Orgo Report post Posted March 19, 2007 (edited) I have a huge FE project happening and im getting to the point i have started to mix.. i started the mixing yesterday and noticed i wasnt getting efes even at the normal rate of ~1 per 5k... instead it took 14k before 1 appeared in inventory... i have spoke with a few others that have also noticed the same thing.. am just wondering if it is just a bad day for us or if there was a mistake in the last update.. EDIT:i finished the first batch of 25k fe with 3 efes... so when i finish my first batch of orders (totaling 175k fes) at this rate i will get about 21 efe and normal i would get about 35 (these are just estimates since even though the rates arent completely random they arent 100% predictable either) ... i will make my second batch of 25k and edit the post further.. Edited March 19, 2007 by Orgo Share this post Link to post Share on other sites
Aerowind Report post Posted March 19, 2007 (edited) I have also noticed this. Once I heard the 100% chance thing I decided to finish the 20k project figuring I could get a few EFE's wait until the rate went down and then sell 'em. Anyways, before the change of everything I got about 1 after 2.5k and then I quit after I had made ~5k. I've now made ~13.3k and I haven't made a single one yet. EDIT: Yep, 14.5k FE and not one EFE. Edited March 20, 2007 by Aerowind Share this post Link to post Share on other sites
sundulana Report post Posted March 19, 2007 8k fes and no efes... i'll try to keep updates. major lags tho Sund Share this post Link to post Share on other sites
Kylara Report post Posted March 19, 2007 I've heard that there are ppl making several EFE in 1k FE, so it all evens out. You guys just had bad luck. Share this post Link to post Share on other sites
agis29 Report post Posted March 19, 2007 8k fes and no efes... i'll try to keep updates. major lags tho Sund When i had arti for a long time i didnt make efe's.Suddenly i made 4 efe's in 700 fes.Maybe thats it. Or if u had artificer you allready have many efe's made so the ratio maybe be bad for you now. Share this post Link to post Share on other sites
Lorck Report post Posted March 19, 2007 Or if u had artificer you allready have many efe's made so the ratio maybe be bad for you now. The chances of making an EFE after making 100 EFEs out of 1k FEs is the same of making an EFE after mixing 10k FEs without making any EFE. To put this in a short way, past events have no influence over future events, where 'event' means a possible outcome of an action (like getting an EFE when mixing FEs). If you think you can't take what i say as true, you can read this, where the math stuff related with what i said is explained. Share this post Link to post Share on other sites
Xeone Report post Posted March 19, 2007 As Lorck has already pointed out the chance to make an EFE is the same for each FE regardless of how many EFE you have already made or not made. (The same goes for any of the special items) The only way you can tell what the rates are is to make massive numbers of FE and take an average, because over a large enough sample the inconsistencies are evened out and you will get a more accurate answer. For example if the rate really is 1 in 5k this would lead to only ~20 in 100k So even if you make 100k FE you can only expect ~20 EFE which is a relatively small sample of successes. Given the low chance to make an EFE each time you make a FE, it wouldn't take many long gaps or short gaps between EFE to adjust this figure in either direction If you want a truly accurate rate try making 500k FE as, at a rate of 1 in 5k, 500k FE will produce ~100 EFE all dotted out at varying distances apart. Xeone Share this post Link to post Share on other sites
eagleprince Report post Posted March 19, 2007 Bah! If you are that desperate for EFEs, buy a stone. . Only thing is, they are probably as scarce as the EFEs themselves! Share this post Link to post Share on other sites
agis29 Report post Posted March 20, 2007 Or if u had artificer you allready have many efe's made so the ratio maybe be bad for you now. The chances of making an EFE after making 100 EFEs out of 1k FEs is the same of making an EFE after mixing 10k FEs without making any EFE. To put this in a short way, past events have no influence over future events, where 'event' means a possible outcome of an action (like getting an EFE when mixing FEs). If you think you can't take what i say as true, you can read this, where the math stuff related with what i said is explained. Maybe you are right or maybe you are wrong.Theory and praxis are different sometimes..\And i am not sure if casino-poker and rouletes are exactly the same thing with assembly,C, C++ code. Share this post Link to post Share on other sites
Schmurk Report post Posted March 20, 2007 Anyway, unless Radu stores the random seed for each object made and for each actor, your past actions have no impact on your future actions. So saying that you'll have more chance in the future because you have bad luck in the past is more related to karma than the way it's coded... Share this post Link to post Share on other sites
Lorck Report post Posted March 20, 2007 (edited) Maybe you are right or maybe you are wrong.Theory and praxis are different sometimes..\And i am not sure if casino-poker and rouletes are exactly the same thing with assembly,C, C++ code. The only way to garantee that what you are saying to have any logic, it would be the server to have a database with all essences/vials/armor/etc made by all characters and the break rates, damage taken, etc, to garantee that they would have the "average luck" in this game. Also, the server would have to run tests against that database for each single item made, damage taken, etc. This along with the database would complicate the stuff and led to a major performance hit on server, and i don't see a reason why someone would implement this stuff that way. Assembly, C and C++ dont magically provides the kind of stuff you mentioned. To make things short: that would be an unnessary complication, dunno why a good programmer would do it. Edited March 20, 2007 by Lorck Share this post Link to post Share on other sites
Pyewacket Report post Posted March 20, 2007 Speaking as a bad programmer, (and on their behalf) I can confidently state we wouldnt do it either Share this post Link to post Share on other sites
Entropy Report post Posted March 20, 2007 Yes, the chances to make a rare item are not dependent on the previous results. Share this post Link to post Share on other sites
sundulana Report post Posted March 20, 2007 The chances of making an EFE after making 100 EFEs out of 1k FEs is the same of making an EFE after mixing 10k FEs without making any EFE. the chances of making 100 efes from 1k fes is (1000 nCr 100)*(p^100)*(1-p)^900 and this is not the same as (10000 nCr 0)*(p^0)*(1-p)^10000. i'm sure that we all know that p is independent. when you said the above, the proposition may not be true. altho it's true if you say that the chance of making an efe from EACH fe remains the same, which might not be what we were trying to get at on this thread. Share this post Link to post Share on other sites
Mugwump Report post Posted March 20, 2007 Maybe the best way to approach the game is with the philosophy of "There is no such thing as luck." Like anything else, for every flip of the coin the odds are replaced with the same odds of the last flip. Every FE has the same chance of getting the EFE whammy as the one before and the one after. If you don't rub the Buddah's belly daily and pee purple on Thursdays, then you ain't gonna make squat. But that's an in-game secret.... Share this post Link to post Share on other sites
Lorck Report post Posted March 21, 2007 (edited) i'm sure that we all know that p is independent. when you said the above, the proposition may not be true. altho it's true if you say that the chance of making an efe from EACH fe remains the same, which might not be what we were trying to get at on this thread. Well, i agree it should be obvious enough. But my post was in reply to someone claiming otherwise, so the intent of my post was to cast some light and insight on it. the chances of making 100 efes from 1k fes is (1000 nCr 100)*(p^100)*(1-p)^900and this is not the same as (10000 nCr 0)*(p^0)*(1-p)^10000. Well, i said AFTER that, the chances of making each individual efe is same, does not matter if you made many efes on the past or not, which should be obvious enough, as i said above. Edited March 21, 2007 by Lorck Share this post Link to post Share on other sites
eagleprince Report post Posted March 21, 2007 I must have missed my way. Didn't realise this was the math forum! Share this post Link to post Share on other sites
agis29 Report post Posted March 21, 2007 A few more maths.I hope u ll survive after that...lol I started wondering how it is possible to repair expensive items to tankel without failure.Tankel gives 60/40 chances to repair iron greaves.The same chances are to repair a col iirc.So i am using this technique.. I am giving tankel to repair greaves.If he fails i am giving him to repair a damaged col.Well i dont know if i was just lucky but tankel never failed this way and he repaired my cols.Probably i was just lucky but this trick is worked well for me many many times... This has nothing to do with the ratio of efes but that was the general idea when i was saying that rouletes and programming are not exactly the same thing... I used the same trick during manufacturing.. One day i had a critical failure with a leather helmet and my needle broke.I felt so lucky and i am not kidding!! After that i went to the temple of the manu god and i made a weapon that was requiring 38 manu level iirc, with manu level 28+8 blessings(36 manu).Maybe i was lucky again About the ratio of efe's i suspect that the chances has to do with the number of on line players that making fe's. But i maybe 100% wrong and i mean it. I am not sure if players should care about those matters but surely its interesting. EG:I know some bad programmers that put lines of assembly code to their programms to make them load faster.And another bad programmer that reverse engineered microsofts networking protocols and he made a ..badwritten programm (j/k) called SAMBA. Share this post Link to post Share on other sites
