Rubbarn tool not working as described - ТЕХНИЧЕСКАЯ ПОДДЕРЖКА / БАГИ ВЕБ-ПРИЛОЖЕНИЙ

#16 Добавлено Bitttymacod 1 десятилетие назад Доклад | Цитировать [en]

Hmmm... I may have crowed too soon, and there may actually be a problem with the description of the Armilo tool. This thread got me to thinking and I started paying attention to my boosts while crafting HA. (The fact that I have a commission probably has something to deal with that as well.) Very preliminary results (only about 100 crafts, so not clearly significant) seem to show that the two "allowed" boosts for HA are only hitting (for either ) at 1/20, not 1/10; i.e. the tool actually hits 2.5% on any given stat increase, and if some increases are blocked the chance doesn't pass over into the other stats, but is just wasted. Looks like that commission is going to take longer than I thought.

Bittty (Associate Researcher, NTCS)

---

Remembering Tyneetryk
Phaedreas Tears - 15 years old and first(*) of true neutral guilds in Atys.
(*) This statement is contested, but we are certainly the longest lasting.
<clowns | me & you | jokers>

#17 Добавлено Trini 1 десятилетие назад Доклад | Цитировать [en]

{translate}

Sywindt

A similar allegation was investigated by us some time ago by checking the item definitions and using large numbers of controlled crafts (~1000 per tooltype), and we did not see any results that would indicate that the percentages are off. In other words, the percentage of resulting boosts were about what was expected.

May I ask if you made the tests on a test server or on the live servers?

Last edited by Trini (1 десятилетие назад)

---

Trini | 'Ys kard' | Arispotle
First and last of the Darkmoor Rangers

#18 Добавлено Marelli(arispotle) 1 десятилетие назад Доклад | Цитировать [en]

{translate}

This reminds me of the lesson I once had about what 'random' is - I was shown various patterns and asked which were random, then was revealed that some were actual random patterns while some were patterns that humans perceive to be random - and sure enough it was those which I had thought were the random ones. The point of the lesson being we humans don't perceive actual random patterns as random at all - we perceive them as being "loaded dice", so to speak. For us to perceive a pattern as random, what it actually has to be is 'fair'.

That being said, computer random number algorithms are far from truly random too.

#19 Добавлено Jola 1 десятилетие назад Доклад | Цитировать [en]

{translate}

Marelli

That being said, computer random number algorithms are far from truly random too.

True enough. They often suffer from serial correlation. This occurs because the outcome of the next event is influenced by the outcome of the previous event. However, just because you get runs of bad (and good) luck in Ryzom, does not mean the system overall is unfair. It remains "fair" so long as the bad and good runs balance out properly. Fairness should be a requirement in computer "randomizers", even if true randomness continues to elude us. :p

Jola(Arispotle) of the ARMAGEDDONS

Last edited by Jola (1 десятилетие назад)

#20 Добавлено Bitttymacod 1 десятилетие назад Доклад | Цитировать [en]

{translate}

A quick report on the use of 14 bags of mats on making heavy armor (approx150-160 pieces).

I got 16 boosts in about equal amounts of hp and stamina. While that is still not a large sample, I believe it to be sufficient to tell the difference between 10% hits and 5% hits, and it appears that the 10% is on the money.

--- Bittty
Associate Researcher, NTCS.

---