Blackballed!

lordpantsington

Long story short: I'm being steamrolled out of a league because the admin has got it in his head that I'm doing something. I was accused of cheating, which is false. That accusation was later recanted, but the witch hunt is still on going. They cannot replicate what they are seeing from my car. I have never seen my car do what they say it is doing. I have seen it on occasion from other cars. My setup has been requested and I have denied (never going to happen mid-season when the admin is also a competitor). I am now being threatened to fix something I don't see, or be banned from the league. We are running the Grand-Slam mod from IDTD3TP.

Evidence:

Doubled and synced as best as possible. You may want to turn the volume down.
http://youtubedoubler.com/?video1=UbvWwUWWOEc&start1=0&video2=cYSt6fZQzEY&start2=2.0&authorName=pants

If you have problems for some reason with that link, here are the source videos.

I am on a 3Mb/512kb connection. My pings are always good, my connection is rock steady.

I thought I remember reading a post (possibly on RSC) where someone else was accused of cheating with these symptoms. A reply given for the odd behavior was something like the way rF handles simplifying physics for transmission to and from the server was at odds with the prediction routines and causing it to happen. I've attempted to convey this information, which was disregarded. I found a link to a CTDP blogger who basically said it was a fault in rF, because they had hard data for the parameters they were using in their F1 mod.

So what is going on?

I would love nothing more then to point the obstinate admin to this thread for some information.
Thanks,
pants
 
I think the visual bug is the result of an exasperated (extreme) setup using a mod having extreme values in the physics parameters.

My guess becouse in some mods, when online, you see cars having a normal behaviour (ie not extreme setup) and cars having that issue.

Btw, we race together in TPTCC and i cant believe you are a cheater :)
 
That happens alot with the mod our league runs, though I'm not sure if I've even seen it tilt from side to side. From what I've seen it usually tilts to one side and stays there, although I may be mistaken. Ill check it out again soon. I'm pretty sure it's caused by setups with really low ride heights and\or really soft springs.
 
Our current F1 mod does something similar, but instead runs either "on the nose" or like a prancing horse. :) In our case it's due to the 3D models not matching the ride height of the physics I think. Increasing CorrectedInnerSuspHeight cured it but messed up the actual ride height instead, so we had to leave it as it was. CTDP06 also had tilting issues but they weren't able to fix it. In their case the car was tilting sideways like your car does. Here's a development blog post on the issue: http://devblog.ctdp.net/?p=681
 
Hehe, seen that in our league about 6 months ago with HistorX, GT40s flying above the track.
WAY above the track.
I forget what was causing it but, it certainly wasn't cheating.
 
With the accusation recanted, I am not so concerned about that as much as an explanation of why it happens. If a setup can trigger it, the admin is sure it is the setup that needs to be fixed/limited. I don't have data about these cars and so I assume the range of setup parameters in the garage is as accurate as it could be made. Likewise I have no choice but to assume physics values are correct. I am hoping this information will avoid a hastily edited version for this league's use. It seems like the intent is to limit one or more of the setup parameters to eliminate the behavior. Some good replies so far, if you can tell a bit of who you are/what you do/it will help give authority to the replies.
 
your not a very good cheater, only running in 4th position.
your showing us two clips, the first clip displaying the oddity, second clip looks to be behaving correct?
same pilot, same setup... different results from one session to another.
if it was the setup i think it would be consistant. (test with default setup)
this looks more connection related to me.

i recall testing an old mod that had such excessive grip, you could reduce the anti-rollbars so far the car would flip over onto it's top exiting the pitstall, but that was constant and could be reproduced.
 
The two laps are the same. The normal, flat view is what he sees, the tilted version is what the rest of us on the server see.
 
I have seen this before simply due to connection problems. Our server is in USA and the guy I am referring to was in Australia. His car would go around a left turn and then a right turn all while driving on the left two wheels. Sometimes his car would break into a turn and then appear to drive on the front two wheels. I think it is just the netcode doing a lot of predictions based on the last data received. You should download PingPlotter and set it up to ping the server every second and let it run for 24 hours. Check the results to see if you have any packet loss. You can have great pings yet still drop packets.
 
JGraf said:
your not a very good cheater, only running in 4th position.
Click to expand...

LOL, and I haven't won a race yet!

Noel Hibbard said:
I have seen this before simply due to connection problems. Our server is in USA and the guy I am referring to was in Australia. His car would go around a left turn and then a right turn all while driving on the left two wheels. Sometimes his car would break into a turn and then appear to drive on the front two wheels. I think it is just the netcode doing a lot of predictions based on the last data received. You should download PingPlotter and set it up to ping the server every second and let it run for 24 hours. Check the results to see if you have any packet loss. You can have great pings yet still drop packets.
Click to expand...

I doubt it is connection. It happens on every turn reliably. Others can replicate the behavior now. Of course they still need someone to tell them when it is happening.
 
If it is connection, It is usually the upload that's causes things like this. If you have a low upload speed or if there is something restricting your upload and are in a full race, you will be limited to what you are uploading to everyone else.

Hypothetically speaking, lets say your upload is a value of 10, and that will allow you to race with 10 other people. If there are 11 other people in the room, your net will now be struggling to keep up, some packets will be ignored others will be banked up and a bottle neck will start to happen, which means packets are being processed later than they should be.

As you go around a corner your car rolls, the other players detect the car rolling and their pc's calculate the movement, there is a slight margin of prediction and this is where this sort of thing happens. Their pc's don't see your car stop rolling and predict your car rolling further than your car actually rolls. In some other games I have seen cars float under the track, go through walls, spin around and do backflips. This is all related to net and prediction.

Do a bit of testing, if you have vista/win7 you can use resource monitor if not you can get a little app called DU Meter, then monitor what you are uploading during a race (probably better in a practice :D ). Try and measure how much you upload in a server with just one other player.

How far away from the telephone exchange are you? Are you on DSL or Cable? Cable will generally not give you this problem but DSL + a long distance from the exchange will give you problems like this. Proxy's, torrents, uploading telemetry, other people on your home net will cause it too. Sometimes even firewall software can also cause this.

pingtest.net will give you a line quality test.
 
*sigh*
If it can be reproduced by everyone, independent of up/down speed, geographical location, and service type/provider, I very much doubt it has anything to do with connection. This is why I believe it to be something in rfactor itself.
 
Just to clarify, you do have the same HAT databases and have disallowed mismatches on the server, correct?
 
.....My setup has been requested and I have denied (never going to happen mid-season when the admin is also a competitor)....
Click to expand...

I dont understand this. I appreciate drivers not sharing setups in a championship, if thats the way of this community fine. But if you are accused of cheating and you know its not the case, why refuse to hand over possible evidence?
Especialy if you feel its RF related, then the admins could possible reproduce the issue with your setup? (although i have never come across such an issue myself, it a wierd case!)

Right now you are facing a ban so its not like you have many other options if you want to stay in this League.

Also what other 'evidence' are they handing out?
Are you unrealisticly fast in laptimes? Do you have wierd lines or braking/throttle points? Are there wierd entry's in the xml result files?
Surely a graphical anomaly is not enough to accuse someone of cheating if this cannot be backed by some performance data?

Luckaly we dont have this allot in simracing (as far as i am aware) but its pretty easy to accuse someone of cheating wheter its based uppon facts or not and its extremely hard to defend yourself and possible ruins your reputation as a driver.
Since we only see your side of the story i'm going to be carefull but i sure hope the admins are responsible about this and handle it in a carefull way. it could very well hurt their league as well.

My advise, hand over the setup.
Or at least let someone outside the league try and reproduce the issue (best on their server) and present that along with your setup.

I'm an admin and race director in my League and i gotta be honest with you, if one of our drivers was 'under investigation' and refused to hand over setups, replays or in general not wanting to cooperate that would only make his case weaker and weaker. By refusing this he becomes even more suspicious.

And on that note, there comes a point for an admin team where the benefit of the community has to become priority. evidence or not, a driver refusing to cooperate whilst he is beeing accused will most likely suffer the consequences.
 
I have seen this happen in 2 mods before, but everyone else saw the same problem with the cars dipping into the tarmac on corners.

To us it was a bug with the mods
 
We @ RACER have had this happen multiple times in our Simdy 500 among other events. I think a lot of it is the soft springs and perhaps the rest is just an oddity in rFactor. It changed from year to year on who had it. In 2007 it was everyone and as late as this year it was limited to just a few people.

But I know you LP, you aren't a cheater.
 
Bee said:
Just to clarify, you do have the same HAT databases and have disallowed mismatches on the server, correct?
Click to expand...

Server is set to kick mismatches at the beginning of race session. I was using the version of Laguna we have always used, 90% sure the HAT matches, how would I confirm that with 100% certainty?
 
The direction this thread is focusing on is not what I intended. I am not looking for community support on the content of my character. It is appreciated, but unnecessary.

What I am looking for is data. The examples thus far of what someone did to get a leaning car have been okay. Slightly hearsay, but not everyone in the community is here/paying attention to this.

Have you ever seen your own vehicle lean?

I cannot get my Grand-Slam Prototype to lean. Come to think of it I can't remember a time where my vehicle in any mod leaned.
 
I've never seen my vehicle lean, but when I'm on a server and other people are using my setups in a certain mod their car leans on my screen.
 
lordpantsington said:
Server is set to kick mismatches at the beginning of race session. I was using the version of Laguna we have always used, 90% sure the HAT matches, how would I confirm that with 100% certainty?
Click to expand...
Have everyone delete them and a new hat file should be created (you could just delete your own and on the server, but it would be better if everyone could start fresh).

I have seen this once before (very similar - some exaggerated roll), in fact it was my own car. The guy running the server sent me a replay to report it as a possible "bug" (my replay looked normal), but I could never recreate it. It could have been due to HAT or mismatch, but it could also be due to me changing my setup after that "strange" race.

I don't even remember what mod it was now because it was shortly after rF was released. One thing I am fairly certain about is that it was not the default content of rFactor. This is quite the exaggerated roll and rFactor doesn't really predict roll like that so I don't really know what's going on here. This is why I wanted to make sure everyone is using the exact same track/HAT version.
 
Bee said:
Have everyone delete them and a new hat file should be created (you could just delete your own and on the server, but it would be better if everyone could start fresh).

I have seen this once before (very similar - some exaggerated roll), in fact it was my own car. The guy running the server sent me a replay to report it as a possible "bug" (my replay looked normal), but I could never recreate it. It could have been due to HAT or mismatch, but it could also be due to me changing my setup after that "strange" race.

I don't even remember what mod it was now because it was shortly after rF was released. One thing I am fairly certain about is that it was not the default content of rFactor. This is quite the exaggerated roll and rFactor doesn't really predict roll like that so I don't really know what's going on here. This is why I wanted to make sure everyone is using the exact same track/HAT version.
Click to expand...

Okay,
I renamed my old HAT, started rf to generate a new. Did a comparison in a hex editor and they match. Admin pulled the HAT from the server, did a comparison and there are parts that match, but large sections of data altered. Not sure how long that file has been sitting on the server. Someone else in the league sent me their file, it matched what I had. I'm going to see if I can get the admin to generate a fresh HAT from the server (if this was old) and see how that compares.
 
Last edited:
Not much help but we've seen every car do a momentary dip every time, up the hill from the Old Hairpin at Donington running the Caterham mod. Nothing happens to your car but everyone around you dips.
 
Yes, the cars tilting at Doningon TL is a track issue and it happens with all the mods/cars, but I guess the issue reported by the OP is a different problem.
 
As far as i recall this has to do with some cars that have low ride height in combination with suspension settings, bump stop values, high values for fast ans slow rebounds and possibly soft springs as well so that conditions are met to trigger this issue online, sorry can't be more specific as I don't know what these exact condtions are.

Though I have never seen the tilt above the ground rather, but sinking into the ground and tilting left and right at an angle, typically one side above the ground and the other sinking into the ground, though pretty sure your issue is the same thing.

We, in CTDP at the time with the F1 2006 mod posted something after some discussion with ISI I believe, this is how I recall it anyway, and one of the devs possibly said it was a bug in rF but ONLY under specific circumstances of some physics values does it seem to happen. though offline is always fine. sometimes it would only be one car and once it starts it would not stop online for that car.

Anyway it is not at all uncommon and when these test were done in CTDP I can assure you no cheaters if that will help your cause at all :)

Another rumour I once heard is that a low poly track surface may contribute, though have always found that hard to rationalise as to why that should be. The last part would be an interesting test by making a long straight piece of track, say several kilometers long just made from only one poly ( two triangles), may be online the car will disappear underground altogether :)

EDIT: regarding the HAT database comments, AFAIK any discprepancy between any of the clients and server will be reported by rF as a track geometry mismatch anyway, so they must match at least if you set it up right to take the appropriate action, i.e. get the boot :) now whether there is a bug in there somewhere that confuses these values somehow is another matter after rF has completed the check . but as it is at least rF thinks all is OK there.
 
Last edited:
Same here, we only experience this with others and they are allways due to lag or packetloss.

Does this occur on every track you guys race on or was it just one track?

I'm not a tech guy so its just a shot in the dark, but at one point we had to alter the A1 Ring to be able to drive there with an F3 mod. turned out the track had invisible obstacles in the tarmack which the low F3's could not handle so we had to remove them from the track.

If there is a difference in 3D surface and graphical surface you might end up with wierd stuff depending on the setup i supose.

Oh and sorry for going too deep into the parafinalia of it all :)
 
I wouldn't confuse what you're seeing here with prediction situations where you see a car that appears above/below the track surface (such as going over swift elevation changes) in a server replay (that doesn't appear in the client replay).
 
I realize that with 3rd party content there is only so much I'm going to get. :(

What else is causing the discrepancy between client/server?

This is the 5th track in which I've leaned. Some of the tracks in this season we've never run before. Those should have fresh HAT files.

If it truly is setup, then why don't I see it from myself?
 
Not sure I understand, but anyway as Bee said per last comment,what you see in client may not tally with server replay. Can't your admins just accept it is just an rF feature :) let them read this thread.

If they are that hung up on it may consider using some client software like some of the more professional leagues do with anticheat detection on the files to make sure files are the same, like FSR do.
 
You must log in or register to reply here.

Back
Top