The Great Camber Experiment: Stage 1 "High Speed Ring" (closed/finished/ended)

Discussion in 'GT6 Tuning' started by DolHaus, Nov 25, 2014.

  1. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Hmmm.... not sure what to do here, I don't want for you to have wasted your time but the consistency is outside of the acceptable range so I can't accept the data at this point. :boggled:

    I have no idea why you've got 8 corners :confused: Maybe the data got corrupted or something? You'd have to ask someone who was more familiar with the software, my understanding is basic at best.
     
    TurnLeft likes this.
  2. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    @DolHaus trust me I understand the numbers are so far off base that they don't add up. I'll uninstall and reinstall Motec and see if that helps. I checked old log files and they are reporting 8 turns as well I also can't pull up a track map it's just a flat line so something is not working correctly in my software right now. I'll upload the ldx files to google and share them with you and you can look at them see if it's my replay data or my Motec.
     
    TurnLeft and DolHaus like this.
  3. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Are the lap times accurate? That's my first point of concern, there are some massive swings in there that suggest there may be external elements affecting the results.
     
    TurnLeft likes this.
  4. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    The camber 6 time was the first time I picked up a ds3 in a month camber 8 I couldn't hold the line in turn 1 and turn 5 hence the slower times. They are true and correct times I was holding th best line I could as close to the driving line as I could on each lap no curb bumping in the laps.
    I uninstalled and reinstalled Motec and the same things are showing up so I dunno what's going on.
     
    TurnLeft likes this.
  5. F1Racer68

    F1Racer68

    Messages:
    1,175
    Location:
    Canada
    The Track Map is usually easily fixed. This is a common issue when MoTeC does not already have a track map built in, such as the High Speed Ring map. I'm not at my PC right now so I can't provide the step by step, but you can generate the map correctly using the G force numbers. Keep in mind that this method will generate a map that is based on your driving line, and so the corners may not be 100% accurate depictions.

    Now, having said that, the fact that your data indicates more turns than are really there may be causing this issue as well. Even if the track layout where "corrected" you would probably see some corners that don't actually exist.
     
    TurnLeft, ALB123 and DolHaus like this.
  6. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    I am prepared to accept odd results with camber over 5.0, by that point the grip is all but gone so odd spikes are to be expected due to mechanically reduced vehicle control.
    The differentials before that are what concern me, you drop nearly a second between 0.0 and 1.0 then level out slighly, then another big swing at 4.0. To me this indicates that you were unfamiliar with the track and car when starting the test, this means you were almost certainly taking dramatically different approaches to more than one corner which compromises the validity of the data.
     
  7. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    I'll rerun 0 camber and bring it in line with 1-3 values, with the 0 run I was not as familiar with the track as I am now and on my wheel my times are a full second lower than with the ds3.
     
    TurnLeft likes this.
  8. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    If you want to repeat the test on the wheel then that's cool but be aware that I can't accept fragmented data so you would have to re-run the entire experiment
     
    TurnLeft likes this.
  9. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    That's the plan for the wheel data. I did load some wheel laps into Motec and got the same wacky max G load in turn one. But the rest of the turns are fine. I'll go do a set at Solverstone and check there see if got a wa my corner as well or something is fishy at HSR on my machine.
     
    TurnLeft likes this.
  10. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    The other alternative is to do it the old fashioned way and use the graphs and the map :tup:
     
    TurnLeft likes this.
  11. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    No useable map....
     
  12. ALB123

    ALB123

    Messages:
    5,162
    Location:
    United States
    I agree that its probably the "track map" that is all screwed up. When I first loaded my MoTeC log file(s) into the i2 software my track map showed up as a straight line!! I had to go into the software and recreate the track map by going into Track Editor and then clicking on Generate Track. Even then, it only had the basic shape of the track laid out. I then had to manually edit the track data to make sure the straights lined up with what Dolhaus defined as a straight and the turns lined up with what Dolhaus defined as a turn (corner). I've never had to do this for a course before and it was a pain in the butt, however, it was well worth it. Now, if I open up a log file from High Speed Ring it always has the course laid out properly in Track Map.
     
  13. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Mine had a map?
     
    TurnLeft likes this.
  14. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    So did mine for the first 6 camber data points I loaded than when I loaded it today no map, 3 extra turns and messed up g load data in turn one
     
    TurnLeft likes this.
  15. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    @ALB123 @DolHaus @ImToLegitToQuit @TurnLeft and anyone else that has used motec with HSR. Could I get the lengths for the straights and turns so I can repair my map? I think that will fix the data problem with Gforce's and turns\

    To get the information. Load a replay with HSR, goto samples->Track Report-Right Click-Track editor you will see a list in the top left with the str and turn numbers, click on one, bottomish left you will see start and end distances. Let me know if it's m, ft or yards and I'll adjust mine to fit yours
     
    Last edited: Jan 14, 2015
    DolHaus likes this.
  16. ImToLegitToQuit

    ImToLegitToQuit

    Messages:
    672
    Location:
    United States
    sorry man, i havent even used it
     
    DolHaus likes this.
  17. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Str 0-1
    Start distance: 11701 ft
    End distance: 1719 ft
    Section length: 2987 ft

    Turn 1
    Start distance: 1719 ft
    End distance: 3568 ft
    Section length: 1849 ft

    Str 1-2
    Start distance: 3568 ft
    End distance: 4776 ft
    Section length: 1208 ft

    Turn 2
    Start distance: 4776 ft
    End distance: 5728 ft
    Section length: 952 ft

    Str 2-3
    Start distance: 5728 ft
    End distance: 6620 ft
    Section length: 892 ft

    Turn 3
    Start distance: 6620 ft
    End distance: 7425 ft
    Section length: 805 ft

    Turn 4
    Start distance: 7425 ft
    End distance: 8320 ft
    Section length: 895 ft

    Str 4-5
    Start distance: 8320 ft
    End distance: 10273 ft
    Section length: 1953 ft

    Turn 5
    Start distance: 10273 ft
    End distance: 11701 ft
    Section length: 1429 ft
     
  18. F1Racer68

    F1Racer68

    Messages:
    1,175
    Location:
    Canada
    Darn it @DolHaus..... you beat me. I was just firing up my laptop to get the info..... :grumpy:
     
    Thorin Cain, TurnLeft and DolHaus like this.
  19. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Bwah ha ha :mischievous: :lol:
     
  20. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    Everytime I input the numbers it changes them to what it thinks they should be :banghead: I can't edit my map for HSR I can't buld one from scratch cause it still does it. Generateing the track is still giving me a flat line for the track and 8 turns and str combos. ugg I think I may have to scrub HSR for data colection unless the ld files I loaded and gav eyou asscess to @DolHaus are working on your end.
     
    TurnLeft and DolHaus like this.
  21. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Just opened 70DS3.ld and it was fine, I don't want to check the whole lot so is there any particular one that is causing issue? I'm sending you a screen grab so you can check if anything basic has changed
     
  22. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    nothiing is diffferent from my screen to the one your sent me other than some displayed data on the right like fast lap but all file info is the same. it just wont generate a correct map for me which I think is screwing with the G force data. The lap times are fine I will go back and rerun 0.0 camber for you to level it out. but the 5+ camber times are correct.

    Edit I generated teh track off a file that had the correct turns and what not alrready. got the map working and distances are still not correct :cheers: but my Gload data is still fubared for turn 1 :odd: attempting to edit distances and seeing. ugg this is pita
     
    Last edited: Jan 14, 2015
  23. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Ok, well get the entire test re-run using the wheel and then send me the data and I will do the analysis :tup:
     
    TurnLeft likes this.
  24. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    There is a god! I got the map edited slowly and surly the distances are off by 1 foot at start and finish I can't them to match your numbers precisely but that fixed the G load data I'll update the Google doc for you @DolHaus with the corrected numbers. and get that 00 camber run done. that took way longer than it should have but it's done and now hopfully it will never happen again *crossed fingers*
     
    TurnLeft and DolHaus like this.
  25. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Just for clarification could you describe your testing process, you keep mentioning re-running 0.0 but I'm not sure what you mean?

    The DS3 data that I have already seen was unacceptable on the grounds of basic consistency so I would need that whole set rerun to ensure its validity. Ideally the tests should be performed over one or two sessions, if done over two or more sessions then I would expect the driver to make sure they could match the times set in the previous session before proceeding with the rest of the test.
     
    TurnLeft likes this.
  26. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    I am rerunning 0-6 now. I smoked my previous times on 0 camber and realized I know the track so well now that all previous DS3 tests are no good. technically all the data was gathered in two sessions but separated by a month and 1000 laps on a wheel so yeah. to be safe I'm covering all the inconsistent data sets and replacing them. 0-6 I think are the ones that are out there. I will than also replace the posted data for 7-10 with the corrected information.
    Sorry for all the headaches on this data I wasn't expecting learning to use a wheel making me so much faster on the DS3. 1:05.701 on 0.0 camber, and on lap 1 with a bad exit from T5 on the outlap I'm leading the 0.0 camber with 1.0 at str 3-4 so it's gonna be lower much lower.
     
    TurnLeft and DolHaus like this.
  27. Johnnypenso

    Johnnypenso Premium

    Messages:
    28,425
    Location:
    Canada
    I am curious about how you guys might explain some of the data that seems to conflict, at least to me.

    0/0 and 3/3 have near identical lap times:

    upload_2015-1-14_12-34-2.png


    And yet 3/3 is faster in every corner but one and a higher average cornering speed:

    upload_2015-1-14_12-36-2.png

    Ignoring the stupidity in the camber thread in the general forum, there have been similar results from testing. Lap times end up equal but cornering speeds with camber were always higher. The difference was in terminal speeds, which were always higher without camber. I think you guys should be looking at terminal speeds as well. The main purpose of cornering in racing is to maximize corner exit speed and get on the throttle as early as possible and camber must be having an effect on these variables as well, otherwise cars that corner faster 5/6 times would have significantly faster lap times.

    Basically it looks to me from this data, that any gains from camber through cornering are lost entirely or almost entirely in the case of 1.0/1.0 and 2.0/2.0, through entry/exit speed. Reporting on terminal speeds might help shed some light on that.
     
  28. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    And my controller died lol forgot my son played angry birds for 3 hours yesterday cause it was too cold and wet to go outside. ugg.

    Edit;
    @Johnnypenso thats an interesting thought. I looked at 6,7,8 data and my top end was the same on all camber levels on str 0-1, but on str 2-3 it got progressivly lower with raised camber meaning acceleration was being hindered. I'll need to load up my new 0.0 run and see what it looks like.
     
    Last edited: Jan 14, 2015
    TurnLeft and DolHaus like this.
  29. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    The data in its current form needs a bit of cherry-picking to remove some odd entries because we are using a relatively small data pool, one large spike has the ability to throw the averages out the window. I am waiting for @Otaliema 's data before making the adjustments (remove highest and lowest data points and average the remaining figures), once this is done the patterns should become more clear.

    My current thought is that the addition of camber might be speeding up the initial transfer of load but is reducing the amount of load the tyre can sustain. I need to compare a few peoples data to check this but I expect to see the peak load occurring earlier in the corner.

    There are plans to review the testing procedures once we have some comparable data from round 2, we are only looking for basic cause and effect patterns at the moment so the practical applications will have to wait until a later stage.
    One thing that I was hoping to get data on in these tests was the effect on longitudinal grip but that has proven inconsistent using this methodology, a simple acceleration and deceleration test should suffice though. The effect on longitudinal grip might hold the answer to if/why exit speeds are apparently decreased. :tup:

    I'm glad that your consistency and speed has improved with the wheel, I've seen your recent seasonal times and they're greatly improved :tup:

    I often found your performance on the DS3 to be err... how do I put this... a little frustrating under testing circumstances :banghead: :lol:

    If you are more consistent and faster on the wheel then just give me the wheel data, I don't need multiple sources of data from the same driver, just give me the most consistent data set you have :tup:
     
    TurnLeft likes this.
  30. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    That would be my wheel I'm running so far with in 0.600 seconds 0-3 camber I deleted the Google doc and am just filling lap times right now you can see it there.
     
    TurnLeft and DolHaus like this.