Please upgrade your browser for the best possible experience.

Chrome Firefox Internet Explorer
×

Beryllius?


Grayseven's Avatar


Grayseven
07.10.2013 , 05:49 PM | #41
Quote: Originally Posted by psandak View Post
what you are describing is akin to a UWT mission generating a schematic. The schematic is a result of a critical success at the mission. A critical success has NEVER guaranteed artifact quality materials only a chance at them, so your reasoning is flawed.
Schematics are not a critical success. I have gotten schematics on less than a critical success.
Sometimes you narfle the Garthok, sometimes the Garthok narfles you.

AlrikFassbauer's Avatar


AlrikFassbauer
07.11.2013 , 08:18 AM | #42
I don't even get critical success anymore. It has become so rare since 2.0 that I just can't get Agrinium anymore.
Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius and a lot of courage to move in the opposite direction. (E.F.Schumacher, Economist, Source)

psandak's Avatar


psandak
07.11.2013 , 12:15 PM | #43
OK so I'm wrong regarding schematics and crits, I'm sorry.

Quote: Originally Posted by AlrikFassbauer View Post
I don't even get critical success anymore. It has become so rare since 2.0 that I just can't get Agrinium anymore.
Last night alone I got crits on UT metals and Sliced Parts. And got four crits out of five crafts on my cybertech crafting ear pieces, crafting five might augments I got eight (three crits), and I REed grade 30 armoring and mod and got the schematics for both (on the first try).

It also took me ten tries to get an RE schematic from green to blue on my artifice (crafted ten enhancements got no extras); Crafted several other augments types (resolve, redoubt, and reflex) five each, only got one extra out of fifteen crafts. Half a dozen companions came back with failed missions (five characters [not all level 50+], 35 missions).

My point is two fold:

#1 - too many forget the really good events that happen with the RNG
#2 - One thing one MUST understand is that random NEVER equates to equal/even.

Grayseven's Avatar


Grayseven
07.11.2013 , 05:12 PM | #44
Correct. However, through sheer volume (I have 8 toons that I constantly keep "on mission") you can more easily see "something" happening that not even the RNG explains.

I've run into the RE issue before...in fact I would swear that my Cybertech and Biochem guys are cursed...but this is different.

Prior to 2.2.1 the ups and downs of the RNG would still even out over time. Post 2.2.1, not so much. The decrease is greater than a RNG statistical anomaly would indicate and it isn't limited to any one Mission skill but something I've seen across all of them.
Sometimes you narfle the Garthok, sometimes the Garthok narfles you.

AlrikFassbauer's Avatar


AlrikFassbauer
07.12.2013 , 03:53 AM | #45
My point was that I got my crits at an *considerably higher* rate than after 2.0.

I do understand that i get them only once in a while - but before 2.0 it was FAR easier for me to get Agrinium. Far easier.

Something must have been changed. In the mechanics. That's the only way I can explain to myself why I get so much less crits than before 2.0.
Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius and a lot of courage to move in the opposite direction. (E.F.Schumacher, Economist, Source)

psandak's Avatar


psandak
07.12.2013 , 09:05 AM | #46
Quote: Originally Posted by Grayseven View Post
Correct. However, through sheer volume (I have 8 toons that I constantly keep "on mission") you can more easily see "something" happening that not even the RNG explains.

I've run into the RE issue before...in fact I would swear that my Cybertech and Biochem guys are cursed...but this is different.

Prior to 2.2.1 the ups and downs of the RNG would still even out over time. Post 2.2.1, not so much. The decrease is greater than a RNG statistical anomaly would indicate and it isn't limited to any one Mission skill but something I've seen across all of them.
And my anecdotal evidence indicates otherwise. And that's the over-arcing problem - the 10 or 20 players who post their data is anecdotal at best. And this has been a problem since UO and EQ - Whenever a random event does not statistically work out for one person that person comes to a forum and claims, "It's broken." You then get a handful of others who claim the same statistical improbability and a handful of those who oppose the argument stating their data shows a different picture.

Only BioWare has the full set of data and they will never release it to the public. Maybe you are right, Maybe I am. We can spout conjecture all we want. Neither side will EVER really know the "truth."

AlrikFassbauer's Avatar


AlrikFassbauer
07.13.2013 , 05:09 AM | #47
Quote: Originally Posted by psandak View Post
And my anecdotal evidence indicates otherwise. And that's the over-arcing problem - the 10 or 20 players who post their data is anecdotal at best. And this has been a problem since UO and EQ - Whenever a random event does not statistically work out for one person that person comes to a forum and claims, "It's broken." You then get a handful of others who claim the same statistical improbability and a handful of those who oppose the argument stating their data shows a different picture.
Technically you might be right - but practically this (and statements like this) aren't going to helo people at all.

What I really hate is people going back into Ivory Tower explanations which have no practical relevance whatsoever.

Statistically the weather was pretty normal here in Germany - if you look at it in the course of several hundred years.

Practically, people have lost houses, furniture, farmers have lost their plants, their cattle, everything they owned in a few cases, due to the flood.
The damages will cost millions. Several millions. And the people are now facing HORDES of mosquitoes because of the flood.

But statistically seen, the weather was prety normal, looking at the last several hundreds of years.

Same goes for the current fires in the U.S. , or the storms like "Catherina". Statistically seen, they are merely irrelevant anecdotes.
Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius and a lot of courage to move in the opposite direction. (E.F.Schumacher, Economist, Source)

Starglitter's Avatar


Starglitter
07.23.2013 , 06:26 AM | #48
Mmm...

Range and I have little love lost among us... In the world of statistics and probabilities is where we all get lost when we discussed Range.

On one hand, each roll of the die, is an independent activity and thus not subject to our emotional need of fairness, "how dare you roll 1 again!" "You have rolled every number but 4!, you jerk"

But in the world of probabilities, things are looked a bit different, its more akin to a gambler's mentallity. What are the odds that I can roll a "1" twice in a row? How much would you bet against my dollar, if I roll a "1" three times in a row?. With such a thought in mind, we can start thinking of statistical distributions, and standard deviations. Lets take a number generator with a range from 1 to 100. If you generate 100 numbers, you would be inclined to think that all numbers would be generated at one time or the other, but in reality there is a chance that will not be the case. If you were to generate 100 sets of 100 numbers and the generator is truly random (fair), you will find that half the time you will get the right mix of numbers, if you were to allow some repetition or some numbers missing say 15 of them, you would be at the 1 standard deviation area and thus you will find that 85% of your sets to be deemed ok, and the same practice is done with 2,3,4,5 etc deviations until all the roll sets are accounted for.

Because the nature of randomness is such, its very hard to really finger "range" as a culprit, on the other hand random number generators are not perfect, and there are documetned cases of this happening for unknown reasons. The first MMORPG that I recall acknowledging this was City of Heroes (RIP), and thus they introduced a "Streak breaker" code to allow for malfunctions of the random number generator. I remember my alt "Stormfront" placing a storm of ice over some baddies, and reviewing my to hit rolls and to my terror so the very same roll result often repeated as many as 5 times in a row, which statistically would be phenomenal, and then saw the sequencing of repeats itself in cyclic manner. This was corrected through the streak breaker code, and to Stormfront's great happiness. Perhaps this could be an issue with SWTOR, but not having access to see what my rolls were, I really can't tell. It would be so cool, if our rolls were displayed as City of Heroes did made available to the player.

Sometimes the roll itself is not biased, but the look-up tables are and thus the issue may not reside in the random roll. For example berylium may only be acquired in a roll of 91-00 while frazium is obtained in a roll of 01-50, and hyper-something in a roll of 51-89...

Just some thoughts

Sue

Brane_Ded's Avatar


Brane_Ded
07.25.2013 , 03:02 AM | #49
Purple crafting components are only obtained from Bountiful and Rich missions if you're not using a 450 crew skill mission. Using a companion with no bonus to the crew skill, Bountiful will yield 2 purples on a crit and Rich will yield 4 purples. 450 crew skill missions will yield 6 purple components however those missions can also crit and yield 8 purples, so make sure to use a companion with full affection and a bonus to crit on the mission you use.

The crit chance is 10% for orange missions and 15% for all others. A companion with full affection gains another 5% crit chance on top of its bonus, if it has one, where the number chance to crit is the percentage (i.e. +2 crit chance = 2% crit chance). Example, a companion at full affection with a +2 chance to crit on a 450 crew skill mission will be: 10 + 5 + 2 = 17% chance to crit.
Raging Elitists
The Harbinger, US-West

Drebble's Avatar


Drebble
07.26.2013 , 08:25 AM | #50
Knowing the crit rate you can calculate the probability of a "crit-less" streak.

Let C be the probability to crit (15% is 0.15).
Let N be the number of attempts.
Then the probability P of not critting on those N attempts is:
P = ( 1 - C ) ^ N
For example, failing to crit even once with 15% crit chance on 7 attempts has the probability
( 1 - 0.15 ) ^ 7 = 0.32057... or aproximately 32% chance.

So, almost a third of all players will have a no-crit streak on the next 7 attempts (if they have no companion bonus).

In fact, one in ten players will have no crits on their next fourteen attempts! If you have a thousand crafters on the server, that's around one hundred people gnashing their teeth and breaking keyboards.

This is what the probabilities looks like:
Code:
N             Chance to NOT crit on N attempts with 15% crit chance
1	        85,00%
2	        72,25%
3	        61,41%
4	        52,20%
5	        44,37%
6	        37,71%
7	        32,06%
8	        27,25%
9	        23,16%
10	        19,69%
11	        16,73%
12	        14,22%
13	        12,09%
14	        10,28%
15	        8,74%
16	        7,43%
17	        6,31%
18	        5,36%
19	        4,56%
20	        3,88%
21	        3,29%
22	        2,80%
23	        2,38%
24	        2,02%
25	        1,72%
I am not a game designer, but I think it would be a good idea to use some kind of mechanic to limit the effects of extremely bad luck. On our hypthetical server of 1000 crafters, 17 people have not critted in the last 25 attempts. That sure does not sound like fun to me.