False advertising sets now are we @ZoS?

Paulington

Ihatenightblades wrote: »

Paulington wrote: »

PlagueSD wrote: »

Random doesn't necessarily mean even. I used excel to roll a D6 5000 times. You'd think there would be an equal change of any number appearing. Well, you'd be wrong.

This is absolutely correct.

Let's say each bonus has a 20% chance to proc (1/5) and you test 100 procs, the probability of you getting 25 or more of that buff is around 9%, that's most certainly not uncommon.

The probability of you getting 30 rolls on the same buff in 100 is about 0.9%, so still not all that uncommon.

As @PlagueSD mentioned, random does not mean even. Random means it will tend to an even distribution over time but even a few hundred procs is most certainly not enough to tell that.

The set is broken as I posted above, but not because the RNG isn't correct. .

Dude that chart is even to me for a random. If u had that same chart for the daedric trickery set it would be soooooooooo different and thats my point.
Its a fact that protection and mending have the lowest proc chance its simple math bud.

Well for the hell of it, let's test.

For this test, I put the set on and stayed in combat for 300 procs of the set.

Numbers at n=100:


Numbers at n=200:


Numbers at n=300:


Now the probability of me having only 20 mending procs at that point is <0.01% but that does not mean a whole lot as the sample size is absolutely miniscule and you'll note that you claim to get Major Expedition most of the time whereas I got Major Vitality. That tells me we're both lying out there on the edges of the curve somewhere but you can see in my graphs that they are becoming more even, and would do so given a long enough test.

Now I do not have the time nor the inclination to do a few thousand procs of this set, which is what you'd need to be anywhere even near representative, but this distribution is exactly what it'd expect from a random proc.

This set is not broken, it is random.

Ihatenightblades

Ackwalan wrote: »

Never count on "random" to be beneficial nor random.

Random seems to be fine to me with every other set.

Paulington wrote: »

Ihatenightblades wrote: »

Paulington wrote: »

PlagueSD wrote: »

Random doesn't necessarily mean even. I used excel to roll a D6 5000 times. You'd think there would be an equal change of any number appearing. Well, you'd be wrong.

This is absolutely correct.

Let's say each bonus has a 20% chance to proc (1/5) and you test 100 procs, the probability of you getting 25 or more of that buff is around 9%, that's most certainly not uncommon.

The probability of you getting 30 rolls on the same buff in 100 is about 0.9%, so still not all that uncommon.

As @PlagueSD mentioned, random does not mean even. Random means it will tend to an even distribution over time but even a few hundred procs is most certainly not enough to tell that.

The set is broken as I posted above, but not because the RNG isn't correct. .

Dude that chart is even to me for a random. If u had that same chart for the daedric trickery set it would be soooooooooo different and thats my point.
Its a fact that protection and mending have the lowest proc chance its simple math bud.

Well for the hell of it, let's test.

For this test, I put the set on and stayed in combat for 300 procs of the set.

Numbers at n=100:


Numbers at n=200:


Numbers at n=300:


Now the probability of me having only 20 mending procs at that point is <0.01% but that does not mean a whole lot as the sample size is absolutely miniscule and you'll note that you claim to get Major Expedition most of the time whereas I got Major Vitality. That tells me we're both lying out there on the edges of the curve somewhere but you can see in my graphs that they are becoming more even, and would do so given a long enough test.

Now I do not have the time nor the inclination to do a few thousand procs of this set, which is what you'd need to be anywhere even near representative, but this distribution is exactly what it'd expect from a random proc.

This set is not broken, it is random.

Im sorry buddy but you are showing what random looks like but you arent showing results of the trickery set.. do you understand this? Do you understand that the graph you show has nothing to do with proving or disproving if trickery set is working as intended?

The graph you show shows that random can pull all types of random numbers different majority's at different times.

But you aren't showing a test of daedric trickery set being proccd.

I come to acknowledge that you arent too familiar with the game or the sets in it to be a insightful critique so il try stick to my statement... set is broken it is not random.. ignoring facts is just delusional...

Paulington

Ihatenightblades wrote: »

Ackwalan wrote: »

Never count on "random" to be beneficial nor random.

Random seems to be fine to me with every other set.

Paulington wrote: »

Ihatenightblades wrote: »

Paulington wrote: »

PlagueSD wrote: »

Random doesn't necessarily mean even. I used excel to roll a D6 5000 times. You'd think there would be an equal change of any number appearing. Well, you'd be wrong.

This is absolutely correct.

Let's say each bonus has a 20% chance to proc (1/5) and you test 100 procs, the probability of you getting 25 or more of that buff is around 9%, that's most certainly not uncommon.

The probability of you getting 30 rolls on the same buff in 100 is about 0.9%, so still not all that uncommon.

As @PlagueSD mentioned, random does not mean even. Random means it will tend to an even distribution over time but even a few hundred procs is most certainly not enough to tell that.

The set is broken as I posted above, but not because the RNG isn't correct. .

Dude that chart is even to me for a random. If u had that same chart for the daedric trickery set it would be soooooooooo different and thats my point.
Its a fact that protection and mending have the lowest proc chance its simple math bud.

Well for the hell of it, let's test.

For this test, I put the set on and stayed in combat for 300 procs of the set.

Numbers at n=100:


Numbers at n=200:


Numbers at n=300:


Now the probability of me having only 20 mending procs at that point is <0.01% but that does not mean a whole lot as the sample size is absolutely miniscule and you'll note that you claim to get Major Expedition most of the time whereas I got Major Vitality. That tells me we're both lying out there on the edges of the curve somewhere but you can see in my graphs that they are becoming more even, and would do so given a long enough test.

Now I do not have the time nor the inclination to do a few thousand procs of this set, which is what you'd need to be anywhere even near representative, but this distribution is exactly what it'd expect from a random proc.

This set is not broken, it is random.

Im sorry buddy but you are showing what random looks like but you arent showing results of the trickery set.. do you understand this? Do you understand that the graph you show has nothing to do with proving or disproving if trickery set is working as intended?

The graph you show shows that random can pull all types of random numbers different majority's at different times.

But you aren't showing a test of daedric trickery set being proccd.

I come to acknowledge that you arent too familiar with the game or the sets in it to be a insightful critique so il try stick to my statement... set is broken it is not random.. ignoring facts is just delusional...

I actually recorded this (or at least ten minutes of it due to Shadowplay restrictions) but I really don't feel that it's worth my effort to upload a 10 minute long 4K video of me stood in front of a target dummy showing you set procs when the results are above.

I have no reason to deceive, merely to show the truth. You come with evidence and you might be worth listening to but for now you have decided that the set is bugged with zero testing and purely anecdotal experience.

I leave with the famous Dilbert comic.

Dracofyre

yes, i get that too often, sometimes i get 3-4 same trait drops in a row, then get other for 3-4 same drops in other copies of traits, getting cycled before getting what you really need.

been getting 4 plague doctor greaves and then later in a week, i get 3-4 helms of plague doctors, 1 purple, 2 blues, and 1 green.

TheNightflame

Someone perform a chi squared goodness of fit test quick! If you already did, much love. Alpha at 0.05 ofc

Ackwalan

PlagueSD wrote: »

Random doesn't necessarily mean even. I used excel to roll a D6 5000 times. You'd think there would be an equal change of any number appearing. Well, you'd be wrong.

Using a program is not a test of random outcomes, it is a test of that program to duplicate a random outcome. A program (currently) can't reproduce a true random result. It will get "stuck" in patterns and duplicate sequences.

HatchetHaro

Epona222

Random does mean that you don't get different things in a strict rotation, or with an equitable distribution of results. As far as anything computer generated can be said to be random anyway (it's more often than not a fake kind of randomisation), you shouldn't expect to get an even distribution of results, especially with a small sample - now if you had been playing for 30 years and it *always* gave you expedition, then that might be cause for concern. I think this is a case of random doesn't mean what you think it means.

Huyen

Flip a coin 100 times. It will never-ever be 50 times heads and 50 times tails. Thats the law of nature.

Tonnopesce

Ihatenightblades wrote: »

Erraln wrote: »

Random does not mean even.

Also, by the sound of it you don't want that buff, so you're likely noticing it every time, and suffering from a memory error concerning its frequency in relation to the others.

No its actually bugged the set isnt working as intended right now i am testing it right now from my 200+ tests so far i have gotten 6% chance on protection. Thats not working as intended buddy

I get 100% vitality.... Nice i need it most

Random is random i have a full medium and a full light of that set and works fine for me, but still il a not reliable set so i've dumped for "competitive" pvp i just keep it for fun

Draqone

Huyen wrote: »

Flip a coin 100 times. It will never-ever be 50 times heads and 50 times tails. Thats the law of nature.

Actually, there is a very good chance that it'll be 50-50.