New patch and GPL hotlapping competitions saved by GPLDump!

I wrote an email to Eric Busch yesterday, and recieved a reply saying that they would not provide a tool to verify hotlaps, so we can tell which holaps where made with either version 1.0.0.0 of GPL or the upcoming 1.1.0.3 Eric responded saying that there was no way possible to verify the laps as no setup data was saved within the replay, and no version number is recorded. I was fearing the worst possible ramifications in the hotlapping competitions....would we see a return of the cheaters...just like cheaters had ruined most GP2 hotlap comps with slo-mo driving....would the same happen with the gpl hotlaps comps because of ppl running 2 versions of gpl. One for low-rider hotlapping, one for "realisitc" on-line racing... But GPLDump is to the rescue!!! With GPLDump u can compare 2 laps...and more importantly, the dynamic ride height from 2 different replays. And there is a significant difference in the graphs :))) So cheaters...beware...we can tell who's been *** and nice ;) To papyrus, I am very disappointed in your inability to provide hotlap verifiers. Hotlapping is an important part of sim racing. (Even if it's a little arcade). You gotta wonder when these developers will start listening properly to their prospective audiences... Maybe I should apply for the tester job? I seem to pick up more bugs than them... Ian Lake /\ Inventor of the low-rider, and happy to see the new ride height /\

Ian... An excellent point. I'll see if I can add that to the future features list of GPaL, as well. Jack Rambo www.RaceLive.com/GPaL

Well to be fair that's not quite what I said. You asked: "A tool is required for hotlappers and hotlap board administrators, that reports the version of gpl, or the setups static ride heights from the replay, otherwise hotlap boards will become a farce. Is there any plans for this kind of tool?" As a direct response: "I'm not sure that one could be made that would be of any real benefit to anyone. You could still load an old replay from version 1.0.0.0 and clip out a fast lap in version 1.1.0.3. In this case even though the lap was done in an earlier version, the replay would be from 1.1.0.3. I don't think the replay has any setup information saved in it. Like the potential use of hex edited setups, I think you're just going to have to rely on the integrity of the drivers to submit valid 1.1.0.3 hot laps." That's not to say a replay validator couldn't be made, just that one with the criteria you gave (i.e. one that checks the replay for a version number or checks the car's static ride height) didn't seem like it would work on first inspection. Eric

Now look what you have done Ian, another month before we get our hands on GPaL! MikeT. -- 22:51 on 06/09/99 From sunny Herne Bay, Kent.UK. mailable also at remove NO-SPAM to mail me.

Quoted text (21 lines)

>> Eric responded saying that there was no way possible to verify the >> laps as no setup data was saved within the replay, and no version >> number is recorded. > Well to be fair that's not quite what I said. > You asked: "A tool is required for hotlappers and hotlap > board administrators, that reports the version of gpl, or the setups > static ride heights from the replay, otherwise hotlap boards will > become a farce. Is there any plans for this kind of tool?" > As a direct response: "I'm not sure that one could be made > that would be of any real benefit to anyone. You could still load > an old replay from version 1.0.0.0 and clip out a fast lap in version > 1.1.0.3. In this case even though the lap was done in an earlier version, > the replay would be from 1.1.0.3. I don't think the replay has any > setup information saved in it. Like the potential use of hex edited > setups, I think you're just going to have to rely on the integrity > of the drivers to submit valid 1.1.0.3 hot laps." > That's not to say a replay validator couldn't be made, just that > one with the criteria you gave (i.e. one that checks the replay for > a version number or checks the car's static ride height) didn't seem > like it would work on first inspection.

Fair enough, but I don't think I took it out of context. At the time I wrote the email to you, I didn't know GPLDump would be useful. Thus the only thing to tell the difference in hotlaps, that I could think of, where the static ride height of a setup or a version number, if they where saved in the replay. As you indicated they aren't. Maybe I should have asked weather there was anything else in the replay that might indicate static ride height, and i'm sorry I didn't ask that, as you may have suggested dynamic ride height on prompting. At that time, you didn't indicated that there wasn't a way possible to verify hotlaps, and that we "have to rely on the integrity of the drivers to submit valid 1.1.0.3 hot laps". I still believe this is a very poor stance to take on such an issue. For many years now there has always been discussions on the need for hotlap verifyers to stop cheaters, yet it still seemed that the developers failed to listen.... Certainly not your fault Eric. It's a design issue that should be sorted out when developing a sim, that certain data should be retained in a replay that will verify a hotlap. And that a verify will be supplied when the sim is shipped. The only tools available today, are those freely developed, and thus only available months after the sim is released, Hof2Lap for GP2, and now GPLDump for gpl. They are in no means perfect, and even the GP2 verifier only just got support for detecting slo-mo drivers. As you have indicated a replay validator could be viable using the car's dynamic ride height (as I have found using GPLDump last night). Does this mean we will recieve such a tool with the patch or soon after? I'm not sure if it's required because GPL Dump does a competent job letting us see who is trying to cheat. It would still be handy if there was a tool that verified a lap simply and easily, instead of having to compare graphs. The last question I have is to ask weather future papy sims will contain a hotlap verifyer? I'm also sorry if it sounded like I was attacking papy or you in my previous post, it was not intended to be taken that way. I just want the software developers to listen to ppl's requirements. Your doing a great job at papy, and papy is the only developer that listens well to their audience, it's just sad when they overlook some obvious issues. Thanks, Ian Lake

I don't know of any sim that contains such a thing, and indeed I remember when the old IVGA developed such a thing there were howls of protest over people's setups being stolen. A part of the checking process the verifier must obviously check for illegal setup settings, and therefore must be able to read the setup used - that is perhaps not what everyone wants. That's the practicality, then too there is the small issue of reality - as a percentage of total sales how many hotlappers do you think there are? And how many hotlappers are there who care about such a thing? Why detract from other parts of the sim by spending time on a verifier when it addresses <1% of the people who bought the game? I woud agree that my own feelings on hotlapping are perhaps a touch harsh since I see it as an artificial and meaningless benchmark (if it is a benchmark at all) since it tends towards benchmarking who spends most time with the sim rather than who is really the best driver. However, when I did do some hotlapping on CompuServe SPRTSIMS the main thing I enjoyed was the trust - you submitted a lap and it was accepted. If you put rules and verifiers it then becomes a competition of who can find the best loophole in the rules or method of cheating the rules - as happened with many IVGA laps, people trying to cheat the checking system simply because it was there. Cheers! John

Dunno if you'd know or not Eric, but anyway. If a replay was imported from 1.0 to 1.1 and then clipped or not, would the lower ride height indeed be detectable in 1.1. I know setups will auto adjust to 2.5 ride height and wonder if the 1.1 replays might display this 2.5 ride height for a setup that was 1.0 ride height?

Quoted text (32 lines)

> Well to be fair that's not quite what I said. > You asked: > "A tool is required for hotlappers and hotlap board administrators, > that reports the version of gpl, or the setups static ride heights > from the replay, otherwise hotlap boards will become a farce. Is > there any plans for this kind of tool?" > As a direct response: > "I'm not sure that one could be made that would be of any real benefit > to anyone. You could still load an old replay from version 1.0.0.0 > and clip out a fast lap in version 1.1.0.3. In this case even though > the lap was done in an earlier version, the replay would be from > 1.1.0.3. I don't think the replay has any setup information saved in > it. Like the potential use of hex edited setups, I think you're just > going to have to rely on the integrity of the drivers to submit valid > 1.1.0.3 hot laps." > That's not to say a replay validator couldn't be made, just that one > with the criteria you gave (i.e. one that checks the replay for a > version number or checks the car's static ride height) didn't seem > like it would work on first inspection. > Eric > > Eric responded saying that there was no way possible to verify the > > laps as no setup data was saved within the replay, and no version > > number is recorded.

I doubt that, since the replay file stores the values as they were when the 'live' racing was going on. When you take a clip from a replay file, it doesn't calculate all the graphics and values for each frame again, it just modifies the headers to tell this replay starts here and includes that many frames (AFAIK). -Juha --

>I don't know of any sim that contains such a thing, and indeed I >remember when the old IVGA developed such a thing there were howls of >protest over people's setups being stolen. A part of the checking I don't think the idea here was to include the setups in the replay. Just a simple system to verify that the replay was indeed generated with GPL 1.1 and not 1.0. A simple system for 'adequate' protection against submitting low-rider 1.0 replays as 1.1 replays. By 'adequate' I mean that the implementation is good enough so that cheaters won't bother. The replay file doesn't need to be RSA encrypted and signed with a digital signature, just a few bytes telling: 1) the version of the program that saved the replay 2) if the replay was originally generated with this version or not Well, you might argue that cheaters will bother anyway. Maybe, but I doubt it, and it'll keep 99% of them away. The idea of the implementation is simplicity. It shouldn't detract from other parts of the game. It's 20 lines of code. Trivial to implement, easy to test. The verifier itself could even be made by the hotlap community. The information just needs to be in the replay file. When looking at the problem from this perspective it kind of feels strange why there should be so much resistance for a solution like this. ---------------- Ok, with that out of the way we can get into arguing about opinions. The reason why usenet was invented anyway ;) If Sergei Bubka (famous WR pole-vaulter) makes a new WR just because he practiced harder than anyone else (which he probably does), do you consider his effort as a meaningless benchmark for others? Just because he pratices more doesn't mean his result isn't commendable. Other people might practice a lifetime and never achieve the same result. The reason he practices much in the first place is probably because he knows he's an excellent pole-vaulter. Same goes for hotlapping. The people who drive those fast laps are all good race drivers. We challenge ourselves and our 'competitors' for the toughest trophy, the no. 1 position. It requires a will to practice and a will to beat records. And it requires skill. It's not simply 'drive around Silverstone for 1000 hours and you'll be unbeatable'. Also if someone does practice for 1000 hours and does as good as the guy who only bothered to pratice for 1 hour I consider both of them equally good drivers. Put them head to head in a race and they will both battle for gold, and in the end that's all that matters. ----------------- And since I'm in a typing mood I'll include a special bonus, free of charge, just for today, where i'll try to explain for non-hotlapping GPLers why this cheat protection would be nice. Imagine GPL 1.1 online servers would allow GPL 1.0 clients to connect to them, with lowrider setups. Would you want to race around in your ultra-realistic 2.5" ride height when others would be whizzing past in their 1.0 low-riders and claim victory and that they were driving 'legally'. Then we might just as well all race online with the 1.0 version and forget 1.1. Or do we have the people just trust each other not to use 1.0? Put 'Only 1.1 GPL players here' in out VROC message and cross our fingers that the hothead cheaters will keep out? Regards, Casper

Luckily the sport of F1 sim racing is a gentleman sport where noone hides their setup ;o), just look at Schubi's hotlap site, almost everybody posts their setup. That's gentleman like to me, everybody saying - I drove this car and with that setup, now go and beat my time :o). IMO, if you need a special setup to be the fastest, you're not really the fastest :o). /Christer > I don't know of any sim that contains such a thing, and indeed I > remember when the old IVGA developed such a thing there were howls of > protest over people's setups being stolen. A part of the checking > process the verifier must obviously check for illegal setup settings, > and therefore must be able to read the setup used - that is perhaps > not what everyone wants.

Im not sure it goes this way... The client/server side of the GPL replay system is much more complex than that. Per example, I remember seeing different wheel and suspension effects between the same replay I had, but with a different *.TRK. In those two replays (one had some weird stuff on the backstretch of a track), you saw the GPL car act different on the same replay in terms of suspension. So from what I speculate... (the GPLDump helps here also) 1) The replay contains essential data. I guess two ways to be done (data position each 1/36th or transfer position for each 1/36th of a second). Not sure but the second is the correct one. 2) The client side of the replay system calculated the suspension movement, engine torque, chassis torque, etc.. from the data it has from the server and also the track data. Some little change like the banking of a track.dat (not a big change, just adding banking on a oval track or a bump on a straight) will make the replay still compatible. Of course I might be totally wrong here. But I guess and hope it uses the way I did, since it means less Hard Disk and a much better compression of the replays, at the expense of more Processor power (client has to do much more tasks). It's the same analogy as a .WAV file compared to a .MP3. -- -- Fran?ois Mnard <ymenard/Nas-Frank> -- NROS Nascar sanctioned Guide http://www.nros.com/ -- SimRacing Online http://www.simracing.com/ -- Official mentally retarded guy of r.a.s. -- May the Downforce be with you... "People think it must be fun to be a super genius, but they don't realise how hard it is to put up with all the idiots in the world."

Quoted text (11 lines)

>> I doubt that, since the replay file stores the values as they were >> when the 'live' racing was going on. When you take a clip from a >> replay file, it doesn't calculate all the graphics and values for >> each frame again, it just modifies the headers to tell this replay >> starts here and includes that many frames (AFAIK). >Im not sure it goes this way... >The client/server side of the GPL replay system is much more complex than >that. Per example, I remember seeing different wheel and suspension effects >between the same replay I had, but with a different *.TRK.

Well as I said, no recalculation of any of the data already stored in the replay file is necessary when you take a clip from it and save the clip. The original poster's worry was, that for example the ride height would alter if someone saved a GPL 1.0 replay with GPL 1.1. It most likely is like you said, that the replay file contains the 'essential' data only and then the fancy stuff like suspension movement is calculated based on that data. Ride height is part of that essential data and the car's movement would change in the replay if you added a small bump into the middle of a straight in the .trk file, since it says in the replay file, that at this position of the track there should be X amount of ride height. -Juha --