|
1 year ago ::
Apr 01, 2012 - 2:00PM
#1381
|
Date Joined:
Sep 18, 2003
|
Daar: The original article is two MTGO versions, at least one developer team, and ten years old now. Things could certainly have changed in the meantime. Not saying that they have, but it is possible.
True enough. However, WotC isn't known for its coding speed, or for quickly fixing bugs. Let alone changing code that's fine. And the complaints about the shuffler we have now are identical to the ones we had in the first year of MTGO. If the shufflercode has changed since then, that would be an argument against it being broken, because both camps agree that shuffler is still the same, and as broken or fine as it ever was. From a coding perspective it's very unlikely that two pieces of broken code give the same result. Two correct implementations of the same algorithm will however produce identical results.
Nice to hear you believe the shuffler is fine though!
"Leagues are coming and soon."
|
|
|
|
1 year ago ::
Apr 02, 2012 - 9:35AM
#1382
|
|
|
The most compelling (if somewhat philosophical) point is that the shuffles doesn't know the difference between any of the cards, so why would it have a bias toward one particular type (ie lands)?...
If the input were ordered (such as lands always at the beginning) and an algorithm flaw caused bias in the ordering of cards at the beginning, then wouldn't there would be a land bias?
Just a theoretical point, I don't think this is actually happening.
|
|
|
|
1 year ago ::
Apr 02, 2012 - 10:09AM
#1383
|
Date Joined:
Mar 20, 2003
|
The most compelling (if somewhat philosophical) point is that the shuffles doesn't know the difference between any of the cards, so why would it have a bias toward one particular type (ie lands)?...
If the input were ordered (such as lands always at the beginning) and an algorithm flaw caused bias in the ordering of cards at the beginning, then wouldn't there would be a land bias?
Just a theoretical point, I don't think this is actually happening.
But the thing is, ppl just as often complain they didn't get red cards, or counterspells, or fill in the blank. I used lands as an example but it's really anything a player happens to be focused on.
Apart from that, even focusing JUST on lands, the amount of land flood complaints about equals the land shortage complaints, and it can't be both consistently if there is a bias.
|
|
|
|
1 year ago ::
Apr 02, 2012 - 10:58AM
#1384
|
|
|
Apart from that, even focusing JUST on lands, the amount of land flood complaints about equals the land shortage complaints, and it can't be both consistently if there is a bias.
Perhaps my decades in writing software makes me realize all the ways it can go wrong...The bias could be such that there's a slight tendency to get "too much" land at the beginning or end of the deck. This would result in either floods or shortages at a non-random frequency.
But, I'm only engineering a situation where the shuffler is less than random. I think a lot of things would have to go wrong (and stay wrong) for this to be true.
|
|
|
|
1 year ago ::
Apr 03, 2012 - 5:25AM
#1385
|
|
|
There is one area where the MTGO shuffler is definitely broken and that is its ability to replicate the shuffling done in paper matches. It is obvious that that was never a requirement for the MTGO shuffler but is it is the reason for many of the complaints against the shuffler.
People are used to the pseudo shuffling done in paper magic which generally yields nicely distributed lands. When these same people start to use a shuffler with real random output, they are not happy and, of course, blame the shuffler.
This is a problem (i.e. difference) that is not going away and will continue to make this thread drag on forever.
|
|
|
|
1 year ago ::
Apr 03, 2012 - 8:20AM
#1386
|
Date Joined:
Mar 20, 2003
|
There is one area where the MTGO shuffler is definitely broken and that is its ability to replicate the shuffling done in paper matches. It is obvious that that was never a requirement for the MTGO shuffler but is it is the reason for many of the complaints against the shuffler.
People are used to the pseudo shuffling done in paper magic which generally yields nicely distributed lands. When these same people start to use a shuffler with real random output, they are not happy and, of course, blame the shuffler.
IF you shuffle properly in paper you should have the same results as you do online. If you don't, you're not shuffling well enough and you are cheating (whether you are intentionally doing so is not the issue here, you most likely aren't). the most common scenario is in casual paper games, where ppl will put their lands into their deck "nicely spread" and then do a couple of hand shuffles. this is *definitely* not sufficient shuffling. Anyone who would want mtgo to replicate is basically disqualified from the discussion. Period.
The problem here is not what you think it is. There ARE two very real differences between paper and digital, but I would certainly not put the blame on mtgo but on paper.
First, there is the impracticality of enforcing sufficient shuffling in a paper environment. If it could be done it would be done, and you would see the same kinds of results as online. Extreme pains are taken to ensure proper shuffling in other card games, but not in magic. The reason is that there are more decks to shuffle (there would have to be a shuffler for every match or one per two players), and that the magic cards are someone's property and valuable at that. In a casino you let the dealer shuffle, but would you let a judge handle your deck the entire time?
Second, in a given amount of time, a typical online player will play far more games online than a paper player. This skews perception. If you play more games per day, you will encounter manascrew etc (and god hands) almost on a daily basis. In paper, you play what, every saturday? Three games per saturday? You'll have it happen once a month maybe. So it SEEMS like you have odd draws more often online, but that is mostly due to the sheer volume of your games.
|
|
|
|
1 year ago ::
Apr 03, 2012 - 1:56PM
#1387
|
|
|
today something unbelievable happened. i was playing an esper control deck with 27 lands. my opening hand :
1xgrave titan (1 in the deck)
1xconsecrated sphinx (1 in the deck)
1xthink twice (2 in the deck)
1xday of judgment (3 in the deck)
1x go for the throat (1 in the deck)
1xmana leak ( 4 in the deck)
1x oblivion ring ( 1 in the deck)
i shipped it back and here is what i got second time
1xgrave tita
1xthink twice
1x doj
1x go for the throat
1xmana leak
1x oblivion ring
same stuff except the sphinx, what is the probability of that?
|
|
|
|
1 year ago ::
Apr 06, 2012 - 10:42AM
#1388
|
|
|
By itself? Extremely low for any one person. Given the number of people who play Magic every day and the number of matches people play, the odds of it or a similar event happening over a certain time period is pretty high. That's why you have lottery winners.
Also, there are a LOT of very low probability events that could happen with any given deck. These numbers are completely made up, but let's say there are 100 "I can't believe I got this opening hand or something very close to it twice!" events in any deck (things like getting 4 of one card, 3 of another, getting all lands, getting no lands, getting all 3 of your legendary creature in two opening hands in a row, whatever makes you stop and think "That's odd!"). Each of those events has a very small chance of occurring. But two things make it more likely that eventually at least one of those events will happen:
1) There's a LOT of very odd events possible when drawing cards in Magic. Compared to the number of events that won't really seem odd it's small, but not so much so that getting ONE odd event is effectively impossible.
2) If you play a lot of Magic, eventually you're going to get an odd event.. in fact, if you get no odd events that's even odder!
And given the number of people who play Magic each day, there are odd events happening all the time to SOMEONE (they just won the lottery), and then they pop up on here and ask how likely it is.
So seydaneen... you were the 'lottery winner'.
|
|
|
|
1 year ago ::
Apr 24, 2012 - 6:07PM
#1389
|
Date Joined:
Apr 22, 2012
|
I just played a draft where I drew 9 lands in a row, surprisingly I lost.
I finished drawing out my hand. 11th draw, spell, 12th draw land, then 13 spells in a row, then 2 lands in the end.
13 spells in a row, in a 40 card deck....
16 lands total, I had 10 on the field, 3 in hand.
Took a screenshot because I was astonished. Blown away.
|
|
|
|
1 year ago ::
Apr 25, 2012 - 8:31AM
#1390
|
Date Joined:
Apr 18, 2007
|
IF you shuffle properly in paper you should have the same results as you do online. If you don't, you're not shuffling well enough and you are cheating (whether you are intentionally doing so is not the issue here, you most likely aren't). the most common scenario is in casual paper games, where ppl will put their lands into their deck "nicely spread" and then do a couple of hand shuffles. this is *definitely* not sufficient shuffling. Anyone who would want mtgo to replicate is basically disqualified from the discussion. Period.
How do you know the computer version of shuffling isn't simply spreading the lands in the deck randomly when assigning the numbers to the cards the same way people shuffle them into their real decks? Wouldn't that mean the computer is doing it the same way we are in real life?
Does anyone know how the land cards are assigned numbers when being put into the shuffler's shuffling? Are the cards assigning different numbers every time I shuffle or do they stay the same with that particular deck for the duration of the match? It seems to me that these things could have sonme sort of effect on the shuffling the same way that humans "shuffle" their decks.
|
|
|
