18.4.2 Update Release Notes

[drops]

That's either randomness, you doing something wrong or you just made that up. Not only those numbers look too "clean", but I think we (as is us, the ones that spend our game time on live) would notice such a high (come on, 33%?) miss rate at full def.

[Xtremtech]

Lol... like the basic games mechanics will change because you're on live...

What actually changes is that you don't really spend your time looking at hit/miss rates... nor counting back through chat logs to establish the exact data. All of which we do on test - because that's what we're there for. The data point may be an unusual one... I just remembered seeing it today and so thought it might be of interest.

All of these are cut and paste directly from the feedback threads in the Test Forum and were carried out specifically on the request of one of the Devs. I'll be doing a load more in the coming days.

Weapon Hitrates:

220 MP using Asp of Titaniush with AR 2154, Full Agg:

Vs 230 Technician: 56 hits (1 Crit), 00 misses = 100% Hit Rate
Vs 231 Technician: 65 hits (2 Crit), 32 misses = 67% Hit Rate
Vs 232 Technician: 58 hits (3 Crit), 33 misses = 64% Hit Rate

220 MP using Asp of Titaniush with AR 2154, Full Def:

Vs 228 Technician: 58 hits (5 Crit), 30 misses = 69% Hit Rate
Vs 229 Technician: 69 hits (1 Crit), 33 misses = 67.6% Hit Rate
Vs 230 Technician: 58 Hits (0 Crit), 29 misses = 66.66% Hit Rate
Vs 231 Technician:
Vs 232 Technician: 74 Hits (2 Crit), 45 misses = 62.8% Hit Rate

219 MP using Asp of Titaniush with AR 2083, Full Def:

Vs 217 Technician: 73 Hits (3 Crit), 53 misses = 57.9%
Vs 216 Technician: 71 Hits (3 Crit), 63 misses = 52.9%

Resistability

220 MP, 2300 NR, Full Def

Vs 228 Technician: 7 Nukes Landed, 1 countered, 3 Interrupted
Vs 229 Technician:
Vs 230 Technician: 7 Nukes Landed, 2 countered, 3 Interrupted
Vs 231 Technician:
Vs 232 Technician: 12 Nukes Landed, 2 countered, 1 Interrupted

219 MP, 2211 NR, Full Def

Vs 217 Technician: 30 Nukes Landed, 5 countered (Interrupt N/A as didn't attack).
Vs 217 Technician: 9 Nukes Landed, 1 countered, 7 Interrupted
Vs 216 Technician: 10 nukes Landed, 1 countered, 1 Interrupted

There's really no reason for me to lie or falsify the data. On test, we're there precisely because we want to generate good data and good feedback... if we massaged the figures that would go entirely against our purpose in being there in the first place.

*shrug* think what you like I guess... but the data above are accurate.

X

[Xtremtech]

I will recheck the original data though - to make sure that there wasn't an error. And I'll try to reproduce it too, to be on the safe side.

We all make mistakes of course.

X

[drops]

I'd say that 100% hit at full agg was a freak accident, look at the 232 technician. 64% hit rate at full agg vs 62.8% at full deff. They're statistically the same.
Edit: You should provide a much bigger data lot. Collect more hit/miss data, like a couple of thousands per mob

[Xtremtech]

Yeah, I'd spotted that higher level mobs had flatter variation - but I'd thought this one could be a threshold datapoint. I even had in mind that agg/def shouldn't affect hit-rate... but the change to 100% was so clear, it persuaded me that I must have remembered wrong. I even counted back twice through the logs to be absolutely sure.

Here's the testing I've just done as confirmation though:

Previous testing was done with a full test on each mob. To rule out potential mob differences, I tried replication with the same 220 MP, fighting 230 Technicians - changing from Full Def to Full Agg halfway through the fight. First one was:

220 MP using Asp of Titaniush with AR 2154 Vs 230 Technician Conf 1:

Full Def: 37 hits (1 Crit), 29 misses = 56% Hit Rate
Full Agg: 44 hits (2 Crit), 13 misses = 77.2% Hit Rate

Which, although it didn't have the 100% hits on full agg, also showed a wide spread between agg and def, of 21% difference.

But then did another test to further check:

220 MP using Asp of Titaniush with AR 2154 Vs 230 Technician Conf 2:

Full Def: 43 hits (1 Crit), 29 misses = 59.7% Hit Rate
Full Agg: 37 hits (2 Crit), 27 misses = 57.8% Hit Rate

which shows them equal... with full agg actually being behind full def. *bows head in abject shame*

All previous data still stands. Just needed more data points. As I said, still got a lot of testing to do...

And drops... Yeah I'd love to do thousands of datapoints, but there's a massive amount of testing to be done in just a few days if they're going to be deciding on changes (which I've heard they are looking into).

I've got 3 different level toons with which I have to cover all changed mobs and all changed missions, including hit-rates (both ways), Damage over Time evaluation, Player counters to Tech Nukes, Mob counters to my casting and my mez pet's casting, analysing technician casting patterns etc... while at the same time, trying to actually play some missions normally in order to get a feel for the flow and speed comparitively on the different levelled MPs.

I'll get as much done as is possible for me - as will the others on test I'm sure. But I'll try to get a spread across conditions first - and then try to get more data-points in each condition.

X

[Xtremtech]

To give you an idea of the amount of work... and because it's germane to the discussion, I thought I'd include the testing I did on mission scaling - though this doesn't include the subjective experience reports, nor the speed of completion etc. There's 400 data-points in this piece alone - and it's not even completed yet.

Summary:

At levels below 220, most players are getting missions with mobs at or below their own level. The missions seem to be spread across something like a normal distribution below 220 too, so that they rarely get mobs their own level and more usually get them a couple of levels below themselves. Also they'll rarely get mobs 5 levels below themselves. The 219 numbers are odd, in that they run to a max of 217, which is 2 levels below their own level (making the missions very easy imho).

At 220 though, there is a sudden jump to mobs being 8 to 12 levels above their level. Given that at 219, the max is 217... the jump is even greater when you ding 220. Also the distribution at level 220 seems to be equal across the board, which means that more players will tend to see the higher levels of missions rather than seeing those in the middle of the spread. When you're only doing one mission a day, some players may only have seen missions toward the top level values.

At least in part, this undoubtedly accounts for the differences that people are seeing. In particular, professions that have toolsets which are more sensitive to higher level mobs - whether by virtue of including level in their defence checks (like some crowd control) or by virtue of being sensitive to higher levels of mob stats (like my mez pets, interrupts on technicians, debuff counters etc etc).

Details

At level 207 I see a spread of mission level (i.e. the level of the mobs in the missions), based on 50 samples for each type, that looks like this:

Terminate Cocoons

203: 12%
204: 50%
205: 8%
206: 20%
207: 10%

Place the Bomb

203: 10%
204: 24%
205: 20%
206: 42%
207: 6%

Salvage

203: 12%
204: 24%
205: 28%
206: 28%
207: 8%

Totals

203: 12%
204: 28.66%
205: 13.33%
206: 25.33%
207: 6.66%

Using the 220 MP, I did the same test, with 50 samples per mission type and got the following:

Terminate Cocoons

228: 20%
229: 20%
230: 20%
231: 28%
232: 12%

Place the Bomb

228: 18%
229: 20%
230: 22%
231: 20%
232: 20%

Salvage:

228: 24%
229: 18%
230: 26%
231: 20%
232: 18%

Totals so Far

228: 20.66%
229: 19.33%
230: 22.66%
231: 22.66%
232: 16.66%

Numbers with 219 MP pulling Daily Missions:

Place the Bomb

213: 14%
214: 24%
215: 22%
216: 24%
217: 16%

Terminate Cocoons

213: 14%
214: 16%
215: 28%
216: 26%
217: 16%

Totals so far

213: 14%
214: 20%
215: 25%
216: 25%
217: 16%

[Xtremtech]

Lainbr,

Just dropping in to say that I've finally been able to replicate an example where the Technician mob nuked for high levels repeatedly. I've been doing some more focussed testing on Nuke Patterns on the technicians, because of the feedback given here and though it's been streaky, it's not been quite as bad as you'd mentioned.

But I just had one Technician that towards the end of a long series of nuking, suddenly produced 5 Cold Nukes in a row (the cold nukes are the heavier ones which have a wide range). The last 4 of those 5 also came at 10 second intervals, which is about as fast as they can come.

This is an exceptional occurence, so you shouldn't see it happen often at all - but clearly shouldn't happen at all.

Just thought you'd like to know.

X

[Tazalanche]

Five of seven posts on this page are yours. You must like to hear yourself talk too.

[Xtremtech]

Hehe... many would say that. There's a reason why I ended up with the highest postcount by some thousands before they switched it off...

But I'd like to think, that I was trying to respond to people's issues and questions and give some useful feedback on some of them. I pick up on issues brought forward here and try to do some sort of formal testing to substantiate them. So, for example, the mission mob level scaling testing done above, was triggered by a post in one of the other threads in this forum - which almost got lost in the noise. And like above, when Lainbr mentioned getting constantly nuked for 4K, I started a whole testing strand on analysing Technician Nuke Patterns - which has turned up some interesting results.

Then I'll usually try to drop back and give some of the results of that testing, whichever way it turns out so that people can make other suggestions to align the results with their experiences or pick holes in the testing procedure - and I'll go off and try to fix it.

It doesn't always work out that way of course... but that's my goal. It's an iterative process that should hopefully, in the end, lead to better feedback to FC and better patches and adjustments.

X

[vios]

Five of seven posts on this page are yours. You must like to hear yourself talk too.

He just writes his walls of text to hide discussions.

http://www.clichequest.com/index.php?pos=393

[Xtremtech]

Yup that's precisely my goal Vios. You got me again.

/sarcasm off

X

[nitr0]

u can be sure about one thing: there are so many nerds playing ao - that being full def resulting in less hits would have been figured out 8 years ago already =)

was just weird u really believed that.

but it's like people ingame not believing yalm ql doesnt matter - because their speed is all the same.
(some ppl can really drive u nuts with stuff like "haha! u so stupid! they have rs! they are not all the same! haha!...^^)

[Demoder]

Some smart person once said that facts always ruin good discussions.

Keep up the good work, X.

[Xtremtech]

was just weird u really believed that.

Yup you're right... it just looked so clear on that one I'd done earlier the same day, that I didn't even really think about it before answering.

Serves me right - but I'm man enough to admit my own foolishness and go back to double-check things.

X

[Mekh]

Some smart person once said that facts always ruin good discussions.

Keep up the good work, X.

"If the facts don't fit the theory, change the facts."

[Tazalanche]

"If the facts don't fit the theory, change the facts."

Bazinga!

[mawu]

Damn FC, after every patch, you almost makes me murder you everytime, 1st login after a patch all my prefs is lost, then i log it again, and its all back to normal, the halloween is over, stop scaring me!

Dont start the game right after patching. Close he patcher and the launcher, dont accept the EULA. Then restart Ao and all will be fine. Reason for what you describe is the elevated user rights your patching session runs under. Your prefs are saved in a different location with user rights elevated.

Alternatively install AO in a folder other than c:\program files\ or change permissions for the AO folder to have write access as a regular user.

[ZenWon]

blah blah! blah blah blah blah blah!! Admit it. You either work for FC or youre the bigest unpaid fanboi in the history of online gaming.

[lainbr]

LOL Thousands walls of text.

Btw, after do some more ai dailies, I realized 2 things:

1 - I was unluck (and got lvl 220+ mob with a lvl 213 toon)

2 - The dificult mechanism is weird and some aliens are really bugged. Ex(based on last Place the Bomb I did):
Scout 1 - 215 on a 215 toon; 700 average dd and low AR. What I call gimp easy.
Scout 2 - 215 on a 215 toon; 1.7k average dd and middle AR. OK and fun.
Scout 3 (following right path) - 217 on a 215 toon; 3.5k average dd (not the crits, glad he didnt critted at all) and high AR. Loads of HP. Coon, BR, mongo, spam layers and I managed to survive with 10% HP. WTF?!!

[Rktim]

As usual FC needs to learn that the formula for fun and challenging goes:
More Dmg, more tricks and LESS hp.. Not more dmg and more HP -.-

What i do like in AI dailys on my 161 MA is that every so often i run in to a mob that goes down in about 5-8sec, now.. All mobs should do that.. Sure they can have high DD and high AR as long as they die fast.

Oh and give X a break... I could have sworn that full agg made you hit more awhile back... Its just flukes tho -.^