|
|
Post by saladdays on Jan 29, 2014 18:12:45 GMT
UrQuan, I don't think you've offended anyone (though it sounds like you might be trying to), but the sarcastic tone of your comments seems to indicate that we have offended you... If so, and especially if it was me - I'm sorry! Internet forums are definitely not the best medium of communication, and the written word omits all kinds of non-verbal cues that make spoken communication more effective.... anyway, I think/hope we can all agree that we are not attacking each other, and we are not saying that you don't know what you're talking about. Clearly, the odds are against anyone finishing such a crazy attempt, but odds are odds - and they must allow for even the very unlikely to happen. There are a number of people who have attempted and succeeded at the Kai run with 5/15 (of which kamikaze was the first to succeed there, as well). I can't help but notice that his run of the Kai series (from 2010) also omitted Mindblast, which is on your list of "must-have" disciplines. That's one of things that makes his runs unique - he likes to experiment! And, in case anyone was wondering - for random numbers, my program doesn't just pick a "0-9" number... it actually picks a random column, and a random row, and uses whatever number is there on the action chart. So, it's not like kamikaze rolled the "same" zero time and again - it just happened that he hit lots of the different zeroes distributed around the table. Same overall effect, I guess, but gets there differently than simply rolling a 1d10. I'm not offended (maybe by Kamikaze's run), but I'm definitely getting frustrated. At this point I'm begging you guys to reconsider. I can honestly say that I was not planning to throw more wood into the fire here. But after reading the post about his run, I was just disgusted. I mean, was he even trying? He's basically just walking around in random sections, getting into unnecessary fights and taking unnecessary EP damage. And then, not only does he get credit for the run, but people outright refuse to believe he could be wrong about anything he does or says. Even after me pointing out the very obvious flaws in his comments about the run (in the other thread). Look, I understand probabilities might be hard to comprehend, but maybe if you try just book 5 yourself without combat skills? You don't even have to go through books 1-4 more than once, just restart at the beginning of book 5. How many times will you bash your head against the Elix before you give up; 10, 20, 100 times? Just the Elix fight at -11 CR (he didn't use Alether here) will take an average of 500 tries. And of course this is absolutely nothing compared to the whole book, or the whole run. Maybe then you will at least get a feel for what type of insurmountable odds we are talking about here. Of course, in the post describing his run he fails to even mention this book (neither does he mention book 6 which is just as ridiculous). I general though, let's be very careful about comparing (any) 5/15 Kai run to longer runs. And if experimenting means deliberately handicapping yourself, then sure. Hopefully no one would argue that skipping Weaponskill and Animal Kinship in favor of 2 useless skills in a Kai run is a good choice though (his original Kai run). Again, do we even know that Kamikaze was attempting to go through all of the "best" routes, etc. through all the books? |
|
|
|
Post by UrQuan on Jan 29, 2014 18:53:18 GMT
I'm not offended (maybe by Kamikaze's run), but I'm definitely getting frustrated. At this point I'm begging you guys to reconsider. I can honestly say that I was not planning to throw more wood into the fire here. But after reading the post about his run, I was just disgusted. I mean, was he even trying? He's basically just walking around in random sections, getting into unnecessary fights and taking unnecessary EP damage. And then, not only does he get credit for the run, but people outright refuse to believe he could be wrong about anything he does or says. Even after me pointing out the very obvious flaws in his comments about the run (in the other thread). Look, I understand probabilities might be hard to comprehend, but maybe if you try just book 5 yourself without combat skills? You don't even have to go through books 1-4 more than once, just restart at the beginning of book 5. How many times will you bash your head against the Elix before you give up; 10, 20, 100 times? Just the Elix fight at -11 CR (he didn't use Alether here) will take an average of 500 tries. And of course this is absolutely nothing compared to the whole book, or the whole run. Maybe then you will at least get a feel for what type of insurmountable odds we are talking about here. Of course, in the post describing his run he fails to even mention this book (neither does he mention book 6 which is just as ridiculous). I general though, let's be very careful about comparing (any) 5/15 Kai run to longer runs. And if experimenting means deliberately handicapping yourself, then sure. Hopefully no one would argue that skipping Weaponskill and Animal Kinship in favor of 2 useless skills in a Kai run is a good choice though (his original Kai run). Again, do we even know that Kamikaze was attempting to go through all of the "best" routes, etc. through all the books? He mentioned how he was starving himself in book 10 to save healing for book 11 (where it still wasn't nearly enough of course). Doesn't really save healing though to fight extra battles. Either way, that was what people were giving him credit for; "knowing the books extremely well", and that his amazing "strategies" somehow helped him do this run. I pointed out how not only does it not help, but as it turns out, he wasn't even trying.
|
|
|
|
Post by saladdays on Jan 29, 2014 19:54:30 GMT
Again, do we even know that Kamikaze was attempting to go through all of the "best" routes, etc. through all the books? He mentioned how he was starving himself in book 10 to save healing for book 11 (where it still wasn't nearly enough of course). Doesn't really save healing though to fight extra battles. Either way, that was what people were giving him credit for; "knowing the books extremely well", and that his amazing "strategies" somehow helped him do this run. I pointed out how not only does it not help, but as it turns out, he wasn't even trying.
I'm not sure if that definitively proves anything though. Can he not do that even if he chose unique routes, etc.? Also, all of this "knowing the books extremely well" sort of stuff comes from Dave. Dave has said Kamikaze helped come up with the awards, etc., and that he has talked to him a lot off-forum. Dave would know about him better than us, unless you are saying Dave is being deceitful in some way. |
|
|
|
Post by UrQuan on Jan 29, 2014 20:25:51 GMT
He mentioned how he was starving himself in book 10 to save healing for book 11 (where it still wasn't nearly enough of course). Doesn't really save healing though to fight extra battles. Either way, that was what people were giving him credit for; "knowing the books extremely well", and that his amazing "strategies" somehow helped him do this run. I pointed out how not only does it not help, but as it turns out, he wasn't even trying.
I'm not sure if that definitively proves anything though. Can he not do that even if he chose unique routes, etc.? Also, all of this "knowing the books extremely well" sort of stuff comes from Dave. Dave has said Kamikaze helped come up with the awards, etc., and that he has talked to him a lot off-forum. Dave would know about him better than us, unless you are saying Dave is being deceitful in some way. Again, what I'm saying is: Did he cheat or did he win the lottery five times in a row? One of those scenarios is almost infinitely more probable than the other. I'm not sure what you mean. If you talk about "proving" stuff, Dave knowing him certainly doesn't prove anything one way or the other. And a lot more people than Dave has praised him over the years for his Hall of Fame runs and his "skill". |
|
|
|
Post by Dave on Jan 29, 2014 22:03:22 GMT
You say "He wasn't even trying" - can you prove this? How can you prove (or disprove) his intention, any more than we (or you) can prove or disprove the validity of his mythical run? Throwing a "win the lottery 5 times" statement out does most distinctly *not* disprove it... despite the odds, it could happen - and it appears that it did, in fact, happen. So, you've got to come up with something that actually proves him wrong, other than just "the odds are so against it, it's impossible." The odds of being struck by lightning twice are so low that no-one walks around in fear of being struck twice (or even once, really). Yes, it *does* happen. Even more astronomical are the odds of surviving twice - and *that* still happens...
Unless/until kamikaze joins the conversation, I'm pretty sure this back-and-forth "possible/impossible" conversation isn't going to make much ground... I'll write to him and let him know about the discussion, but I do know that some recent events in his real life are taking priority, so he may not be able to join in for some time. Regardless, I'll send an invitation.
|
|
|
|
Post by UrQuan on Jan 29, 2014 23:25:59 GMT
You say "He wasn't even trying" - can you prove this? How can you prove (or disprove) his intention, any more than we (or you) can prove or disprove the validity of his mythical run? Throwing a "win the lottery 5 times" statement out does most distinctly *not* disprove it... despite the odds, it could happen - and it appears that it did, in fact, happen. So, you've got to come up with something that actually proves him wrong, other than just "the odds are so against it, it's impossible." The odds of being struck by lightning twice are so low that no-one walks around in fear of being struck twice (or even once, really). Yes, it *does* happen. Even more astronomical are the odds of surviving twice - and *that* still happens... Unless/until kamikaze joins the conversation, I'm pretty sure this back-and-forth "possible/impossible" conversation isn't going to make much ground... I'll write to him and let him know about the discussion, but I do know that some recent events in his real life are taking priority, so he may not be able to join in for some time. Regardless, I'll send an invitation. Well, either he wasn't trying or he doesn't know the books as well people like to think, or both. I pointed out some pretty glaring inaccuracies in his run earlier. Very unfortunately I cannot prove beyond a doubt that his run was fake, but as I've said before, there is an almost infinitely larger chance that he cheated. Hmm, I really don't want to start this statistics tug of war, but just so that we don't accidentally muddy the waters here: "The odds of becoming a lightning victim in the U.S. in any one year is 1 in 700,000. The odds of being struck in your lifetime is 1 in 3,000 ... About 10 percent of lightning-stroke victims are killed, and 70 percent suffer serious long-term effects."Which means the odds of being struck by lightning twice in your lifetime is "only" (1/3000)*(1/3000)=1.11111111e-7 Funny enough this is still a larger chance than winning the lottery! (not on a like-for-like basis of course) About 400 people survive lightning strokes in the U.S. each year. This means that on average once every 7.5 years someone gets struck by lightning twice in their lifetime (and survives both times). So turns out the odds are not that astronomical at all! Compare this to winning the lottery five times in a row, which happens every... I don't know... quadrillions of times the age of the universe? Source: National Geographic
It would be very interesting to have Kamikaze join the discussion. Speaking of muddying the waters though, that's something he would definitely try to do if his run indeed is fake, but still I would like to hear from him  |
|
|
|
Post by saladdays on Jan 30, 2014 16:29:21 GMT
You say "He wasn't even trying" - can you prove this? How can you prove (or disprove) his intention, any more than we (or you) can prove or disprove the validity of his mythical run? Throwing a "win the lottery 5 times" statement out does most distinctly *not* disprove it... despite the odds, it could happen - and it appears that it did, in fact, happen. So, you've got to come up with something that actually proves him wrong, other than just "the odds are so against it, it's impossible." The odds of being struck by lightning twice are so low that no-one walks around in fear of being struck twice (or even once, really). Yes, it *does* happen. Even more astronomical are the odds of surviving twice - and *that* still happens... Unless/until kamikaze joins the conversation, I'm pretty sure this back-and-forth "possible/impossible" conversation isn't going to make much ground... I'll write to him and let him know about the discussion, but I do know that some recent events in his real life are taking priority, so he may not be able to join in for some time. Regardless, I'll send an invitation. Well, either he wasn't trying or he doesn't know the books as well people like to think, or both. I pointed out some pretty glaring inaccuracies in his run earlier. Very unfortunately I cannot prove beyond a doubt that his run was fake, but as I've said before, there is an almost infinitely larger chance that he cheated. Dave already said he doesn't believe that Kamikaze knows how or would have gone to the effort to actually change the appropriate file to cheat. Can Dave be sure of that? No. But I think him saying that has at least some weight. Yes, I understand that the odds are definitely against Kamikaze to do what he did. But unless we have some evidence other than just low probability, why go to such lengths of to accuse him of cheating? |
|
|
|
Post by rmsgrey on Jan 30, 2014 18:04:20 GMT
Taking the 1e-37 figure as accurate then, to put it into some context, if everyone on Earth played one game with that setup, it's more likely that next time a well shuffled pack of playing cards is dealt out into four piles of thirteen, each pile contains the whole of one suit (~1e-26) than that there is at least one person who gets through book 11 without needing to restart.
If every human who ever lived played one game every second for their entire lives, the chance of at least one no-death run through books 1-11 is still less than 1e-15 (if I refined my estimates, it'd probably land between 1e-16 and 1e-17). If I bought two tickets for the UK National Lottery (for different draws) it's more likely that both of them would win the jackpot (~5e-15) than that every human being to date making a run each second, every second of their entire lives would produce at least one successful run through books 1-11. It's still more likely that I could pick up the phone, dial digits at random, and call any given UK phone (1e-11 for most numbers).
In comparison, suppose there's only one person in all the world with the skills to fake such a log, and the personality to both do that and keep up an acquaintance with Dave in which they don't appear to be that sort of person. then it would be roughly a 1e-10 chance that kamikaze is that person - or 1e27 times more likely than such a run occurring on the first attempt.
Of course, there will have been more than one attempt at a 5/15 hardcore run, and there may not be anyone in the whole world with the right combination of skills to fake it, but if you assume that there have been a million such runs started, and there's only a one in a million chance that there's anyone in the whole world who could fake the run, you still have another factor of 1e15 to account for.
To have an even chance of coming across one 1e-37 event in your lifetime, you need to encounter 1e27 events every second - that's an average of 1e16 events per second for each neuron in your brain.
It's overwhelmingly more likely that the 1e-37 figure is completely wrong than that kamikaze actually made the run - possible reasons include errors in computing it, bugs in Seventh Sense or the underlying random-number generator (a random number generator with a 64-bit seed can only produce less than 1e20 different sequences, so no outcome from such an RNG can actually be less likely than 1e-20; a 32-bit seed would be worse than 1e10, etc)...
|
|
|
|
Post by saladdays on Jan 30, 2014 21:03:53 GMT
It's overwhelmingly more likely that the 1e-37 figure is completely wrong than that kamikaze actually made the run - possible reasons include errors in computing it, bugs in Seventh Sense or the underlying random-number generator (a random number generator with a 64-bit seed can only produce less than 1e20 different sequences, so no outcome from such an RNG can actually be less likely than 1e-20; a 32-bit seed would be worse than 1e10, etc)... And UrQuan did initially state that there could be some bug or another problem that Kamikaze had nothing to do with, but the latter posts (at least to me) seem to indicate UrQuan believes that Kamikaze somehow cheated. |
|
|
|
Post by Dave on Jan 30, 2014 22:10:34 GMT
Are you guys suggesting that my program might have bugs?!? Inconceivable!
|
|
|
|
Post by Honza on Jan 31, 2014 5:59:59 GMT
Are you guys suggesting that my program might have bugs?!? Inconceivable! I support you, Dave. I'm playing Seventh Sense since version 1.4, and I've never came across any bug. Ever. All this is just nasty slandering! |
|
|
|
Post by UrQuan on Jan 31, 2014 16:57:46 GMT
Taking the 1e-37 figure as accurate then, to put it into some context, if everyone on Earth played one game with that setup, it's more likely that next time a well shuffled pack of playing cards is dealt out into four piles of thirteen, each pile contains the whole of one suit (~1e-26) than that there is at least one person who gets through book 11 without needing to restart. If every human who ever lived played one game every second for their entire lives, the chance of at least one no-death run through books 1-11 is still less than 1e-15 (if I refined my estimates, it'd probably land between 1e-16 and 1e-17). If I bought two tickets for the UK National Lottery (for different draws) it's more likely that both of them would win the jackpot (~5e-15) than that every human being to date making a run each second, every second of their entire lives would produce at least one successful run through books 1-11. It's still more likely that I could pick up the phone, dial digits at random, and call any given UK phone (1e-11 for most numbers). In comparison, suppose there's only one person in all the world with the skills to fake such a log, and the personality to both do that and keep up an acquaintance with Dave in which they don't appear to be that sort of person. then it would be roughly a 1e-10 chance that kamikaze is that person - or 1e27 times more likely than such a run occurring on the first attempt. Of course, there will have been more than one attempt at a 5/15 hardcore run, and there may not be anyone in the whole world with the right combination of skills to fake it, but if you assume that there have been a million such runs started, and there's only a one in a million chance that there's anyone in the whole world who could fake the run, you still have another factor of 1e15 to account for. To have an even chance of coming across one 1e-37 event in your lifetime, you need to encounter 1e27 events every second - that's an average of 1e16 events per second for each neuron in your brain. It's overwhelmingly more likely that the 1e-37 figure is completely wrong than that kamikaze actually made the run - possible reasons include errors in computing it, bugs in Seventh Sense or the underlying random-number generator (a random number generator with a 64-bit seed can only produce less than 1e20 different sequences, so no outcome from such an RNG can actually be less likely than 1e-20; a 32-bit seed would be worse than 1e10, etc)... Thank you. I have desperately tried to find errors in my calculations. Now all the major combats have even been cross-checked with a different independently programmed combat calculator. The only minor inaccuracy I could think of is that Seventh Sense uses book-specific Random Number Tables, and none of the calculators I used has that. Should be a very small difference in the grand scheme of things though, especially since the amount of 0's is around 10% anyway (9 and 10 in book 11 and 12). I might have given him the wrong 2-handed bonus in some of the minor -11< CR combats in the last post, but all -11 CR combat should be correct. This is one of the combat calculators I used: www.lw-oasis.org/study/profiler/As saladdays said I did not explicitly mention cheating at first, but after originally reading this thread and finding no inconsistencies in the combats (other than ridiculously good rolls exactly where he needed them), but finding inaccuracies in the route used, I started to lean heavily towards cheating. |
|
|
|
Post by UrQuan on Feb 5, 2014 6:55:34 GMT
Hmm. Yes. I tried out this Seventh Sense stuff. I got past the bridge in book 1, even! But then I thought I too would try to follow in Kami's (his friends call him Kami) footsteps.  Oh, goodness me! I stumbled into quite a ruckus here. This was not part of the plan but I'm glad I scraped by alive! The main problem I had was book 8 in this run. First, I mistakingly ate the soup offered to me in this one temple. Hmm, I thought it was very hospitable at first, but the soup wasn't very nice I don't think. Tasted like it was made out of Black Lakeweed! Very unfortunate! Anyway, I really ran into trouble at the end of the book. Even with perfect rolls, I cannot pass this battle!  Probably didn't help that I somehow stumbled into a couple of permanent CS losses along the way. Ah, very unlucky!  Well, that was the end of my run. Kami really bested me here! I thought I would still try out the last book though. This time I was very lucky, scraping through at 1 EP for most of the book. And then I got to fight this giant fly at the end.  Hmm. This was a very hard fight, yes. I took one damage every turn, from using this one weapon I stumbled upon. Not a very nice weapon. What helped me out though was my trusty quarterstaff, considering I rolled a 3 at -45. Couldn't have done it without it! resume1.dat (20.92 KB) Disclaimer: I'm sorry. |
|
|
|
Post by UrQuan on Feb 14, 2014 16:16:34 GMT
After daily activity in this thread, there has been silence for a over a week now since my last post. I sort of thought I would stir up a hornet's nest with that one. Maybe you are all silently irritated at me. Maybe you just don't care either way, but I care.
Now that I have even recreated a fake run, painstakingly, without any programming experience, what more "evidence" do you want? I know I won't convince Dave, so I'm guessing the run will never be "officially" discredited, but does someone side with me here?
Again, as rmsgrey said, the point is that any other imaginable scenario is absolutely overwhelmingly more likely than that this was a legit run. And of these, cheating is again overwhelmingly the most likely, but that's beside the point. A lot of events that that we in daily speech call "impossible" are in fact infinitely more likely than this. This means that the constant argument "it's possible though!", is very murky in this context. It's just not relevant in any way. Why is it that an event, completely unprecedented in the history of the universe, is somehow the "default" option here?
At this point I've really burned all the bridges, so I will argue this thing till the end. Kamikaze's fake run completely discredits every run ever made, before or after it. Hardcore 1-20 is an absolute joke, in comparison.
Well anyway, I rest my case.
|
|
|
|
Post by t0pel on Feb 14, 2014 21:25:22 GMT
After daily activity in this thread, there has been silence for a over a week now since my last post. I sort of thought I would stir up a hornet's nest with that one. Maybe you are all silently irritated at me. Maybe you just don't care either way, but I care. Now that I have even recreated a fake run, painstakingly, without any programming experience, what more "evidence" do you want? I know I won't convince Dave, so I'm guessing the run will never be "officially" discredited, but does someone side with me here? Again, as rmsgrey said, the point is that any other imaginable scenario is absolutely overwhelmingly more likely than that this was a legit run. And of these, cheating is again overwhelmingly the most likely, but that's beside the point. A lot of events that that we in daily speech call "impossible" are in fact infinitely more likely than this. This means that the constant argument "it's possible though!", is very murky in this context. It's just not relevant in any way. Why is it that an event, completely unprecedented in the history of the universe, is somehow the "default" option here? At this point I've really burned all the bridges, so I will argue this thing till the end. Kamikaze's fake run completely discredits every run ever made, before or after it. Hardcore 1-20 is an absolute joke, in comparison. Well anyway, I rest my case. While I am more or less your side, I would say that you created this resume file using the copy paste trick. Dave specifically said that he checks if the file has been copied during playthrough. |
|
