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

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

  1. Thorin Cain

    Thorin Cain Premium

    Messages:
    2,262
    Location:
    Scotland
    I agree. I think this is a much better way to collect and perform a good analysis on the lap times. Using the eclectic times should paint a better picture of where the camber is helping or hindering the car's performance. It does help reduce the impact on the results which are due to the inconsistencies of us humans.

    Oh, @DolHaus I have a confession to make. I was only using the standard version of this software and was actually analysing the graphs rather that relying on just a data sheet (I think you get a better picture of what is happening if you can actually see the picture;).) I also suffer from man disease and tend not to use instruction manuals too much, so I don't even know if this was in the standard version. But I have the PRO version now so it's all good.

    And speaking of which....

    I booted up Motec when I got back in from work and checked the Track layout and the time data sheet. Here are the corresponding sectors. I have also noted the camber settings which were fastest through each sector, the time differences between the 1st - 2nd fastest setting and 1st'last

    Track Sectors----Camber @ ---- Diff. from 2nd ---- fastest-slowest

    1: Start - Turn 1 ---- 2.0/2.0 ---- 0.076s ---- 0.143s

    2: Turn 1 ---- 2.0/2.0 ---- 0.006s ---- 0.009s

    3: Straight 1-2 ---- 1.0/1.0 ---- 0.001s ---- 0.003s

    4: Turn 2 ---- 2/0/2.0 ---- 0.050s ---- 0.165s

    5: Straight 2-3 ---- 1.0/1.0 - 1.5 1.5 ---- 0.004s ---- 0.017s

    6: Turn 3 ---- 1.5/1.5 ---- 0.026s ---- 0.068s

    7: Turn 4 ---- 1.5/1.5 ---- 0.015s ---- 0.042s

    8: Straight 4-5 ---- 1.5/1.5 ---- 0.041s ---- 0.076s

    9: Turn 5 ---- 2.0/2.0 ---- 0.027s ---- 0.209s

    10: Straight 5 - Finish ---- 2.0/2.0 ---- 0.029 s ---- 0.096s

    Banked corners (and sectors effected by banking):
    Start-turn 1, Turn 1, Turn 2, Straight 2-3, Straight 4-5, Turn 5 and Straight 5-Finish.

    Flat corners (and sectors un-effected by banking): Turn 3 and Turn 4

    So what do we learn from that data?

    From the data of @LeoStrop's Eclectic Lap-time, we can clearly see that camber set at 2.0/2.0 was the big winner in all of the banked turns 1, 2 & 5 as well as the main straight and approach to turn 1 (as these are heavily influenced by the exit of the banked Turn 5 and the banked approach to Turn 1.

    But as expected on the two flat corners Turns 3 & 4, a lower camber setting was quicker. With both of these achieved with 1.5/1.5. You can also see that this setting was faster in sector 8 (Straight 4-5) This is due to this not really being a straight. After the tunnel there is a right hand bend towards Turn 5 (Yes, I know you all know this, but you never know who reads these.), this bend has a camber change at the apex as the banking starts and you do record a reasonable amount of Lateral G's here too. Also was quicker in Straight 2-3 (Tied with 1.0/1.0).

    For the remaining two sectors Straight 1-2 and Straight 2-3 the 1.0/1.0 setting was recording the best sector times although not by much ( Tied with 1.5/1.5 in Straight 2-3) Again these sectors are also effected a little by the tracks banking.

    The really obvious thing from the table is that at no point was 0.0/0.0 the fastest camber setting. Although being fair it was real close to getting it's name on the board, only missing out by a mere thousandth in sector 3 (Straight 1-2).

    Sorry guys. I was just going to post my answer to Hami's suggestions and get the track sector info but I couldn't help doing a little analysis of my own. Hope you don't mind:guilty:. And sorry too for not using any crayons (That one's mostly for @Lionheart2113, I know how much you like them:(), I am having issues with my laptop's office programme:banghead: and haven't got round to fixing it yet.

    Hope I didn't miss anything! (I didn't do any difference % calculations as I thought that would just be too much. (And I wanted to leave some of the maths just in case someone got upset...Yes, I mean you @TurnLeft:lol:)

    Have a good day everyone, I've really got to go to sleep now, that's hour 23 just ticked by.:cheers:

    Edit: I need to get that office programme fixed...Why does this thing not allow multiple spaces:banghead: :mad: :crazy:
     
    Last edited: Dec 28, 2014
  2. LeoStrop

    LeoStrop

    Messages:
    890
    Location:
    Brazil
    @Motor City Hami, you are correct with every issue you pointed out, lap times don't tell half the story. Max Gs can really have spikes on it: my max G was 2.38, with camber at 1.0, but the second max with the same camber, was 2.21, at the exact same place of the track(trail-braking for turn 2). So i guess the average tell us a better story.

    The last picture i posted, showing the comparison of all sets, are only for the fastest laps, i think it might be better to sellect all 9 laps, and then analyse each set and get the max. min. and avg. I did that, but there is not a big difference for this test in particular.

    Maybe the most reliable information should be the average cornering speed, this is what it looks like for the sets:

    0.0 - 194.0 km/h
    1.0 - 195.1 km/h (not on the fastest lap)
    1.5 - 195.8 km/h
    2.0 - 195.8 km/h (not on the fastest lap)

    Max cornering is not important on this test, bacause it was on turn 1.

    There are also some differences in lap times from the data, compared to in game:

    Here you can see i did 3 laps exactly the same, 1:12.900, with camber at 1.5:

    [​IMG]

    But in game, they were > 1:12.904 , 1:12.898 , 1:12.907
    Well, they are still the same physically, but i just wanted to point that out.


    @Thorin Cain, good observations! And you are very right about turn 3 & 4, my max g, in turn 3,was using 1.0 and 1.5 of camber, they both gave -1.65, with 0.0 it was -1.64. And the lowest max value, at the same place, was with camber at 2.0, wich was -1.61.

    Take a couple of minutes to watch the training video, is very helpful. :cheers:


    @Motor City Hami and @Thorin Cain, thanks a lot for your time to make these observations and posts! If you need the data, just send me a msg that i will sent it right away.
     
    Last edited: Dec 28, 2014
    DolHaus, TurnLeft and Thorin Cain like this.
  3. Thorin Cain

    Thorin Cain Premium

    Messages:
    2,262
    Location:
    Scotland
    :lol: I will do! I'm sure it will be, I'm going to be doing a lot of testing now we have the tools and I have the taste for it. So it is needed to get the best out of it. :tup:

    And you are most welcome. Thank you for putting in the time and contributing the data:cheers:
     
    DolHaus, TurnLeft and LeoStrop like this.
  4. Motor City Hami

    Motor City Hami

    Messages:
    4,801
    Location:
    United States
    Adding some real world perspective (flame suit engaged). At Mid-Ohio my Miata likes a lot of camber. In fact, it likes as much as I can put on the front. My camber settings there are 3.2 front and 2.7 rear. There is much elevation changes at Mid-o, some banking and really high speed corners.

    At super flat circuits like Waterford Hills Raceway and Gingerman, the Miata likes less camber, like 2.5 front and 1.8 rear. There is no banking on either track. Waterford is also very tight with lots of slower speed corners. My best times were with these settings; 2.5 LF, 2.0 RF, 1.8 LR, 1.5 RR.

    If GT6 only had independent wheel settings and tire temp/pressure data, there would be no argument.
     
  5. Highlandor

    Highlandor

    Messages:
    2,385
    Location:
    United Kingdom
    I think people have forgotten about the one variable that changes consistently throughout the test.

    The main test is segmented into a specific car on a certain track, this is then broken down into the variable being tested (camber) into smaller tests, where everyone uses the same values of camber, but repeats the run with the next value of camber. So all is well and good here, there's consistency, apart from one aspect - the driver.

    Taking an average, and also splitting the track into segments is again all good, but without cross referencing the 'average' against the one consistently changing variable (driver), this undermines the results.

    It all depends on a big 'if'; if the average matches differences between drivers i.e. someone who feathers the throttle and trail brakes, compared to someone who uses 100% throttle & brake, or someone who is smooth with corner entry compared to someone who 'throws' the car into the corner, by testers from each 'style' getting both higher and lowers values than average - then all is well and good. The average can be used as the definitive data for people to analyse for conclusions as it's been proven the changing variable doesn't have an impact.

    But if one 'type' of driver or driving style etc is consistently different i.e. either higher or lower values than the average, especially if an opposite style gives similar results in the opposite way, then this shows the changing variable does have an impact. Therefore, the average would be better suited as a base line reference to compare against the results of similarities between the changing variable (ie driver, driving style), as it's been proven that the changing variable does have an impact.

    Not sure I explained that in a way everyone will understand. :indiff:
     
    Last edited: Dec 28, 2014
  6. ALB123

    ALB123

    Messages:
    5,162
    Location:
    United States
    I agree with what you are saying @Highlandor, and in a laboratory setting, I think you would be on the mark. However, I disagree with you for one simple reason, "camber" should be effective regardless of driving style. To bring credence to your point, our varying driving styles will possibly end up showing that a driver who throws their car into corners benefits more (or less) from camber than someone who takes wide, swooping, smooth turns. Or perhaps, those who use 100% throttle & brake benefit more from camber than trail brakers. Regardless, I think the results should still show whether there is an overall benefit to using camber or not. However, you are right, if we only take the "average of all the data" as our definitive answer, then we are getting a blend of all driving styles results.

    Basically, what it's going to come down to, once this experiment is finished (and I hope this ends up being an exhaustive study) is if we see Thorin Cain, for instance, always benefiting tremendously from camber settings and no one else is THEN we must analyze the difference in driving style. Otherwise, I contend as I did in a previous point, that if we are all going to agree on ONE driving style, ie., we all agree to brake at the 50M marker and take a mid-corner apex one turn one, full throttle to turn two and brake between 50 & 100M marker but take a late-apex, etc.. then why would we really need more than one driver?

    What I'm saying is, I think right now the differences in driving styles is actually going to be beneficial to the experiment even if we don't fully understand why at this moment. At the same time, I DO understand what you are saying and your point is valid, as contradictory as my statements might seem. I just don't think that greatly different driving styles will negatively affect the eventual conclusion of this experiment. I think it will have an impact, yes, but I think that's why we need to collect many measurements, with as many drivers as possible, and then that impact will actually be a positive one.

    To use @Motor City Hami as an example. He just told us how in the real world, he needs more camber on Mid-Ohio than he does Waterford. Now, if you or I were racing him in our Miata's, with possibly different driving styles, I'm sure we would probably end up using different values of camber than he does, but camber nonetheless. Because we know in the real world that camber is going to be beneficial no matter what in that style of circuit racing.

    Again, by taking an average measurement we are only seeing one answer and if we regularly see certain drivers benefiting from more camber than the rest of the testers, or vice versa, then we probably need to start investigating what the heck is going on between driving styles....At least, these are my thoughts on the matter. Perhaps I am well off base. I certainly don't mean to be. :tup:
     
  7. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Not sure what you are trying to say there?

    1) I agree that looking at sector times could prove fruitful and insightful.

    2) Again I agree that average data might hold better results, I wasn't aware that this was possible using the software when the experiment started and so the peaks were used. I would still like to keep the peaks documented as well though

    3) Wasn't really looking at steering angle to be honest, if everyone was using the same means of control then I might consider it but I don't know if we will get usable results with both DS3/Wheel being used.

    4) Same situation as 2)


    At this stage I would rather concentrate on data from a single lap rather than a collection purely for accuracy. There are a lot of small things that can really throw the data and as a result I would have to demand a squeaky clean run over X laps which would put more stress on the testers. With a single lap it is much easier to control the variables.
    This will of course be reviewed as a testing method later on in the experiment once we start homing in on the exact effects but at this early stage I don't feel it is the right approach :tup:

    Very well put, we are looking for the trends that affect everyone rather than the isolated improvements/losses encountered by individual drivers. :tup:
     
    TurnLeft likes this.
  8. Ridox2JZGTE

    Ridox2JZGTE

    Messages:
    11,648
    And this was on racing slick and not so high spring rate, correct ? More grip, more roll, more camber if I say that correctly, with street tires or semi slick and stiffer suspension, less camber would have been preferred, especially on cars with a lot geometry changes when driven hard ( toe and camber ).
     
    Thorin Cain, TurnLeft and DolHaus like this.
  9. Highlandor

    Highlandor

    Messages:
    2,385
    Location:
    United Kingdom
    No worries..

    To be honest, if the impact of driving styles or how these can impact the data aren't clear, then I don't know what more to say..

    As always, it was just a suggestion, so it doesn't have to be used :tup:
     
    TurnLeft, Thorin Cain and DolHaus like this.
  10. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    Its a fair point, driving style almost certainly will have an effect on results but we can't isolate or identify the difference at this early stage. Keep working on these thoughts so we can investigate them at a later stage :tup:
     
    Thorin Cain and TurnLeft like this.
  11. Lionheart2113

    Lionheart2113

    Messages:
    1,485
    Location:
    United Kingdom
    I'm not sure if it's because I have the basic MoTeC or what, but when I look at the steering input it shows me turning the wheel to the right at Indy.:confused:
     
    Thorin Cain and DolHaus like this.
  12. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    This goes against everything I know about oval racing (drive real fast and turn left sometimes) :lol:
     
    Thorin Cain likes this.
  13. Lionheart2113

    Lionheart2113

    Messages:
    1,485
    Location:
    United Kingdom
    That's what I was thinking. Keep things simple! That's why I only have raced at New Paris speedway...1/4 mile bowl(maybe closer to 3/8). Left turns only! And we had severe front right camber.
    Just concerned about using steering data if they can't even show me turning the correct way. Unless I'm the only one who has this issue.
     
    Thorin Cain and DolHaus like this.
  14. LeoStrop

    LeoStrop

    Messages:
    890
    Location:
    Brazil
    It's inverted for some reason, i have that in pretty much every data, but the angle should be correct.

    Also, your in-lap and out-lap always have some issues, so make sure you don't select neither of those parts, on the graph.
     
  15. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    I think it could be one that needs to be put on the back burner until later, at this stage I don't think it will give us relevant data. There will almost certainly be a correlation between camber and wheel angle but there are a lot of areas of uncertainty at the moment so I can't trust the data.
    The main area of concern is the difference in function between DS3 and a wheel, if tested in future these results may have to be separately analysed :tup:
     
    Thorin Cain and TurnLeft like this.
  16. Thorin Cain

    Thorin Cain Premium

    Messages:
    2,262
    Location:
    Scotland
    When someone with a wheel finally puts up some data that is. ;)
    Not having a go at anyone, just saying. I know it's on the way :tup:
     
    TurnLeft and DolHaus like this.
  17. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    @Motor City Hami has a wheel not sure he used it for the lap data he posted I just sorta skimmed the last twenty posts.
    Has anyone found away to export a set up from Motec so we all use the exact same Motec output so the data is the same.
    I did some reading on the Motec gt6 forums the lack of suspension and wheel data is PD's doing the are not exporting it to the file so Motec can't display it so we are again waiting on PD to allow us access to that data.
     
    DolHaus, Thorin Cain and TurnLeft like this.
  18. Thorin Cain

    Thorin Cain Premium

    Messages:
    2,262
    Location:
    Scotland
    Fair point my good man! And I'm sure he probably did.

    But I was referring more to the original test that is part of the experiment, where all 3 sets of results have been gained by DS3 users (last time I checked at least.) Of course all test data from all sources are valid and welcome. The more light that can be shed from various other tests the better as far as I'm concerned.

    As I said, I was not having a go at anyone and I apologise if I caused anyone any offence. That was not my intention at all.

    Just a light hearted observation was all.

    And no, sorry no idea about exporting set-Ups from the Motec software.
     
    TurnLeft and DolHaus like this.
  19. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    I guess I can have a look and try to write a "How To" on how I want the software set up and the data collected, that would probably be the easiest way :tup:



    Notice to Everyone:
    Data outside the parameters of the main test is welcome to be posted and discussed but will have no bearing on the outcomes of the experiment. We cannot gain any accurate conclusions from a single point of information no matter who the source is :tup:
     
  20. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    There has got to be away to locate the worksheet file
     
    TurnLeft and DolHaus like this.
  21. LeoStrop

    LeoStrop

    Messages:
    890
    Location:
    Brazil
    If you go to >Layout > Layout editor(Ctrl + F7) , you can export each individual worksheet, or the entire workbook. But i don't know if the settings remain the same, or if it only saves the locations and elements of your worksheets.

    Mine is here, if someone wants, i think all the worksheets can give you a good idea of your lap.
     
  22. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    Perfect the worksheet is the hard time consuming part. Changing setting to display mph and feet are easy todo so @DolHaus can set up a global sheet we use to get the same data out every time
     
    Thorin Cain, LeoStrop and DolHaus like this.
  23. Motor City Hami

    Motor City Hami

    Messages:
    4,801
    Location:
    United States
    I use the G27 90% of the time.

    I have the flu again so my game play time over the past three weeks has been very limited.
     
    TurnLeft, Otaliema and DolHaus like this.
  24. Otaliema

    Otaliema

    Messages:
    5,489
    Location:
    United States
    Get well soon sir.
    With how persistent this flu has been you might want to look a herbal supplement called Vircaid it's about 30 bucks a bottle and will knock down any bug with in 2 days, so normally two bottles if your up for 15 hours a day.
     
    TurnLeft likes this.
  25. LeoStrop

    LeoStrop

    Messages:
    890
    Location:
    Brazil
    Thanks for allowing us to share our findings, Dolhaus. These unofficial tests help us visualize the best way to make the official test, and what we need to be concerned about.

    Also, i did some quick tests online with tire wear set to "Very Fast", using SH, at Monza 80's. I was faster using 1.0 of camber(+-1sec), which was the last set i tested. But for the tire wear, camber had no advantage, the front tires drop to 9, almost at the same time, but with 1.0 of camber they dropped a bit earlier, maybe because i could get more out of the car, and i was also a lot faster. So, in my opinion, camber doesn't affect nothing on tire wear.

    I will do this test again and complete, i was just playing around with my impreza, i know that i need to repeat the bmw 135i test online, so we can compare it to arcade mode, but i was getting bored of HSR. :p
     
  26. ALB123

    ALB123

    Messages:
    5,162
    Location:
    United States
    I found your post to be perfectly clear, and I agree with it...I just don't know what we can do about the different driving styles. When you are running an experiment to test ONE thing, ideally you want every single other "factor" to remain consistent/constant in order to get the most accurate scientific results. Unfortunately, I can't even begin to understand how we can control that "driving style" variable. Personally, I only know how to drive the way that I drive. :grumpy: Maybe that's why I'm always so far behind the really good GT6 racers... :lol: :D In all seriousness, I really want this experiment to be valid, but I'll confess I am not 100% confident in some of my suggestions. At the same time, I don't want this to be a nightmare for all of those involved. I am confident that things will work out, in the end, once all the particulars are settled.

    Thank you! I saved a copy and I look forward to using it later on... :tup:
     
  27. TurnLeft

    TurnLeft

    Messages:
    1,370
    Location:
    Canada
    I personally don’t worry much about the different drivers and their styles, if we were to test the differences between the tire compounds available we would all get different lap times but there would be a constance that would show some types of tires are faster than others, same for the camber, yes certain settings will favor a certain type of driver, but there should be a constant pattern that will show up, and that’s what we have to look for, not the exact value. What is important is that for the following tests the drivers who start them will finish them, that way we will have our first constance and the rest will show up in numbers, graphs or however we want to show them. I do hope you will be onboard with me for the next tests, I don't want to be the only one dodging moose's.
    [​IMG] [​IMG]
     
    Thorin Cain, LeoStrop and DolHaus like this.
  28. ALB123

    ALB123

    Messages:
    5,162
    Location:
    United States
    Yeah, I think I said something similar in a previous thread. I firmly believe that in this particular experiment the varying driving styles will not devalue the results. As you state, IF we start seeing a particular pattern with one driver that varies wildly from the rest of the testers we can investigate what's going on later. As long as no one deletes the replay files, we can always go back to examine the driving style if someone is benefiting or suffering at a rate wildly inconsistent with the rest of the group.

    I'm 95% sure I will be contributing to this experiment starting with round 2, whenever that may be. Has the course, vehicle and tune been determined yet? I hope that wildly powerful cars aren't used (like Red Bull x2014) because I struggle with my consistency with those...GREATLY! :banghead: :lol:
     
    Thorin Cain, TurnLeft and DolHaus like this.
  29. TurnLeft

    TurnLeft

    Messages:
    1,370
    Location:
    Canada
    Don't worry, I've borrowed all the keys from those type of cars in DolHaus's garage, and he's only getting them back after all this is done, also took the keys to the Stratos, so were clear believe me. :lol:
    [​IMG], don't lose sleep over that. [​IMG]
     
    DolHaus likes this.
  30. DolHaus

    DolHaus

    Messages:
    3,865
    Location:
    England
    The car will be the same as this first test (Jag XKRS) to gain direct comparison data, SSR5 Clubman is looking likely at the moment but I haven't tested yet :tup:

    I wouldn't choose a car/tune that was beyond the skill of any driver as it would be counter productive when trying to control variables
     
    TurnLeft and Thorin Cain like this.