Notes on new team difficulty settings
By
Abigail Frost,
Posted on: 2009-08-26 15:02 in The Market and Inventions
Did Cimerora random missions today.
Settings: +0/6/yes/no
Character: lvl 50 wp/ss tank
As you can see I had a dismal drop rate. I did nothing fancy except running various random missions, some blinkies, some defeat boss etc. No fighting outside in-between missions. Note that salvage drops are spot-on while recipies are... sigh.
----------------------------------
Archie Gremlin's Drop Stats v0.3.1
http://www.glasspaw.com/dropstats/
----------------------------------
Type DropRate Drops Expected Mobs
--------------------------------------------------------
Pool A 0.71% 10 27-51 1411 minion equivalents
Salvage 7.65% 97 84-122 1269 minion equivalents
--------------------------------------------------------
(Table excludes mobs of unknown rank and their drops.)
(1 in 20 runs will be outside the expected range.)
961 rewarding mobs defeated
110 bosses and elite bosses
233 lieutenants
615 minions
3 rewarding mobs of unknown rank
Lucius Lucilius (1)
Soldier of Romulus (1)
Vibius Calpurnius (1)
986 inf/debt rewards
5633689 inf
11 Pool A recipes (Mob Defeats)
2 uncommon and rare recipes
Blood Mandate: Acc/End (Recipe)
Kinetic Crash : Knockback/Dam/Endurance (Recipe)
9 common recipes
97 salvage
3 rare
18 uncommon
76 common
11 single origin enhancements
96 inspirations (7 tier 3, 35 tier 2, 54 tier 1, 0 special)
0 other drops
Member of Top Ten since 2005
Arc 98079: Hay Fever hits Heroes Hard
Arc 53357: Family Business is not Men-only
I noticed the change on my 1st mish after fixing my power colors.
Will download parser and deliver data.
Just a quick observation, but it seems, outside of a couple of people, everyone is testing drop rates with their diff set at +8 team. For completeness' sake, are we making sure that the other settings are getting looked at as well? It may or may not make a difference, but it might also point Synapse in the right direction if certain settings were giving odd numbers and others weren't.
Loose --> not tight.
Lose --> Did not win, misplace, cannot find, subtract.
One extra 'o' makes a big difference.
Quote:
This is what he said when he posted his data in this thread. He only sent me data for the first five, which I assume were on the internal server, since he said something about then heading to Test to verify what he'd seen.|
Originally Posted by Canis_Lupus
OK.... so, Synapse's testing was all done on either Training room or an internal only server, is this correct?
|
As far as I'm concerned, his having done some portion of his drop analysis on the Test server doesn't invalidate anything; even if he used [The I Win Button] to clear the last 10 maps, the fact that his results were so incredibly consistent (and above expected values) that it looks like Internal-Test is a reasonable translation. As to Test-Live, my piddly drop rates began on Test when I started in the Beta, and carried over pretty exactly to Live. If he'd only done his testing on the internal server I'd see that as a meaningful criticism of his testing, but he didn't.
That being said, it can't hurt to replicate on Live, and I really hope he takes someone up on the opportunity to run some maps with them. Even if he's there as a non-teamed police drone, it'll give him a chance to observe first-hand what the drops are like. We know the drops generated for the entire team were once put into their own channel in the chat tab, so that datastream still ought to exist.
My postings to this forum are not to be used as data in any research study without my express written consent.
The one thing that I do wonder about with the Test vs Live vs Internal testing is whether there could be any sort of server load factor causing a problem. If a large amount of processing on the server side is having an effect on this, the question arises, would we still see the same results on test as were observed during Beta?
Outside of that, I agree, I can't see how his running on test would be a problem, given that the testing/data collection that you've done started there.
Quote:
It had crossed my mind that one of the elements of the RNG seeding might have to do with the server clock and the creation of the mission instance, but given how consistent my findings have been on specific maps, I tentatively rejected that idea. The different clocks, and anything else that might be different across servers when maps were generated, were things I thought about, but the fact that I can still get back on test and get the same drop rates I got all the way through beta, and still get on live, leads me to tend to reject that - though I'll admit that the motivation to run a lot on test is kind of low for me right now. |
Originally Posted by Canis_Lupus
The one thing that I do wonder about with the Test vs Live vs Internal testing is whether there could be any sort of server load factor causing a problem.
|
My postings to this forum are not to be used as data in any research study without my express written consent.
I am sticking with my zoning bug idea lol.
Lots of stories about crappy drops while running in coop zones. Someone try logging a character out for the day(long enough for the server or whatever to "reset" the toon, I dont recall how long it takes for that) in a zone with no coop, no entry to a coop, no wentworths, etc. Inde port? then tomorrow log in, dont go anywhere, and try some radio missions. no zoning, just loading into the mission. no pocket D, no SG base tp's, etc. just log in, radio, mission.
the control for that would be to have a toon logged off in like pocket d, or wherever, then zone through pocket d, or a SG base, and then go to inde port and run radio missions.
there are a lot of areas that have different interaction with your character when you zone. WW is cross server, arenas and pvp zones and coop zones, AE, midnighters, pocket D. Inde port is the only place I can think of that has pretty well none of those, and it still has an AE. Maybe the test needs to be done in a hazard zone like creys or dark astoria with street sweeping. best bet would be to have any characters testing be 100% free of any missions, so the diffy settings have a clean slate/have no effect on the test.
If a character that logs directly into a zone after the "reset" period is shown to have a normal drop rate, and a different character that traveled to the zone has a crappy one, then the process of elimination could begin as to what zone or combo etc would cause it.
provided of course that I am not totally grasping at straws. the reports given here of complete lack of phat l00tz bothers me, but at the same time I can not bring myself to spend my few hours a week I play to test these. My new scrapper calls my name when I log in lol.
Liberty server
Eldagore lvl 50 Inv/ss, co-founder of The Legion of Smash
3.5 servers of alts....I need help.
May the rawk be with you.
Arc #'s
107020 Uberbots!
93496 A Pawn in Time
So, if the drops are granted based on a series of rolls per kill, would it be possible to have the numbers visible, a la ToHit?
This is slightly reminiscent of the situation that necessitated the addition of the "streakbreaker" factor to combat rolling. Ugh...I remember logging data and actually having a night's average sit at a value of something like only 52% of my ToHit-capped attacks successfully striking. Due to this, I've never been a huge fan of the RNG used by this game since, in just a quick&dirty and anecdotal observation, it seems like mobs have a tendency to "roll low" and PCs to "roll high", ignoring whether or not the attacks are successful or not (I don't exclude perception but I have tanked Recluse on the STF too many times to not notice his love of averaging between 15 and 25 on his rolls).
'Course, whether or not that is relevant. But hey.
Blue: ~Knockback Squad on Guardian~
Red: ~Undoing of Virtue on [3 guesses]~
Quote:
That's interesting because I've actually noticed an improvement in performance since the release of I16. I don't know if I'm playing when my roommates aren't also trying to use up some of the bandwidth from our wireless router or if it's the game servers, but I've been chewing through things quite easily and have only had a small differential in my drops as far as what I think should be expected. In other words, I am only slightly off what the drop rate should be on recipes. Salvage drops are right within the targeted range, but the recipe drops still don't seem to be flowing like they should. I could care less about enhancements as they don't make any money on the market.|
Originally Posted by WoefulKnight
With the other complaints of additional lag I wonder if maybe it isn't communications related.
|
Where I'm still getting salvage drops as expected, I am utterly befuddled as to why recipes seem to be lacking. I wonder if it has something to do with the weighting formula...but those are only for reward rolls at the merit/ticket vendors........
/head scratch
Quote:
|
Originally Posted by Captain_Denmark
As you can see I had a dismal drop rate. I did nothing fancy except running various random missions, some blinkies, some defeat boss etc. No fighting outside in-between missions. Note that salvage drops are spot-on while recipies are... sigh.
|
Quote:
OK, just ran my Save Carnival Victime's mission again this evening. Large map, set at +0 (lvl 50), for 5 people, no bosses. Had a total of 2 recipe drops, 1 accuracy, 1 Thunderstrike (yellow), 50+ salvage, plenty of inspirations. But 2 recipe drops?|
Originally Posted by Synapse
Thought I would chime in here and let you know what we've done to test this. The data has been checked, the code has been checked. There has been no change to either. That said, none of us would be satisfied until we tested this for ourselves. I knew this issue would haunt my dreams until I dedicated some time to testing. So, I went and did some pretty comprehensive testing today. Both internally and on the training room.
The results are pretty consistent and drop rates appear to be correct. The way I tested this is load up the first mission given out by Abyss (as a number of players mentioned this in their tests). I then set my difficulty to 8x and +0. I then went through this mission 5 times internally and 10 times on the training room and the results were very consistent: -14 to 19 recipes per mission -5 to 8 enhancements per mission -30 to 51 salvage per mission So, for the time being I am calling this a perception issue, that said I will keep an eye on this. If there is in fact a bug at work here then there will be reproducible steps one could take to get this bug to happen to anyone. If anyone does in fact nail down possible repro steps for this bug, please PM me or post them in the forums. I know this isn't the answer many of you were hoping to hear, but I will keep an eye on the issue. Again, please contact me via PM if you have any possible reproducible steps. Lastly, I want to clear up any rumors of a "stealth nerf" or drops being different per map. Drop rates are on a per class basis. All minions, lieutenants and bosses have the same drop rates as all other minions, lieutenants and bosses regardless of what map they're on. The only difference is what items they drop (magic salvage vs. tech salvage, different origin enhancements and such) Regards, Synapse |
Before issue 16 I used to use this map filled with 3 or 4 people and would have to delete recipes, salvage, etc. Sorry Synapse, this has been my experience with issue 16 since Beta and still not having any luck. I have yet to fill my recipe tray with any mission set for multiple players when I used to fill it with 4 fillers each time.
Something definitely doesnt seem to be working at least for me.
I'm running +3/+4heroes setting and my drops have been wanting. Alot of salvage drops (i filled up my salvage in five missions on Sunday) but only three recipes drops. And all three were common IOs.
Other code has changed, it is perhaps interfering with the rewards related delivery, recipe only, in an unanticipated way. There also may be some interaction between the live database, the server environment there, and a host of other factors that are not the drop code itself. But the end result is toons aren't getting rewards
ooo..nasty thought, if BIG IF, but if they've added some new recipes to the game but haven't activated them yet, couldn't that cause issues?
The more people I meet, the more I'm beginning to root for the zombies.
Updated composite stats. As before, the Expedient Reinforcement isn't recognized properly because it doesn't have (Recipe) at the end of the name. But both it and the Mako's Bite are Pool C drops from a boss, and so shouldn't be counted. That gives me a total of 9 Pool A Recipes for 2295 minion equivalents, or 0.39%. That's about 1/7 of the expected rate.
For all of you people commenting on the test server, all of this testing was done there. It's far better than the drops I was getting on live under slightly different circumstances (1 recipe in 6000 mobs). So hopefully Synapse can reproduce it.
"Defeat all Clockwork at factory" from Unai Kemen, gotten from him directly (no Flashback), solo, difficulty +1/8/yes/no, level 50 Stone/Stone Tanker, mobs are level 51, most spawns have 2 bosses, but some are all Minions and Lts.
Code:
----------------------------------
Archie Gremlin's Drop Stats v0.3.1
http://www.glasspaw.com/dropstats/
----------------------------------
Type DropRate Drops Expected Mobs
--------------------------------------------------------
Pool A 0.44% 10 48-78 2295 minion equivalents
Salvage 8.11% 166 141-189 2048 minion equivalents
--------------------------------------------------------
(Table excludes mobs of unknown rank and their drops.)
(1 in 20 runs will be outside the expected range.)
1822 rewarding mobs defeated
165 bosses and elite bosses
400 lieutenants
1000 minions
257 underlings (XP only)
1822 inf/debt rewards
10898152 inf
44000 debt
10 Pool A recipes (Mob Defeats)
3 uncommon and rare recipes
Blood Mandate: Dam/End (Recipe)
Enfeebled Operation: Acc/End (Recipe)
Mako's Bite: Acc/End/Rech (Recipe)
7 common recipes
166 salvage
3 rare
27 uncommon
136 common
16 single origin enhancements
546 inspirations (14 tier 3, 177 tier 2, 355 tier 1, 0 special)
1 other drops
Expedient Reinforcement: Resist Bonus Aura for Pets
Goodbye and thanks for all the fish.
I've moved on to Diablo 3, TopDoc-1304
Just ran 2 oro arcs, both with AT only powers, Hatreds Hungry Heart, and The Organ Grinders. Got a total of 7 recipes. 3 were mish comnpletes. 4 set pieces and 3 Commons, so it's affecting Oro too.
The more people I meet, the more I'm beginning to root for the zombies.
Okay, the sampling is rather small, but these numbers actually look correct. I have tried to be as detailed as possible with my records. Please let me know if anything is lacking:
Code:
Diff: +0/8/yes/no
Char: Level 50 Kat/WP Scrapper
Mission 1: Server: Freedom Zone/Mission: SmoothCaves_30_Layout_08 - Rikti
Type DropRate Drops Expected Mobs
--------------------------------------------------------
Pool A 2.78% 3 1-7 108 minion equivalents
Salvage 4.95% 5 4-15 101 minion equivalents
--------------------------------------------------------
(Table excludes mobs of unknown rank and their drops.)
(1 in 20 runs will be outside the expected range.)
99 rewarding mobs defeated
13 bosses and elite bosses
13 lieutenants
43 minions
Mission 2: Server: Freedom Zone/Mission: V_Abandoned_Office_30_Layout_05 (8:16)
Type DropRate Drops Expected Mobs
--------------------------------------------------------
Pool A 1.71% 2 1-9 117 minion equivalents
Salvage 10.62% 12 5-16 113 minion equivalents
--------------------------------------------------------
(Table excludes mobs of unknown rank and their drops.)
(1 in 20 runs will be outside the expected range.)
117 rewarding mobs defeated
16 bosses and elite bosses
9 lieutenants
51 minions
Mission 3: Server: Freedom Zone/Mission: V_Caves_30_Layout_02_01 (8:35)
Type DropRate Drops Expected Mobs
--------------------------------------------------------
Pool A 2.74% 2 0-5 73 minion equivalents
Salvage 7.14% 5 2-11 70 minion equivalents
--------------------------------------------------------
(Table excludes mobs of unknown rank and their drops.)
(1 in 20 runs will be outside the expected range.)
67 rewarding mobs defeated
8 bosses and elite bosses
6 lieutenants
37 minions
Mission 4: Server: Freedom Zone/Mission: SmoothCaves_30_Layout_01 (8:50)
Type DropRate Drops Expected Mobs
--------------------------------------------------------
Pool A 2.82% 2 0-5 71 minion equivalents
Salvage 6.09% 4 2-11 66 minion equivalents
--------------------------------------------------------
(Table excludes mobs of unknown rank and their drops.)
(1 in 20 runs will be outside the expected range.)
59 rewarding mobs defeated
6 bosses and elite bosses
9 lieutenants
35 minions
Totals:
Type DropRate Drops Expected Mobs
--------------------------------------------------------
Pool A 2.44% 9 5-17 369 minion equivalents
Salvage 7.44% 26 19-39 350 minion equivalents
--------------------------------------------------------
(Table excludes mobs of unknown rank and their drops.)
(1 in 20 runs will be outside the expected range.)
342 rewarding mobs defeated
43 bosses and elite bosses
37 lieutenants
166 minions
28 underlings (XP only)
68 rewarding mobs of unknown rank
Chief Mentalist (25) failed heuristic rank check
Guardian (43) failed heuristic rank check
0 pets and 40 greys
Quote:
It looks like those numbers are fine. I've been seeing problems at +1/8/yes/no, so maybe you could try that. If the only difference is the mob level, perhaps that'll tell Synapse something.
|
Originally Posted by AgentMountaineer
Diff: +0/8/yes/no
|
Goodbye and thanks for all the fish.
I've moved on to Diablo 3, TopDoc-1304
Quote:
The other thing that makes this problem insidious is that Synapse himself demonstrated that drops can be above average, a number of us have demonstrated that in specific testing situations (specific characters on specific maps - or, in BillZ's case, pretty much everywhere) the drop rates are abysmally low - and sometimes, everything will work as advertised. |
Originally Posted by TopDoc
It looks like those numbers are fine. I've been seeing problems at +1/8/yes/no, so maybe you could try that. If the only difference is the mob level, perhaps that'll tell Synapse something.
|
The thing is, I don't think the overall distribution of drops is normal right now. I suspect that it's rectangular, which would still keep the mean and median right where they're supposed to be.
My postings to this forum are not to be used as data in any research study without my express written consent.
Is anyone else testing with paper/scanner missions with a level 50?
If this is a case of the universe deciding to shove probability out the window just for me and just for recipe drops then I should start buying lottery tickets for the karmic payback.
I have to reiterate: My drops have sucked since I-16 hit beta with my level 50 scrapper running scanner missions in PI. How many weeks have passed in that period?
The odds of me maintaining such low drops rates purely due to some statistical aberration in the random number generator over this much time approaches nil.
Be well, people of CoH.
Quote:
Reasonably so. I think many feel the same way...|
Originally Posted by Bill Z Bubba
The odds of me maintaining such low drops rates purely due to some statistical aberration in the random number generator over this much time approaches nil.
|
=!+.+!=
Ignoring anyone is a mistake. You might miss something viral to your cause.
Quote:
I took Smurphy up on his offer and ran Liberate TV with 5 real fillers for two runs (I bore easily). SRSLY Settings: +0x0/no/no|
Originally Posted by Smurphy
I haven't personally tested the problem and I'll admit I haven't bothered look into it in too much detail but...
... if you had the difficulty set to +0/x0 and 5 real players on your team standing outside (or 3 real players if you want a 4 man spawn) would, in theory, the drop rate act as normal? I ask because I can provide whoever wants the test the fillers necessary to do this test. |
Code:
---------------------------------- Archie Gremlin's Drop Stats v0.3.1 http://www.glasspaw.com/dropstats/ ---------------------------------- Type DropRate Drops Expected Mobs -------------------------------------------------------- Pool A 1.85% 17 16-35 917 minion equivalents Salvage 7.11% 56 49-79 787 minion equivalents -------------------------------------------------------- (Table excludes mobs of unknown rank and their drops.) (1 in 20 runs will be outside the expected range.) 727 rewarding mobs defeated 1 bosses and elite bosses 194 lieutenants 526 minions 6 rewarding mobs of unknown rank Modified Fake Nemesis (6) 0 pets and 2 greys 727 inf/debt rewards 2520178 inf 44000 debt 17 Pool A recipes (Mob Defeats) 1 uncommon and rare recipes Nightmare: End/Fear (Recipe) 16 common recipes 58 salvage 0 rare 9 uncommon 49 common 1 single origin enhancements 134 inspirations (1 tier 3, 50 tier 2, 83 tier 1, 0 special) 0 other drops
edit: NM
Ignoring anyone is a mistake. You might miss something viral to your cause.
Quote:
What difficulty level are you running them on Bill? I'll run a few tonight after work.
|
Originally Posted by Bill Z Bubba
Is anyone else testing with paper/scanner missions with a level 50?
|
Quote:
I have been testing with scanners in PI as well as Oro. All set to +0/8 Bosses no AV's. I said I was done testing but since I run Herostats I'll post last night's 1.5 hour radio mission runs below. |
Originally Posted by Bill Z Bubba
Is anyone else testing with paper/scanner missions with a level 50?
If this is a case of the universe deciding to shove probability out the window just for me and just for recipe drops then I should start buying lottery tickets for the karmic payback. I have to reiterate: My drops have sucked since I-16 hit beta with my level 50 scrapper running scanner missions in PI. How many weeks have passed in that period? The odds of me maintaining such low drops rates purely due to some statistical aberration in the random number generator over this much time approaches nil. |
635 Enemies defeated
47 Salvage
59 Inspriations
9 Recipes (Including Rope-a-dope: Acc/Stun and Thunderstrike: Acc/Dam)
Do players that solo 8 man spawns deserve the drops equivalent to an 8 man team?
Who do I have to *&^% around here to get more Targeted AoE recipes added?
Arc Name: Tsoo In Love
Arc ID: 413575
Quote:
If they do the work, why shouldn't they get the reward?
|
Originally Posted by EnigmaBlack
I
Do players that solo 8 man spawns deserve the drops equivalent to an 8 man team? |
With the other complaints of additional lag I wonder if maybe it isn't communications related. In that perhaps the Live Server communication net has no check to ensure that the reward processed to the client. Where if you had yellow or red lines on your netgraph you lost the critical packet saying you got your reward. Of course this wouldn't explain the only getting common recipes symptom. Perhaps the RNG result for the second roll (after recipe RNG result is yes) going to the Uncommon/Rare table is being lost in communications.
In the Test and Internal Server environment these communication networks would be relatively closed circuits but the Live environment is not so limited and much more subject to volume issues.
Protector Server
Woeful Knight (BS/Regen/Body Scrapper)
Kevin Christian (MC/FF/Primal Controller)
SilverCybernaut (Eng/Dev/Munitions Blaster)
Apixie OhNo (Fire/Fire/Pyre Tanker)
Y'ru Glowen (Rad/Rad/Psy Defender)