Viewing Page 48 of 53 (Total Entries: 5262) |
![]() |
|
Aug 31st 2005 at 09:30:05 AM |
|
Name: |
Dmitriy |
Comments: |
hey, how about to implant the mouse chip to the brains? the wire should use the usb protocol for the fastest communications... |
![]() |
|
Aug 31st 2005 at 08:40:02 AM |
|
Name: |
AreOut |
Comments: |
Dmitriy what if I connect wire from right mouse button to wire from left mouse button and therefore have double clicking speed in NF?! You probably can discover all software cheats but no single hardware cheat... ![]() |
![]() |
|
![]() |
|
Aug 31st 2005 at 07:59:41 AM |
|
Name: |
Elmar |
Comments: |
@James C: The difference between my method and setting a button to repeat is that I control every single click that is done directly. And I never made it a secret. I shared my idea with Détrusor (who btw didn#t reject it) half a minute after I had it, long before I finished my first board with it! |
![]() |
|
Aug 31st 2005 at 07:54:40 AM |
|
Name: |
Serpenter |
Comments: |
if there really would be an invalid who could compete with us using the default Windows(R) help_control engine, would this be consideret as a cheat? i mean a handicapped, without finger for example. He used ight hand for moving mouse, left for clicks. Is my method cheat ? |
![]() |
|
Aug 31st 2005 at 07:23:22 AM |
|
Name: |
Serpenter |
Comments: |
[Let's say you set [any] key for instant repeat] - if you hold key this utility doesnt repeat key press. |
![]() |
|
Aug 31st 2005 at 07:16:49 AM |
|
Name: |
James C |
Comments: |
On another note, I got the run-time overflow error on the clone. It said my history file was renamed to history_corrupt and a new file would be created. Does anyone know if Rodrigo can fix this? And on even another note, I passed the 10.00 total 3bv/s mark last night. I didn't really get that excited about it, but apparently it is some sort of milestone. I got a 4.60 on a 17 for a new beginner best of 4.72 3bv/s to put me at 10.13 total. |
![]() |
|
Aug 31st 2005 at 07:11:03 AM |
|
Name: |
James C |
Comments: |
Actually, you could click faster with the keyboard. Let's say you set [any] key for instant repeat, then set it for as fast as it is capable. Then you use the mouse to guide the cursor, and press the key for an onslaught of clicks. You would be able to clear a row of 10 squares as fast as you could move the mouse down the line. This method, although exaggerated greatly, is exactly what Elmar did. Before he told the truth about his video, I was actually willing to try to defend him. I played the board several times, and managed a 13 with 65 clicks by clicking continuously and guiding the mouse. I thought, "Well, since he is much faster than I, maybe he would be capable of that time using a similar, although bizarre, method." I think the fact that he didn't immediately admit what he was doing when he started his trial of this method supports the fact that the 11 should not be accepted. The fact that he probably could have made an 11 on the board anyway just makes it a little more sad. Bottom line is this: If he had in fact made a 10 using this method, would it have been accepted by the community as a new legitimate world record? I think the answer to that is NO. |
![]() |
|
Aug 31st 2005 at 06:52:39 AM |
|
Name: |
Elmar |
Comments: |
@Rilian: I would appreciate if you stopped spamming with overlong articles and let other people share there opinions. And yeah, I can't prove I didn't use a solver so my method is cheating? Great reasoning! ![]() ![]() To comment on Serp's post (without answering his question), I tried using the keyboard as he proposes and I can't imagine anyone doing faster or even decent times with that. For one, you can't really click faster. And on the other hand you introduce the addtional difficulty of coordinating your left and right hand. Just to calm down this whole "what-if-people-used-that-and-that-setup-hysteria"... ![]() |
![]() |
|
Aug 31st 2005 at 06:36:26 AM |
|
Name: |
Serpenter |
Comments: |
I have found, that it is possible to transfer left (right, both) clicks on the keyboard without use of any programs, only through standard utilities. My question is not about useful. It is cheat ??? |
![]() |
|
Aug 31st 2005 at 06:33:02 AM |
|
Name: |
Dmitriy |
Comments: |
YES, you are right absolutely, Christoph. IMO the vid should be legit if every one (and not more than one) of these events is produced by one movement of the players hand (or any other part of his body): like, you press a mouse button down (sending MB_LEFTDOWN winAPI message) and mouse controller MUST deliver the OnLeftMouseDown event AND NOTHING MORE. The working events are: LButtonDown, LButtonUp, RButtonDown, RButtonUp, MouseMove and MButtonUp and MButtonDown. Otherwise it IS cheating. Everything that may be applied to the concepts, listed below MUST be considered as cheating. there are ways to detect winAPI mouse-related messages, so, this is not the problem at all. the problem is that some minesweeper buddies like to think that this should give them the legal (for the Respect Minesweeper Community) rights to have such video file (in any format, but preferrable internal-logging one) as a legal hi-score. i do not call Elmar a cheater, because his new vids are not accepted for bestever, but i should note, WHY the hell WE NEED to have a bunch of such vids and share the semi-computer-produced files? i may produce clone/recorder/camtasia video files with the limits about int 2 sec and exp 15 sec, but do anybody need them, do anybody need to waste his time to share opinions like "i have done int 5,23 seconds!" and "haha, and i did a 5,22! i am the best"?? @Elmar: 1. what's the purpose of trying your method?? 2. solver is involved in A case, because you cannot proof that that your vid used just logithech or other firmware but not the primitive solver, which better should be called CLICKER @Christoph: you know, i thought that your tel file had a format like "(1) mouse moved to x.y;(2) left click done;(3) cheat option "infinite lives" used;" (due to the caption "text eventlog") when i posted reply to you, but some later i downloaded your vc6 program and saw the real situation. i do not think that your 1996 replay engine used something new for the game replay recording, as, for example, chess gameplay recordings were knows yet since the Middle Ages. i think i can't supply you with the avf specifications. as i wrote, i do not see the reason of implementing pure cheat options support to the arbiter video file, but i may produce the "semi-legal" ".tel" file recording if a group of sweepers would see any reason to do it. |
![]() |
|
Aug 31st 2005 at 05:50:48 AM |
|
Name: |
Elmar |
Comments: |
@Dmitriy: As Christoph already pointed out, the difference between A and F is obviously the solver involved in F. @Christoph: Your definition also excludes using a third button for for double-click (l+r) as it produces two down-actions when pushing and two up-actions when releasing. However this method has been considered as legit for years. @At all: Has anyone even tried the method before refusing it? |
![]() |
|
Aug 31st 2005 at 05:26:40 AM |
|
Name: |
Christoph |
Comments: |
@Dmitriy: Your red button theory is known and discussed. We call such programs SOLVER. And yes, it's very easy to program a SOLVER if you know how to handle mouse messages. But it doesn't help to control the complete system. You still could have a computer on the other end of the mouse cable which pretends to be a mouse. With a webcam you could fix the rest. E.g. we discussed such a case here in vienna at the meeting. So if there's anyone who really wants to cheat he can, no matter what we try against (as long as we cannot send a judge to every player to watch him sweep ![]() @all: The working events are: LButtonDown, LButtonUp, RButtonDown, RButtonUp and MouseMove. I think, that a game should be legit if every one (and not more than one) of these events is produced by one movement of the players hand (or any other part of his body). If I understood everything right Elmars setting produced a LButtonDown and a LButtonUp when pressing the button down (two events with one movement). So in my humble opinion I wouldn't call the technique legit. (I do NOT claim Elmar to be a cheater, but I think we should talk about what we accept as legit mousefunctions.) BTW (@Dmitriy again) I wouldn't be that sure that I can't handle your format - cou could give it a try ![]() |
![]() ![]() |
|
![]() |
|
Aug 31st 2005 at 04:29:47 AM |
|
Name: |
Dmitriy Sukhomlynov |
Comments: |
WARNING! HACK FOUND! (nice caption, isn't it?) ---------------------------------------------- First, i ask to read this article the Elders of the community. Please, read ALL this text and only then write an answer. Else, please, pay attention on the CAPITALIZED words! DEFINITIONS: ----------- 1)you - a sweeper who agrees with the cheeting nature of so called "elmar technique" without prior knowledge of the next article. 2)you - a sweeper that uses NF-style sweeping or going to change his style to NF AND accept "elmar techinque". 3)you - a sweeper who finds "elmar technique" funny/UPK and do not think it will fit him. 4)you - a sweeper who does not fit the previous definitions. driver - an application that used to control the hardware. 5)XMan - imaginated sweeper 6)Company - imaginated driver producer. single programmer or a software engineering group. 7)mouse - a hardware that may be supported with a driver application (program) ACKNOWLEDGEMENTS ---------------- 0. i suppose you are sure, , if you will set a hi-score using this method, it SHOULD be accepted. 1. i understand very well this "technique". you press left mousebutton Down over a non-bomb-cell, then you move mouse to the next non-bomb-cell and do a mousebutton Up. with this "technique*" you may at least double your clicking speed if you will fit with your new mouse setup. this was intro. then, imagine: A) a man called XMan WRITES a program called Cheat_1.exe, that, __CONTROLLED_BY_HUMAN__ ACTION (LET US SUPPOSE THIS ACTION WOULD BE "PRESSING SPACE BAR (generally, any other button you wish except Reset, may be programmed for this purposes)") allows for an NF-sweeper to do SOME EXTRA (ADDITIONAL) mouse clickings over the target application - minesweeper. AND - this is important - this Cheat_1.exe sends the winAPI_system_message_LeftMouseButtonDown and winAPI_system_message_LeftMouseButtonUp simultaneously to the mouse CONTROLLER CHIP - thus --> the Cheat_1.exe program is a DRIVER - an application used to tweak the hardware functions. So, all the hi-scores produced in this way, should be accepted Taking this concept as the default value, just imagine that Cheat_1.exe uses not THE ONLY keyboard's "Space Bar" button, but EVERY OTHER BUTTON. let's call this program Cheat_2.exe --> so, this XMan just need to move his mouse cursor to a correct position and use all the available computer's buttons to produce leftclicks. And no matter how many hands does he use. B) Case is the same, but a Company supplies a XMan with a mouse and DRIVER working in the way of a Cheat_2.exe program C) Company supplies a XMan with a mouse and DRIVER working in the way of a Cheat_1.exe program. as you see - this is the "elmar technique" case, which is currently accepted by some sweepers. there IS NOT the concept difference between the ways Cheat_1.exe and Cheat_2.exe work, so... let us call this "technique" a "SILENT CHEATING TECHNIQUE" D) this is more ""interesting" case. I) the same Company supplies XMan with mouse and a Cheat_3.exe DRIVER. Cheat_3.exe program generates an interruptable thread of LeftClicks when XMan holds mouse button pressed and interrupts the LeftClicks thread when the mouse button released. II)the same Company supplies XMan with mouse and a Cheat_4.exe DRIVER. Cheat_3.exe program generates an interruptable thread of LeftClicks when XMan holds ANY button pressed and interrupts the LeftClicks thread when this button released. anyone see the difference between I) and II) ? i do not. this is a "SILENT CHEATING" - quite obvious one, isn't it? E) Company supplies XMan with mouse and a Cheat_5.exe DRIVER that recognizes the board and, if needed - if a cursor if over the non-bomb-cell (this is calculated by a probability calculator) - sends one or more winAPI_LeftMouseButtonClick messages to the controller for the producing of the leftclicks. Along with this, normal left clicks of the mouse are fully supported. it is obvious (at least to me), that: I)an AVI video, produced in this way may not be proofed as a cheating. II)video, produced in this way by a Recorder (any one) may not be proofed as a cheating without prior defense concept knowledge and full winAPI system messaging logging. so, i would not call this "SILENT CHEATING" because it IS CHEATING. F) finally, here is a piece secret "Brutal Overmind" concept, that may annihilate the Respect Minesweeper Community. I constantly did not want to write this, but you hitted me to do this. i am holding the finger on the BIG RED BUTTON. I)Company supplies XMan with mouse and a Cheat_6.exe DRIVER that recognizes the board and, if needed - if a cursor if over the non-bomb-cell (this is calculated by a probability calculator) - sends one or more winAPI_LeftMouseButtonClick messages to the controller for the producing of the leftclicks. Along with this, normal left clicks of the mouse are fully supported. II) DRIVER recognizes the board and, if needed - if a cursor if over the bomb-cell (this is calculated by a probability calculator) - sends one or more winAPI_RightMouseButtonClick messages to the controller for the producing of the rightclicks. Along with this, normal right clicks of the mouse are fully supported. III) DRIVER recognizes the board and, if needed - if a cursor if over the known non-bomb-cell rounder with the exact amount of flags (this is calculated by a probability calculator) - sends one or more winAPI_DoubleMouseButtonClick messages to the controller for the producing of the doubleclicks. Along with this, normal right clicks of the mouse are fully supported. THIS IS THE **** ****** CHEATING! WHY THE HELL THIS "ELMAR TECHNIQUE" MAY BE ACCEPTED BY SOME SWEEPERS IF THERE IS NOT MUCH CONCEPT DIFFERENCE BETWEEN A) and F) CASES??? THINK YOURSELF, WHILE IT IS NOT TOO LATE. THANKS FOR READING. |
![]() ![]() |
|
![]() |
|
Aug 31st 2005 at 03:44:29 AM |
|
Name: |
Elmar |
Comments: |
@Damien: Yeah, I also used to play solitaire with that setup (Best48sec ![]() ![]() About the clicks-count on the 11. I did it too, but I only found 3 or 4 clicks done by releasing. I guess the rest must have been during the wildly clicking at the start. Btw, it's really funny to watch the vid at 4% speed and to see the clock tick passed 10. ![]() I think for the 49 the ratio should be similar. I only took a look at the first clicks because it would take about 20minutes to analyse it completely. Allthough, I'm pretty sure I didn't use it at all during the last 15-20 secs. ![]() |
![]() |
|
Aug 31st 2005 at 02:59:27 AM |
|
Name: |
AreOut |
Comments: |
Ive seen your profile on planet minesweeper site and there is nothing about my 41. ![]() |
![]() |
|
![]() |
|
Aug 31st 2005 at 02:08:29 AM |
|
Name: |
damien |
Comments: |
@everyone commenting on elmar's methods: i used to play solitaire a lot on Win98...to put a card on the foundation you had to doubleclick it...so, i set the right mouse button to doubleclick, and my times improved by 5-8 seconds. All the best players did so. XP now automatically makes the right button a doubleclick. I initially thought elmar had set both buttons as 'left clicks' so was playing with 2 fingers. ![]() Although most mouses don't use doubleclick the way elmar's logitech does, it is still a legitimate mouse function and not a written program. If anything, elmar's very high in-efficiency shows that it is going to take a lot of skill to master the technique...if it can be done. (elmar did 31 downclicks and 9 upclicks that cleared squares...but 80 clicks overall. you have to be careful not to release by accident) just food for thought ![]() |
![]() |
|
Aug 30th 2005 at 08:12:51 PM |
|
Name: |
Dmitriy |
Comments: |
congrats. no, that night i did 3 24sec games, so having 24x7 now. grr! by the way, i had 25x6 before my sub25 dream realized :) so, if you see the same symptomes like mine, your next sub26 should be 24.xx game :) happy sweeping! |
![]() |
|
Aug 30th 2005 at 07:54:23 PM |
|
Name: |
Arjádre |
Best intermediate: |
25x5==>25x6 |
Comments: |
My first 25 on Arbiter- 25,72 on a 42 board. It seems Dmitriy and I have the same problem- just a second apart ![]() |
![]() |
|
![]() |
|
Aug 30th 2005 at 05:09:24 PM |
|
Name: |
Elmar |
Comments: |
PS: Nobody ever called Matt a cheater. |
![]() |
|
Aug 30th 2005 at 05:06:51 PM |
|
Name: |
Elmar |
Comments: |
@James C: Please try the method before you get all judgemental. You must not forget that I got these scores on really nice boards. Maybe it doesn't even have a big impact. That's why I asked you to form yourselves a picture, before you comment. That way we can have a well-founded discussion on this subject. ![]() |
![]() |
|
Aug 30th 2005 at 04:16:48 PM |
|
Name: |
Dmitriy |
Comments: |
@AreOut: i have some ideas how ur 41 was done. i may talk pm about that. and read my ar profile for acknowledgements of ur methods. haven't seen you at #minesweeper today. |
![]() |
|
Aug 30th 2005 at 04:15:32 PM |
|
Name: |
James C. |
Comments: |
Okay, so let me get this all straight... One guy unintentionally memorizes a board. He didn't practice the board on another program. All he did was watch a video a few times. (We all watch our own videos after all.) So he gets the board, is able to recognize it, has very quick reaction times, plays through the board, has the hand-eye coordination to place nearly every click exactly where he wants to in a remarkably low time. All of this was achieved by the power of his own brain and his own hand-eye coordination. Yet he's branded a cheater. Another guy uses a program to slow down the clock on purpose to get "better" times. He's also labeled a cheater. (Which in this case is appropriate.) But I just have to know how in the HELL that is any different than a program that allows you to click faster than what would normally be considered humanly possible. Seriously? After all this crap about UPK and you post your new "best" times with a ![]() |
![]() |
|
Aug 30th 2005 at 04:13:21 PM |
|
Name: |
AreOut |
Comments: |
@Christoph: doesnt work and doesnt matter, I will look at it in some Internet Caffe @Dmitriy: if everyone on chat except me think that Arbiter isnt cheatproof because I did 41 on it you better stop further development... That is "community" ![]() |
![]() |
|
![]() |
|
Aug 30th 2005 at 03:33:31 PM |
|
Name: |
Dmitriy |
Comments: |
Elmar, try to create such a video on arbiter. i 98% sure that logitech hardware would not be detected, but the engine 100% WILL shouw your double-left-clicking easily. Just move the replay speed tracker to 0.01 sec position and see the clicking counters result. As well as engine can see the one-moment doubleclicking with the same mouse button. The same for Sorin's Recorder, which shows the clicking period time length for every replay. |
![]() |
|
Aug 30th 2005 at 03:17:34 PM |
|
Name: |
Elmar |
Best expert: |
48 |
Best intermediate: |
11 |
Best beginner: |
1 |
Comments: |
Ok, as Roman already pointed out there is something odd about the 11 and the 49 I achieved last night. ![]() I wanna briefly explain what I did so you can all tried it out yourselves and decide what you think of it. ![]() I used LogitechMouseware to reconfigure my mouse in a way that the left button now does doubleclicks (2 left clicks). Logitech Mouseware can be downloaded, just google for it. I don't know if it only works for Logitech Mouses or for all brands. If it doesn't I'm sure there are similar programs that work for the others. The effect is that now my left button does a left click when I push it and a 2nd left click when I release it. This allows me to click faster while I still control every single click myself. And that is why the clicks counter shows such high values. ![]() Apparently this method works pretty well for me. To get a better idea of what I'm talking about I'd like you all to test it and see how it works for you. Once you've made yourself a picture feel free to post you comments here. ![]() |
![]() |
|
Aug 30th 2005 at 03:17:29 PM |
|
Name: |
me again |
Comments: |
i am very sorry for excess posting. @Christoph: your minesweeper emulator has been ranked at my Minesweeper Defense Ranking. as your project just started, and i think you should know the result (20 of 23 points). |
![]() |
|
Aug 30th 2005 at 02:49:16 PM |
|
Name: |
Dmitriy |
Comments: |
omg! i just invented the Blood minesweeper feature... hoho! this would be horrible! coming soon. you will see more when i release the next arbiter version. hmaaha. i am too nervous! can't wait to apply it! |
![]() |
|
![]() |
|
Aug 30th 2005 at 02:22:28 PM |
|
Name: |
Dmitriy |
Comments: |
@everybody and mainly Christoph As you all know (i hope), Arbiter CAN capture videos in CUSTOM mode, which is still "cheat" in the Clone. this means, you can do for custom mode: record/save/load/replay video files (avf) as well as saving/loading the exact board (available for finished and blasted games - this means that you can't save board while the timer running and then review this board and thus use upk for the remaining game). please, note, that i do not see the reason to provide full-featured _video_-recording for the pure cheat mode with, for example, "infinite lives" or "show mines positions" options. as for avf format usage, i may tell that 99% you could not handle it because of the extra large amount of blocking and iterational-style checksum protection.. i mean only Arbiter process can at least read it and analyse.. i even do not know how how your program could use above 800 variables for generating checksums.. (btw, avf handling source code takes about 200KB) but, i should tell that this is a nice idea to store a the replay in the event list. i do not know when you had invented this idea, but should note that my cheat-analysing tool (not releaseed for public usage) generates for me all the game event lists, including every game-parameter changes, allowing me to reveal all the avf creation process. ok, tell me your tel format specifications and in the next arbiter version i will include the "tel" public storage option. anyway, you may ask me everything about minesweeper-related program creation. i always help ppl if i can by some of my principles, like enlarging the planetary knowledge base concept. lol. and assept all good ideas as well. as for general_upk ideas, i have started writing the Minesweeper Board Editor Pro (plugin for Arbiter). as for base format i took Yoni's MBR format (Arbiter uses it, btw, saving files with .avf extension) questionmarks are for losers. my personal opinion is that it would be better to exclude them at all and do not waste some kilobytes of source code and winPE space. @Stevan: i can't why don't you download latest FireFox version and forget about browsing problems?? @hey, everybody, there still unclosed poll (at Minesweeper-Addicts Yahoo Group) about Arbiter acceptance. do not forget to add your vote. |
![]() ![]() |
|
![]() |
|
Aug 30th 2005 at 01:43:27 PM |
|
Name: |
Dmitriy |
Comments: |
i may tell i am having some problems with french and german translations too, but i had found a very good online text translator, very fast, i must say, wich allows to translate the texts 0.5KB size at once. multi-directional translation for most popular european languages .here are the links: http://www.online-translator.com/text.asp?lang=en http://www.online-translator.com/text.asp?lang=fr http://www.online-translator.com/text.asp?lang=de http://www.online-translator.com/text.asp?lang=ru good luck |
![]() ![]() |
|
![]() |
|
Aug 30th 2005 at 01:43:08 PM |
|
Name: |
Christoph |
Comments: |
There was something wrong with the direct link ![]() |
![]() |
|
![]() |
|
Aug 30th 2005 at 01:40:55 PM |
|
Name: |
Christoph |
Comments: |
I like this UPK board idea. But it again reminded me, that clone cannot capture vids in cheat mode. Thatwhy I made some changes in a clone I wrote some time ago. You can download it from my site (linked below). It captures vids as a text event list (*.tel) so I'd suggest to use this format for cheat-games ![]() I also added the feature that you not need to reload the board everytime you want to play it but that you can restart it by clicking the smiley (that works only if you loaded a board, otherwise you'll just get a new board). Since the program uses the MT-randomizer it's VERY probaly that it can produce nearly every board, so I added 3bv-limits for normal games (4-27-100). There's also a flag-analytics and a middle button=F2 feature. @AreOut: If you still cannot display my site, try to write a ?style=2 after the index.php (if you follow the link below &style=2 after the end). @Rodrigo, Sorin & Dmitriy: I'd like to include your video formats to my Video-Player. Would be great if you could hand me over the formats, I promise to treat them trustworthy. @all: Is there anybody who uses Questionmarks? |
![]() ![]() |
|
![]() |
|
Aug 30th 2005 at 01:28:29 PM |
|
Name: |
Jan Dubois (Iguana) |
Best expert: |
60 |
Best intermediate: |
13 |
Best beginner: |
1 |
Comments: |
Hello Everybody! I'm really bad in Englsih so I hope you will understand me. I know that Hopsing Elmar AreOut Detrusor Dion had a discussion about the new records of Elmar. Elmar say to me his tips and I try it since saturday night. I love this double-clic and I recommand everyone to try it, before to say that's doping cheat or whatever... PLEASE TRY IT and we will talk about this double-clic after these tries. I wish that the best NFers will try too, and I'm sure that they will love this tip. ![]() |
![]() |
|
![]() |
|
Aug 30th 2005 at 10:25:04 AM |
|
Name: |
Arjádre |
Comments: |
UPK Records can now be sent to the email address below ![]() |
![]() |
|
![]() |
|
Aug 30th 2005 at 10:15:46 AM |
|
Name: |
AreOut |
Comments: |
Yeah Roman now I understand ![]() |
![]() |
|
![]() |
|
Aug 30th 2005 at 10:15:12 AM |
|
Name: |
AreOut |
Comments: |
Yeah Roman now I understand ![]() |
![]() |
|
![]() |
|
Aug 30th 2005 at 09:54:37 AM |
|
Name: |
Arjádre |
Comments: |
Wow, Elmar- congratulations... ![]() The news has been updated at my site |
![]() |
|
![]() |
|
Aug 30th 2005 at 03:47:49 AM |
|
Name: |
Roman Gammel |
Comments: |
@Areout: Look at cl/s, you'll understand... But I'm sure you think it's ok ![]() |
![]() |
|
Aug 30th 2005 at 03:28:34 AM |
|
Name: |
AreOut |
Comments: |
Roman what is there not to like?! ![]() |
![]() |
|
![]() |
|
Aug 30th 2005 at 02:57:56 AM |
|
Name: |
Elmar |
Comments: |
What's there not to like? ![]() No UPK, no stupid guessing,reasonable risk, pattern recognition is ok, speed is good, times are excellent! ![]() |
![]() |
|
Aug 30th 2005 at 12:21:52 AM |
|
Name: |
Roman Gammel |
Comments: |
@Elmar: I don't like the way you achieved this results. ![]() ![]() ![]() |
![]() |
|
Aug 29th 2005 at 06:09:01 PM |
|
Name: |
Dmitriy |
Best intermediate: |
24x4 ( 42bv ) --> 24x5 ( 47bv ) --> 24x6 ( 51bv ) --> ?? |
Comments: |
Feels like todays night is the HI-SCORE NIGHT!!! DO NOT MISS YOUR CHANCE :) |
![]() ![]() |
|
![]() |
|
Aug 29th 2005 at 05:40:24 PM |
|
Name: |
Minesweeper Community IRC Channel Promotion |
Comments: |
hey, do not forget about the community IRC channel!!! Join us at channel: #minesweeper server: irc.initialized.org port: 6667 nickname: yourname validity: 7/24 ircbot: accepted param: value password: tell us |
![]() |
|
![]() |
|
Aug 29th 2005 at 04:55:33 PM |
|
Name: |
Elmar |
Comments: |
I uploaded the videos to the yahoo group. Go there or just follow the linkg below. ![]() |
![]() |
|
![]() |
|
Aug 29th 2005 at 04:37:07 PM |
|
Name: |
Elmar |
Best intermediate: |
NF: 14--> -->11,28 |
Comments: |
Unbelievable!!!! ![]() 32 3bv, stupid hesitation at the end made me miss a possible 10!! ![]() I had the feeling there was more to come... ![]() |
![]() |
|
Aug 29th 2005 at 04:14:58 PM |
|
Name: |
Elmar |
Best expert: |
NF: 57 ---> ---> 49,62!!!! |
Comments: |
Yabadabadoooohh....!!! ![]() 3bv119, I could've even been quite a bit faster if I hadn't been so nervous and the finished hadn't been that nasty!! And nobody's on the Chat to share it!! ![]() I'll supply the vid as soon as Détrusor gets his the ftp of his forum fixed. ![]() ![]() |
![]() |
|
Aug 29th 2005 at 02:08:58 PM |
|
Name: |
Dmitriy |
Comments: |
else, if you could remember the time length of the syndrome, this would be some better |
![]() |
|
![]() |
|
Aug 29th 2005 at 02:01:10 PM |
|
Name: |
Dmitriy |
Comments: |
24x4 --> 24x5 (max 3bv=47) Well known symptomes... Btw, almost reached 200 sub30 int games barrier (with 2300 finished games) :) Else... you remember, the initial Syndrome list was created... I will/may manage this project and appreciate much your help: Please, send me all the maximal syndromes you remember (for all modes)! |
![]() |
|
![]() |
|
Aug 29th 2005 at 10:57:00 AM |
|
Name: |
Arjádre |
Comments: |
Just got a 3bv 22 or 23 board on winmine int that doesn't appear to be a dreamboard shift... ![]() Download mbf version below |
![]() |
|
![]() |
|
Aug 29th 2005 at 09:42:05 AM |
|
Name: |
Rasmus H. Jensen |
Best expert: |
71,54 |
Comments: |
Just finished a 71,55 ![]() Soo close... |
![]() |
|
Aug 29th 2005 at 08:25:13 AM |
|
Name: |
Rasmus H. Jensen |
Best expert: |
2,63 ====> 2,68 3BV/s |
Comments: |
81,94 on a 217... |
![]() |
|
Aug 29th 2005 at 07:52:36 AM |
|
Name: |
kkx3 |
Best expert: |
207 |
Best intermediate: |
50 |
Best beginner: |
8 |
Comments: |
A good game. |
![]() ![]() |
|
![]() |
|
Aug 29th 2005 at 05:38:50 AM |
|
Name: |
Elmar |
Comments: |
Thanks, Damien, Arjàdre and Jan. I have the feeling there is yet more to come. ![]() About the UPK, IMO the boards (especially int) could nevertheless be a bit more complex. I mean, something you couldn't do a sub4 on... ![]() |
![]() |
|
Aug 29th 2005 at 03:47:27 AM |
|
Name: |
Jan Dubois |
Best expert: |
60 |
Best intermediate: |
13x2 |
Best beginner: |
1x7 |
Comments: |
Wow, congrats to you Elmar! 57.88 @ 2.95 in NF! ![]() ![]() |
![]() |
|
Aug 28th 2005 at 07:31:55 PM |
|
Name: |
Arjádre |
Comments: |
You no longer need to subscribe to get the UPK boards via email. I uploaded the boards at a different host and provided links at the main UPK page. ![]() ![]() |
![]() |
|
![]() |
|
Aug 28th 2005 at 06:47:32 PM |
|
Name: |
Oli |
Comments: |
just did a 12.06 on a 43 3bv board and something is wrong with the gb didn't let me post |
![]() |
|
Aug 28th 2005 at 06:47:31 PM |
|
Name: |
Oli |
Comments: |
just did a 12.06 on a 43 3bv board and something is wrong with the gb didn't let me post |
![]() |
|
Aug 28th 2005 at 05:13:30 PM |
|
Name: |
Dmitriy |
Comments: |
Arbiter 0.42 released, among with some interface and links fixes, i had implemented: Index of Speed calculator for the statistics files. There is a "Index of Speed" button in the Statistics-Seearch window, wich allow to calculate average index of speed for the defined (meant, you define, the percentage of excluded games) - searched! - statistics part. Note, that you may sort the games data by decreasing of the 3BV/s before the index of speed calculation for the presize result. Among the index of speed, the IOS normal distribution calculated. As there is no sense to make forecasts for beginner mode (guess why), forecasts are done for both expert and intermediate modes simultaneously, so, just search one of these modes and get your foresast. but anyway, do not forget that probability of setting the highest (forecasted) records is very little. Ah, i forgot! you may easily load your Clone history file and work with it using Arbiter stats/search engine! Plugins support! you know, what a plugin is. so, on the startup, arbiter searches for available plugins in the /plugins/ folder and creates a drop-down list in the options_menu--plugins-- just select proper plugin and get a simbiotic application. Currently available the ported from my exe "Minesweeper Calculator", and in the nearest future i am going to create a Minesweeper Board Editor (pro) which would work with Reader/Clone/Arbiter files (ho, format is the same - MBF (minesweeper Board File)). As you know, Arbiter may save the replays of a custom games, so, this UPK tournament would be quite funny soon (remember Stephen's suggestions). if you have any plugin idea, just tell me; and, by the way, i released a demo plugin source code, so, if you know Delphi or C++ Builder, you may try to make some. |
![]() ![]() |
|
![]() |
|
Aug 28th 2005 at 04:05:35 PM |
|
Name: |
Arjádre |
Comments: |
@Elmar: Congrats on your record... I wish I could upload the boards as .mbf files, but my host won't allow it. (it only allows the most common types of files to be uploaded, such as .doc, .jpg, .html, etc) @All- But in response to this problem, I will allow people to "subscribe" to the contest. Subscribers will automatically be emailed the boards as an attachment. Sign up at my website's guestbook with your email address. ![]() |
![]() |
|
![]() |
|
Aug 28th 2005 at 02:50:50 PM |
|
Name: |
Elmar |
Best expert: |
NF: 59-->57,88 |
Comments: |
3bv168, so I beat my 3bv/s record on the same occasion with a 2,95 ![]() @Arjadre: Love the UPK idea! But can't you make the boards available for download as clone board files? I'm sure that would make a lot more people play. ![]() |
![]() |
|
Aug 28th 2005 at 02:50:47 PM |
|
Name: |
Elmar |
Best expert: |
NF: 59-->57,88 |
Comments: |
3bv168, so I beat my 3bv/s record on the same occasion with a 2,95 ![]() @Arjadre: Love the UPK idea! But can't you make the boards available for download as clone board files? I'm sure that would make a lot more people play. ![]() |
![]() |
|
Aug 28th 2005 at 01:46:38 PM |
|
Name: |
Arjádre |
Comments: |
Announcing new weekly contest at my website- remember UPK? |
![]() |
|
![]() |
|
Aug 28th 2005 at 12:57:26 PM |
|
Name: |
Dmitriy |
Comments: |
Hi all! I had release my new program - Minesweeper Calculator. This small and useful tool allows you easily calculate almost all known minesweeper speed/quality measurement values, such as 3bv/s, Index of Speed, RPQ (Rapport Qualite Prix) and inverted RPQ - the value suggested by James. Check it out at the downloads page of my site. If anyone wants to add his desired value, just tell me the formula. |
![]() ![]() |
|
![]() |
|
Aug 27th 2005 at 04:35:33 PM |
|
Name: |
Dennis Lütken |
Comments: |
Something's still terribly wrong with the clone rankings... Mads Dørup is in 1st position on the 3bv/s ranking with a beg 3bv/s of 35.56!!! That value is nowhere to be found in his stats so I don't know how this has happened but it's nice to see the Danes rule the 3bv/s rankings for NF AND flagging! ![]() |
![]() |
|
Aug 27th 2005 at 04:29:31 PM |
|
Name: |
Dennis Lütken |
Best expert: |
59x2 ---> 59x3! |
Best intermediate: |
13x4 |
Best beginner: |
1x3 |
Comments: |
59.42 - 140 3bv - NF, of course! No new record but still nice. I finished 8 NF games in less than 1½ hours and for me that's extremely good! ![]() @AreOut: Thanks, man! ![]() |
![]() |
|
![]() |
|
Aug 27th 2005 at 04:16:08 PM |
|
Name: |
Arjàdre |
Comments: |
sorry- bad link |
![]() |
|
![]() |
|
Aug 27th 2005 at 04:15:16 PM |
|
Name: |
Arjàdre |
Comments: |
I added another page to the site, but had to remove/replace some personal information. My parents may or may not allow me to update it in the future. ![]() ![]() @Dmitriy: I hope this answers your question |
![]() |
|
![]() |
|
Aug 27th 2005 at 02:09:58 PM |
|
Name: |
AreOut |
Comments: |
Dennis you really rock that NF ![]() |
![]() |
|
![]() |
|
Aug 27th 2005 at 12:43:42 PM |
|
Name: |
damien |
Comments: |
may i be first to congratulate you, elmar! it's been a long time in the making! ![]() |
![]() |
|
Aug 27th 2005 at 10:35:30 AM |
|
Name: |
Elmar |
Best expert: |
60-->59,01 3bv155 |
Comments: |
every day a new record! ![]() |
![]() |
|
Aug 27th 2005 at 07:43:23 AM |
|
Name: |
Dennis Lütken |
Best intermediate: |
4.20 3bv/s |
Comments: |
I just missed another >4 3bv/s on int NF... 18.42 when the timer stopped, 72/74 3bv solved and I click the wrong because of clumsiness... I only had two easy squares left to the left of the mine I clicked and I knew that. Too bad. It was est. 4.13 3bv/s so it wouldn't have been a new record but since my second best is 3.97 it still would have been pretty sweet. Anyway, it's nice to know that it was not just once I was lucky or good enought to get above 4 3bv/s! ![]() @James C: I actually like your suggestion and I think it's so nice to see you contribute something regardless of your best scores! Good job! ![]() ![]() |
![]() |
|
![]() |
|
Aug 27th 2005 at 04:05:53 AM |
|
Name: |
Dmitriy |
Comments: |
concerning index of speed/quality debates, i am going to ad a "calculator" to my program. You just input the Time and 3BV values, select the desired "Index" from the drop-list and get the value ![]() total stats index calculator already added and tested (my exp index seems to be 1,09 ![]() |
![]() ![]() |
|
![]() |
|
Aug 27th 2005 at 03:57:52 AM |
|
Name: |
Traian |
Comments: |
@Arjádre: Great idea, great site. Keep it up! ![]() |
![]() |
|
![]() |
|
Aug 26th 2005 at 11:19:43 PM |
|
Name: |
James C |
Comments: |
Roman, I do understand that post much better. You'll have to forgive, as I said, it has been 13 years since I have done more math than I can reasonably do with a piece of paper or a generic calculator. I also never doubted the accuracy of your indexes, and I realize they would be different for different players/games. I think yours is probably a better way to go for a large database such as that Detrusor maintains. I do like my method, for some simplicities sake. I can do most of it in my head for starters, and even the square roots are not much of a problem because of the wide range of results that will all yield the same number on the timer. I am not at all trying to discredit your post, and I do appreciate the extra effort you made to explain it. I just wanted to throw a simple idea out there, and see what people thought. |
![]() |
|
Aug 26th 2005 at 11:08:22 PM |
|
Name: |
Arjádre |
Comments: |
My site is now complete and will hopefully be updated regularly. ![]() |
![]() ![]() |
|
![]() |
|
Aug 26th 2005 at 10:57:33 PM |
|
Name: |
Roman Gammel |
Comments: |
@James C: Probably you didn't understand the idea of index. You're talking about 3BV/s^2 instead of 3BV/s, i think this value is not better than 3BV/s. I'm suggesting the value 3BV/s^{degree X} where X is not 1 or 2. This X is personal number of each player - index to compare their speed. If you will look again you'll see that from formula Index=log_{time} 3BV you can obtain 3BV=time^index - that's what i'm talking about. All math stuff I wrote was how to count it in general, that is most accurate way, but you can count it at home for each of your games, main value is just average of values of games in your history. For example: my exp record is 43,73 on 3BV 135, log_{43,73-1} 135 = 1,3064 - even better than Oli's average, this value is greater than my average, but for me probability to get such hi result is <0,01, almost all my games with very high probalitity are concentrated in segment [average-0,01;average+0,01] - this is the main idea, this value is good statistical index because it's almost constant!! |
![]() |
|
![]() |
|
Aug 26th 2005 at 09:23:24 PM |
|
Name: |
Arjádre |
Comments: |
Thx Dmitriy @all: my site now has four separate pages... are there any active Jewish players out there? ![]() |
![]() ![]() |
|
![]() |
|
Aug 26th 2005 at 09:22:43 PM |
|
Name: |
Arjádre |
Comments: |
Thx Dmitriy @all: my site now has four separate pages... are there any active Jewish players out there? ![]() |
![]() |
|
Aug 26th 2005 at 09:03:46 PM |
|
Name: |
Robert Benditz |
Comments: |
Furious Jim, chill .... ![]() I never meant to offend you. |
![]() |
|
Aug 26th 2005 at 07:48:40 PM |
|
Name: |
David Tucker |
Best expert: |
130 |
Best intermediate: |
40 |
Best beginner: |
5 |
Comments: |
OK, so I haven't played the game as much as most of, if not all, of the players in the record books, but I am not too shabby at it either. Maybe, with some additional practice, I will get good enough to make the records myself. I'll keep trying now that I know what the best times are. |
![]() |
|
![]() |
|
Aug 26th 2005 at 06:09:52 PM |
|
Name: |
damien |
Comments: |
i find some of these ideas quite exciting... on a different note: has anyone noticed that jon's 19s board from vienna has 79 squares in identical positions to the dreamboard? in all, it is 30.86% identical (to the DB 'map' i created) |
![]() |
|
Aug 26th 2005 at 05:01:10 PM |
|
Name: |
James C |
Comments: |
@Elmar: So what do you think of the validity of it? I never really played with the numbers in the inverse, but I can see how they may be more accurate because you might save a small percentage when rounding. I kinda liked the 3bv per second per second thing myself because it's sounds similar to acceleration. Either way, I was just trying to offer a simple math solution for the do-it-yourself stats geeks out there like me. |
![]() |
|
Aug 26th 2005 at 04:40:33 PM |
|
Name: |
Elmar |
Comments: |
@James: At the Détrusor's french forum we have introduce a value similar to yours. It's called Rapport Qualité Prix (Quality Price Ratio) and its defined as time/3bv/s, so it's basicly the inverse of yours. ![]() |
![]() |
|
Aug 26th 2005 at 04:31:08 PM |
|
Name: |
James C |
Comments: |
See this is why I stopped posting so much. All I was trying to do was offer my little effort at something to the community. I believe I even asked for other input... Yet I somehow cannot resist the temptation to get in an argument with someone over something stupid. Maybe when you sober up, schoolboy, you'll realize that I: a) said the lower score was better anyway, and b) never said anything about Dion's 18 (from the example) being "as good as" Oli's 11. Nor did I ever propose that this be an "end all, be all" method to compare two players, but more rather it be a method to compare your own games and possibly gain some estimate to your potential. The comparison between the 18 and the 11 was intended to show that a player who could do either should potentially be able to do both, a fact that Mr Tiu has already shown us all. Lot's of people talk about 3bv/s like it has some meaning, when really it is arbitrary, and, IMO, not as accurate as my own theory. You see people all the time post news of a new record, then immediatly complain that "the 3bv/s was so low, I could surely do better." But can they really? There's no real way of knowing, so I try to offer a simple way to at least do a reality check. My own personal best on intermediate may have looked a little frustrating to me for the 3bv of the board. But when I crunch the numbers based on my own stats, it is actually a better than average game. Either way, I am sure this is not going to find a timeless place in the community, but I really don't think it should be up to Mr. Hopsing to decide that. To those of you that agree, then I am sorry to have even bothered you with my efforts. Maybe when I get to be a "sub-60" player people will take me a little more seriously. |
![]() |
|
Aug 26th 2005 at 03:57:43 PM |
|
Name: |
Elmart |
Best expert: |
NF: 5x62 -->60,65 |
Best intermediate: |
NF: 3,89 3bv/s -->3,96 3bv/s |
Comments: |
141 3bv and 90 3bv respectively. ![]() |
![]() |
|
Aug 26th 2005 at 03:36:53 PM |
|
Name: |
Dmitriy |
Comments: |
yeah, this idea was suggested for Reid but he is still silent.. btw, i have some ideas how to estimate board difficulties and sweeper's speed too. but, anyway, i find Rogen's way quite good. I even implemented this IOS-calculator to my stats page and found my expert Index of Speed = 1,08xxx ![]() |
![]() ![]() |
|
![]() |
|
Aug 26th 2005 at 01:42:41 PM |
|
Name: |
AreOut |
Comments: |
Youre just banging your heads for no reason. 3bv 50 int board could be theoretically easier than 3bv 20 so no point. Only thing I can tell is to count shortest mouse path and minimal number of clicks (R D L) and mix it. ![]() |
![]() |
|
![]() |
|
Aug 26th 2005 at 01:25:34 PM |
|
Name: |
Robert Benditz |
Comments: |
matt's eleven was an worthless effort because he already knew the board, whatever it may be according to your formula. That's why we use the clone, to prevent that ppl score low times on original ms due to upk. Everyone of the top players who still plays original ms can easily be accused of cheating if he comes up with an 11 or whatever. |
![]() |
|
Aug 26th 2005 at 01:20:00 PM |
|
Name: |
Arjádre |
Comments: |
@all: check out my new website! ![]() |
![]() |
|
![]() |
|
Aug 26th 2005 at 01:19:45 PM |
|
Name: |
Robert Benditz |
Comments: |
hmm I dont see the point :S Why should we have a certain formula to compare games like olis 11 and dions 18 or 19. What you did is finding a formula that pretends to compare times. Everyone can make up such a formula, just trying out a lot of them until one thinks now it looks good to me. that is highly subjective. what is, on the contrary, important is examining a way how to define a boards difficulty for flaggers. I mean, u r saying with your formula that olis 11 is as good as dions 18 or 19. i dont really care what u think. for me olis 11 is much better. but who am I to judge your idea? A drunk german schoolboy. have fun ![]() |
![]() |
|
Aug 26th 2005 at 11:14:05 AM |
|
Name: |
James C |
Comments: |
I wanted to comment on Roman's recent posts. I did well in math, and I am usually no slouch in the smarts department, but there are a few facts I have to face. I never took advanced math past high school and it has indeed been 13 years since I even did that. While I have no doubt at all about the accuracy of his calculations, they did nothing but confuse me. I am sure Gregoire can make use of them in his ranking system, but what about me sitting in my own home? I had been thinking for a while of a system to easily compare games with simple math. It's easy to see where games with similar 3bv/s or times rank, but what about my 19 on a 47? Is that better than a 22 on a 66? The short answer, for me at least, is that the lower time is always better, but how can you really tell? Well here is what I have come up with for a speed index to compare various games. 3bv/s^2 or 3bv per second per second. First off let me say two things about my examples. 1 - this is not a sample of all my games, just a few notables 2 - I am in the process of rebuilding my computer, so I had to round off a number here or there So here is what I have come up with for my top 5 [notable] games using the 3bv/t/t method where t of course equals time. t 3bv index 19 52 .144 22 66 .137 19 47 .133 21 56 .130 20 51 .128 Now my 3bv/s record of 3.05 (24 on 71) comes in a little under those at .127 and my time record of 17 (34 3bv) is lower still at .119. For comparison's sake Dion's 18 at 5.03 3bvs comes in at .280 slightly rounded up and Oli's 11 (Congratulations) also comes in at .280, slightly rounded down. Until today I had not considered using this method to estimate times, but here is what I have come up with. Est t, in real seconds, not scores uses the formula t = square root of (3bv/i). From that you can also estimate 3bv values where 3bv = t^2i. So here's what I came up with using this tiny sample of numbers. Based on my last known averages on the clone of 25.88 and 2.32 3bv/s my typical index is .093. Using that index my average estimated time for a 30 3bv board would be 17.96 seconds, or an 18. This seems correct, although I have never actually done it. My best time estimate for a 30 3bv board, based on my .144 index would be 14.35 seconds or a 15. This also seems reasonable, or at least I would like to think so. Remember, that that would be my potential BEST performance, so it is still against a big stack of odds. Using Dion and Oli's index of .280 both would take an estimated 10.35 seconds to complete a 30 3bv board. Now, I have no way of knowing if these are their best performances, so I wouldn't rule out the possibility of a 10 for either one of them. Another way I tried to verify the numbers was to estimate the 3bv for a typical 22 second game. I came up with 45.01 which I can say from experience is about right. One problem I did encounter was when you take it to extremes. Scoring a 10 for me would require a 14 3bv for my best idex and a 9 3bv for my typical index. That's mathematically correct, but jeez, I hope I could do a little better than that. Either way, you can see that these are rough numbers, but the math is simple enough that anyone could do it at home in just a few minutes. Also, don't forget the fact that 3bv in itself is actually a variable because there are other factors of the board that relate to how easy it is to complete. One other thing that I wanted to note. Matt's 10, while debunked for various reasons, was not exactly an unrespectable game, with an index greater than or equal to .300. So what do you guys think? |
![]() |
|
Aug 26th 2005 at 09:09:28 AM |
|
Name: |
Arjádre |
Comments: |
@Detrusor: Oli is moved up to second place on bestever, but is still listed with a 12 on int... |
![]() |
|
Aug 26th 2005 at 05:59:40 AM |
|
Name: |
Roman Gammel |
Comments: |
Example of application (best time for 3BV 120, best 3BV for 50s game, best 3BV/s for 3BV 250): Dion Tiu: (34; 210; 4,49) Oliver Scheer: (36; 186; 4,09) Roman Gammel: (39; 165; 3,71) Elmar Zimmermann: (41; 154; 3,51) Dan Cerveny (only nf): (43; 145; 3,34) Grégoire Duffez: (44; 143; 3,30) |
![]() |
|
Aug 26th 2005 at 04:32:52 AM |
|
Name: |
Roman Gammel |
Comments: |
(Continue) About excluded games: the reason why 25% is current average 3BV/s, that's hard to understand what's your current 3BV/s, I asked Detrusor and Elmar about the level of current 3BV/s how they feel it, comparing this result with mine I decided to use 25%. Fun thing is excluded games do not contradict this model because of gamma distribution of the value - if we are counting most possible borders for 3BV as [time^{ios-3sigma},time^{ios+3sigma}] left part of this segment ([time^{ios-3sigma},time^{ios-2,5sigma}]) is not filled by values of best 75% 3BV/s games, here you can find other games ![]() |
![]() |
|
Aug 26th 2005 at 04:23:18 AM |
|
Name: |
Roman Gammel |
Comments: |
Index of speed for some players: Dion Tiu - 1,3283 Oliver Scheer - 1,2917 Roman Gammel - 1,2693 Elmar Zimmermann - 1,2437 Dan Cerveny (only nf) - 1,2301 Grégoire Duffez - 1,2240 |
![]() |
|
Aug 26th 2005 at 04:09:56 AM |
|
Name: |
Roman Gammel (Rogen) |
Comments: |
@Oli: Congratulations, I'm glad you're still in group of competitors for 10s ![]() ![]() @All: We all know that 3BV/s depends on 3BV and for high 3BV it is easier to get high 3BV/s and harder to get low 3BV/s. It means that best 3BV/s and average 3BV/s are only indirect indexes of speed. I'd like to suggest a value, which doesn't depend on 3BV and could be real index of speed. Maybe this value is far-fetched and needs some calculations and statistics, but it gives possibility to forecast your best 3BV/s and to explain your current games and statistics. The value is average log_{time-1} {3BV}. Explanations: at first I'll describe what games I included in the list of the games. I think most players improve their speed all the time, that's why for example avg 3BV/s is increasing, if we need index of speed we should work under assumption of presence of some number of games with almost constant index of speed, because of this I excluded from alltime list of the games 25% worst 3BV/s games (25% I took approximately, I think this is exact number I need). Now we have a list of the games, for each game it's easy to count log_{time-1} (3BV) and then to count average. Next question is why this value is good. The answer is sqrt(dispersion)/(avg value)<=0,015 for most players (sigma=sqrt(dispersion)). If the distribution of this value is normal this equation means that the value is almost constant - index of speed. But... the distribution is not normal... it looks like gamma distribution with some parameters, i counted parameters for Me, Dion, Oli, Detrusor and they were the same function from dispersion. Moreover the famous rule for normal distribution - P(x<=M+3sigma)=0,997 works here too!!! Thus the value is almost constant (with probability almost 1). Applications: ios=index_of_speed, 3BV=time^{ios}, ios is > 1 for all players in bestever, 3BV/s=time^{ios-1}, so 3BV/s is increasing if the time is increasing and at the same time 3BV is increasing - explanation of the famous rule. With probability 0,99 time^{ios-3sigma}<=3BV<=time^{ios+3sigma} - this means that you can forecast your best 3BV/s for each length of the game and using this best 3BV/s you can get, first calculations showed that best 3BV/s is predicted very accurately. Another application is best possible time - we will assume that 3BV of your best game will be greater than 120, you can easily count the best possible time... I think this post will generate lots of questions, I hope I'll be able to answer. ![]() |
![]() ![]() |
|
![]() |
|
Aug 26th 2005 at 03:28:49 AM |
|
Name: |
Dmitriy |
Comments: |
yeah, need to cut away some seconds.. ![]() |
![]() |
|
Aug 26th 2005 at 03:17:30 AM |
|
Name: |
Robert Benditz |
Comments: |
@Arjádre. Fortunately Arbiter scores don't mean jack for bestever. ![]() |
![]() |
|
Aug 26th 2005 at 02:56:40 AM |
|
Name: |
damien |
Comments: |
congrats, oli! well, at least i got to be #2 for a few months....although i'd still be second if you had been 1/100th slower ![]() |
![]() |
|
Aug 26th 2005 at 02:05:39 AM |
|
Name: |
Christoph |
Comments: |
@AreOut: That's very strange. You can try ms.dotti.at/wien0508all.php or /wien0508report.php But I don't think that that will make any difference. At the moment I cannot imagine what causes this error. But I think your browser must have any problems with stylesheets. What do you use? You could try if the same problem occures if you safe the page on your PC and open it offline. A file style1.css should automatically be saved in the same folder. You can try if you can view the page if you remove it. |
![]() ![]() |
|
![]() |
|
Aug 26th 2005 at 01:38:06 AM |
|
Name: |
AreOut |
Comments: |
I see just "christoph minesweeper page" and 2 smilies and updated 20.8.2005. Nothing to click on. Give me some other direct link to tourney page. |
![]() |
|
![]() |
Viewing Page 48 of 53 (Total Entries: 5262) |