Viewing Page 3 of 27 (Total Entries: 2658) |
![]() |
|
Dec 8th 2007 at 08:29:38 AM |
|
Name: |
daniel |
Comments: |
just my first 20 after 7 sub20 and 19 times 21^^ |
![]() |
|
Dec 8th 2007 at 12:53:59 AM |
|
Name: |
Christoph |
Comments: |
I did just activate the IMC poll: minesweeper.cc->scores&people->elections->poll->2007 IMC election You can vote and change your vote using this site. The poll will be open until midnight (GMT) on the 21st of december (unless I should fall asleep in which case it will be closed the following morning ![]() |
![]() |
|
![]() |
|
Dec 7th 2007 at 10:54:55 AM |
|
Name: |
KAmil |
Comments: |
yes ![]() |
![]() |
|
Dec 7th 2007 at 10:21:19 AM |
|
Name: |
Christoph |
Comments: |
@KAmil: got my last mail? |
![]() |
|
Dec 7th 2007 at 10:00:16 AM |
|
Name: |
KAmil |
Best intermediate: |
NF 12x4 |
Comments: |
Ive just checked my clone stats and noticed that I passed 1256 int NF games with average time 19,958s ![]() ![]() |
![]() |
|
Dec 7th 2007 at 09:09:26 AM |
|
Name: |
Tommy |
Best expert: |
42,16 |
Comments: |
... As a matter of fact, I seem to know my decimals better than my integer times :-P my old exp score was a 46, decimals were .69. |
![]() |
|
Dec 7th 2007 at 09:07:30 AM |
|
Name: |
Tommy |
Best expert: |
46,16 |
Best intermediate: |
12,00 |
Best beginner: |
2,44->2,13 |
Comments: |
Oh yeah, forgot, new beg rec - vid linked. (Neither did I chris ;-)) |
![]() ![]() |
|
![]() |
|
Dec 7th 2007 at 09:01:32 AM |
|
Name: |
Tommy |
Best expert: |
76.76 |
Comments: |
76,76... ... On the minesweeper x skin I linked :-D Try it! ![]() I agree with damien. And I would think that the IMC is for people taking the minesweeper seriously and regard it as a sport in the same way as many other computer games. It may not be well-known, but that doesn't matter IMO. And if we do that properly, we have better chances of being taken seriously. We wont get far with an attitude that states that we don't take what we do seriously ourselves, and I think we need people who take minesweeper and the IMC seriously inside the IMC. This is a community of progamers. Face it... |
![]() ![]() |
|
![]() |
|
Dec 7th 2007 at 09:00:19 AM |
|
Name: |
WP |
Comments: |
@Christoph: I'd like it to be the 5th of July! It'll be the first weekend of my summer holidays as well. (and besides, Joni has a point about the room prices) It would be nice to confirm the date pretty quickly...before I buy my ticket home to Malaysia :P |
![]() |
|
Dec 7th 2007 at 09:00:17 AM |
|
Name: |
Michael |
Comments: |
Forgot to mention - I totally disagree with Jan. What determines a sport is the way it is played. It's competitive, and it involves physical skill and precision. Which is not to say it is *always* played as a sport. But a lot of us *do* play it as a sport. The argument of "hey, 46.01 and 46.99 are both 46, so what?" is ludicrous, why not just say "hey, 40 and 49 are both sub-50, so what?" And classify everyone as either "sub-50 player" or "non-sub-50 player". Or don't bother with times at all, just create a list of everyone who has ever completed a board. The point is that while we're using time as a method of sorting players, we could make it a little more accurate. It's not the end of the world, sure, if we can make it more accurate, why not? |
![]() |
|
Dec 7th 2007 at 08:49:53 AM |
|
Name: |
Michael |
Comments: |
Joni - yeah, I realised what I said could be interpretted that way after I came offline. Perhaps what I should have said is "it's a list of the best personal bests". I still think it's true though - minesweeper doesn't tell you how skilled you were, just your raw time. *We* infer some sort of skill level from that, the game makes no such assumptions. Neither does a best ever list, it just tells us best times. That's why I think argument against decimal from the point of view of skill levels are redundant, because skill has never been an essential part of the rankings. Cristoph - without looking, my scores are 57.6 (57.9 on the rankings), 13.6, and 2.5. ![]() On another issue, I've never been to a tournament, but I quite like the idea of non-standardised tourneys. Brings a different element to each one, perhaps lets different players flourish. I, er, think. ![]() |
![]() |
|
Dec 7th 2007 at 08:49:13 AM |
|
Name: |
KAmil |
Best expert: |
NF 54x2 -> 53,962 |
Comments: |
on 132 ![]() |
![]() |
|
![]() |
|
Dec 7th 2007 at 08:01:48 AM |
|
Name: |
damien |
Comments: |
@Christoph: Just so you know, Gergely and I have both investigated non-profit organisations (in Hungary and Britain). I'm glad there seems to be more interest in this. Obviously, we need to agree on the IMC and its function before we can choose the proper structure from the most useful country. @Joni: The .00 after Winmine scores is a side-effect of me cutting Clone scores to only 2 decimal places. I'll try to fix this during the weekend, so people realise the correct accuracy. All the tournaments in 2005 and 2006 were either best 5&5 or 6&6 on Clone, which made it easy to compare. Now Shanghai was 3&3, Budapest2007 and Vienna2007 were on Viennasweeper, and Vienna2007 was a play-off format. Christoph is inventing a new format now....perhaps the IMC will consider standardising tournaments? ![]() |
![]() |
|
Dec 7th 2007 at 07:37:39 AM |
|
Name: |
joni |
Comments: |
Christoph, i think things will get really expensive if we mix up with the Euros ![]() ![]() @Jan: heya! nice to hear you writing a couple of words in here ![]() ![]() |
![]() |
|
Dec 7th 2007 at 07:06:30 AM |
|
Name: |
Christoph |
Comments: |
I'd like to pick up another important topic: I thought long and hard about the worldchampionships next year and I think most issues are sorted out. (The system I'm favoring will be tested on the next austrian championships, but you'll hear about that anyways - it will not mainly be about completion as it was in last vienna tourny). But there's one thing I'd really like to get the communitys attitude on: the date. I've three choices left: 1st: 14th June (Saturday) EURO 2008 games in Vienna: 12th Austria-Poland, 16th: Austria-Germany 2nd: 21st June (Saturday) EURO 2008 quarterfinals in Vienna the day before and after 3rd: 5th July (Saturday) first week of summer vacation in austria+first weekend after the EURO. please post or mail me your comments on this. |
![]() |
|
![]() |
|
Dec 7th 2007 at 07:01:20 AM |
|
Name: |
KAmil |
Comments: |
We should do ranking like that: Dion: 0,7 - 9,67 - 36,77 SUM=48 ... Damien: 0,67 - 11 - 39,34 SUM=52 Real times for clone, scores for winmine, SUM by scores ![]() |
![]() |
|
Dec 7th 2007 at 06:22:20 AM |
|
Name: |
Bertie |
Comments: |
@Jan: Thanks for your post, it was really neccesary. It's good to get a perspective from someone whose not so close to the issue. I actually agree with everything you're saying, but I think that if we're gonna do a ranking we might as well do it right. (Just think of it as another fun way to pass time, coz we all like building something) @Christoph: I don't really know what the procedures are in this sort of arena, but I do agree that we need to somehow ground the IMC to the real world. Oh, and @Schu: The IQ thing is something I also noticed. The worst thing I noticed was that the highest score any of my "educated" university friends (70+ of them, barring my one friend who studies with me coz we kinda have an unfair advantage with our math training) could muster was 137! Not that I'm saying that the facebook test is even a valid IQ test, but it sure shows a pattern. And btw some gamers get paid big money to play, that just makes me sad. |
![]() |
|
Dec 7th 2007 at 06:00:58 AM |
|
Name: |
Christoph |
Comments: |
If your record is at 17.32 say and you get a 16.95 you are happy because you got a 16, and you don't care about the decimals, right? If you get an 16.29 you're still happy because of the 16 and will not really care about the .29, right? And when you're than asked about your record either way you'll say 16, right? So could we agree that the seconds are what really matters - at least in the minds of the most of us? You know your integer records, but how many of us can tell their decimal scores without having looked them up in the last two weeks? - I can't. Could we now discuss something that makes more sense than decimal rankings? i.e. should we make the IMC official and how? I think yes - but how is the question. The central question is where the seat of the IMC should be. Already a year ago I made a little research about how to register it in Vienna (that would be pretty easy). The only concern I had was that all the paperwork would be in german - of course no problem at all, everything's easy to translate and the language these things are written in doesn't really matter, but: if there should come a year noone from vienna or even worse noone who speaks german is elected to the IMC little difficoulties may arise. But that's somehow the same problem whereever we would pick the seat to be. However, I would offer to do all the organisatoric work if we'd choose the seat to be vienna. |
![]() |
|
Dec 7th 2007 at 01:53:23 AM |
|
Name: |
Schu (Andrew McCauley) |
Comments: |
@ Jan: I'm a sweeper too, so yes, I did have to include it. I think you'll see that there are a lot of people that will disagree with your "not some kind of a sports discipline" comment. Otherwise why try to rank people at all? And other computer games take high scores seriously enough to get them into the guinness book of records. You may find it a fun little game, but you can't speak for everyone. |
![]() |
|
Dec 6th 2007 at 09:42:12 PM |
|
Name: |
Jan Parucka |
Best expert: |
57 |
Best intermediate: |
16 |
Best beginner: |
2 |
Comments: |
I read this guestbook from time to time in the form of lurking nothingness but today I kinda got the urge to write a little bit too. I really don't like recent discussion about decimals and so on. To me it seems like some of you try to give the game more importance than it deserve. Difference between 46.99 and 47.01 with the same 3Bv game is pathetic in its own and even more pathetic is to think that the player who got the better time is better than the other one. The point is one got 46 while the other one 47 and that's it. Such is the nature of the game. The same applies to for example 46.01 and 46.99 games. Both got 46, so what. Minesweeper is a fun game and NOT some kind of a sport dicipline. Everyone has different circumstances while playing and no one can be controlled in a way sportsmen are. In, for example, athletic disciplines there's no such thing as world record accomplished during training, at home etc. So, please, leave the decimal rankings for tournaments and don't succumb to illusion that minesweeper is something more than a fun little game with clever design. Don't make it sport. Keep the world rankings simple as they are now. To me it seems that some of you feel somehow guilty (probably more subconciously than otherwise) of dedicating much time to playing the game, that you are afterwards trying to justify it by making the rankings seem more serious and "professional" so that you can feel it wasn't just a fun way of killing time. Personally while I am playing minesweeper I conciously don't concentrate at the game at all, instead of that I am solving entirely different issues (work, composing music etc.) and I'm sure lot of people do it the same. If you play the game just concentrated on breaking your records, it's you loss. Don't try to justify it by proposing the changes that would make the game seem more important than it is. I wish new records to all of you, but don't take it serious (double meaning :-)). I apologize for my poor english as well. Happy sweeping! As a side note - to Schu: as for that IQ test results, you really had to include your own score (which was absolutely unnecessary regarding the subject), hadn't you? |
![]() |
|
Dec 6th 2007 at 07:00:37 PM |
|
Name: |
Schu (Andrew McCauley) |
Comments: |
For something a little lighter: On the facebook IQ test (yes, I know, IQ is a flawed measurement, blah blah blah, but it's still fun), showing the IQ's of the friends of mine that have done it, the ranking is: 1. me - 147 2. Bertie - 146 3. WP - 145 4. Arj - 144 5. Rogan (not Rogen), one of my old school friends - 143 6. Ian - 143 7-9. Ben and Aaron, friends from the geek club (science club) and sasha, friend from choir - 141 10. Another Ben, also from choir. 11. Detrusor - 138 12. Damien, a friend formerly from that choir - 137 13. Joni - 137 14. Gergely - 137 15-28 - all friends in person - lowest is my brother on 112, another school friend on 114, everyone else is above 120. It says a lot that my minesweeper friends, making up so little of my friends list, dominates the IQ list so well, making 5 of my top 6 and 9 of my top 14. It's not like my friends are unintelligent by any stretch of the imagination. I just invited all the rest of the sweepers on my facebook to use the IQ app. Anyone else, "website" is the link. |
![]() ![]() |
|
![]() |
|
Dec 6th 2007 at 02:29:51 PM |
|
Name: |
Cristian |
Best expert: |
65 |
Best intermediate: |
16 |
Best beginner: |
1 |
Comments: |
time and then rqp are my preferred indexes. By the way, I came here to announce a test of writing fastly... in which position are you? There are around 250000 possible positions at this moment hahaha, see my website url. (I found it looking in google) |
![]() ![]() |
|
![]() |
|
Dec 6th 2007 at 02:23:59 PM |
|
Name: |
Cristian |
Best expert: |
65 |
Best intermediate: |
16 |
Best beginner: |
1 |
Comments: |
it would be interesting to rank all aspects of the game (rqp, efficiency, ios, time, 3bv's, etc), but if it's too much work then you could restrict them to some requirements of sum of times, for example |
![]() |
|
![]() |
|
Dec 6th 2007 at 11:58:43 AM |
|
Name: |
joni |
Comments: |
quote from Michael "In my mind, the best ever list is a ranking of the best ever times, not necessarily the best ever players" i think that would be a very interesting ranking! ![]() ![]() BTW michael, i see you refer to Damiens ranking. The discussion last year focused on the BestEver, which was sort of the most well mantained ranking, and a bit closer to the IMC. This year, BestEver has gotten into a coma, and Damien's updates have become much more frequent. Besides them there's even a third one, the IMC's own (trying to incorporate both rankings and using a ranking style similar to BestEver). Damien has tried almost all possible methods of ranking on his, at one time or the other, always changing at his own discretion. the method he's using now, as you point out, is quite confusing, having decimal times and summing the integers... besides that winmine games are not represented correctly. If a winmine game has ended and the timer displays 41, it should be represented on the ranking as an integer SCORE of 41 (writing it down as 41.00, or 41.99 adds fake precision to it) anyway, that was to remind that nothing bounds Damien to follow any suggestions from the IMC, and this discussion is not about that ranking. |
![]() |
|
Dec 6th 2007 at 11:41:01 AM |
|
Name: |
Thomas Kolar |
Comments: |
I have to say that I absolutely agree with michael, and that I dont agree with berties resolution argument - if the resolution is so high that it is indistiguishable from absolute precision, you can seperate two arbitrarily close players. This is of course in practise impossible. BUT... In a ranking where multiple players with the same total are to be expected, the resolution is definetly too low. I agree that from a certain point onwards we have sufficient precision to seperate at least most players. So we must also ask ourselves how many decimals we need. And getting back to what christoph posted about decimal accuracy, maybe we should use one decimal. That would seperate most ties in the ranking, but still not add too much "information overkill". Another reason I don't like integer scores is that in one second you can do a lot, a second is a period of time within the game where the mouse can move all the way across the board, and open a couple of 3bvs to boot. In a tenth, you can either move the mouse a little, or maybe open a 3bv if you're fast. In less that a tenth you can't do a lot more. So, a couple of hundredths more/less make little difference, I think we can almost neglect hundredths. We should IMO not, however, neglect tenths. On a last note, sorry for writing so much. I just really wanted to make all that public, and also explain why that was my opinion - which I don't think I did well at. So I think I will reformulate my ideas in a more clear way over the course of the next year in the shape of an article. Hope a few people were able to make sense of it nevertheless! One last note on how I stand on other issues: - I am for registering the IMC/making it official. IMO that is important if we want to get sponsors/organize international events, and the like. And it does no harm. - I am for informing everyone of the IMCs' discussions/votes - however, I think that there must be non-public discussions as well. For example, discussions about vulnerabilities in clones should only be open to the IMC, and the creator of the clone. I say make everything open that can't be misused to harm us by cheaters. For more information on my opinion, email me, I am happy to answer questions. ;-) And as I think I've probably said everything I wanted to say - happy sweeping, and good luck to the other candidates! |
![]() |
|
![]() |
|
Dec 6th 2007 at 10:18:46 AM |
|
Name: |
KAmil |
Best beginner: |
3000 sub5 |
Comments: |
1x7 2x218 3x1210 4x1568 ![]() I think that discussion about decimal ranking starts to be boring ![]() |
![]() |
|
Dec 6th 2007 at 10:08:26 AM |
|
Name: |
WP |
Comments: |
Good point, Gergely. But no deal :P The reason is that I find decimal rankings ugly. And I just thought of this: it will also make people try to beat their record by 0.01 second instead of passing from X to X-1. And that, to me, spells "argh!" (and I mean only to me, it isn't meant as an argument) And Bertie: No, I'm not going to throw anything on your head. Not this time. ![]() |
![]() |
|
Dec 6th 2007 at 09:50:57 AM |
|
Name: |
Bertie |
Comments: |
Okay, I've got a few things to say which might be trampling a few toes, so take it with a pinch of salt if you feel offended. I think WP hit a very true point. The extra decimal information is, in a sense, "information overload" but not at all in the way Micheal interpreted it (You completely misunderstood WP point) The problem isn't that we will have to process more info, the problem is in the idea of "accuracy". Everyone is throwing the term around, but we don't even exactly know what we are measuring. Yes we are measuring time, and we all know that this reflects skill in some unknown way, but that's about it. We can't even define skill! This leads me to the conclusion that to change the ranking from an integer to decimal system would only make the ranking work more fluently and give ABOLUTELY NO greater accuracy to the ranking. It would be like changing the resolution on your PC screen, you can't discern any increase in detail because our eyes aren't fine tuned enough (And please, don't take this resolution thing any further. Read my next point) Thus I would say that the only merit of changing the ranking to decimals is that it would eliminate the quite valid problem Schu brought up, and NOTHING further. This next bit is on the art of constructive argumentation and isn't meant to offend, but only as constructive critisism. @Gergely: There aren't two groups of people in this discussion. As Einstein said: Simplify the problem as far as neccesary, but no further. Not one among us can say that they don't care about their ranking. And not one among us can say that their ranking is the alfa and omega. So let's not fall into the trap of starting to talk of "them" 's. That's how war's are started. (I know for a fact that that is how apartheid started (the bad part of it )) Okay, I'm gonna go get my helmet so long for the rain of backlashes I'm probably gonna get for this one. ![]() ![]() |
![]() |
|
Dec 6th 2007 at 06:58:29 AM |
|
Name: |
Michael |
Comments: |
And "either side", not "other side". Crikey, bad day. |
![]() |
|
Dec 6th 2007 at 06:57:23 AM |
|
Name: |
Michael |
Comments: |
By "higher scores" I meant "better scores", obviously ![]() |
![]() |
|
Dec 6th 2007 at 06:56:08 AM |
|
Name: |
Michael |
Comments: |
WP - you say "precision is good if you like it, but otherwise it's just information overload" My reply to this would be - look at the current rankings! It's already displayed in decimal! So a decimal ranking requires no more information than we already have displayed. It simply makes better use of it. And Gergely makes a good point. If people really don't care about whether they are, say, 33rd or 37th, then why would they mind a change in system? In my mind, the best ever list is a ranking of the best ever times, not necessarily the best ever players. It seems crazy to me that someone can improve their best time from 51.99 to 51.00 and nothing happens, yet improve by a further 00.01 and they move up a few places. The boundary between 50.99 and 51.00 is just as arbitrary as any other. We should be giving higher places to the higher scores, not the scores that happen to fall on the better side of arbitrary boundaries. I think this also deflates that argument that weshouldn't use decimal because it could be very slightly inaccurate on other side - if this is the case, that inaccuracy could surely push somebody into a different second boundary altogether, so the inaccuracy people are afraid of won't be any more than what we're used to. |
![]() |
|
Dec 6th 2007 at 06:07:40 AM |
|
Name: |
manuel heider |
Comments: |
why not just create a new ranking, call it decimal ranking an leave it to the people which one they prefer (like country ranking....) the integer score ranking stays the official, while on the decimal thing there are notes like "to old scores there are 0,99 s added" and "this ranking does not pretend to be accurate" ;-) OR just create an up to date "clone ranking"! there u'll have ur decimal ranking, old players without history file could be also in this ranking (+0,99s or whatsever) but without stats... mfg manu |
![]() |
|
Dec 6th 2007 at 05:26:35 AM |
|
Name: |
Gergely |
Comments: |
This decimal ranking debate gets interesting. For me, it comes down like this: - there are some people who want as precise a ranking as possible. They care about their position in the rankings let that be either 3rd or 133rd. - there are some other people who don't want any twist of the current rankings and they say they don't care about their position, only about absolute performance (i.e. time). So why do this latter group care about how the rankings are made? If they don't care about position, then let the others have more precise rankings so they can decide their position. Deal? ![]() |
![]() |
|
Dec 6th 2007 at 05:06:13 AM |
|
Name: |
WP |
Comments: |
You said it, Schu, SOME people define their progress by rankings. And some people don't. Which is why this issue is pretty much at a deadlock; your argument of decimal rankings making it possible to define a rank might be an advantage for you, but not to me. So about the top 10: And how many people don't really care whether they're in the top 10 or not? How many people never put it as their new year's resolution? The argument works both ways. Another example: if we want to argue that we are ranking by time and not skill, and that we might as well make it precise...there are also two points of view: precision is good if you like it, but otherwise it's just information overload. Again, it depends on each person. The purpose of this post is to show that this argument is pretty much stuck where it is. We can find lots of other examples to support our points, but in the end, it's a matter of preference. I shall try not to post any more arguments on this, because I shall certainly be repeating myself. So I'll just conclude with my opinion: Let's leave the integer rankings alone, with beginner included. |
![]() |
|
Dec 6th 2007 at 03:39:02 AM |
|
Name: |
Schu (Andrew McCauley) |
Comments: |
@ Joni: I can understand that point of view entirely - but the fact is that some other people define their progress by rankings and it is important to them. |
![]() |
|
![]() |
|
Dec 5th 2007 at 06:44:44 PM |
|
Name: |
joni |
Comments: |
thank you manu. now it comes back to me, i quote "the BestEver is simply a list of minesweeper players [with a total under 100] sorted by time"- horst2104. that's a hell of a definition, IMHO. You can use decimal if you want/can, as long as you write that at the top and don't call it a Ranking. @schu: for the reason mentioned on the previous post, I've never placed much importance to placements. My resolutions have always involved time objectives... @tommy, sorry but i can't read that |
![]() |
|
Dec 5th 2007 at 06:02:12 PM |
|
Name: |
Schu (Andrew McCauley) |
Comments: |
@ WP: It's not just me that attaches importance to being in the top 10 - every sport with any statistics at all does it, hell, every endeavour does it too. Humans just like top 10 lists. How many people have wanted to be on the top 10? Maybe even put it as their new years resolution? |
![]() |
|
![]() |
|
Dec 5th 2007 at 04:28:25 PM |
|
Name: |
Michael |
Comments: |
Ian - you're right that people play different boards and therefore it isn't a level playing field, so to speak. But think of the number of boards we all play. I'm guessing almost everyone in the top, say, 100, will play hundreds or thousands every week. It's not like I play an easy board every day, whereas someone else plays a difficult board every day. We all get some of the easy boards, some of the difficult boards, and a lot of the mediocre boards. Minesweeper is weird, definitely - like a racing driver trying to beat his best lap time, but doing each lap on a different circuit. I think the amount of boards we play eliminates that. And the only reason *not* to use decimal is if you want to get rid of time altogether as a statistic. As long as we use time as a measurement, there's absolutely no reason to be *less* accurate than we can be. Sure, a 42.2 might be less skillful than a 42.6. But a 61.9 might be more skillful than both. That's just the nature of the game. |
![]() |
|
Dec 5th 2007 at 03:16:15 PM |
|
Name: |
Bertie |
Comments: |
Sorry Tommy, but I kinda miss what you're trying to tell me. I think I might have misrepresented my point a bit, but enough for now. I do however have something to say about the skill in rankings thing. If we were to compare minesweeper for a moment to sprinting, this is how it pan's out: A sprint race, like the olympic final, isn't comparable to the normal conditions in the minesweeper landscape for exactly the reasons stated earlier. In a sprint race all the competitors are subjected to the SAME conditions and the one able to perform best in those conditions will win. The only equivalent case in the minesweeper world is minesweeperlive.com, where two players pllay identical boards and the winner is the one who finishes the board the quickest. What we represent on rankings is then something completely different. The times given on the rankings are the times reached when all the factors worked best in favour of that particular player. I don't mean just luck, I mean things like a players 3bv niche, mine being between 140 and 155. This is the range where I get my best times usually. If we look at two times on a ranking like the ones you guys have mentioned, let's say a 50 on a 140 and a 50 on a 195 board, these are the times these two individual players were able to get in a game where all the factors worked in their favour. Put another way, we can't say that the player with 50 on 140 is less skilled than the player who got 50 on 195 because if we where to give the the 50@195 guy the same 140 board (in some way ensuring that he has no UPK) he would probably not score as well. This is because his style is more suited to higher 3bv boards (among other things). If you look at my argument above it becomes apparent that the way we rank at the moment (by best time) actually provides a leveler playing field than what we have in sprinting. As example, take the personal best times of all the competitors in the last olympic 100m final and see how they stack up. Chances are that the outcome will widely differ from that of the final race. This brings the minesweeper tournaments into the discussion. These tournaments are a mix between these two extremes where we try to get the best possible (5) times on a smaller variety of boards. (I was lucky to get a board on which I could break my record in Budapest ![]() anyway, that's the can of beans. Please tell me if I'm typing too long. BTW Isn't it kinda strange that we keep stroking our ego's by comparing ourselves to these top sprinters. Perhaps I should state that all the sprint races I mentioned earlier are actually highschool athletics races ![]() |
![]() |
|
Dec 5th 2007 at 02:15:50 PM |
|
Name: |
WP |
Comments: |
Oh wow...I read quickly through all the posts, but I don't believe that there are many new arguments (we've said everything in the past year or so :P) Besides, I like what Kat said: "Change the game too much and you will have no claim on the name Minesweeper." And it brings me to what I've always said: Let's keep things simple. @Schu: Why attach so much importance to the top 10? 10 is just a number we use often. In a case where several people are tied for 8th, 9th, 10th, ...15th place, we can always say that there is no top ten, but there is a top 7. |
![]() |
|
Dec 5th 2007 at 02:04:48 PM |
|
Name: |
Christoph |
Comments: |
Who else thinks, it would be a good idea to have a maximum of characters/IP/day? ![]() |
![]() |
|
Dec 5th 2007 at 01:26:31 PM |
|
Name: |
Jon S |
Comments: |
I just scored 61 on a 144 3BV board. The board is ok, but it had 17 openings. I had a nice int board that I should have finished in 13 or 14 seconds, but I missed a square and didn't find it before the timer reached 16. I compensated for that failure by scoring 15 seconds on a 50 3BV board. Some interesting discussion. I'm afraid I didn't get to read all posts, since the guestbook doesn't fit in my browser now... |
![]() |
|
Dec 5th 2007 at 01:05:52 PM |
|
Name: |
Thomas |
Comments: |
I absolutely agree that to really rank by skill, time is suboptimal. However, what is skill? The problem is that skill is something highly subjective. What defines a skilled player? Low times? high 3bvs/s? good completion? high efficiency? solving capabilities? good reflexes? high clicks/second rate? ![]() The problem with skill is that it is neither measurable nor definable, because everyone has a slightly different definition of skill, and because it is emergent, meaning that it can not be evaluated without taking into account every factor that contributes to a players skill. Also, players can be suited better or worse to certain boards or situations, NF/F being a classical case, for example on some boards efficient chording can be much more effective than on others,where NF is actually quicker than F. Int vs exp being an extreme example. I would like it if there were one, but I'm afraid I do not believe there is a stat that reflects skill in a way that is equally satisfying for everyone. If we find one, fine... :-P On the other hand, imagine winmine hadn't come with time stats, but stats that rewarded efficiency, completion, and so on, but not low time. Would our definition of skill be the same? Most probably not. We would frown at every click that was not thouroughly planned/had no purpose. chording on squares before we even had time to see what number was on it, just because the chance was high that it would open more squares and speed up the game would have been noobish. Being able to get through the game fast might have been something that was appreciated, but replaceable by other qualities and a matter of style, as efficiency is today. Maybe we would have needed clones to give us time, not 3bv. we might have discussions among the lines of "OK, we have two players. The first solved a 183 3bv board with 181 clicks, and in 75 seconds, the other only needed 180 clicks but took 100 seconds. Isn't the first score a lot more impressive? Let's round 3bv to the nearest ten." ![]() ![]() And dion would get straight IOEs of at least 1.01. I think that the definition of skill we have is strongly based on being fast, primarily in terms of a low time, but also in terms of 3bvs/s (which I think is why it is the second most popular score stat.) So, when defining skill, we actually have two options: Defining skill the way I demonstrated is hard to do, and the way I understood most people here define skill is something like that, the ability to play impressive/difficult feats. There is also the possibility of defining skill as the ability to get a high score as defined by the rules of the game. And winmine does not make a lot more clear than that, the object of the game is to clear a board in the least possible time. So according to that definition, yes, 49.99@100 is better that 50.00@250. And that is IMO not quite as bad as it seems. In the example above, I would consider it very strange if the second player didn't get a low sub50 time after a while, while player ones' score is pretty likely to be a fluke. Also, remember that some players actually are better on medium low 3bvs than very low 3bvs. One mans garbage can be another mans treasure, and that goes for boards as well. But board luck distribution fortunately levels out over time. Board difficulties may vary, but on average and in the long run, we get the same chances at a good game. Ian, try normalising boards in a way that is completely accurate. IMO it's not possible, because board dificulty is subjective. daniel, what does it change about gameplay exactly how time is displayed? In my opinion, that is entirely part of the interface. And we have changed the interface of clones compared to the original by a long way. think of counters windows. I would say tenths/hundredths in the timer are a thousand times more problematic - here, the timer moves permanently, which adds a psycological factor. Gergely, why limit the area for which scores are to be decimal? If it helps at a high level, why shouldn't it on a lower level? Bertie, the distribution thing is an interesting point, the problem as far as I can see now is that the curves don't have to be between two integer times or anywhere where categorizing everything between these two points necessarily makes sense, they can be on arbitrary points on the "time line". and that means that while some distributions will be sliced exactly through the middle. I hope you can follow me on that bertie, and apaologize for not formulating that more clearly - I will have to take a look at that at school. On another and completely unrelated note, I think that there should be clear standards as to what criteria there are for clones that are to be accepted. I propose: -mimic the behaviour of winmine as accurately as possible -save videos in an encrypted format that should not be made public, but known to the IMC. -In general, the IMC should have access to all specifications of the program, including all parts of the source code. -By contrast, the source code should not be made public, because this would enable cheaters to find vulnerabilities more easily. -there should be a way to watch videos played on the clone without downloading it. Especially regarding the format of the clone from 2006 onwards. I will risk getting emotional now, but that absolutely piszes me off and is unnecessary. I do not want to be limited to youtube for world records (even though I must say that schu does a great job in maintaining that, and that I am extremely thankful to be able to see them at all.) Sorry manu ![]() And LOL @ VC -think the guestbook is trying to tell me something? VC=MU7E |
![]() |
|
![]() |
|
Dec 5th 2007 at 11:44:45 AM |
|
Name: |
damien |
Comments: |
i agree with kat. (nice to see you back!). manu...gregoire gave me permission to start clone rankings and sent me his database, but joni and i didn't have time to write all the code from scratch (and there were some errors in the database). it is too bad we have missed this for 2 years already. i would love to get it going again. |
![]() |
|
Dec 5th 2007 at 11:15:57 AM |
|
Name: |
Katrina Cassar |
Best expert: |
61 |
Best intermediate: |
16 |
Best beginner: |
2 |
Comments: |
Change the game too much and you will have no claim on the name Minesweeper. I believe the IMC should pay more attention to endeavors like Gergely's getting a recognised organisation and Andrew's GBOR. These will not take obscene amounts of intellectual debate, just a lot of hard work. @Gergely:If the IMC was a recognized body it would make it much easier to hold tournaments all over the world. There would be an authority to support and advise less experienced organisers. Good luck with your goals! |
![]() |
|
Dec 5th 2007 at 10:41:29 AM |
|
Name: |
manuel heider |
Comments: |
"the original goal of the game is time -> why change this??time doesn't measure skill..." as long its is clear that the BE or damiens world ranking have not that much to do with skill, we can leave it as it is plus create sth like a new clone ranking, i'd love to see these stats up to date ! mfg manu |
![]() |
|
Dec 5th 2007 at 10:02:48 AM |
|
Name: |
daniel |
Comments: |
i agree with kamil and andrey another point, for me, is that we ever tried to imitate the original minesweeper and only created clones to exclude some cheating possibilities - never to change rules. so why don't stay at this method and go on with adding a second and without ranking by decimals? if someone prefers other ranking methods he can look at other values - whatever they may be, 3bvs, rqp, ios...but the original minesweeper gave us the time to measure and i cant find a reason to chagen this. 3bv eg is a measure for the speed of boards but not for the difficulty (whichs would have to include the number of guesses i think) to put it in a nutshell, dont change anything, everything has been well for some years (i believe, im not that long her ![]() |
![]() |
|
Dec 5th 2007 at 09:27:07 AM |
|
Name: |
Ian |
Comments: |
In a sprint, all racers compete under equal circumstances, which may not be equally favourable to all (perhaps some work best at different temperatures for example) but there is a good chance of some time/skill correlation. In minesweeper, this is not the case, because players play different boards, so we need to normalise this factor to get a true representation of skill. Of course the original goal of the game is time, but the original goal of any 'world ranking' is to rank in terms of skill. |
![]() |
|
Dec 5th 2007 at 07:20:16 AM |
|
Name: |
andrey |
Comments: |
the original goal of the game is time -> why change this??time doesn't measure skill, so we could develop a complicated index that takes into account 3bv, luck etc but who's gonna understand this? even if you take 3bv/s. can you imagine sprinters ranked by their time divided by their body mass index? why not keep simple things simple? |
![]() |
|
Dec 5th 2007 at 06:42:52 AM |
|
Name: |
Schu (Andrew McCauley) |
Comments: |
I'm ever so slightly drunk, so forgive my manner... Joni: Well, what you say is possibly true, but you might as well say that Dion might not be in the top 10. It really doesn't matter that there might be x many people with scores of whatever - first of all, if they're not there to be ranked, they don't count, and secondly, most people only get REALLY good scores AFTER they find the rankings. And can anyone justify not having 10 people in the top 10? |
![]() |
|
![]() |
|
Dec 5th 2007 at 04:15:36 AM |
|
Name: |
Gergely |
Comments: |
Kamil has some good points about this topic, I think. IMHO the decimal scores would be necessary in case of scores below 45-50 (arbitrary limit to be set). In this range small differences do tell the story, i.e who were the faster? Skills cannot be evaluated by this properly, I agree, but the world rankings don't have such purpose. That's better done at tournaments. The beginner debate is interesting, but I think it has too much of a luck-factor that it should be excluded from rankings. Anyone can get a lucky board here and 3bv-limits do not apply for scores before IMC. So that would do the same as decimal scores so that early scores would be excluded because those do not comply with more recent technical features. |
![]() |
|
![]() |
|
Dec 5th 2007 at 04:03:22 AM |
|
Name: |
manuel heider |
Comments: |
two words ... @beginner discussion: if beginner is excluded, i suggest to use 17x15 for int with 43 mines, 17x31 with 103 mines for exp ... why ? -> why not ? ( i think that this problem should not be decided by the IMC alone, it would be a too powerful desicion^^) beginner is part of the original game as i said years ago AND it needs its own skill and style like int and exp do . but this skill isnt displayed on rankings like BE or damiens, so imo this discussion goes in the wrong direction. i do NOT care about rankings sorted by best time because they say NOTHING about skill/luck/finish rate ... we can not measure the skill of someone. we can get an idea of someones skill by watching his/her videos, look at the style and the finishrate, are the finished games a normal graph or is there one (the highscore) that doesnt fit to it... so imo theres no need to discuss wether to use decimals or not, because it both makes no sense to me. the question that must be asked is: how can we rank people fair? this is a very difficult question and never cant be answered correct, so we need an approximation, and now the task is to find one.(if we realy want to rank fair...) i personaly prefer sth like the clone ranking with lots of stats, people sorted e.g. by time to give a direction, but not to rank them. @newbs discussion: when i was new to minesweeper i only looked on the ranking and thought sth like oh, they are **** fast^^ and not whos ranked where and why and so on, i only knew theyre around 40 and thats all. @thommy please try to keep ur posts short, i doent have the time to read them all^^ mfg manu |
![]() |
|
Dec 5th 2007 at 03:24:35 AM |
|
Name: |
KAmil |
Comments: |
If we want to have decimal ranking, we would must to remove all players who played on winmine. Of course we can't remove them, so we wouldn't have decimal ranking. No decimal ranking => no problem with beginner real time => Beg won't be removed from rankings ![]() |
![]() |
|
Dec 4th 2007 at 08:28:46 PM |
|
Name: |
joni |
Comments: |
it's 5 a.m. i'm back from a party with the student theater group i'm part of. We had our premiere last night, and this year I had the extra responsibility of being the lighting technician. It was simply amazing!!! That is the main reason why i have no free time, and why i do not intend to run... @Schu: it is even possible that none of them is in the top 10... I remember you mentioning you were once in a three-way tie for 98th place, and you were angry you couldn't say with absolute accuracy if you were in the top 100 or not... Well i can tell you that you WERE NOT! If you ask the same question today (you said your position was 44th in some list) I'd say it is quite probable you are in the top-100 in the world... but it might quite well not be so... (there are over 6 billion people on this planet). I liked the way Damien used to say, there are X people KNOWN to have sub-50d ;D I'd really like to hear two words from Manu in this guestbook, preferably a response to Tomas :D (i can tell he's quite good at reasoning when he's not spamming with his games...) P.S. I'd like to hear Banzhafs ideas also! /joni. always a good memory for remembering useless things. |
![]() |
|
Dec 4th 2007 at 05:44:23 PM |
|
Name: |
Bertie |
Comments: |
I say it again: I'm loving this! Everybody is giving their unbridled opinions while KAmil just keeps on breaking his records!!! ![]() ![]() Just a point of a more technical nature. I think it was Tommy who mentioned that luck's beneficial effect is spread 50/50, sometimes giving the better player the edge, sometimes the worse player. This is true. It is however not the whole story (and quite frankly not the important part of the story). Let me draw you guys the picture I have in my mind pertaining to this little dilemma. If you where to graph the extent (it doesn't matter exactly how you measure it) to which a player is helped (+) or hindered (-) by luck, you would probably get something akin to the Normal function. This is because of the statistical nature of the luck beast. (Google "Normal graph" for a picture, it looks like a bell) The important part of this is however not that it is balanced, but the important thing is the spread of the graph (go to answers.com for lingo). The "spread" of our luck help vs. occurence mental graph is the thing which we are trying to better reflect skill in a ranking. I hope that someone has been able to keep up with me. Having said this, I have to also say that I don't know whether us using integer scores is masking the problem, or inflating the problem. Thus, I'll stick to my previous vote: Status quo on the integer thing, for now. P.S.: I agree that time should remain our metering rod. But we should however strive for an optimal representation of the time to reflect skill. To those who got lost in the lingo, I say to what was said to the great and wise Marvin: Sorry for the inconvenience. ![]() |
![]() |
|
Dec 4th 2007 at 05:26:19 PM |
|
Name: |
Schu |
Comments: |
@ Christoph: I don't recall integer time always being found the best call over and over - it's always been pretty deadlocked. Also, reviewing videos won't be so bad - you don't have to judge by total time - just by where we consider the fraction of a second to be. I think we all agreed that while camtasia did slow down the timer a tiny bit, it also slowed down mouse reaction and the speed with which squares are uncovered, so if the timer shows 52, say, and there are .37 seconds after that 52 mark on the video before the game is finished, I say we award the player with a 52.37 Like others have been saying, just because time may not be the ultimate expression of skill, doesn't mean we have to resign to it being imprecise to only the second. I mean, why not round it to the nearest ten then? As for old scores, winmine scores are few and far between and have been disallowed as new records by the IMC. Most videos we can have a look at and "guestimate" a relatively accurate decimal time. The rest, that are done by screenshots, are SO few and SO far between that I don't think we're going to worry anyone by penalising by a fraction of a second on a program that had so many bugs, some associated with the timer, and a cheat etc. @ WP: that's different, that's casually telling the time. But, for important events, like, say, new year, you count the seconds. And you're hardly going to say when someone's running an 100 metre race, that they started at 10:20 and finished at ..... 10:20.... And you really think it's acceptable that we could have more or less that 12 people in the top 10? No, we can't definitively say that anyone's faster because of a few milliseconds, but nor can we with integer time, can we, even with actual integer differences. And especially not in integer time if their scores are tied. It's just an indication, just like integer time is. But decimal time has the advantage of separating almost all times. Guys, beginner should already be a moot point. The IMC has deecided, as I said. Just no-one seemed to notice and post "oh look, there's the 4th vote". Just so you all understand, some of the people I'm most likely to vote for are the people I most vehemently disagree with - but people that use intelligent arguments. I urge you all to take the same attitude and NOT vote based on the issues, but based on the strength of their argument, whether or not you agree with it. And of course, vote based on how likely you think they are to be an active participant. |
![]() |
|
Dec 4th 2007 at 04:52:29 PM |
|
Name: |
Jason K. |
Comments: |
Thomas hit it right on the head w/ his post, if not in the most concise fashion. ![]() There really are 2 separate issues here. I think people need to forget about the argument that a .02 second difference isn't an accurate representation of who is more skilled. That's not the point. The point is that TIME is, and has been, the measuring stick for rankings. Is a 50.5 on a 160 more impressive than a 50.4 on a 120? Hell yes, but it's irrelevant. As long as "time as an absolute" is what matters most, then decimals should be used in the interest of accuracy. |
![]() |
|
![]() |
|
Dec 4th 2007 at 03:21:28 PM |
|
Name: |
Christoph |
Comments: |
Well, alpine skiing: they did take the time to thousends for a while and decided to go back to hundredths (well, more precise: internly they take thousends, but they only rank by hundreth - so why shouldn't we take hundredth and rank by seconds?) However, we discussed that over and over - and integer rankings always turned out the best call. Some of may like it, some not - but we cannot even be sure of the second decimal (since the timer precision gives app +/-0.004) so where's the point in rank by something we're not even sure of? We have way more important things to do than to waste our time on that. |
![]() |
|
Dec 4th 2007 at 02:56:20 PM |
|
Name: |
Michael |
Comments: |
I agree with Thomas's last post. If we're going to rank by time, we may as well make it accurate. You won't see *any* other sporting world records that round to the nearest second. Can you imagine that in the Olympics? |
![]() |
|
Dec 4th 2007 at 02:24:23 PM |
|
Name: |
WP |
Comments: |
@Schu: Why do we say "it is 10.20pm" instead of "it is 10.20pm, 22 seconds"? Because the extra information is useless. The same goes for decimal rankings (in my opinion) (I agree with Joni on this, as you can probably see ;P) And about your top 10 argument...I think they can all be considered as being in the top 10. Why not? Decimals will make one person a few milliseconds "faster" than another, but can we really say who's the better player? @Thomas: by adding 0.99 to scores, we wouldn't be making rankings more accurate... I don't like taking beginner off the rankings because it sends a message that beginner is not important. And each player should be able to decide by him/herself whether a level is important for him/her. (I'm only replying about stuff I feel pretty strongly about, 'cos I'm pretty busy these days...darn you people who make me pass time reading all these arguments :P) |
![]() |
|
Dec 4th 2007 at 01:40:26 PM |
|
Name: |
Ronny |
Comments: |
I think the integer times + 1 are just part of the game. If it makes people happier thinking they actually are 1s faster, that's their choice, but that doesn't change their official score. |
![]() |
|
Dec 4th 2007 at 01:27:46 PM |
|
Name: |
Ian |
Comments: |
I'd have to agree with Thomas: there's not much point fiddling around with fractions of seconds when the time rankings aren't an accurate representation of skill anyway. In terms of making a better one, I think 3BV/s is a good starting point to generalise off, basically in 3BV (as we all know, but for ease of extension): Openings are worth 1 Zeros or squares next to zeros are worth 0 Other squares are worth 1 And of course, it doesn't take very long playing to realise that this is horribly unbalanced. Areas with squillions of mines are much harder to clear than places with 1 mine per 9 squares. So, we could extend it in this way: Openings are worth 1 Zeros are worth 0 Other squares are worth how far away they are from a zero. (Good luck with coding that ![]() Probably this won't be perfect. But it's an idea, and maybe in this way a better index of skill can be established. |
![]() |
|
![]() |
|
Dec 4th 2007 at 12:34:23 PM |
|
Name: |
AreOut |
Comments: |
Doh, same old story again...keep integer times and get 0.99 off decimal ones we use now, thats it :) |
![]() |
|
Dec 4th 2007 at 12:14:06 PM |
|
Name: |
KAmil |
Best intermediate: |
NF 12x3 -> 12x4 |
Comments: |
37 3bv in 12,506 @Damien: I changed something in my bio, but not in Word, submitted and it broke down again ![]() ![]() ![]() |
![]() |
|
Dec 4th 2007 at 11:56:14 AM |
|
Name: |
Thomas |
Comments: |
I disagree christoph, there are ways to get decimal times from integer times without reviewing the vids. First possibility is to add .99 seconds, which would seem fair to those using accurate clones. It makes sense to say that with clones, we can be sure that the decimals we know are accurate. With integer times, we only have evidence that the time was less than what the timer said. We could also round old winmine times down. This makes sense because new players will not use that to their advantage as winmine is now illegal anyway, and it seems fair to give the players who still used winmine credit for getting their scores at a time when the community was younger and less experienced. Also, these players are much more likely to be retired, while new players are likely to have a chance to catch up. So I think it might be fair to give older players the benefit of the doubt. I am not sure about this yet. However, I think that not using a decimal ranking just for the reason that we can't find out every score as a decimal score is absurd. It may be that we have to use some inaccurate method to make the old scores "compatible", but I believe that the inaccuracy we inflict when we do that is a lot less relevant than the inaccuracy we have now. @luck/decimals: yes, some games with similar times may be more lucky than others. However, you can carry the argument even further. What is better, 49.87 on 106 oder 50.43 on 157? Most probably the 50, but the 49 is better in an integer ranking. So, assuredly, the integer ranking gives us a false sense of security - better to adopt a ranking where scores are approximated to the nearest ten seconds (we'd have to add ten, of course ;-)). Then again - these two scores are still in two seperate leagues (for the ranking we now have is no more accurate than a league system now) - better add another power of ten, approximate scores to the next factor of 100 (of course, again, adding 100 seconds). That is ridiculous enough, but one more step and all we have is a ranking that reflects whether or not a player has beat "anonymus" on all three levels or not. astonishingly, ranked players will still seem to have a worse total than that, by 3000 seconds to be exact. What we should accept is that what we use to determine a players rank is time, nothing more and nothing less. The stat we use to determine the skill of a player is, and has always been, time. According to this criterion, yes, 49.87@106 IS better than 50.43@157, and 55.27 IS better than 55.99, no matter what the 3bv difference is/how many forced guesses a game contains, and even if more skill may be required for it. Let me repeat this. At the moment, we do not rank according to skill. We rank according to time. If we want to continue ranking by time, we might as well do it right. If we want to abandon that system and try to find a way to rank players according to their skill, we might as well try to find a better representation of skill than distorted time. Anyone who thinks he is solving the problem of representing skill with a not entirely related stat by further distorting it to take into account the fact that a better result on one scale may be worse on the other is deluding himself, as I am about to demonstrate. The standard argument so far is a slightly worse time on a much worse board. What about a slightly better time on a much worse board? The argument does not take into account that the luck element can just as well benefit the player with more skill. Luck would here be evenly distributed, meaning that a decimal ranking will not be less accurate than an integer ranking - a decimal ranking may favor the less skilled player ~50% of times, but in the other ~50%, the more skilled player will be favored. We therefore have no overall decrease in accuracy when it comes to comparing results from two players with similar scores on boards with different dificulties. However, the more equal the boards get, the more accurate is the decimal ranking. Assuming that two players complete two boards with the exact same dificulty, but with slightly different times, a decimal ranking will always be more exact (excepting the situation where both decimal times are the same, but here the two rankings are equal, and the decimal ranking leaves us with less doubt as to whether or not players really scored equally or not). We therefore have improved accuracy for similar boards, while retaining the accuracy of the integer ranking. So, if we go for a time ranking, decimals are better, to the best sensible degree of accuracy (which at the moment is two decimal points if you believe christoph, he gave a very good explanation to me and some other sweepers on the subject, and I agree with him. The reason was as far as I can remember the fact that mouse click event processing time varies so strongly that computing more than two decimal points is impossible. I hereby entitle him to hit me around the head should my explanation be inaccurate, I do not assume to be at this point ;-)) To be continued, dinner is on here ;-). |
![]() |
|
![]() |
|
Dec 4th 2007 at 11:38:16 AM |
|
Name: |
damien |
Comments: |
I would love decimal rankings...but it is not possible unless you delete old scores or add time penalties to old scores or create a new ranking. If the ranking stays integer it should stay +1. If the ranking goes decimal, it must become -1. Beginner should definitely be included for rankings. Your Beg time is less than 5% of your sum. (3bv=1 is already banned). Schu: on a decimal ranking, your 1.9 would only have changed your rank by 1 or 2 places. I think we can now agree that it is possible for more than 7 people to have good ideas each year! The new IMC should allow people to submit opinions to it (like Joni sending his Arbiter43 investigation). Just some thoughts! |
![]() |
|
Dec 4th 2007 at 11:36:08 AM |
|
Name: |
KAmil |
Best expert: |
NF 54,484 -> 54,07 (54x2) |
Comments: |
![]() |
![]() |
|
![]() |
|
Dec 4th 2007 at 08:36:03 AM |
|
Name: |
Christoph |
Comments: |
I'm just gonna drop my oppinion on the decimal rankings: There is NO WAY to replace old integer scores by decimal ones. Even if we would get the decimals from avi files for all the games on the ranking, we'd have a major problem: We would certainly find out, that some of the scores are 1 or 2 secs slower than belived - so we have no guarantee at all that this game is the actual record of this sweeper (and may be he has another equally good game we're just not able to the the vid of). The timer slip may even have denied this sweeper a better score - and we'll never find out that he would have a better score, but we could have to give him a worse one... No. Scores which were previously accepted can only be changed if they turn out to be obvious cheat or very very wrong. Moreover I think the point of view Joni described is very right. |
![]() |
|
Dec 4th 2007 at 06:37:50 AM |
|
Name: |
KAmil |
Comments: |
I have 2,333 on 7, so Oli's 2.2 on 6 is ok ![]() |
![]() |
|
Dec 4th 2007 at 05:44:26 AM |
|
Name: |
Bertie |
Comments: |
WOW! I've honestly never seen the guestbook so alive! I think, in light of this, a new IMC should be elected every month!! ![]() Just kidding. But I really am loving this heated debate. So her's my 50c. @The +1 issue: Tommy made a very strong case in favour of it, and WP an co. a strong counter (no pun intended). This said, I have to reluctantly agree with the "realtimer's" on this one. Even considering the hassle involved, in the long run changing the timer is the best thing to do. It's like ironing a shirt: No-one will really notice the difference, but you'll know you've got a nicely ironed shirt on. The newbies won't care about what system is used in a ranking, because to someone getting a regular 160 on expert will be equally impressed, if not more impress, by a 49rt and a 50. Also, we are in a good time to change it because the WR on expert (the big cahuna) isn't near a seudo-important barrier, meaning that we're not giving Dion a sud30 by doing this, just a 36. Int of course is another matter, which brings me to an issue not yet touched in this discussion. In the history of the game many barriers have been broken, two of the more significant being the sub50 and sub40 barriers on exp. Changing the timer could have a huge effect on who was the first to break the barrier. Same for int and sub10. @begginer: I think beg should stay. Yes, it requires more luck than on other levels, but it has a much smaller effect on a players sum. On exp, typical jumps are in the order of 1 to a few seconds, which has real tangable effects on the players position on the ranking. When you go down the order to int and beg games, the level's effect on the sum goes down in quite good proportion with the luck factor going up. This said, Schu has a good point on the top ten thing. I however think that the solution should be found somewhere else, because the effect of this is a bit too global for my taste if you consider the drastic drop in the amount of people who would play beg. This doesn't sound like too bad a thing, but it would most probably lead to a much lower average exp and int score on the rankings, which I think could serve as demotivater for newbies. I think Schu meant that the inaccuracy would be Less. Which I don't agree with. If you consider all the variing factors involved in a player's sum, then one quickly gets the feeling that all these factors don't add up linearly. (Don't take the add up too seriously, but take linearly seriously) This means that the loss of beg times in the ranking would most probably have quite a different effect. @rankings: I vote to keep them integer. I agree with Joni on this one completely. (For my opinion, refer to Joni's post coz my fingers are tired now! ![]() |
![]() |
|
Dec 4th 2007 at 04:42:35 AM |
|
Name: |
Schu (Andrew McCauley) |
Best expert: |
52 |
Best intermediate: |
14 |
Best beginner: |
1 |
Comments: |
@ Joni: I know it is inaccurate with or without beginner, but: 1. it is LESS inaccurate with beginner. 2. It is better for the game of minesweeper to ignore beginner and not have people obsessing over getting their stupid 1 which always needs a huge amount of luck anyway, much more than the other levels. If people like beginner, there should be another ranking to facilitate their scores being displayed in some way, but no on the main ranking. With decimal timing - first of all, we time to these decimals, so why not use them? What actual HARM does using the extra precision do? A "false sense of accuracy"? No. Everyone knows luck is involved. Increasing precision would not change that knowledge. What good does it do? Well, simple. It separates those with similar scores. This is important for many reasons, not the least of which is, imagine 5 people were tied for equal 8th spot (a very possible scenario with integer time). Which people can actually claim to be in the top 10 (and how many people have wanted to be in the top 10 in the world? Honestly? Even written it as a new years resolution?)? Either they all can, in which case there are 12 people in the top 10, none of them can, in which case there are only 7 in the top 10, or we find a way to separate them. and my suggestion is that decimal time is the best for that. @ Kamil: If I remember rightly, he had a very low 2 (2.2 I think) on a 6 3BV, which is pretty impressive, even if not a world record. .................. I got another 15 NF not long ago, and am getting decent (and really strangely efficient, even for me) expert scores. |
![]() |
|
![]() |
|
Dec 4th 2007 at 03:54:04 AM |
|
Name: |
KAmil |
Best intermediate: |
4,186 -> 4,44 !!!! |
Comments: |
on 92 ![]() |
![]() |
|
Dec 4th 2007 at 03:34:24 AM |
|
Name: |
KAmil |
Best expert: |
3,117 -> 3,612 !!!! |
Comments: |
223 3BV in 62,742s ![]() ![]() ![]() |
![]() |
|
![]() |
|
Dec 4th 2007 at 03:19:01 AM |
|
Name: |
joni |
Comments: |
@flashback.ru: hahaha, typically russian to always overfulfil the plan. Championship Done= 108%. ![]() ![]() i have been part of this debate before, but i feel forced to write (as clearly and shortly as I can) my opinions on the issues for those who haven't heard them @decimal rankings: IMO they give a false sense of accuracy to the rankings. we are trying to rank the BestEver sweepers by using their fastest time on ONE particular game. this is the method we have chosen, although it is dependent on many other factors rather than sweeper skill (guesses?, openings?, moonlight?,football match? who was on the phone with you...etc). Example (three sweepers beat their records): Sweeper A gets his 4th 55, 55.47 on a 111 3BV board with no guesses. Sweeper B gets a 55.27 on a 150 using 7 unforced guesses, Sweeper C gets a 55.99 on a 147 with 3 forced guesses. can we say with absolute accuracy that B is better than A who is better than C (or faster or more skillful or whatever adjective you want to use)? By rounding all to 55, the current ranking adds a bit of a blurr saying: "all three are capable of getting at least a 55, that's all I can say". @beginner: It is one of the three original levels of the game. I guess that's why it plays it's part on the (overall) ranking system we use. (@ Schu)if I remember correctly your expert was at 58 for a long time (6 months-1 year?) and you fell back 30 places. and then you got a 52 or 54... and jumped 30-50 places ahead. Does it mean you didn't improve your skills for one year, but suddenly improved them (became a faster sweeper) overnight? your ranking IS INACCURATE with or without beginner, you have to accept that, removing it won't improve anything. @Tommy: there are already way too many rankings and counters out there... @+-1 i don't have any strong opinions on this. |
![]() |
|
Dec 4th 2007 at 12:11:27 AM |
|
Name: |
Arsen |
Comments: |
@Michael +1000 Just too much mess in changing the +1 thing. For newbies twice as much. That's the main point. We definitely don't need any "new" or "alternative" rankings, it just leads to less transparancy and hence less attractiveness of the game. |
![]() |
|
Dec 3rd 2007 at 11:35:28 PM |
|
Name: |
KAmil |
Comments: |
@Schu: "Oli should be given regognition for some of his 2's that were on quite high 3BV" how high 3bv? I have 3x 2on10 and 2x 2on11, most times NF ![]() @Rankings by decimal sum: sime old players like Lasse would have problems, because they played on winmine. Lasse's 41 could be 40 real time or 40,(9), so If we want to have decimal ranking we would must write 40,5 to him ![]() ![]() |
![]() |
|
Dec 3rd 2007 at 07:18:56 PM |
|
Name: |
Michael |
Comments: |
Re: the +1 thing. I imagine it was put into the original minesweeper because the time was rounded down to the nearest whole second. So as such, it would be possible to get a zero second time on beginner. But now that we all have more accurate times, it shouldn't be an issue. We can see that what minesweeper tells us is one second, is actually 0.7 or whatever. To be honest, I don't think it's a big issue either way, but changing things would be a lot of hassle - not to update the rankings, but to let everyone know the new procedures. And any newbies using the minesweeper that came free with their computer might not realise the differences. I can't really see that it's something in desperate need of change. |
![]() |
|
Dec 3rd 2007 at 06:09:39 PM |
|
Name: |
Schu |
Best expert: |
52 |
Best intermediate: |
14 |
Best beginner: |
1 |
Comments: |
@ WP: as a counter-argument, those unofficial rankings will already have scores missing/old scores etc. so it's no big deal that some might still have times +1. We're already thinking "why did Donner and Johnson add the +1 in the first place" so in say, 3 years, we'll probably be thinking "why didn't the IMC change it" if we don't change it. BUT!!! Beginner! This is one of the few issues which the IMC did actually decide, though I didn't realise it until now. There were 4 votes for "remove it entirely", 2 votes for "keep it as it is" and one abstain (grr, stupid abstained votes). Don't get me wrong, I'm all for a beginner ranking, maybe even an official one, but not alongside other levels. Beginner skill can't be judged by lowest score - for example, Tam should get recognition for his ridiculous number of 1's. Oli should be given regognition for some of his 2's that were on quite high 3BV. When I got my 52, I got to #45 on the record list, but had I not gotten a 1.987 (.013 seconds from being a 2) in which most of my clicks were barely a pixel away from a mine, and I got a 50/50 guess right, again by a pixel) then I would have gone all the way down to equal #55. Fair that I beat these people on such a technicality? No way. I know other levels involve luck, but not to the same degree. Not even close. |
![]() |
|
![]() |
|
Dec 3rd 2007 at 05:49:11 PM |
|
Name: |
Michael |
Comments: |
I decided I might as well run. Nothing to lose, and I think I have the enthusiasm to complement the greater knowledge and experience of others. |
![]() |
|
Dec 3rd 2007 at 02:21:19 PM |
|
Name: |
WP |
Comments: |
sigh sigh sigh...we've been through this so many times before! In the end it's just a matter of preference. Those who love playing with numbers will want "real time" (I'm putting it in quotation marks 'cos there is no 100% precise time anyway), those who want to keep things simple (like me) will prefer keeping things as they are. There are no clear advantages or disadvantages either way; what someone names as an advantage could be considered a disadvantage by someone else! I think my best argument for NOT using "real time" is that you will NEVER be able to change ALL the rankings in the world. You might be able to change all official rankings, but when a player stumbles upon a ranking, he doesn't want to have to check which system that ranking uses! (I know I don't...) @Thomas: You can see how complicated "real time" is just by looking at the length of your post! ![]() ![]() |
![]() |
|
Dec 3rd 2007 at 02:12:22 PM |
|
Name: |
damien |
Comments: |
@thomas: see www.minesweeper.info/tournaments/ranking.html for a tournament ranking (the individual tournaments won't be finished until next weekend, and I haven't added Moscow yet, and I'm allowing ViennSweeper scores to count). Subtracting 1s is simple for my site..although I feel funny saying I have 39.34, because I have used the old way so long ![]() |
![]() |
|
Dec 3rd 2007 at 01:49:27 PM |
|
Name: |
Rupert Foggo McKay |
Best beginner: |
2.270 |
Comments: |
New best beginner time, STILL not a 1 second beginner despite trying since before the big bang. Nonetheless I'm currently standing at 106th in the world and that one second shaved off (previous beginner: 3.13) might just be enough to get me to 100th or better. |
![]() |
|
Dec 3rd 2007 at 01:14:08 PM |
|
Name: |
Thomas Kolar |
Comments: |
@ time+1: I don't see what is so complicated. You say that a ranking without the additional second is incompatible with rankings now. But that is simply not true. Of course a score that is a 51 in terms of now would be a 50 in terms of then. But... There is absolutely nothing that stops us from converting those old scores to realtime scores. We do not have to kick out any old scores, we can simply take the old scores and subtract one second. No rankings would have to be broken. And the one second does matter. Not if you are linearly comparing times, because the additive constant does not influence the difference between two scores if you compare them. But for 3bv/s we are in fact already using realtime, because here time is used as a factor, meaning that an additive constant such as the additional second makes a difference. If we calculate 3bv/s, we subtract 1 sec from the timer time and divide the number of 3bvs in the board by it. And if we calculate estimated time, we take the timer time, subtract one, use it, and add one to the result. If we have to build cosmetic additions such as these into our formulas to account for the timer time not being the time in which the game was done, it should be appearant that what we use to measure time is a misrepresentation of the time we actually took to play a game. We don't need more rankings/new rankings for the purpose of distinguishing between time/time+1 scores. We just have to convert the old ones. Of course we can continue to use time+1 in IMC-endorsed rankings for a while, but IMO time+1 is a design error that should be designated as deprecated and given up sometime in the near future for all rankings that are to be accepted by the IMC. Of course we can have two different measurement standards, one being the other one +1 - but why the f***? It is absolutely redundant. If you ask me, having time+1 as a measurement for time is complicated. You could argue that it costs time to convert/adapt to the new standard. So where would it cost time? You would have to update rankings - if scores are stored in a database that is designed in a reasonable way, I could write a 10-liner in python that does that in less than half an hour. For rankings that are stored in HTML format - regular expressions are there to deal with situations exactly like that. Clones that deal with old videos should have no problem displaying these videos according to new standards. And everywhere where conversion is impossible, one could add, for example, a -rt (for realtime) suffix. For example, as video files often have time in their filename, files with realtime times in their filename might display the time as xx.xxrt (instead of xx.xx). Hands up everyone who thinks that this would obstruct him, but give a reason why. In my opinion, realtime is a game stat, and realtime+1 is the current interface we have to it - all we have to do is change this interface, which might be something to get used to, but definetly nothing major. Also, clones could offer the option to display times the old-fashioned way for nostalgic players. However, the rankings should be in realtime. Why? You could, at this point, of course ask what good converting would do even if it is not much of a change. My answer is, precision and transparency. Transparency for non-community members, displaying time as realtime+1 is counterintuitive. You can argue about precision. There is often the argument that we can't have absolute precision anyway. My answer is that while there can be no absolute accuracy, that doesn't mean that we can't try to get as close as we can. Adding a second to the time we measure is not accurate. There is the explanation that time+1 represents the time IN which the board completed, representing the minimum possible value the time could have had in terms of the smallest units used for measurement. However, this definition is flawed because if this were the way the winmine timer would operate, excact integer times would be themselves and not themselves+1. A 1 3bv board on winmine solved on the first click yields a one second time and not 0 seconds (try it, make a big custom board with 10 mines). So, face it, winmine adds one second. Another good reason not to add the second: If I as a newb got a 99.xx realtime on expert, I would not want it to count as a 100 if I knew the time was actually sub100. And this counts for every time barrier there is... @ rankings: I think that the IMC should offer many different rankings, one of them being the classic ranking we have today. But apart from that, we could have for example int+exp, beg,beg9*9,classic+beg9*9,exp,int, each as an own "discipline". Also, a championship ranking would be interesting if difficult as truly international tournaments are impossible at the moment because players have to pay for trips themselves. VC=H4FT |
![]() |
|
![]() |
|
Dec 3rd 2007 at 12:12:40 PM |
|
Name: |
Tibor |
Best beginner: |
1.958 ---> 1x2 |
Comments: |
today at university while i was showing how to play ms to my friend i find a 2 3bv board so i could get 1 sec again... in few minutes i reached every times 1,2,3,4,5,6,7,8... cool, but i learned that beg 9x9 is too easier than beg8x8... wr now already counts decimals(comparing expert times)... if someone want more precision is possible to compare arithmetic sum between exp and int,considering beginner(8x8 if possible) score only with its integer part... |
![]() |
|
![]() |
|
Dec 3rd 2007 at 11:40:55 AM |
|
Name: |
daniel |
Comments: |
i totally agree with you, arsen! it would be much more complicated with two rankings and it doenst matter whether you save one second or not. |
![]() |
|
Dec 3rd 2007 at 11:34:52 AM |
|
Name: |
KAmil |
Comments: |
There can be also 2 rankings: 1 with scores and 1 with real times (timer starts @ 0) |
![]() |
|
Dec 3rd 2007 at 11:14:09 AM |
|
Name: |
Arsen |
Comments: |
Guys, maybe I didn't cacth the clue, but I see you speaking about starting timer from 0 instead of 1. I totally disagree with that for the following reasons: 1. this would make new and old results incomparable. 2. this would make results from clone and winmine incomparable and if you take into account the fact that we want to engage more people who usually start playing on winmine, this is not a good idea. Finally, does it matter for you if it is 50 or 51 in the end since all of us are in equal conditions? Time is rather conventional thing. I think it is better not to change this tradition. |
![]() |
|
Dec 3rd 2007 at 10:27:15 AM |
|
Name: |
WP |
Comments: |
@KAmil: Nah, too complicated :P @Gergely: So are you going to be a candidate? I hope you will. ;) |
![]() |
|
Dec 3rd 2007 at 10:12:44 AM |
|
Name: |
KAmil |
Comments: |
The best idea is to write beg time in rankings, but only int and exp are sum to ranking ![]() |
![]() |
|
Dec 3rd 2007 at 09:06:37 AM |
|
Name: |
Daniel Brim |
Comments: |
The solution to that is clearly to remove beginner entirely ![]() |
![]() |
|
Dec 3rd 2007 at 06:08:11 AM |
|
Name: |
KAmil |
Comments: |
I also agree with that timer should start from 0 (not +1), but I don't agree with rankings with decimal scores (it would be good for int and exp, but not for beg: my 1,668 on 3 isn't worse than Tam Minh Bui/s lucky 1,1 on 2. I think that 1,7 on beg is as good as 1,999, but 12,999 on int is worse than 12,7). |
![]() |
|
Dec 2nd 2007 at 04:26:40 PM |
|
Name: |
Jason K. |
Comments: |
Ok, so I suppose if I'm going to run for the IMC, I should actually state my opinions here seeing as how I rarely (if ever) chime in on issues, preferring rather to sit back and observe discussions. Plus, I definitely didn't take the process seriously on the .cc page, painfully obvious if you read my "vote for me because". ![]() Like many others, I'm in favor of the IMC being more transparent. Perhaps even more so than others. I believe transcripts of IMC discussions should be made public, as should votes and the like. I agree with Schu that the +1 timer should be removed. I agree with Tommy that decimal rankings should be used as well, as they are used in nearly every major sport in the world where lowest time is the goal (track and field, swimming, etc.). The other point I would make (and it might not be a very popular one) is that I believe ego and fear sometimes enter into decisions involving acceptance of scores and accusations of cheating. Understandably so, it's a lot tougher for a Top 20 player in the world to come to terms w/ someone rocketing up the ranks and potentially passing them by, sometimes in a very short time period. In cases where acceptance seems to be a judgment call, I believe players should be given the benefit of the doubt first. Requesting other videos as proof of skill is ok, as long as it isn't the default reaction to someone making an incredible score. Amazing things happen in all walks of life, including Minesweeper, and I believe skepticism sometimes runs too rampant in this community. Anyways, that's where I stand on a couple of things. Good luck to everyone running! |
![]() |
|
![]() |
|
Dec 2nd 2007 at 04:23:27 PM |
|
Name: |
Jon S |
Comments: |
@Damien: The game I sent you the other day (int 81 mines) is from 26.11.2007. I think there's something wrong with the link in the custom record list, so I uploaded it to my folder too. I'll try to make a profile next week. @all Regarding the video of my int game with 81 mines: http://www.minesweeper.info/videoindex.php?dir=JonSimonsen/ I think this board was quite straightforward. There are "only" three 50-50 guesses once I hit the opening. The initial click that revealed a 6 probably helped a bit. I thought the pattern containing the last three mines were a bit interesting. It's obviously best to start by opening one of the squares in the 16th row, because if you hit a mine it would have been a 50-50 situation anyway. This is the case on the actual board. I looked a little more closely at the possible mine configurations of the last three mines. It seems like there are 15 configurations, and optimal strategy lets you solve the pattern 10 out of 15 times. I don't know if this is pretty obvious for those of you that are solving in density mode. Has anyone published any kind of strategy for density games? I know some strategies that ordinary minesweepers would never need that are quite useful in density games. I might consider to publish an article one day, but I'd like to know if anyone would be interested before I start working on something like that. |
![]() |
|
![]() |
|
Dec 2nd 2007 at 01:44:58 PM |
|
Name: |
daniel |
Comments: |
oh i narrowly missed the deadline...next time i'll hurry ![]() |
![]() |
|
Dec 2nd 2007 at 12:59:27 PM |
|
Name: |
damien |
Comments: |
Website (.info) updated. Roman and his friends are probably still drunk....results: http://flash-back.ru/ms/ |
![]() |
|
Dec 2nd 2007 at 12:58:36 PM |
|
Name: |
daniel |
Comments: |
![]() ![]() ![]() ![]() 70,08nf->65,04!!!! this time flagged! flagged? used only one flag ![]() ![]() just kinda happy ![]() |
![]() |
|
![]() |
|
Dec 2nd 2007 at 10:41:14 AM |
|
Name: |
Bertie |
Comments: |
Well, now it's official, I'm running! Thanks for the nudge Tommy. I see that the discussion has become somewhat more heated since I last posted. I would've posted sooner, but I'm home from uni and thus didn't have a connection till I figured out how to connect using my mobile. ![]() I see that the main points been thown up are transparency and the effectiveness of the IMC. One issue, in my opinion, might have been partially inflamed by the other. If transparency had been maintained from the start the IMC might not have seemed so inactive. We will never know. Micheal hit the nail on the head about the transparency. I think, judging by the amount of cries that have been uttered for it here, this issue will probably be dealt will quickly. On the minesweeper front. I got a 52 in last week which would've gone nicely with the 17 I had lined up for the period, but sadly my submission ussing my mobile (not as modem) didn't go through. It wouldn't have lowered my effective score, but it would've been nice not to have that ugly +5 next to my name. ![]() Anyway. Happy hunting! |
![]() |
|
Dec 2nd 2007 at 10:34:49 AM |
|
Name: |
Gergely |
Comments: |
Hello! The run for the IMC seems to be quite flat this year. I hope more people will join in next week. So, anyway it's time to make a point for my candidacy ![]() Plus, I'd like to work on making minesweeper more famous. We are good at it, so let's make the world hear somehing about us. We are fast. We have fantastic logic and mouse handling skills. We deserve some spotlight ![]() ![]() Hope you'll join me - I haven't quit on this case and won't ![]() |
![]() |
|
![]() |
|
Dec 2nd 2007 at 10:33:47 AM |
|
Name: |
Ian |
Best intermediate: |
13.39 --> 13.35 |
Comments: |
I beat leet! Hurray! And my syndrome is extended to 13x6 ![]() |
![]() |
|
Dec 2nd 2007 at 10:27:32 AM |
|
Name: |
damien |
Comments: |
hey kyle...just right-click on the video and copy the shortcut and paste here in the 'website' box. i tried your game...and lost the first 20 times, so i quit! i tried to play it without upk, so i guessed wrong in the same place the first 6 times. ![]() the site will be updated (www.minesweeper.info) at 20:00 GMT tonight. |
![]() |
Viewing Page 3 of 27 (Total Entries: 2658) |