Home | TMF: Stats & Records | TM2: Stats & Records | Forum | Login |
Dedimania forum
Feel free to express yourself on your favorite dedicated related tool.
You are not logged in.
#1743 2022-04-28 12:24:06
#1744 2022-04-28 13:21:28
Re: Report strange & suspect TMF records here
tophie_tm wrote:
Login: truma123
Track: Q8a5f_wrv62UWk1HPB1RLm2FGp
I don't believe that it is possible to go this much faster on this map, and the end sector is almost definitely impossible.
It seems possible (but extremely difficult/lucky) to make the last jump left towards/on top of the wall, and from there into the finish. The login has no other suspect entries as far as I checked, so I'm giving this one the benefit of the doubt.
Offline
#1745 2022-04-28 13:31:48
Re: Report strange & suspect TMF records here
dalbor93 wrote:
Login: iveashittypc
Uid: UoN0ki7KnZKFgv67cbVKMW7ZTl3
Completely impossible finish time.
Also just have 5 Dedimania-record, two of them are world record.
Hmm, a .14s difference on a 16+ second track does not seem so clearly impossible, but another entry was definitely cheated. Housekeeping performed, thanks.
Offline
#1746 2022-04-30 13:13:49
#1747 2022-04-30 14:22:04
Re: Report strange & suspect TMF records here
miszko333 wrote:
login: sign.br
Uid: Ttu0CvYbiTAckqPyKUtQ45SvyI7
Faking sigN cheater, trying to somewhat hide it but you can see his cps being off on techmaps
Yes, typical cheaty patterns on 5 tracks. Banned and all deleted, thanks.
Offline
#1748 2022-05-04 14:58:19
- ???????» ????
- pro_94_
- New member
- Registered: 2012-07-29
- Posts: 8
Re: Report strange & suspect TMF records here
I'm not 100% sure about this but something sounds quite off to me, also sorry in advance for the long post but I want to write down a lot of thoughts about it because I spent a lot of time into this but I can't still explain how this was achieved.
login: aston_one_77
Uid: bkID8AItneTaEZHph5FjOyIE5xm
TMX: https://tmnforever.tm-exchange.com/trackshow/5788102
I've hunted this track for a few hours because I thought there was a cut near a loop, and yes there is but it's very precise and you can still get caught up by the intended route before the end since it's a speed map with no slow downs except the end.
After beating the WR offline by a few hundreaths I checked the checkpoint times to see if anybody else has had this idea. Apparently everyone on the leaderboard has followed the intended route but the times on dedi2 are a little bit off: it has a slow start but it gains a lot on the last two checkpoints.
At first I thought it was made with another "cut" where you can take a booster on the restart route after a checkpoint to gain more speed (you can do that because the booster is placed after the checkpoint). I tried a little bit but I didn't get anywhere close to have more speed with that. Then I had a closer look to those checkpoint times and I was probably wrong: there's no sign of that booster, the entire save time cames in the last two checkpoints.
The turn before the second to last checkpoint is a dirt turn where you'll eventually slow down even if you drive full throttle, I leave an example video here from one of my runs. Before this turn this dedi2 is 1s slower than the WR and at least 0.3s slower than anybody else in the leadearboard. After that there's a time save of over 0.3s which sounds really suspicious considering how popular this map is and how many players have tried it multiple times. But the biggest time save it's right after where the remaining gap is closed to less than a tenth and that part consists of only a loop and a wide range turn.
The only way to gain time there is just to exit from dirt section with a better speed but the amount of speed you'd need to have to be that fast is just ridiculous. There's a "cut" (quotes because it's actually slower) on the dirt turn where you can drive on the inside and cut a portion of that segment, as it's shown in this other video still made by me.
This can explain the 0.3s gain on the second to last checkpoint but the problem is that with this route you have way less speed, in fact the run in the video lost 1.3s afterwards. If someone wants to be really creative, they might think about a bugslide on dirt to mantain a lot of speed but I still have doubts about it:
1. First there are cacti in the middle of the road which force you to take that exact line;
2. The only way to trigger that is to climb a little bit the dirt hill and hope for the car to get in the correct position. However the hill is extremely bumpy and I haven't gotten even close in all of my attempts;
3. Even if you get it, you still have to re-align your car which makes you lose a little bit of speed.
Being unable to find any reasonable explaination I went to check the other records of this player: having only 20 dedi and some of them in highly rated speed maps doesn't really sound as a good beginning, but I didn't find anything weird about the chckpoints, except for one entry:
login: aston_one_77
Uid: n9HldgiX3citQZMjg5CcGPWGw_b
This, like the first map, is a quite known speed map but it's way more straightforward without any known cut. The end is a drop down from the road into the grass and here's a video example from my dedi. My end is not perfect but it's solid, the time between the last checkpoint and the finish is 4.05s, 9th fastest on the dedis. The second fastest time is 3.97s which is reasonable with a better ending and a slightly higher speed into the turn meanwhile the aforementioned player sits on top with only 3.73s in this segment.
I thought a bit on this end but again I couldn't find an explaination for such a time, even a grasslide seems out of question because you don't have a sharp angle to get an advantage from it and given the high speed it's less effective anyway.
My conclusion is that these two records are cheated (the others look fine) using a tool which gives you road grip on other surfaces: this explain why the strange behaviour is detected only in grass/dirt sections (to be fair the first map has other dirt transitions but they are full speed and they do not affect your speed).
Given the low number of records in relation to the competitive level of such maps this is probably a second/alternative account of some speed player (or someone coming from Tm2).
The two records were set only one day apart meanwhile the 20 dedis spawn over a course of 3 years, so it might just be a case of a player trying to use that kind of cheat just out of curiosity.
As I said in the introduction I'm not 100% sure about this and it could just be a series of unfortunate coincidences but my hopes on the latter statement are getting slimmer and slimmer the more I try the maps.
Offline
#1749 2022-05-04 21:07:04
Re: Report strange & suspect TMF records here
pro_94_ wrote:
My conclusion is that these two records are cheated (the others look fine) using a tool which gives you road grip on other surfaces: this explain why the strange behaviour is detected only in grass/dirt sections (to be fair the first map has other dirt transitions but they are full speed and they do not affect your speed).
Given the low number of records in relation to the competitive level of such maps this is probably a second/alternative account of some speed player (or someone coming from Tm2).
The two records were set only one day apart meanwhile the 20 dedis spawn over a course of 3 years, so it might just be a case of a player trying to use that kind of cheat just out of curiosity.
As I said in the introduction I'm not 100% sure about this and it could just be a series of unfortunate coincidences but my hopes on the latter statement are getting slimmer and slimmer the more I try the maps.
Agreed, housekeeping performed, thanks.
Offline
#1750 2022-05-29 16:15:38
- ??? | Nicklander
- tm2:nicklander
- New member
- Registered: 2018-05-18
- Posts: 4
Re: Report strange & suspect TMF records here
login: syos.
uuid: LQDjcEZvTyJSUTi7aAGjxz1_dkl
challenge: Kackiest Kacky #–150
record: 00:16.64
Offline
#1751 2022-05-30 11:59:40
Re: Report strange & suspect TMF records here
tm2:nicklander wrote:
login: syos.
uuid: LQDjcEZvTyJSUTi7aAGjxz1_dkl
challenge: Kackiest Kacky #–150
record: 00:16.64
https://tmnforever.tm-exchange.com/trackshow/8550332 has 16.67 and 14.30 records, so anything in between can hardly be suspect on a CP-devoid luck track.
Offline
#1752 2022-05-31 04:26:38
- ??? | Nicklander
- tm2:nicklander
- New member
- Registered: 2018-05-18
- Posts: 4
Re: Report strange & suspect TMF records here
xymph wrote:
tm2:nicklander wrote:
login: syos.
uuid: LQDjcEZvTyJSUTi7aAGjxz1_dkl
challenge: Kackiest Kacky #–150
record: 00:16.64https://tmnforever.tm-exchange.com/trackshow/8550332 has 16.67 and 14.30 records, so anything in between can hardly be suspect on a CP-devoid luck track.
I can't prove it in this case but he's a known cheater and has already been banned on the official kacky servers. There's no way he beat -150, it's one of the luckiest kacky tracks to date, only a very few people managed to complete it so far.
Last edited by tm2:nicklander (2022-05-31 04:29:28)
Offline
#1753 2022-05-31 08:40:48
- Vrba
- tm2:vrbnjak
- Moderator TM²
- From: Slovenia
- Registered: 2015-01-26
- Posts: 351
- Website
Re: Report strange & suspect TMF records here
Login: swag-_
UID:TPh4CTxmJR2gmQ4wfaqSdFQ6KLk & kzLKqYdW4RdyQR8s6gX93SFjhai
Offline
#1754 2022-05-31 08:43:04
- Vrba
- tm2:vrbnjak
- Moderator TM²
- From: Slovenia
- Registered: 2015-01-26
- Posts: 351
- Website
Re: Report strange & suspect TMF records here
Login: p15t0n1
UID: 7ONiDCiWxV4Z1zqVU_KAqb8e1D9, 9UMotR19_KqRJD9YymNtcTszp17 & yrV8pg7GkM8iXbaukdUUWvmKRJ1
Offline
#1755 2022-05-31 09:32:19
#1756 2022-05-31 09:48:19
#1757 2022-06-20 21:04:01
- ??? » yoy0
- yoy0
- Member
- Registered: 2011-12-25
- Posts: 33
Re: Report strange & suspect TMF records here
siema_eniu_kupe_lat
VxXBS_QPhFcbqMgUPN3a4f44luf
yes short map but with these tight records and basically everyone has the same CP time but this guy is -0.14
only a few dedi records by this guy and no other 1st but this one just looks off imo
Offline
#1758 2022-06-21 09:26:40
Re: Report strange & suspect TMF records here
yoy0 wrote:
siema_eniu_kupe_lat
VxXBS_QPhFcbqMgUPN3a4f44luf
yes short map but with these tight records and basically everyone has the same CP time but this guy is -0.14
only a few dedi records by this guy and no other 1st but this one just looks off imo
Yup, housekeeping performed, thanks.
Offline
#1759 2022-06-28 18:14:35
- po?tato
- heilelbergensis
- New member
- Registered: 2022-06-28
- Posts: 2
Re: Report strange & suspect TMF records here
sklave69
o_iW9NHnLWyRZNpMseFXFpclgpe — Mini fs hunt 02
mega.nz directory containing the validation inputs
The sectors are, by themselves, not obviously suspect, yet I am pretty confident some of this player's records are illegitimate. Please bear with me here.
A couple weeks or so ago, I clicked to "Validate" sklave69's record in-game on Gold Digger (FS) (rBNVjrrcmrwF6sUn1H0gn0iyAxa, which he overwrote with a faster time later). The game reported the race as an invalid, unfinished race. When I re-ran the validation inputs in TMInterface, sklave69's car appeared randomly to veer off in the second-to-last sector. After another player and I asked sklave69 about the invalid race, he left the server without a response.
Could have been a random glitch, I thought, but kept it in mind.
Later, I met him again, and clicking the "Validate" button produced an invalid race report once more, on Mini fs hunt 02 with a Dedimania time of 0:22.04. I asked him about it as before, and he suggested that this may be because he used ds4windows.
(Side note: I can't think of how this would matter. An invalid race is a mismatch between the inputs and the car's recorded positions. TrackMania still has to receive the inputs generated by ds4windows, and then run the physics engine to generate the positions. The source of the inputs shouldn't make a difference. A mismatch, then, comes from the physics getting directly altered in-memory, such as via a speed boost. Perhaps more knowledgeable players such as donadigo could think of other mechanics.)
Then, I requested that he upload his autosave to TMX, informing him of the path where they are saved, and of the fact that he can simply sort by most recent to find the necessary file. He said he would look for it, but left the server and joined half an hour later (!), with this TMX replay uploaded as the sole replay under a fresh account. The filename matched the autosave schema. Thus, we know that he found where the autosave directory was located.
The TMX time did match Dedimania's 0:22.04, too. Alright! But then I compared that replay against the inputs from the validation save. The validation save veered off again, but what was more interesting is that his—half-an-hour late—TMX replay had 0:05.86 at the first checkpoint but the validation run and Dedimania both had 0:05.87 at the first checkpoint. The rest of the checkpoints did not match the Dedimania checkpoints either, nor did the driven lines, as one can see in the TMInterface recording.
(EDIT: If you look at the TMInterface recording of the validation run, it crashes right before the finish, and so we have the opportunity to rule out a ds4windows issue even harder. I edited the inputs extracted from the run so that I could take over in the loop at 19 seconds in, and finish manually. Could he have gathered enough speed to catch up to the TMX run? No, it finishes 0.03 behind no matter what. I suspect some kind of brief speed boost was involved.)
(When I still thought it could have been a legitimate run,) I logged into the server's SQL console to check if any of his online runs reached the first checkpoint at 0:05.86 and the finish at 0:22.04:
Code:
MariaDB [mini]> SELECT score,date,checkpoints FROM rs_times WHERE playerId=(SELECT Id FROM players WHERE Login="sklave69") AND challengeId=(SELECT Id FROM challenges WHERE Uid="o_iW9NHnLWyRZNpMseFXFpclgpe"); +-------+------------+------------------------------------+ | score | date | checkpoints | +-------+------------+------------------------------------+ | 22320 | 1655393647 | 5880,10650,14080,17380,20070,22320 | | 22170 | 1655393695 | 5870,10620,13990,17260,19930,22170 | | 22180 | 1655393720 | 5860,10620,14010,17300,19950,22180 | | 22160 | 1655393745 | 5870,10600,13990,17280,19930,22160 | | 22110 | 1655393825 | 5880,10600,13990,17250,19890,22110 | | 22220 | 1655808993 | 5870,10620,14010,17310,19980,22220 | | 22200 | 1655809043 | 5870,10600,14020,17300,19960,22200 | | 22100 | 1655809068 | 5870,10600,13960,17230,19880,22100 | | 22090 | 1655809107 | 5860,10580,13930,17200,19850,22090 | | 22070 | 1655809290 | 5870,10580,13940,17200,19840,22070 | | 22050 | 1655809315 | 5870,10580,13960,17200,19830,22050 | | 22110 | 1656424246 | 5870,10600,13950,17210,19880,22110 | | 22100 | 1656424271 | 5870,10590,13950,17210,19870,22100 | | 22040 | 1656424296 | 5870,10570,13940,17200,19820,22040 | | 22040 | 1656424361 | 5870,10570,13930,17190,19820,22040 | +-------+------------+------------------------------------+
Nope. Which means he spent the half an hour driving that TMX record offline to match the Dedimania time. As soon as I explained what I found and asked him about it, he left.
The background is also a little suspect: a recently registered player whom nobody had heard of before driving decent lines and times, with an in-game country (Angola) not matching the IP's location (Germany), using an edgy nickname ("negroskipper99") previously.
Thank you.
P.S.: Do you, by chance, happen to have an SQL query to prune all of a player's XAseco records?
Last edited by heilelbergensis (2022-06-28 21:56:07)
Offline
#1760 2022-06-29 09:59:38
Re: Report strange & suspect TMF records here
heilelbergensis wrote:
sklave69
o_iW9NHnLWyRZNpMseFXFpclgpe — Mini fs hunt 02
The sectors are, by themselves, not obviously suspect, yet I am pretty confident some of this player's records are illegitimate. Please bear with me here.
Sorry, this is way too elaborate and convoluted to wrap my brain around in what little time I still have available for this.
I could delete just the record in this one track but kind of fail to see the point for a #18 entry.
heilelbergensis wrote:
P.S.: Do you, by chance, happen to have an SQL query to prune all of a player's XAseco records?
Dunno, been out of that way too long. DELETE FROM records WHERE PlayerId = blah? Same for rs_times table. YMMV.
Offline
#1761 2022-07-04 13:12:13
- Vrba
- tm2:vrbnjak
- Moderator TM²
- From: Slovenia
- Registered: 2015-01-26
- Posts: 351
- Website
Re: Report strange & suspect TMF records here
Login: geedy12345
UID: xB1KfspjvmreJD2ppn7DKT6IKkd
Impossible 2nd sector
Offline
#1762 2022-07-04 13:26:41
#1763 2022-07-13 18:49:17
- Vrba
- tm2:vrbnjak
- Moderator TM²
- From: Slovenia
- Registered: 2015-01-26
- Posts: 351
- Website
Re: Report strange & suspect TMF records here
Login: Cooper13__
UID: uuAtL0HjFCwV5A0cdnACROr5W0g
impasibl taim!
Offline
#1764 2022-07-14 08:47:23
#1765 2022-07-14 21:50:27
- Vrba
- tm2:vrbnjak
- Moderator TM²
- From: Slovenia
- Registered: 2015-01-26
- Posts: 351
- Website
Re: Report strange & suspect TMF records here
Login: Pajos02
UID: L2deNfpe81ZmPVFyEmX8uCjtqlk
Offline
#1766 2022-07-15 10:07:20
Re: Report strange & suspect TMF records here
tm2:vrbnjak wrote:
Login: pajos02
UID: L2deNfpe81ZmPVFyEmX8uCjtqlk
This May 2011 entry fell through the cracks but made the login a repeat offender after the March 2011 warning. Banned and all deleted, thanks.
Offline
#1767 2022-07-18 13:24:46
- Vrba
- tm2:vrbnjak
- Moderator TM²
- From: Slovenia
- Registered: 2015-01-26
- Posts: 351
- Website
Re: Report strange & suspect TMF records here
Login: bobanboom1
UID: z0gq0kBibh03t8qQEDwg2Ihd_G7
Offline