Crafting Success Rates: Possible Bug

I wanted to report a potential bug regarding the success rates.

ISP:
AT&T U-Verse Highspeed ADSL

Type of Internet Connection:
ADSL

Internet Connection Speed:
12MBPS DOWN / 1.75 MBPS UP

Date & Time:
PST Region. Since the start of Early Access.

Frequency:
Sometimes

Character Name:
Shizuo Nahga, Kizuna Aikawa

Race:
Lalafell M, Miqote F

World:
Masamune

Class/Level:
Armorer 1-50, Leatherworker 1-50, Blacksmith 1-28

Area and Coordinates:
Any area that allows crafting

Party or Solo:
Solo

NPC Name:
None

Monster Name:
None

In-Game Time:
All time.

Steps:
1. Crafting/Synthesis commenced.

2. Using actions such as "Basic Touch", "Standard Touch", "Advanced Touch". Sometimes paired with Steady Hand to Increase success rate.

3. Problem begins when actions fail in succession.
Examples: Basic Touch, Basic Touch, Basic Touch with three failures at 90% with Steady Hand. Cases where actions fail at high success rate thrice in a row are common (or more). This problem has persisted from leveling through 1-50 on multiple crafts, even with different characters. It is more than likely to receive all but 1 failure in actions that increase Quality.

4. I have well over 20k shards, in addition to receiving an abundant amount during levequests, I have kept track of all my crafts. Over the course of approximately 6000 crafts, the rate of occurrence would be 2/3.

5. HQ success rates when the craft is finished have also been very abnormal. When finishing a craft with 71-98% Success rate in HQ, there is a moderate chance the craft will become NQ. The opposite is also true where 2-20% Chance will more than likely succeed.

Additional thoughts and notes:
As I mentioned before I am reporting a possible bug due to the sheer amount of times that this highly unlikely scenario has happened. If the system is indeed working as intended, I would like to give feedback in that the current system does not feel good to craft. The numbers are more akin to 40-60% without "Steady Hand", whereas if the actions were labeled as such I would not feel as horrible failing successive actions in streaks, happening in every 3-5* crafts.

The craft itself will not fail in terms of progress. It is the quality actions that are my main concern. I consider the system to be relatively balanced, but it does not mean it is fun to see high % rates fail. I would prefer the system be adjusted accordingly to reflect realistic rates.

I actually expect 70% actions to fail once per craft, possibly twice in a row every so often. Any farther than that though causes frustration. I have been crafting since Early Access. My crafting sessions are roughly 4-6 hours each. I have only listed a handful of examples, and am hoping to record a crafting session to further help visualize the problem at hand.

EDIT: I actually took some time out of my day to look up different RNG's.
- http://en.wikipedia.org/wiki/RANDU
- http://physics.ucsc.edu/~peter/115/randu.pdf
- http://www.purinchu.net/wp/2009/02/0...ber-generator/

This sounds strangely similar to what I have to deal with, it doesn't have to be RANDU, but it sounds like a horrible RNG was implemented.

I sincerely hope this is accepted as a bug. I put my 30 GSM on hiatus for over a week since I had 3 synths at 90% or higher to HQ and not one proc'd. I understand every time is a new ~10% chance to fail, but that coupled with the random strings of failures of quality boosting (which I have experienced as well) make me not the nicest person on our Vent server.

I've noticed this, too, but I developed a superstition over it. The longer I craft without taking a break, the more it seems I also hit that 1 in 10 chance of failing with Basic Touch/Steady Hand. I eventually thought, "Maybe the wear on my gear is affecting the success rate". I don't craft in the nude; I believe in making HQ items ALL the time, if only for the 300% exp bonus.

Like I said, it's only superstition to me, and I have no proof to back it up, nor can I support that repairing my gear helped in any way. RNG tends to hate on people with green hair (me being superstitious again).

This is getting ridiculous. On top of the occasional inconsistencies I just had every single action of mine used with "Steady Hand" fail. 14 times @70% or 90%. All of them failed. And then I had 5 back to back crafts at 85~% all become NQ. This is definitely not normal. Just another one of my experiences with the system, and I have had the 70% actions fail 3-4 times in a row on a consistent basis now. It seems to occur after a handful of crafting attempts. "Steady Hand" makes no difference in crafting unless you raise the % to 100%. If this is so-called luck, I'd like to know where the good luck streaks are, because I have not had any I did not have to literally 100% to achieve anything. I've spent 4000 crystals since posting this thread. About to spend another 4000 to see if my "luck" changes.

This is of interest to me, so I used the log parser at http://xivparser.com/summary/ to see what the actual rate of success is. I measured success rate using Hasty Touch buffed by Steady Hands II, which has a supposed 80% success rate. Here are the results:

http://pastebin.com/nDfPZme3

In summary, I got 938 successes and 216 failures out of 1154 attempts, which works out to a 81.3% success rate. It's actually surprising that it's higher than 80% because I was forced to use Steady Hands I for one single Hasty Touch step in a few of the chains which dropped the success rate down to 70%.

It's disheartening to get 5 failures in a row but keep in mind you tend to notice extreme cases more than mundane ones. I believe this is a true RNG and not a pseudorandom number generator which biases your chances based on past performance, so don't fall prey to the gambler's fallacy. If you think you are getting particularly bad rolls consistently, and that it might be a bug, then try parsing your logs to see what success rates you're actually getting and maybe submit them as part of the bug report.

Stating that you use a 3rd party program puts you at risk of being banned. Please be careful. SE is very strict about this.

It's a program that parses the log files the game generates. Has nothing to do with the game itself. Maybe if SE made a log window I can copy out of I wouldn't have to resort to a 3rd party program.

Its RNG.. I've leveled GSM/BSM/ARM and now LTW to 50 and its all RNG. The numbers for success are accurate.

Yes, it's RNG, but it doesn't mean it's particularly a good one. If you have anything to add please do. I've also leveled multiple crafts to 50 now and don't see how this is relevant.

And Sinbios, I have gotten "Steady Hands II" and have used it. I think it's accurate and you are correct. My problem was with "Steady Hands" (V.1). The buff itself misleads you with 10% when the buff is cast, on your status effects bar, but supposedly gives you 20%. Based on what I have seen and heard I believe there is an actual problem here, at least with "Steady Hand".

Yes, this does suck, but no, I'm not biased. I have had decent hits and am aware of this. It does not excuse the horrible streaks that shouldn't happen more than it has to. Crafting is already easy as it is, but as stated in my original post it feels horrible, and I'd rather have lower chances of success with greater rewards if this is intentional. Actually I'd just be happy if they adjusted the "Steady Hands" buff to reflect 10%, that would make a lot more sense.

Note: I use the Hasty Touch -> 100% method.

Oh, I see. Try the log parser and see what you get, or an easier way to test is use steady hands I and see if you ever fail standard touch.

Sinbios...I totally agree with you. I just know they banned people for their use in FFXI. Just looking out for ya. "Not like I have no clue what you mean." I'm not disagreeing about parsing.

I must say it's a little naive to think the RNG in this game (or any MMO) is truly random.

I have done 100+ dungeons (have achievement), about 70 of those are ifrit HM, not one time have I seen the weapon drop for me, in addition I have 5 other ifrit weapons that I can't use or sell. apparently the chance of not getting a 1/8 (since there are only 8 classes) roll 70 times in a row is 0.00872%, according to the accumulative binomial probability formula.

evidently, I have triggered a 0.01-% situation with a 12.5% RNG. on less statistical front, numerous people have told me they have seen the monk weapon drop few times in one day, all with not a single monk in party. I can't help but think somewhere in the code there is something deliberately making the game harder than it says it is, therefore require more play time ($$$).

likewise, your craft succession rate is certainly subject to the same "randomness", I guess the game wants to encourage you to get 100%, but by what I feel slightly shady methods.