.
Career Officer
Join Date: Dec 2012
Posts: 679
# 22
01-24-2013, 11:35 PM
I'll do some tests tonight once I'm back from work.

Food for thought for Borticus: Is FAW currently using the beam or the cannon accuracy formula? (http://hilbertguide.com/blog/#2012-08-18)

Oh, and another thing: As I expressed in all my bug report threads, it could also be that FAW is not ignoring gear accuracy, but only using a wrong multiplier, so that a +10% [Acc] counts only for +1% or +0.1%. (We had wrong multipliers on the Plasma-Disruptor Hybrids, so it's not such a wild idea.) If this were true, it would explain why the +1000 Acc gear tests gave perfect accuracy. At this level, it doesn't matter if the multiplier was wrong.

(If it is a wrong multiplier, then it is impossible to determine with certainty for the players without access to dev tools, because a +1% change in accuracy disappears in the background noise of the random number generator and can't be reliably concluded from test data in the limited acc/def range we have at our disposal.)
http://hilbertguide.com

Last edited by mancom; 01-25-2013 at 12:35 AM.
Captain
Join Date: Aug 2012
Posts: 984
# 23
01-25-2013, 12:32 AM
Barely related but a thread would probably be overkill.

The devs should add phaser beam detonation like we see in balance of terror. It obviously couldn't be used like depth charges as shown (unless a cloaked ship was near your target) but it would be a cool weapon against carriers and difficult to hit escorts. It could do more focused damage than faw but less singular damage than beam overload or something.

Good luck with faw and thanks for the updates.
The Somraw, K'tinga, D'Kyr, D7, Kumari, Xindi carrier, Xindi escort, and the T'Varo are all older than the Constitution Refit and yet they are tier 5. The Constellation is made up primarily of Connie refit parts and it is tier 5, there is no logical reason whatsoever for the no tier 5/6 connie rule.
Empire Veteran
Join Date: Jun 2012
Posts: 6,671
# 24
01-25-2013, 01:33 AM
well i parsed some random pug matches and got some data. no surprises lol, it had terrible accuracy and misses everything equally. it does not discriminate between spam and player

pug matchs with FAW3 at global thanks to tech doffs, activated as soon as it came off cooldown every time, on a fleet excelsior. it had 28.4 bonus accuracy thank to a vet boff, omega deflector, and accurate trait. test conducted with 8x [Tetryon Beam Array Mk XI [Acc]x2 [Dmg]]

https://i.minus.com/i1odg7MuTJzmX.JPG
https://i.minus.com/ibvaBSw5HTDemE.JPG
https://i.minus.com/i0mIRLaxlvNbo.JPG

these were all hectic matches full of targets at variable speed and plenty of spam with high evasion.

https://i2.minus.com/idS4Pa8eLATd1.JPG

this was not. this was a duel with 1 person, no spam. we were both moving at normal speed the whole time and he said his defense score was 73, quite high. FAW has its worst showing yet here.


so FAW is an ok garbage man skill. sure, it will hit spam most of the time and clear it, but it turns off any illusion of damage dealing wile its on. even in the duel with FAW on i could tell he was taking less damage then he was when my beams were just hitting him unbuffed.

beams already have a large damage deficiency vs cannons, and cannons have CRF that increases their rate of fire and damage dealing significantly on top of it. i can do a ton of damage in pvp with kdf cruisers, they are very effective tactical platforms. fed cruisers with their terrible turn rate and loathsome beam arrays cant even deal effective pressure damage anymore. beams need a rapid fire skill too, or they need to have a FAW like fire rate all the time, and faw can just be random target hit mode.

Last edited by dontdrunkimshoot; 01-25-2013 at 02:16 AM.
Captain
Join Date: Jan 2013
Posts: 3,144
# 25
01-25-2013, 02:11 AM
I don't think there actually is a Cryptic combat parser, but Bort's test seems to be rather dramatic: With enormous values like that, if the modifier were even slightly wrong, you'd get a massive gap between the defense and attack values, and you'd still thus have a misstastic lightshow.

It therefore seems likely that one of two things is occurring:
1. The combat parser in use is not correctly interpreting the data.
2. FAW fires upon bad targets in combat usage, thus resulting in higher levels of misses because those targets are not hittable to begin with.

Since I can't see any obvious flaw in Bort's testing methodology, and yet these two dramatically differing observations exist, there must be some missing link that explains this discrepancy, and one or both of those two seems to cover it.
Career Officer
Join Date: Jun 2012
Posts: 880
# 26
01-25-2013, 05:17 AM
Quote:
Originally Posted by doffingcomrade View Post
1. The combat parser in use is not correctly interpreting the data.
2. FAW fires upon bad targets in combat usage, thus resulting in higher levels of misses because those targets are not hittable to begin with.
Empirical evidence from 1v1 private matches suggests that there is more to it though.

I would put forward two other possibilities:
  1. FAW calculates hits in a strange way -- two examples that have been brought up include the possibility that it is using hit calculations for cannons instead of beams, or that it's faster firing rate (and/or other unique properties) is having unintended effects on the algorithm that determines whether an attack hits.
  2. Borticus' test all but ruled the problem being a simple multiplier as such a modification would have been immediately noticeable after adding 1000 defense and ACC, but didn't rule out that the base accuracy isn't somehow lower. For example, if FAW suffers from an ACC penalty of 20, the results will look the same whether you're comparing 20 to 40 or 980 to 1000.
Starfleet Veteran
Join Date: Jun 2012
Posts: 295
# 27
01-25-2013, 05:34 AM
I would like to ask Borticus to try the test a different way. Adding +1,000 Accuracy in a magical dev way vs using FAW in the 'real world' could be causing the difference. Wasn't the main issue that the acc mod on the weapons themselves were not being recognized or that they were being misinterpreted somehow? An overall/general boost to acc that bypasses the weapons may be missing the whole problem, neh? And even if the test focuses on wep mods only, if the problem on the weapon mod is off by only .2 (just an example), boosting acc to +1,000 would kinda drown out and make differences difficult to discern.

I don't claim to be experts like many of you, but it doesn't seem like the test was an apples to apples one.

But I do appreciate a test being done. TY Borticus. but please don't stop here.

Last edited by sonulinu2; 01-25-2013 at 05:37 AM.
Career Officer
Join Date: Jun 2012
Posts: 1,348
# 28
01-25-2013, 06:04 AM
Quote:
Originally Posted by borticuscryptic View Post
This may be a side effect of each pulse hitting two targets?

A regular beam fires 4 times in a cycle, hitting one target between 1 and 4 times.
A fire-at-will firest 5 times in a cycle, hitting up to two targets between 1 and 5 times each.

Is it possible that whatever parsing tool is being used is counting the misses on Fire at Will, but not counting the correct number of potential activations, therefore driving the Hit:Miss ratio off-center?
Given that information you aren't forced to only account for hits and misses. But sure the guys could try. You could also just check to see if FAW is adding as much damage as it should be.

We know what normal phaser fire does. We know how many activations FAW will trigger given a known amount of targets. Check to see if the incremental damage dealt by FAW matches those expectations. If it does, it could just be a reporting issue. If it doesn't, it would be a FAW issue.

If you're having difficulty with a metric, and you think the metric itself may be the issue, try a different one.
If I don't respond to posts on this forum don't be offended. I don't sub or follow any of them.
Career Officer
Join Date: Jun 2012
Posts: 1,348
# 29
01-25-2013, 06:12 AM
The hit/miss is based off of the difference between the two values. So the first test would have had a difference of 1000. Presumably pushing the hit rate to its lowest. The second test would have had a difference of 0, putting the hit rate at its highest. So. Say acc mods of any type were not being accounted for. That was the premise. The second test would have had virtually the same results as the first and Bort wouldn't have needed to do any heavy lifting on the percentages. Issue resolved. If there WAS some deviation from expectations (never missing) but it wasn't catastrophic then we would have a measurement of a partial change in the acc or def mods or even base values.
If I don't respond to posts on this forum don't be offended. I don't sub or follow any of them.
Captain
Join Date: Jun 2012
Posts: 1,015
# 30
01-25-2013, 06:36 AM
Quote:
Originally Posted by borticuscryptic View Post
Good News
Here's how I was testing:

1) Give an NPC a +1000 Defense buff.
So maybe try shooting at 10 NPCs? People are reporting poor accuracy, but I'm betting they're using FAW against, you know, multiple targets...

Maybe there's something being introduced when there's more than one. Like it starts using the wrong formula for later targets.
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off


All times are GMT -7. The time now is 12:29 AM.