Authoritative Minesweeper

On 2007-11-21 Gongor sent Recorder videos of 5.37 - 30.35 - 97,68 made on XP.
On 2007-12-26 he sent Recorder videos of 1.96 - 18.90 - 63.96 made on XP.
On 2008-11-26 he sent Clone2007 videos of 1.921(7)- 10.584(49) - 37.386(142).

All the XP videos look okay to me, except I did not allow the 1.96 video because it is corrupt and does not work.

Do the Clone videos look good to you guys? They are played in normal (not UPK) mode and the videos are work.
The surprise is the high 3BV on each level and how fast Gongor improved.

The videos are attached.

my 1st impression after watching the clone videos is that it might have been speeded up. in the exp game he suddenly leaves a pattern unsolved (the 5 above the 3) and starts solving from another point. i don't think that any sweeper who has sub50 would not recognise that this pattern is still solvable.
on int he does only 2 initial clicks and then stops very suddenly. compared to the speed of the mouse movements during the 1st ~0.5-1s the rest of the board seems to be solved at an oddly high speed.
the beg game is suspect to me too because of the too high 3BV

i would like to see him getting a decent time on some of the other clones.


the fast improvement isn't that strange - eg it took me also only a very short time to get my current scores (only ~1 year from sub90 to sub50)

I think the 1s is most evident that it are a speed-up videos. If you slow it down (i think 4-5x is correct speed) it does start to look like a normal newby again.

If you look at his first clicks for instance and compare it to all the fastest players in the world, you will always that if they quickly start a board with 4 clicks (spread over the whole board), you will always see the mouse press, some dragging over a few squares and then a release. This Gongor dude start a 19 clicks/s and has almost no drag at all, meaning he breaks, presses, released AND changes direction in a staggeting: 0,04s on a 1 square area. Try clicking 4 squares like him yourself in a similar pattern and you'll instantly see the difference (besides the fact that you don't even get the 19 clicks/s)

If you want a good laugh, try playing against him on minesweeperlive (and don't forget to record it, so we can laugh too)

(Lol where's the edit button....I'm helpless without )

he plays like almost like me, makes the same stupid wasted clicks... it looks like it's jsut sped up on the exp

i upk'd his exp board and got 54 in 1st and 49 in 2nd try, being not at all warmed up - this board is a fast one. also my ioe was with 0.6 a bit above my average. but still i'm convinced the original video was sped up

same goes for the int board: 17 in 1st and 15 in 2nd try upk. the same story as the exp board

the beginner board is impossible to upk in less than 2.2s, this is evidence enough for me that this guy is a cheater.


i'm sure that these scores can not at all be accepted, but what is going to happen with his other scores? is there a place for cheaters on the worldranking??

Check out the cl/s in the beginner game... versus the incredibly low cl/s at the beginning of the expert game. Maybe this is just careful playing.
Also a huge number of right clicks in the int and exp vids compared to the actual number of flags... and tons of wasted flags if you look at the "events".

I agree about minesweeperlive/other clones. It might also be worthwhile to request his history.

And he just *happens* to have tied every single one of Dion's integer scores.
Looks very suspicious... possibly real but much much more likely sped up.

My biggest complaints to say it is sped up are.
1 the beg game 10 NF clicks/s I don't think it is humanly possible to to 10 left clicks/s.
2 the int game, on two occasions he uses a flag for a pattern that could have been solved faster NF.
3 the exp game basically too many Hesitations, if you watch Dion's vids, or that of any time sub 45 you will notice the player almost never hesitates, and again he does over 5.5 clicks/s.

Also all 3 videos would be RQP world records, or close to it, and I can't imagine that happening.

Luke

Clicks/s Limitation: Time between every two clicks must not be less than 100ms.(For example)

I think that obviously cheated records shouldnt be mentioned at all

If I look at the vids, he probably made his 1s vid when he was still a beginner and made the intermediate and expert vid a lot later when he was more experienced. I have an idea how he might have made his vids (although I haven't verified if it's easily possible), but I won't post it here to avoid bringing guests on bad ideas. Even though this case is pretty obvious, in future cases of WR claims from new players that never participated in any tournaments, I think a live demonstration on minesweeperlive (or otherwise live webcam) is the best way to verify a players capability.

I asked Gongor for more videos. On the 6th he sent 2, 3, 3, 11, 13, 14, 14, 14, 38, 41, 41, 49, 52. The 11 on Int is made at 5.911 3bv/s.

On the 7th he sent a 33.836 on Expert made at 4.143 3bv/s.

All videos attached.

the 33 says everything....tell this cheater that we want to see him getting such a score on arbiter, msx and vsweep! also i'd like to play a few matches against this guy on mslive. can you arrange that, damien?
i mean, when we see his mslive results we'll know it anyways...

I wanna see him on MSlive
Also What I think is the bigger issue that needs to be raised is just how hack proof clone is.

as I already told, stop talking about him, it gives him pleasure that his scores are mentioned

LTsully wrote:I wanna see him on MSlive
Also What I think is the bigger issue that needs to be raised is just how hack proof clone is.

i can't help it but this aspect kinda worries me - this time we have a noob cheating by increasing the replay's speed, but what can be done against a cheater who is able to get times around 15 and 50? i guess kamil's 40s wouldn't look too odd if played at 1,1x speed and then it would be only 36s long...i'd really like to hear rodrigo's opinion on this cheat videos: is there another way to proof that he cheated other than watching the videos?

KamilSaper wrote:as I already told, stop talking about him, it gives him pleasure that his scores are mentioned

and it would give us pleasure to kick his ass on mslive

And this is the reason why having many clones is a good thing. What if there IS a bug in the clone that makes it possible to speed up vids? Would suck to have only the clone.
Monoculture is always vulnerable, because any vulnerabilities are global.
As it is, we can request videos from other clones as evidence of a players skill. If the issue gets serious, we can suspend the clone for new scores until the bug is fixed, and use the other clones in the meantime.
Cracking the clone is one thing. Cracking the clone, vienna, arbiter and msx is another. Cracking all of these clones so that you can speed up vids on ALL of them is an even more different story, and you need to do that for consistency (a cheated upk score looks different that a cheated sped-up video).

EWQMinesweeper wrote:
i can't help it but this aspect kinda worries me - this time we have a noob cheating by increasing the replay's speed, but what can be done against a cheater who is able to get times around 15 and 50? i guess kamil's 40s wouldn't look too odd if played at 1,1x speed and then it would be only 36s long...

It's good that you bring it up...

I asked Gongor to play on MSX...two days later (today) he sent a 39 on Expert. Attached.
He just asked to join the forum, so he will be here soon I think. He is learning English, so be nice!

@Tommy: many years ago I was going to make a good Camtasia video and carefully delete frames and edit the timer. I was going to email Dan and Lasse and pretend I made the world record....but I was too lazy It is easy to tell when a bad player fakes videos, but a good player...

Twp other things I've noticed with his videos,
1 He wastes almost no clicks, this is more noticable in the newer ones
2 His mouse distance is amazingly efficient, which leads me to suspect maybe a good solver program, or maybe sped up UPK.

@Damien: ask him to make vids on Arbiter, Viennasweep and luckysweeper Im just wondering which of this clone he wont hack

Janis wrote:he plays like almost like me, makes the same stupid wasted clicks... it looks like it's jsut sped up on the exp

kinda take back what i said there.. i just compared myself with him.. and yeah he plays a lot better.. if this is sped up he's probablr normally around 60-70 in exp

i already sent him a pm, telling him that i want to play against him on mslive - i gave him 2 days to answer

EWQMinesweeper wrote:i already sent him a pm, telling him that i want to play against him on mslive - i gave him 2 days to answer

From what Damien said about MSX, 2 days may be too long

don't mix things up. so far all we've seen were altered video replays. a match on mslive is totally different. even if all he does is running so ram-intensive programms that the clones' counters are slowed down (this occurred to one of my friends playing arbiter - the replay seemed pretty normal, but in fact the real game was played in much more time) it wouldn't work on mslive because in the case that i should be able to complete the board before he does, normally his game ends too right at same moment. if it goes on i'll know he cheats.