eso_lytw8 wrote: »

EmEm_Oh wrote: »

Does the system check your inventory on your toon...then drops. Would like to know if possible.

It could, but this is highly unlikely....this would fall in line with conspiracy.

The real answer is much simpler, all random number generators are imperfect, this we know for certain. Humans and human developed technology can not actually generate truly random data. They build algorithms that use a seed (starting number, often date/timestamp) and create a set of numbers that is one 100% predictable. The deal is once the generator creates the same number again the sequence starts over and you get...THE EXACT SAMES NUMBERS.

The possibility we are exploring is that each character has a seed and a predetermined set of random numbers they may or may not include many numbers needed for a certain weapon. This is why people will see 3 ice staff in a row and see far fewer of something else, hopefully not the item you want. It is not until this seed resets that you get a new set of numbers (3-6 months).

There are tests people use to determine if an RNG is good enough for it purpose or not. But to understand all its uses is easy to miss something especially in a game as complex as this. Like across many characters the drops rate appear to be just as desired, but not so within characters or the drops too frequently generate similar numbers (i.e 3 ice staffs or no shields) but across a lot of people or time they look fine.

So no it is not conspiracy it is the fact we humans are imperfect and we build imperfect systems.

This is not how random number generators are seeded these days.

It is VERY easy to add a bunch of other variables that make it very close to random for the purposes of a game.

Your starting point here is really flawed because you are assuming "you can create a set of numbers that are 100% predictable". That is just not how it works.

Again, all the experiments with REAL DATA on vMA show that you are incorrect in assuming this. There is just no way around it.

It is fun to create theories and try to game the system, but here we have proof that it doesn't work that way. You can try and imagine all the ways ZOS is trying to manipulate you but, once again, run 1000 runs and you will see without a shadow of doubt that all weapons drop close to the same no matter what your main is, resources, day of the month and all this other superstitious stuff.

eso_lytw8 wrote: »

[quote="Calm_Fury;c-6281423This is not how random number generators are seeded these days.

It is VERY easy to add a bunch of other variables that make it very close to random for the purposes of a game.

Your starting point here is really flawed because you are assuming "you can create a set of numbers that are 100% predictable". That is just not how it works.

Again, all the experiments with REAL DATA on vMA show that you are incorrect in assuming this. There is just no way around it.

It is fun to create theories and try to game the system, but here we have proof that it doesn't work that way. You can try and imagine all the ways ZOS is trying to manipulate you but, once again, run 1000 runs and you will see without a shadow of doubt that all weapons drop close to the same no matter what your main is, resources, day of the month and all this other superstitious stuff.

This is a great counter argument, thanks so much for offering it up. Examples I have provided are from 137 runs I made and logged into a spreadsheet. I have not see any other examples of "real" data so would be interested in hearing from anyone else who actually logged data.

I also agree that over large samples like 1000 the data will likely appear as expected. My concern is that it does not over a short period and what we are seeing is lots of bad short periods that sum up to reasonable data overall. So it really impacts some players badly.

Let me provide one more example....back in the veteran days monster helms were a rare drop and could not be traded. The generally accepted drop rate was 25%. Well I had one Vet 16 character, one V5 and one V3...anyone remembering those days knows how hard it was to max a character back then. I was moving my V16 to a tank so I wanted a bloodspawn helmet. I ran spindleclutch day and night trying to get one...not a specific one mind you, but anyone. It took me 44 tries before a single helmet dropped. I asked myself how is this even remotely possible? What caused me to really take notice was during this time I could run it on my V5 and V3 characters and get helmets of course the wrong level. One day out of the blue a helmet dropped, by then I was numb seeing person after person with a helmet for months on end. So I ran it again and what would you know another helmet, from that point on my V16 got a helmet just about every 4th time.

Now I acknowledge that event caused so much distress that I sought an answer of how that was possible and brought to my current theories today. It might be that I experienced bad luck beyond anything that one could imagine and that all the other players that report run after run with incredibly bad luck...like 1 in a million bad luck is just that. Or its possible the generator has a flaw.

One last thing, pay attention to reports of new mass refinement. This is one case where we can see mass random generation over a short period of time. If players begin the report anything beyond reasonable refinement rates with large data sets it might help prove or disprove the appropriateness of the RNG being used.


[/quote]

I believe we have now 17 kinds of weapons (might be wrong). That means 137 runs is not nearly enough.

I do understand how you feel. My first 30 or so runs in vMA I mostly got Ice Staves which are all but useless. Still, I felt like I had bad luck and such but in the end it is just RNG.

About the helms, I cannot say because I didn't play the game back then and I don't have the information to argue on that point at all so I'm not.

Again, your point is always about very small sample sizes. If you throw a coin, you can indeed get the same side 3, 4 times in a row even though the probability is very low. The reasons we are already coming back to the argument of bigger sample size is that those outliers will eventually even out.

If you want data, you can search the forums from when ZOS changed the drops in vMA so that every weapon unique type dropped equally instead of being tied to the general type first, then specific type. I remember a bunch of people logged their runs and shared their data and am pretty sure someone built a basic addon to track it. After a lot of runs were logged, all weapons dropped at around 6% of the time, if I remember correctly. The number after a lot of runs was getting closer and closer to the mathematical chance calculation given the information we had. ZOS said they evened out the numbers, people tried and confirmed that this was the case. I think that this is more than enough to squash out any other theories.

About the refinements, it is the same issue. Lots of people saying they refined lots of mats and got zero gold, but the only people actually posting videos and logs and images are confirming that the ratio is exactly the same. And a lot of people are refining one or two stacks and thinking that it is enough.

In summary, because I think this discussion has reached its end (at least for me), the ONLY way to reverse-engineer those kind of odds is to have loads and loads of data. Anecdotal evidence does not count, unless it is proven and added with a huge number of runs.

Taleof2Cities wrote: »

I can’t get dual wield in vMA if my life depended on it ...

You and a lot of others just like you report the same thing. If my theory is correct the generator sequence will reset at some point releasing you from your current chances and you will see them drop. Or they may still drop on your current character just with low frequency, i.e. you need a lot of runs. Alternatively I'd recommend trying it on another character if that is a possibility.

Or I might be totally wrong and everything is working as intended.

I scored 26 daggers alone across my runs (1 was slashing), I know it could be just as intended but it does not feel right.

Sorry for your pain.

the funny thing here is my main is a sorc and i run vma only on my mag sorc, in about 150 vma runs i have gotten 2 ice staffs 2 lightning 3 inferno 14 bows 1 2h sword 1 2h axe 1 2h maul and over 100 drops between resto dw and sword shield. Thats over 30 of 3 drops, the 3 inferno staves i got during this event just last week 3 runs in a row, like they were rigged to finally drop after not being able to get one over years