3BVS ranking rules

Suggest ideas for the World Ranking
Post Reply
thefinerminer
Site Admin
Posts: 136
Joined: Tue Jan 08, 2008 3:33 pm
Location: UK, Scotland
Contact:

3BVS ranking rules

Post by thefinerminer »

There have been many talks about 3bv limits or other rules for the 3bvs ranking. It would be nice to solve this problem finally. I can not remember all the places we discussed it, so am starting a new topic to collect all opinions. There are a few things to discuss:

1. A 3bv limit? Obviously no 'Lucky' games. (If you want limits, I like 10-40-99). What do you think about someone who does 5 lucky clicks and 1 solve to get a high 3bvs?
2. A time limit - eg, no 1s games? (I used to apply this 'rule', but the 'Lucky' rule eliminates most of these).
3. Weights - should 10.65 Beginner be given the same weight as 3.70 Expert?
User avatar
Harryck_Repse
Posts: 24
Joined: Tue Apr 14, 2009 5:49 am

Re: 3BVS ranking rules

Post by Harryck_Repse »

It's tough to say what should be allowed on 3bv/s rankings and what shouldn't. I think it's pretty obvious for expert and intermediate but beginner is tough... personally I'd like to see beginner removed entirely from all rankings but I don't think that's going to happen any time soon. When you mention beginner requirements it brings Chinese player Wen-Xiong Mo to mind. He has a non-lucky 0.95 on a 10 3bv board for world record 10.53 3bv/s (as well as two lucky sup10 5click 5 3bv games), so when you try to add his achievements to the 3bv/s world ranking it skews his total greatly. Right now, Wen-Xiong Mo is ranked first on saolei.wang among Chinese players for overall 3bv/s, and if his 10.53 3bv/s game were to be accepted on minesweeper.info it would put him at second in the world, behind only Kamil. This becomes a problem because he's barely good enough to sub60 on expert and doesn't even have sup3 on expert. So my point of all this is that the weighting idea is probably the best, so as to not completely ignore the impressive feats of such players but also so that true skill is ranked. Intermediate and expert should be kept the same but beginner should be divided by 2 or 3.

btw, what does the second black number on 3bv/s ranking mean?
babz
Posts: 21
Joined: Mon Jul 25, 2011 6:36 pm
Contact:

Re: 3BVS ranking rules

Post by babz »

The 3BV limits would be great for beginner, for intermediate and expert it won't make any difference since all records are made on high 3BV board. Think to the intermediate, on 30 3BV board, to get a sup4 you need to almost break the world record...

The 10 3BV limits seems rigth for beginner.

The first number in the ranking is the normal sum, and the second is the weighted sum with the number of mines in each level.
EWQMinesweeper
Posts: 419
Joined: Sun Nov 30, 2008 11:50 pm

Re: 3BVS ranking rules

Post by EWQMinesweeper »

harryck: that is a weighted sum

10/149 beginner + 40/149 intermediate + 99/149 expert

a ranking based on this formula would look like this:

1. (1.) kamil - 5.711
2. (2.) dion - 5.087
3. (4.) manu - 4.971
4. (5.) pavel - 4.842
5. (10.) [name removed]- 4.812
6. (3.) tian - 4.766
7. (7.) zhang - 4.722
8. (6.) damien - 4.687
9. (9.) ian - 4.602
10. (11.) mao - 4.489
11. (8.) tam - 4.480
12. (21.) reid - 4.465
13. (30.) lukasz - 4.461
14. (17.) roman

this ranking would still take all levels into account, but 1.00 3BV/s improvement on beginner has the same effect as 0.25 3BV/s on intermediate and 0.1 3BV/s on expert (namely 0.06 points inmprovement), so that the players' main focus is on expert and intermediate as it should be ("just play beginner until you have a 1s game").

iirc this formula was the result we came to when the topic was discussed on irc for the first time.

this weighted ranking is my favourite solution.

however, a formula based solely on the number of mine is a bit arbitrary, although avg 3bv, avg zini, number of mines would lead to similar results.
„Das perlt jetzt aber richtig über, ma sagn. Mach ma' noch'n Bier! Wie heißt das? Biddä! Bidddää! Biddddäää! Reiner Weltladen!“
KamilSaper
Posts: 149
Joined: Mon Dec 01, 2008 7:16 pm

Re: 3BVS ranking rules

Post by KamilSaper »

I also like weighted ranking, but I also dont want to allow Mo's 1 on 10. 3BV limit for beg doesn't solve the problem, as this guy showed that u can get lucky 10 3bv game. I think we should set totally different rule: at least 50% clicks for beg record must be made on purpose. It's obvious that 10 3bv solved using only 2 clicks on purpose can't be considered as a world record (3bv/s is supposed to show player's speed).
0.49 - 7.03 - 31.13
NF: 0.49 - 7.03 - 31.51
qqwref
Posts: 125
Joined: Thu Sep 23, 2010 4:17 pm

Re: 3BVS ranking rules

Post by qqwref »

I do actually think beginner 3BV/s matters - OK, maybe more so for NF players like me, but it definitely does show something worth looking at (essentially, max burst 3BV/s). So I wouldn't really want to remove it. In the normal timed event a low beginner time just means you got lucky, but in this case, a higher beginner 3BV/s really does mean you played a better game. The weighting is fine and I wouldn't complain if the ranking was ranked based on that alone, but the 3BV total is useful and should be kept there.

I'm really not sure about the 3BV limit though. My current best beginner 3BV/s is on 11 3Bv with only two 3BV opened randomly at the start (including the opening) and I don't feel like this should be just barely legitimate. I think it would be better to just have a lucky rule that takes into account the number of 3BV gotten with lucky clicks (1/2? 1/3?). For any other difficulty I'm not sure taking luckiness or 3BV into account is worth the effort, since I've never really seen a truly lucky board outside of specially created cases (like the IOE WR boards).
NF player. Best scores 1-10-39.
User avatar
Tjips
Posts: 74
Joined: Sat Apr 18, 2009 1:15 am
Location: South Africa

Re: 3BVS ranking rules

Post by Tjips »

In both the beg and exp cases the 3bv/s ranking bumps its head against effects that only become important at the extreme ends of the 3bv domain. For beg, games become solvable solely by luck, and for exp, the measure of work itself (3bv) breaks down and loses its meaning. As [name removed] pointed out in a GB post, it isn't the high 3bv that allows the 3bv/s record on exp, it is the relatively low zini. On low 3bv boards (~100 - 120) the difference between the zini and 3bv is around 20 clicks (or an over estimation of the total work on the board by around 20%), while on high 3bv boards this difference goes up to around 80 clicks (or just under 40%). Granted, neither 3bv nor zini are absolutely on the money when it comes to the amount of work on the board, but the zini of a board is definitely a better reflection of it. In all honesty I'd rather see a ranking based on the zini-like-thing/s scores of the individual players alongside a 3bv/s ranking allowing only NF scores. But, as this isn't going to happen, we should get to work on painting over 3bv/s' flaws.

Here's a way-off-the-grid suggestion for exp that we could paste over [name removed]'s 1-4-10 weighting system (which, for the record, I like atm): Adjust the exp 3bv/s scores according to the nominal 3bv-to-zini ratio for the specific 3bv the score was attained on. That way you take out at least one of the statistical elements at play...

On int and beg that idea is less likely to work as the outcome of the first 3-5 clicks have a bigger effect on the time than they do on exp (with beg being almost entirely reliant on a good start).

So, atm I'm favouring the weighting option as it seems to get closest to putting a reasonable ranking out there. This said, any scheme you guys could come up with which would put me on the top of the ranking would also be fine by me ;)

Oh, and it's 7am, and I haven't slept. So if this post is complete gibberish, I'm sure you'll forgive me :P...
The number of minesweeper boards:
Exp: 140055249834355336357264746443955277014822625680974475320364702381803619892657792049596418323789908370400 (1.4e104)
Int: 13115156192346373485000211099954895788134532256 (1.3e46) &
Beg: 18934455246 (1.9e10)
:D
one7th
Posts: 2
Joined: Thu Dec 22, 2011 11:17 am

Re: 3BVS ranking rules

Post by one7th »

Something from saolei.wang

:arrow: someone suggest that we should use "3bv-solved-on-purpose" rule as the lucky rule when discussing the 3bvs ranking rules :
"at least 50% 3bv must be solved on purpose. And if the first click hits a opening, this 3bv should be taken away when counting lucky-solved 3bv since it can not be avoided.

I personally believe that it is better than "click-on-purpose" rule since we are talking about 3BVS but not CL/s, and the more wisely you play minesweeper,the less click you use to solve a specific pattern. the click rule just violates the efficiency principle."'


:arrow: There is another one's word about MO Wenxiong. And I agree that we should try to understand "beginner-player" instead of simply label them "luck-seeker":

"it seems that the focus of this issue is MO and his 10.53.i have no idea about ranking rule,but i have something to say about MO.i do not want to criticize anyone,i just feel disappointed about the bias on beginner-players and the view some people hold towards MO.

i still can not believe that world's best sweeper like EWQ try to make everyone focus on expert or intermediate as IT SHOULD BE.the word sounds ridiculous to me.is it WRONG to play beginner level?it is MINESWEEPER,TOO.weight ranking is not bad,but the attempt to change beginner-players into the shape you want may not work because players care more about fun than record,don't you think so? the force you put on them can only make them sad because they are looked down upon.

and the sterotype towards beginner-players is also ridiculous.DON'T jugde a beginner-player by its expert records.otherwise you have to draw a conclusion that one of my friend must be a cheater.(he got beg time 1s*17 and 3bvs 8.01 while his exp time 81s and 3bvs 2.11)

he can easily beat most players in China on beginner while it reversed on expert(judging by AVG time and AVG 3bvs).he devote himself into beginner level,and i think MO did the same,too.MO payed his efforts just as you do on exp level,and he deserve his luck(just as your luck when you get the board on which you broke record).

and i want to figure it out that actually MO's quick-reaction and quick-click account for the record.he got 3bvs10*3 and countless 1s(once he won 1s*2 just in an hour/minute, sorry for lack of detail),which witnessed his strength rather than luck. in other word,i believe 10.53 is his speed.the board is lucky but just as lucky as your record board.seriously,perhaps MO can beat most of you on beginner.

PS:it's clear that what some people are talking about is just to find out a way to sweep MO's 0.95s away legally.so seriously,why not just ban videos made by clone0.97? that is the simplest way to get rid of the trouble,and then nothing will bother you. XD"


Given the example of someone with beg 3bvs 8.01 but exp time 81s, maybe we can't take it for granted that a player being good at beg must be good at exp.Then the problem that MO is not good enough on exp is no longer a problem because it has nothing to do with his fantastic performance on beg. He shows us how charming that beg can be, like an artist; so I want to allow his 10.53 as a record.

PS: Sorry for the flaws in translation ~.~ I'm sure you'll forgive me......
EWQMinesweeper
Posts: 419
Joined: Sun Nov 30, 2008 11:50 pm

Re: 3BVS ranking rules

Post by EWQMinesweeper »

one7th wrote:"3bv-solved-on-purpose" rule as the lucky rule when discussing the 3bvs ranking rules :
"at least 50% 3bv must be solved on purpose. And if the first click hits a opening, this 3bv should be taken away when counting lucky-solved 3bv since it can not be avoided.

I personally believe that it is better than "click-on-purpose" rule since we are talking about 3BVS but not CL/s, and the more wisely you play minesweeper,the less click you use to solve a specific pattern. the click rule just violates the efficiency principle."'
1 click that was made on purpose serves well as a rule since it only takes out the 1s games that are 100% lucky.
one7th wrote:There is another one's word about MO Wenxiong. And I agree that we should try to understand "beginner-player" instead of simply label them "luck-seeker"
who said luck-seeker? general view on beginner is that once you have a 1s game you can stop playing. 3bvs ranking is nothing more than an additional ranking. be assured that we understand all known aspects of the game.
one7th wrote:it seems that the focus of this issue is MO and his 10.53.i have no idea about ranking rule,but i have something to say about MO.i do not want to criticize anyone,i just feel disappointed about the bias on beginner-players and the view some people hold towards MO.
we don't hold any view. and you are criticising. if you have nothing connstructive to say, maybe better don't say anything at all. the ranking includes all levels and a broader view on the several rankings obviously reveals that expert requires more skill than intermediate and intermediate requires more skill than beginner. thus it is just reasonable to mainly focus on the bigger levels.
one7th wrote:i still can not believe that world's best sweeper like EWQ try to make everyone focus on expert or intermediate as IT SHOULD BE.the word sounds ridiculous to me.is it WRONG to play beginner level?it is MINESWEEPER,TOO.weight ranking is not bad,but the attempt to change beginner-players into the shape you want may not work because players care more about fun than record,don't you think so? the force you put on them can only make them sad because they are looked down upon.
the world ranking is what matters most. get over it. see the above paragraph for explanations. i am in no way forcing anyone to play a single game. it is my damn right to express my opinion on which levels i think should matter the most. anyone can get a sup3 on beginner quite easily - sup3 on int and expert ain't that easy, or am i mistaking? can't you just accept that?
one7th wrote:and the sterotype towards beginner-players is also ridiculous.DON'T jugde a beginner-player by its expert records.otherwise you have to draw a conclusion that one of my friend must be a cheater.(he got beg time 1s*17 and 3bvs 8.01 while his exp time 81s and 3bvs 2.11)
this doesn't make any sense. beginner is part of minesweeper, thus all levels matter.
one7th wrote:he can easily beat most players in China on beginner while it reversed on expert(judging by AVG time and AVG 3bvs).he devote himself into beginner level,and i think MO did the same,too.MO payed his efforts just as you do on exp level,and he deserve his luck(just as your luck when you get the board on which you broke record).
you don't get the point. i don't know how much luck is involved in his games but the way i play luck, in the sense you are using it, does not play a role. in this context beg and exp can in no way be compared. avg time and avg 3bv alone don't mean much. can he beat me or kamil to fastest time for 10, 100, 1000 beginner? how many games does he complete? actually, i don't care for answers on that question. see them as an impetus to think more about the game as a whole. to clarify: it takes damn much practice to get a sub50 - even the biggest noob can get a 1 on beginner.
one7th wrote:and i want to figure it out that actually MO's quick-reaction and quick-click account for the record.he got 3bvs10*3 and countless 1s(once he won 1s*2 just in an hour/minute, sorry for lack of detail),which witnessed his strength rather than luck. in other word,i believe 10.53 is his speed.the board is lucky but just as lucky as your record board.seriously,perhaps MO can beat most of you on beginner.
and? to be honest, i don't really care as long as he hasn't completed 100 beg in under 10min and 1000 beg in under 1:40.
one7th wrote:PS:it's clear that what some people are talking about is just to find out a way to sweep MO's 0.95s away legally.so seriously,why not just ban videos made by clone0.97? that is the simplest way to get rid of the trouble,and then nothing will bother you. XD
are you aware of what you are saying? well, in this context your whole post becomes almost completely irrelevant. can we take you serious? obviously not.
one7th wrote:Given the example of someone with beg 3bvs 8.01 but exp time 81s, maybe we can't take it for granted that a player being good at beg must be good at exp.Then the problem that MO is not good enough on exp is no longer a problem because it has nothing to do with his fantastic performance on beg. He shows us how charming that beg can be, like an artist; so I want to allow his 10.53 as a record.
personally i don't care for the sup10. i just want the weighted ranking.
one7th wrote:PS: Sorry for the flaws in translation ~.~ I'm sure you'll forgive me......
thanks for the translation.
„Das perlt jetzt aber richtig über, ma sagn. Mach ma' noch'n Bier! Wie heißt das? Biddä! Bidddää! Biddddäää! Reiner Weltladen!“
KamilSaper
Posts: 149
Joined: Mon Dec 01, 2008 7:16 pm

Re: 3BVS ranking rules

Post by KamilSaper »

"and i want to figure it out that actually MO's quick-reaction and quick-click account for the record.he got 3bvs10*3 and countless 1s(once he won 1s*2 just in an hour/minute, sorry for lack of detail),which witnessed his strength rather than luck. in other word,i believe 10.53 is his speed.the board is lucky but just as lucky as your record board"

His 10x2 were totally random, his 3rd uses 2 clicks on purpose to open 10 3bv, so telling that it's more skill than luck is just funny and shows your disknowledge about the game. And so what he won 1x2 in 1 minute/hour? I won 1x2 in 1 minute too, probably a few times if I remember correctly, or 1x4 in 1 hour. If he wants to claim for beg 3bv/s world record, he should get better than me and beat my record in non-lucky way.
0.49 - 7.03 - 31.13
NF: 0.49 - 7.03 - 31.51
thefinerminer
Site Admin
Posts: 136
Joined: Tue Jan 08, 2008 3:33 pm
Location: UK, Scotland
Contact:

Re: 3BVS ranking rules

Post by thefinerminer »

Hello Pengyun, I am glad you joined the Forum. We are very interested in what the players at saolei.wang think - it is unfair we make all the rules without hearing your ideas too. (This week 5 players from the China ranking joined this Forum, so that is a good start).

You are correct that some of the best players do not like Beginner. After you get 1s there is almost no reason to play, unless you want to join the 3bvs ranking. The problem with Beginner is that you can get 1s or 3bvs=10 with only 1 click on purpose. It is hard to argue this is the same skill as getting Sub40 on Expert. I agree you can specialise and be really good at one level. (I used to be the best Beginner player in the world). I think most people agree Beginner is good on the World Ranking now that we use decimal time and Lucky games are banned.

In my opinion, the longer you keep a high speed the more skill. Getting 6.00 3bvs on 3bv=30 is better than a one click on purpose 3bv=4. This is not a problem on Intermediate and Expert because you need to do at least 30 or 100 clicks, this deletes most of the luck. Your idea to not count the 1st click is good (we already do this with the "Lucky" rule, the 1st click has no skill). The idea to make a "clicks done on purpose" limit (instead of 3bv limit) is very interesting.

If you did not understand, [name removed]was talking about his Beginner skill: he won 1000 Beginner games in 2:02:26 (but has 6.56 3bvs record).

We have already banned Clone 0.97 for everyone except China, because of the problems every player at saolei.wang knows.

We talked before about Mo's 10.53 and the decision was to allow it. The purpose of this talk is to make clear rules, because we do not want the ranking to become stupid. (Banning "Lucky" games was a good rule, for example, because now people will not click randomly to get 0.10 games and make Beginner meaningless).
EWQMinesweeper
Posts: 419
Joined: Sun Nov 30, 2008 11:50 pm

Re: 3BVS ranking rules

Post by EWQMinesweeper »

thefinerminer wrote:If you did not understand, [name removed]was talking about his Beginner skill: he won 1000 Beginner games in 2:02:26 (but has 6.56 3bvs record).
i feel this needs further clarification: there are different aspects you can be good at. a player who wants to be the best should be good at all aspects. 6.56 3bvs is very good. it is by far better than most other sweepers will ever get. however, i give a flying f*** about my beginner 3bvs, since i refuse to go above a certain cls, as that would be mean taking irreasonable risks to permanently damage one's wrist and hand.
„Das perlt jetzt aber richtig über, ma sagn. Mach ma' noch'n Bier! Wie heißt das? Biddä! Bidddää! Biddddäää! Reiner Weltladen!“
one7th
Posts: 2
Joined: Thu Dec 22, 2011 11:17 am

Re: 3BVS ranking rules

Post by one7th »

Er...... I was surprised that so many of you response to my post~ I think it is a good start, too.

:arrow: And I have just got the reply to [name removed]from that one [name removed]argued with, still in English.

"after finishing reading his comments, i find that i did not get myself across completely so there were some misunderstanding. i will try agian to make myself clear to others. i discovered that when i saw EWQ responsed in a serious way to my joke of banning clone0.97 videos. that was only a joke and nothing more( i used to use "XD" as a sign of a joke but it failed to function this time......) and the way he responsed to a joke shows that he may be annoyed by my words. i apologize for that, i did not mean to do that.

in fact when i mentioned EWQ, i only disagree about his opinion that once you have a 1s game you can stop playing. that make sence only to players who only focus on records. but many of us are playing for fun and beg is still fun even after you get 1s. i find it obvious that if i play beg for records i will get bored soon but i still like to play beg because i find it fun. i do not play minesweeper for records so i do not like the idea of focusing on expert or intermediate since all levels share the same fun of minesweeper. so i think beginner-players will not change by the ranking rules because they enjoy the fun from playing beg but not enjoy the records. that is all i want to say about EWQ. i did not mean that there was bias on beg in the aspect of skill or something like that. EWQ used his right to express his opinion, and thanks to 7[me^.^--one7th] i just use my right to say that beg is always worth playing.

as for the weight ranking...... that sounds reasonable to me, and i think i did not express any opposite view in my previous word. but if anyone think i did, i apologize and i will work harder on English......

another main point is MO, his strength and luck. i mentioned him because though i know it for sure that his 10.53 was accepted but , there are people including kamil ( but not you, EWQ, i know it for sure too) still trying to deny it. and as far as i know, they hold this opinion because:
1-the board is too lucky;
2-MO is not good enough on exp;

and the aim of the rest part of my previous word only wants to convince them that:
1-the board is lucky, but the same luck as the boards on which they got BEG(i left this three letters out last time, my fault....i did not compare his beg to others' exp,just beg VS beg.) record, his quick-minded action is the main factor of high 3bvs;
2-despite his exp performance, he is one of world's best beg players. i know that AVG scores alone don't mean much, i mention them just as examples. the fact is that-- judging by record or AVG, time or 3bvs or rqp or the whole time to win 10En(n>2) begs, anyway you name it-- the one i mentioned before as my friend is at the top in china and in all aspects.

and he is an exception of "all levels matters". before i met him, i hold the same idea as you; however, after watching him playing, i have to change my mind. the contrast between his performance on beg and exp makes him just more unbelievalbe than MO! if you watch him play, i guarantee you will also admit that there CAN be little direct relationship between the performance on exp and on beg. so in comparison, i do not feel any strange about MO's performance. it seems impossible to you, but it CAN happen and it did happen.

BTW,i mentioned MO can beat most of you but not ALL of you. i do not know the answer if in comparison to [name removed]or kamil judging by fastest time for 10, 100, 1000 beginner since they are just beyond godlike holyshit( to clarify: i mean no offence.....just an expression in DotA.), i am longing for the result. however, i am sure that MO can be at least TOP10 in fastest time for 1000 and TOP5 in 10000 beginner and, my friend may win the fastest in 100000 beginner since MO is busy at working now...... their craziness is beyond my imagination......but they really treat beg in that way. if you see the way they play, you will understand what i said. they are "insane" in all aspects.

PS: the "click-on-purpose" rule 7 mentioned when we talking about "3bv-solved-on-purpose" rule refers to kamil's 50% clicks rule but not the 1-click rule.[My bad>_< I should have pointed it out.--one7th] i agree that "1 click on purpose" works well, but kamil's suggestion sounds unreasonable since it violates the efficiency principle as mentioned before.

MO talked to me in person too. He said he think little of record or ranking, and it was his honor to make kamil try to modify the rule again and again.( I don't know whether I got it right......)


BTW, it is my first time to use English as a lingua franca , first time to communicate with people with different culture backgroud outside classroom, and what is more, i am glad that i can share ideas with world best minesweepers.so much for the reply. i made every effort to express myself clearly. if anyone misunderstand or feel annoyed this time, i apologise in advance; but i can not spare any time to reply them because the final exam ..... so .... merry christmas and happy sweep ...."

:arrow:And players from saolei.wang began to join this Forum because we did not find the access to register until this week>_< I believe that further communication can help narrrow the gap between two sites. The gap appeared when clone2007 could't run on the operating system of Chinese edition , no matter XP or VISTA or WIN7; Framework installed or not. I don't know the reason why it can't run on Chinese operating system only, but I believe that if we work together we can work out an adjust version of clone2007 which fit the Chinese operating system soon, and that is every player in China hope to see. Maybe I shouldn't talk too much about that since it is not related to the topic.

And..... merry christmas and happy sweep too~.~
EWQMinesweeper
Posts: 419
Joined: Sun Nov 30, 2008 11:50 pm

Re: 3BVS ranking rules

Post by EWQMinesweeper »

@7: thank you very, very much for the fast and good translations. much respect for your efforts.
one7th wrote:after finishing reading his comments, i find that i did not get myself across completely so there were some misunderstanding. i will try agian to make myself clear to others. i discovered that when i saw EWQ responsed in a serious way to my joke of banning clone0.97 videos. that was only a joke and nothing more( i used to use "XD" as a sign of a joke but it failed to function this time......) and the way he responsed to a joke shows that he may be annoyed by my words. i apologize for that, i did not mean to do that.
jokes might not always work when there's a language barrier. what you intended to be a joke, seemed to me as a cynical remark. thanks for the apology. i apologize too, if i went too far with my first reply. (i first read it 20min before a class at university started and was in dire need of sleep - today i've had more than enough sleep and see this discussion in a different way than yesterday).

one7th wrote:in fact when i mentioned EWQ, i only disagree about his opinion that once you have a 1s game you can stop playing. that make sence only to players who only focus on records. but many of us are playing for fun and beg is still fun even after you get 1s. i find it obvious that if i play beg for records i will get bored soon but i still like to play beg because i find it fun. i do not play minesweeper for records so i do not like the idea of focusing on expert or intermediate since all levels share the same fun of minesweeper. so i think beginner-players will not change by the ranking rules because they enjoy the fun from playing beg but not enjoy the records. that is all i want to say about EWQ. i did not mean that there was bias on beg in the aspect of skill or something like that. EWQ used his right to express his opinion, and thanks to 7[me^.^--one7th] i just use my right to say that beg is always worth playing.
ah, thanks for clearing that up. now i see that there was a major misunderstanding. for me and many other sweepers our main focus is indeed on time records that count for the world ranking. but of course we still do enjoy playing beginner. not always for speed (we had tournaments in which the fastest 10 to 20 beg games mattered too) but always just for fun.
one7th wrote:as for the weight ranking...... that sounds reasonable to me, and i think i did not express any opposite view in my previous word. but if anyone think i did, i apologize and i will work harder on English......
great. this means that basically we are yet to hear anyone disagree about a weighted ranking.
one7th wrote:another main point is MO, his strength and luck. i mentioned him because though i know it for sure that his 10.53 was accepted but , there are people including kamil ( but not you, EWQ, i know it for sure too) still trying to deny it. and as far as i know, they hold this opinion because:
1-the board is too lucky;
2-MO is not good enough on exp;
well, yeah, i'm not participating in that specific discussion. if there should be a vote upon it, i'll follow damien's arguments first. oh, wait. if i'm not going to be discussing it, i shouldn't vote on it, right? but about 2): i honestly doubt that it plays any role, especially since with the weighted ranking it would no longer be maintainable to hold as a position.
one7th wrote:PS: the "click-on-purpose" rule 7 mentioned when we talking about "3bv-solved-on-purpose" rule refers to kamil's 50% clicks rule but not the 1-click rule.[My bad>_< I should have pointed it out.--one7th] i agree that "1 click on purpose" works well, but kamil's suggestion sounds unreasonable since it violates the efficiency principle as mentioned before.
my bad, i totally forgot that he suggested that. (joke)probably my brain dismissed the idea, before bothering remembering it(/joke). i spent quite some time thinking about how to treat beginner time high scores, but haven't bothered much with 3bvs records yet.
one7th wrote:BTW, it is my first time to use English as a lingua franca , first time to communicate with people with different culture backgroud outside classroom, and what is more, i am glad that i can share ideas with world best minesweepers.so much for the reply. i made every effort to express myself clearly. if anyone misunderstand or feel annoyed this time, i apologise in advance; but i can not spare any time to reply them because the final exam ..... so .... merry christmas and happy sweep ...."
good luck int the exam
one7th wrote:And players from saolei.wang began to join this Forum because we did not find the access to register until this week>_< I believe that further communication can help narrrow the gap between two sites. The gap appeared when clone2007 could't run on the operating system of Chinese edition , no matter XP or VISTA or WIN7; Framework installed or not. I don't know the reason why it can't run on Chinese operating system only, but I believe that if we work together we can work out an adjust version of clone2007 which fit the Chinese operating system soon, and that is every player in China hope to see. Maybe I shouldn't talk too much about that since it is not related to the topic.
use arbiter ;)


new suggestion: i think it would be better to split this thread into 2 threads. the current one is about finding a new way to determine how to compare players' 3bvs records for the ranking. the sup10 has to be treated like any other score in this context. if people feel the need to discuss more about the sup10, i think this belongs into a new thread.
„Das perlt jetzt aber richtig über, ma sagn. Mach ma' noch'n Bier! Wie heißt das? Biddä! Bidddää! Biddddäää! Reiner Weltladen!“
User avatar
Tommy
Posts: 257
Joined: Mon Dec 01, 2008 9:22 pm
Location: Vienna

Re: 3BVS ranking rules

Post by Tommy »

Hm. I actually started replying to this topic once, I never got my reply done though, and I forgot to submit it.

Welcome here one7th :) If I may say so, your english is awesome given the circumstances ;) I really appreciate seeing someone from saolei.wang here, it would be great if those two communities worked together more.

I don't think we should stop talking about Mo's game here, as it is a real extreme case that we can use as a benchmark, here is the video (thanks to aradesh for uploading it):
http://www.youtube.com/watch?v=Q0Y5TroCUFo

Some facts to note:
- The mouse path clearly indicates that all moves before the flag are setup moves. Look at that perfect circle, and take into account that there is such a thing as reaction time. Also, the last click is redundant (unless there is another opening there, but the circle is not broken in any way there).
- six setup clicks are made, four of which contribute to the game (the last one technically opens a 3bv, but the chord would have opened that square, as well).
- Execution is perfect. As soon as Mo reacts, he makes the one move that he has to make, and lightning fast at that.

In my opinion, this score involves both a great deal of luck _and_ skill. The reflex is insane. I might have taken .10 or .20 more to react to that.

Can you maybe get Mo to stream himself playing beginner for a while? I'd be really interested to see that :)

Anyway. We should really discuss what we are using 3bv/s as a benchmark for.

In general, 3bv/s is a measure of speed over time. 3bv/s highscores (usually) reflect how good a player is at clearing many squares at a time continuously, while reacting to the board (most of the time, 3bv/s highscores are achieved on high 3bv boards).
On beginner, evidently, 3bv/s highscores aren't always a measure of general playing speed - they can also be a measure of single reaction speed. This is not the case on any other level.

The question is: Do we want a "speed over time" ranking, or a "true 3bv/s" ranking.

I personally want a "speed over time" ranking. Beginner is just one level, and strange stuff that happens on it shouldn't affect the entire ranking, as everyone seems to agree (which is why we removed completely lucky fast beginner games). And therefore I am in favor of some kind of regulation that rules out single-reaction games like Mo's. I don't want to discount the skill involved there, it is an achievement in its own right, I just don't think that it belongs on that ranking.

How about "a 3bv/s highscore needs to contain a reaction to information that was not gained by setup clicks alone, where it must be obvious that a given click is a setup click to apply the rule"? This may be pretty hard to apply, but fits what I would like to achieve pretty well. I don't think a 3bv limit would be very useful (Mo's game has a 3bv of 10), and if we use one, it should only apply on beginner, where this kind of strange stuff actually happens.

And I am also in favor of a weighted ranking - 10-40-99 are great nothing up my sleeve numbers IMHO, 64-256-480 (ie, 2-8-15) also works and has the benefit of not nerfing int quite as much. Or we could round 10-40-99 to 1-4-10.

Mainly first thoughts, mostly not definitive opinions. More input would be very welcome.
Don't anthropomorphize computers - they don't like it.
Post Reply