[Fishing] Changes in difficulty of MiBs?

[sos440]

Hello everyone! I'm a very-recently-returning player.

As far as I remember, and all existing internet sources indicate that, MiBs require GM fishing skill. But in my personal experience, they even begin to appear around 85 fishing now.

Do you have similar experiences?



p.s. I am now doing Fish Monger's quests, and am recording every quest as a spreadsheet data.

I'm assuming that the reward depends only on the total point computed by the items in the given order, and want to know this relationship.

This data is still far from yielding a reliable prediction, but even at this point, I can see that over the half of the reward is gonna be bait, regardless of the difficulty of the order.

I want to see if there are similar projects going on or done. If anyone knows about them, please let me know!

 

[DreadLord Lestat]

all existing internet sources indicate that, MiBs require GM fishing skill

You can start getting them at 75 as shown here on Stratics in the
[format=h5]FAQ - HOW TO GAIN FISHING, REWARDS & QUESTS! (NOW WITH HIGH SEAS & NEW MIBS DATA!) on post 3: [/format]
[Fishing] - FAQ - How to Gain Fishing, Rewards & Quests! (Now with High Seas & New MIBs Data!)

The fishmonger rewards for the most part, the more types of fish needed at one time and the higher amount of fish needed will be a higher reward. The more quests you complete the better the next one should be but the RNG is a real beast so a good portion is luck with the rng lol.

 

[sos440]

You can start getting them at 75

Thank you for clarification and also for that awesome link!

The fishmonger rewards for the most part, the more types of fish needed at one time and the higher amount of fish needed will be a higher reward. The more quests you complete the better the next one should be but the RNG is a real beast so a good portion is luck with the rng lol.

Yes, so far I got 3 powerscrolls, 105, 110 and 115 each, in nearly 130 trials. Sometimes a small order gives me a scroll, while on the other hand an order of 6 crustacean & dungeon fishes gives me a bait RNG is a true beast.

 

[DreadLord Lestat]

Yes, so far I got 3 powerscrolls, 105, 110 and 115 each, in nearly 130 trials. Sometimes a small order gives me a scroll, while on the other hand an order of 6 crustacean & dungeon fishes gives me a bait RNG is a true beast.

Wow I am jealous. I have not counted but I must have done close to 100 with nothing more than bait but I also know that the RNG hates me in all games lol. Congrats and I am sure that 120 will be coming along soon for ya!

 

[Sarak]

last novembr i did 350 quests, got two oracles of the sea, 35 uses chardybolis bait, five 105s seven 110s three 115s adn four 120s three of the 120s happened with in 35 quests two of which were back to back and were large orders. i have received thousands of uses of various baits, and the rare part 6 book a couple times along with dozens of the other parts

 

[Sarak]

now for sos's i am finding that 1 in about 15 are ancients uoguide says 1 in 12; but here recently i have received 545 mibs and 45 were ancients. i guess that is 1 in about 14 not 15 sorry.

 

[sos440]

last novembr i did 350 quests, got two oracles of the sea, 35 uses chardybolis bait, five 105s seven 110s three 115s adn four 120s three of the 120s happened with in 35 quests two of which were back to back and were large orders. i have received thousands of uses of various baits, and the rare part 6 book a couple times along with dozens of the other parts[

now for sos's i am finding that 1 in about 15 are ancients uoguide says 1 in 12; but here recently i have received 545 mibs and 45 were ancients. i guess that is 1 in about 14 not 15 sorry.

350 quests and 545 MiBs... You must be a true fisher!

So far I have only open up like 150 MiBs and 5 of them were ancients. Maybe 150 is still statistically small samplings or again a RNG beast thingly... or luck may play on this lottery.

 

[Basara]

Even 500 would be a statistically too small sample, the way the RNG is set up.

At one point Draconi was going to introduce a new RNG to the game, after myself and others had proven there was a flaw in the existing one - but he was let go before he could do more than internally test it, and his successors might not have even known about the issue.

BTW, the fault with the existing RNG is.

Over millions of samples, it is fair.
BUT
Over small samples taken quickly, there is a tendency of number clustering.

You can see this sometimes on smelting (where there's too many failures in a row for an unbiased sample - and some times too many successes, but that's harder to spot), BOD pulls (where you go through 6 different BOD runners, and you end up with 4-6 BODs for the same item (though usually different counts and/or material), when there should have been about a 2% chance of getting the same item again for each BOD runner on 1 pull, 6% from 3 pulls). As the number clustering is happening at ALL points on the curve, when the standard tests for fairness are ran, they are big enough that all the short-term anomalies themselves average out "fair" - but that doesn't help the guy who just failed 10-12 times in a row smelting Valorite with 105 skill (or worse, gold - that has a 85% success chance at 105, IIRC).

One of the "tricks" with this broken RNG I've seen mentioned for SOS was "small batching" - essentially, open only about 5-10 SOS every few hours, but have the others on you. If in the 5-10 you get an ASOS, open 5 more as quickly as you can, and there's a chance you'll hit a number cluster and get several ASOS out of a small sample. If those next 5 don't get an ASOS, stop there, and wait for your next scheduled stop.

It's not much better than superstition, but I do know and have seen the basis for it. I personally opened up 50 MIBs this weekend, and THREE ASOS occurred in a 5-bottle stretch - and were the only ones I got out of the lot; they were also in the mid-teens among the randomly mixed MIBs I'd collected, in the bag I had arranged them in for easy opening and deposition into my Davies Locker. There is something to it, but in a way that's almost impossible to deliberately game.

 

[sos440]

That is a really interesting observation.

Considering that even 32bit PRNG would possibly be costly for a massive server, I would not be surprised if they are using a less-costly workaround at the expense of producing significant statistical bias.

Anyway, if such a 'clustering' is indeed a thing, it sounds really a good tip for me. I don't want plain SOS cluttering up in my house.

 

[Basara]

Part of the problem is that UO is using a PRNG based on a college-student-level public domain/free PRNG (literally, something that students are taught how to recreate in higher-level college programming classes, the way they are taught to create bubble sorts, etc. in intro classes), which is what most cheap/free games (like most games on Facebook) use. IIRC, Draconi was trying to make a replacement that wasn't as vulnerable to seed-based streaking, that would have belonged to UO/EA had it been completed since part of the work was done in his position as Lead at the time (and therefore would have cost UO nothing, since EA was already paying him).

 

[marcbobb72]

Great Thread !!

 

[SugarMMM]

Welcome back to the game!

I don't want plain SOS cluttering up in my house.

If you would like I bet I could find an extra Davey Jones Locker on my account to help you keep your house clutter free.