Example of MoTeC data analysis

Discussion in 'Gran Turismo 6' started by ALB123, Nov 20, 2014.

  1. TCSdisable

    TCSdisable

    Messages:
    359
    Location:
    Hong Kong
    I sure someone have problem track map and Wrong Lap time on Sierra and Suzuka (Maybe only issued on 2014).
    The problems is produce by some beacons were place wrongly.
    Wrong Beacons.jpg

    Found out the wrong beacons ID
    Wrong ID.jpg


    Remove them from Tools > Lap editor to to correct Laps time of Telemetry.
    Make sure at least one lap time telemetry could set up as "Trusted Lap" on "Laps" Tab > "Edit".
    Lap editor.jpg

    Them let the Track editor using the correct telemetry source to draw a map, and make sure kind of circuit, for Goodwood or any Point to point layout should pick as "Open Circuit", remember, no GPS data was recorded on GT6 MoTeC right now !
    Track generator.jpg

    Them track map would be corrected.
    Full.jpg
     
    Last edited: Dec 1, 2014
  2. YZF

    YZF

    Messages:
    820
    Location:
    Lithuania
    True, however some drivers during accidents have very high g forces, yet they survive...
    "At the United States Grand Prix, Schumacher suffered serious injuries in an accident that occurred on the ninth lap of the race. The deceleration was measured at 78 g (765 m/s²), resulting in a concussion as well as two minor fractures to his spinal column. Schumacher was forced to miss the next six races."

    http://en.wikipedia.org/wiki/Ralf_Schumacher
     
    super_gt likes this.
  3. cutmeamango

    cutmeamango (Banned)

    Messages:
    672
    There are four factors (This is a quick simplification, excluding all biomechanical and vehicular structure factors) for tolerance of acceleration:
    - The onset of the acceleration, in this case the time it takes to reach peak acceleration (the jerk @Griffith500 mentioned).
    - The duration, which is referred as sustained (linear acceleration over time).
    - The vector, up/down (head<>foot), front/back and left/right, that can all happen concurrently.
    - Acceleration magnitude.

    ---

    What we have with the X2014 are two lat G scenarios:

    1: High G onset (35G/s if I don't suck at math), high magnitude acceleration peak (21G), but low duration (~1.1s for the whole acceleration and deceleration).
    = This jerk is somewhat similar to what it is experienced in some types of parachute opening, but laterally here. At the 2nd link at the bottom it says impact tests up to 21.6G (much higher onsets) didn't reach a tolerance threshold.
    *edit* Adding: The whole section of decreasing jerks last 8secs, don't know if it would constitute a dangerous vibration. Could be the most dangerous part of it all though.

    2: High avg. sustained acceleration (~9.8G during ~1.2s, repeatedly).
    = Griffith pointed out this could be the troublesome scenario and it would probably be in the long run in terms of discomfort and muscular fatigue. At the first link at the bottom there's a graph about design guidelines of acceleration x time design limits/tolerance for those unconditioned, and this scenario is just slightly above the threshold.

    ---

    Both cases are certainly much above what F1 drivers are exposed, but they are not above thresholds of tolerance (or injury), and that's without accounting the use of specialized designs of anti-G suits, HANS devices, seating and body restrainers for attenuation.

    For those interested in better summaries, here and here. Cool article about how HANS work and seatbelts physics. You can too search for free fall survivability studies to see the physiological effects of really high acceleration impacts.
     
    Last edited: Dec 1, 2014
    Griffith500 likes this.
  4. ALB123

    ALB123

    Messages:
    5,162
    Location:
    United States
    The human body is capable of handling very high G loads for incredibly short periods of time. I'm talking fractions of a second.
     
    Rotorist and B4NDiT like this.
  5. Elevenfive

    Elevenfive

    Messages:
    6
    This is very cool! I need to try this out.

    What would be the best track to use with the track overlay? I would like to see the exact line I'm taking, along with braking/ accelerating points.

    I gather some tracks aren't as accurate as others for track overlay and require some tweaking?
     
  6. ALB123

    ALB123

    Messages:
    5,162
    Location:
    United States
    The software uses precise GPS coordinates to map out the track. I don't think any tracks are going to be "more accurate" than others, however, I haven't even tried 10% of the layouts so I could be wrong.
     
  7. MatskiMonk

    MatskiMonk

    Messages:
    12,842
    Location:
    United Kingdom
    GPS data only exists for some tracks, the track that the software creates is based on your racing line, not a pre-defined track map as such. I could be wrong about this.
     
    GregOr1971, Elevenfive and ALB123 like this.
  8. ALB123

    ALB123

    Messages:
    5,162
    Location:
    United States
    Well, there ya go! I stand corrected. :lol:
     
  9. MatskiMonk

    MatskiMonk

    Messages:
    12,842
    Location:
    United Kingdom
    Back a few pages Example of MoTeC data analysis the screen caps I posted of Tsukuba show that it does have GPS data, but you can see from the track editor map how it only actually follows my line, hence the corners are smoothed out, and there's a kink in the line after the first hairpin...
     
    Rotorist likes this.
  10. Elevenfive

    Elevenfive

    Messages:
    6

    I see thanks for the info, I need to play around with this software when I get a chance.


    It would be great to see a track overlay on top of the GPS data, to visualize what line is faster or compare braking points etc
     
  11. YZF

    YZF

    Messages:
    820
    Location:
    Lithuania
    To find out which line is faster, you use ghost of your own previous best time
     
  12. MatskiMonk

    MatskiMonk

    Messages:
    12,842
    Location:
    United Kingdom
    In game data logger is probably best for this, as it will sync up to comparative laps, and display the racing liners over the ingame track map, you can then look at the brake pedal trace, the steering angle trace, and the speed trace, on the entry, apex, and exit of the corner respectively, for two laps at the same time. You might be able to do this in the MoTeC software too, but it still doesn't have a track map that will show you two lines that you used against an actual map of the track.
     
  13. Johnnypenso

    Johnnypenso Premium

    Messages:
    28,465
    Location:
    Canada
    You know what's a great tool, simple to make, wouldn't cost more than a day's worth of programming and would be hands down, without a doubt, the single most effective hotlapping tool? A session delta app. I found it in Assetto Corsa and it simply tracks in real time, the gap between your current lap and your fastest lap. So if you enter a corner say 0.300 down on your fastest lap and you want to experiment with late braking and a late apex, you do your stuff, and as you exit the corner you glance at the counter and see which way it's going and where you end up entering the next corner. No need for a ghost to interfere with your vision or concentration and of course a ghost isn't effective if it's behind you or you fall a second or two behind after an error. It's been absolutely invaluable for me and tells me in real time, the exact effect of my driving inputs down to the millisecond. I'm watching it sometimes and getting mad because I lose 1/10th between corners!!
     
  14. YZF

    YZF

    Messages:
    820
    Location:
    Lithuania
    Yes, indeed. F1 drivers (IRL) have this kind of system. I have noticed it when watching onboard videos. This should be really invaluable for hotlapping
     
    Rotorist, Johnnypenso and super_gt like this.
  15. Elevenfive

    Elevenfive

    Messages:
    6

    I'm aware of the ghost and use it, but I'd still like something that I can analyze and look at while not in the middle of driving, something that I can go back to and compare each lap

    Something like the attached picture, I think would be helpful, for me at least
     

    Attached Files:

  16. yuc2002

    yuc2002

    Messages:
    220
    Location:
    United Kingdom
    Does this do overlays?
     
    Rotorist likes this.
  17. Dyemonic

    Dyemonic

    Messages:
    23
    Here's a few resources that may help you guys out:


    This is a video I made last year for my school's FSAE team. It's long and covers more stuff than what GT6 generates.


    MoTeC's video tutorial. Much shorter, but having not watched it I don't know what it covers and what it doesn't.

    And attached is a pdf seminar.
     

    Attached Files:

  18. azidahaka

    azidahaka

    Messages:
    5,536
    I saved a race replay to see how a certain someone was way too fast >_>

    and... wait a second, the export just shows your own datas? :banghead::banghead::banghead::banghead::banghead::banghead:

    Please someone tell me i'm an idiot and that there's a way to compare a race's best laps from different drivers :/
     
  19. Elevenfive

    Elevenfive

    Messages:
    6
    Thanks for this, your video helped me to figure out how to get the view I wanted
     
  20. Elevenfive

    Elevenfive

    Messages:
    6

    I believe it only records your data not other drivers data
     
    Rotorist likes this.
  21. azidahaka

    azidahaka

    Messages:
    5,536
    Damn i feared this... It lowers usefulness a lot :/
     
  22. Griffith500

    Griffith500

    Messages:
    8,633
    Data recorded from online opponents is very low resolution anyway, and the physical movement of online cars is produced differently. So the data would have gaps and discontinuities, as well as lots of unrealities inherent to the latency compensation mechanisms and low bandwidth requirement for peer to peer communications.

    I'm not sure what practical benefit analysing the AI's driving would bring, but it would be fascinating.
     
  23. ALB123

    ALB123

    Messages:
    5,162
    Location:
    United States
    You would have to ask the person you're interested in analyzing for a copy of his MoTeC i2 data files. Then you can open up HIS log file in addition to YOUR log file in the same project and overlay the data.

    Of course, if the person forgot to save his replay file or refuses to give you the files then that is a moot point.
     
  24. azidahaka

    azidahaka

    Messages:
    5,536
    Yup, would have been great for cheaters spotting and finding out show some guys are that fast...

     
  25. Noname0529

    Noname0529

    Messages:
    186
    Location:
    Canada
    Ok, so this is a thread about GT6 and the MoTeC i2 pro tool.
    Post your laps, compare with others, and have fun. :)
    Download link and Steps are here: http://www.motec.com.au/gt6/gt6downloads/
    Also if you're gonna post your laps, make sure to include the name of the car and the track.
    Here's mine:
    Vehicle: Honda RAYBRIG HSV-010 '12
    Track: Red Bull Ring Short Track
    rbr lap.png rbr lap 2.png
    rbr lap 3.png
     
    Last edited: Dec 15, 2014
    ProjectF and nuerphil like this.
  26. nuerphil

    nuerphil

    Messages:
    48
    Location:
    Germany
    This is the thread I looked for! In the next weeks I´ll create a video tutorial how to use it and how to get better by comparing laptimes! I probably will use the 370z on Stowe to help new GT Academy candidates and I´ll try to transfer what I learned ;)
     
    Rotorist and ALB123 like this.
  27. Noname0529

    Noname0529

    Messages:
    186
    Location:
    Canada
    That fantastic!:tup:
    I look forward to seeing the video!:sly:
     
    nuerphil likes this.
  28. YZF

    YZF

    Messages:
    820
    Location:
    Lithuania
    As we were discussing high speed acceleration analysis and issues in this thread, I made new (interesting) test regarding top speed issue in GT6. I don't want to duplicate, so I'll just provide the link
     
    Johnnypenso likes this.
  29. ALB123

    ALB123

    Messages:
    5,162
    Location:
    United States
    It's funny. I created another thread that provided a link to a magazine article written in 2003 that showed a hot lap around Nordschleife in a BMW M3 and the article provided an overhead view of the entire track spread across dozens of pages and it showed the driving line, braking points and exactly which gear the driver was in and where gear changes took place.

    For kicks, I bought a 2004 BMW M3 and downgraded the tires to CS from SH and took it for a rip around Nordschleife in stock form - no practice laps. I wasn't looking to set any records. I wanted to drive fast, but clean and realistically - no 45 degree slides around corners with the tires screaming like theyre about to explode. Wouldn't you know that the driver in the magazine posted an 8 min 22 second lap time and my hand to god, I posted an 8 min 22 second lap time?! The gear changes match up remarkably well - although I tended to hold 5th gear longer and/or not even get into 6th gear on a few occasions that the BMW Test Driver did enter 6th gear.

    I provided my MoTeC Track Report that shows which gear I was in and what spots on the track (I hid the speeds as that wasn't pertinent to my objective) in that thread. My point is, while Gran Turismo gets a lot of things wrong - I don't think anyone would say they don't - they do tend to get a lot of things right. I was blown away at how close our laps matched up.
     
  30. LeoStrop

    LeoStrop

    Messages:
    890
    Location:
    Brazil
    I just started using this toll this week, what an amazing feature. Overlaying laps and changing the graph from time to distance, shows you exactly what you are doing right, or wrong.

    I don't think people are giving enough credits for this, even without the suspension data, you can read your lap in different ways and overcome its flaws. Looking forward for more testing, please keep this post updated with new discoveries and tips, thanks!