RaceChrono v8.0 beta

2

Comments

  • edited June 2023
    @ConteDj no, the RaceBox Mini needs to be configured as "RaceBox Mini". If you configure the wrong device type, it will continuously reconnect.

    If it's not found then make sure to disconnect the device from any other phone and any other app. It cannot be connected anywhere before connecting with RaceChrono.
  • Hi

    i have ios and Vlinker MX+

    hiw I can connect to beta test ?

  • Hello,
    my RC Pro got updated yesterday on Google Play, I now have v8.0.2. Isn´t v8.x still Beta? I cannot remember I´ve opted for the Beta-testing...
  • It's 100% in production since yesterday.
  • Thank you for the welcome news of the official release of v8.0.2 for Android and v8.0.3 for iOS.

    I'll be able to meet the next beta or production version (e.g. v8.0.4) soon, right?
  • edited June 2023
    @ShoRang yes, v8.0.4 beta will be soon on both platforms. They will make to production after couple of weeks testing in beta.
  • edited June 2023

    I just tried the new version with some sessions I have previously imported from gopro. The new analysis features are awesome! I can't wait to try it with better IMU 🙃 (no, my car is not capable over ~1G)


  • the traditional way of calculating drift angle is with 2 gps antenna, front and rear. We use oxts ins boxes at work but they are not really hobby-budget compatible.

  • Hello!
    For some reason, when I overlay devices on the video, when saving the video, it does not display what I specified in the settings. Only speed and overload. When I go into editing, everything is fine there.
  • @Igorka it hides the gauges with channels that do not exist in the session it self.
  • @ShoRang v8.0.4 beta for Android now uploaded (may take and hour or two to appear)
  • edited July 2023

    Thank you! I tried v8.0.4 Beta for Android right away, and I'll give you quick feedback.


    First, In v8.0.4, there are 'shadows' that have never been seen before in all letters and numeric fonts, such as digital gauges and arcs and timers, etc.

    This is determined by the 'Background alpha' value, and this unintended 'shadow' is emerging with subtle changes in font size at alpha values below 20%. At alpha values above 30%, there is no such phenomenon at all.


    The 'shadow' phenomenon in the above pictures and videos was not seen in previous versions (v8.0.3 and below).

    In the video below, you can see that the same 'Background alpha' change in v8.0.3 for iOS does not create a 'shadow'. This was the same v8.0.2 and below for Android.


    Second, Among the input channels of the digital gauge, too many 'Time' channels were created in OBD-II reader channels. Please look at the picture below.


    Hope it's always helpful to you and RC team! Thx :)

  • edited July 2023
    @ShoRang

    OK, I screwed up font initialisation apparently. Did some related fixes, and then this happened :blush: I will fix the glitch with the OBD-II time too.

    I will make a hot fix ASAP. Thank you for testing it before I pushed it to production!
  • @ShoRang please see the new v8.0.5
  • edited July 2023
    Just released v8.0.6 beta. Edit: Also released v8.0.7, busy day :smiley:
  • Hi, I really appreciate the updates of v8 regarding more customizing of overlays.
    But on more request from my side: I would like to have simple vertical bars to show throttle and brake side by side to speedometer. Right now I can only find the circle gauge. Would this be possible to implement?
  • edited July 2023
    @WernerTiffy Yes this is something the new overlay editor allows us to do, and we're planning to do it.
  • Hi, I also like the idea with an option to have bars. Would it be possible with the new layout editor to also add "ON/OFF LED's" for sensors that only has two states?

    Thanks,

    /Per

  • edited July 2023
    Hi everyone!
    Great improvements in progress, really appreciate that!

    Last time on track I noticed that the red dot showing the position in the video overlay is somehow shifted a few pixels (or a few meters) off the white line.
    I don´t know how to paste images or videos here. Hope you can see what I mean with this video link:

    "www.instagram.com/p/CuxHO1NroZc/"

    Best regards,
    Josef
  • @GiuseppeBinomi yep, we got couple reports like that, and seems they are accurate. Will be fixed soon.
  • So I am primarily a road course guy, but I have been doing a lot more drifting recently. Would be great to help add drift angle. Is there something I could do to help? I saw the comment about tradtionally it's done with 2 GPS (front and rear) and I guess you 'simply' calculate the angle from the 2 positions and the heading. The racebox mini has an IMU, and high refresh rate, which makes me wonder if it's possible to get a decently accurate angle without 2 GPS units.

    I can play around with post processing the CSV. I suspect it will have somethign to do with taking the last 1-2 gps locations to calculate the motion heading (direction of the car, but not necessarily the car's direction), and somehow determine the direction the car is facing. I do not know if the headine parameter is calculcated from previous GPS points, or if it's derived from IMU data. The IMU can detect the rate of rotation, it's very obvious when looking at that data to see the car transitioning, but I'm not sure if it can be used to accurately determine which way the car is pointing.

    Drift angle is just the difference between the direction of car's instantaneous motion vs the direction the car is pointed... Seems straight forward on the surface, but then you start to think about how you determine which way the car is facing, and it gets hard haha
  • So I spent a few hours in python trying to math this. I don't have a perfect solution, but I am able to sort of estimate angle... No idea if the process is correct, but essentially what I do
    1) I have the vehicle's rate of rotation (RoR) from the IMU
    2) Calculate the vehicle path's RoR by calculating the change in bearing over time (comes out to some deg/s)
    3) Get the difference of the 2 to determine how much of the total vehicle rotation is due to slip/angle instead of simply turning the car. Let's call this one slip-RoR
    4) I integrate the slip-RoR to estimate the slip angle/drift angle of the car, but since integration will sum all errors in, I have a few conditions to reset the slip angle to 0 such as
    - No lateral g's (abs(lat accel) < 0.05)
    - Low speed (probably not drifting at < 10 mph)

    I'm realizing it's kind of difficult to determine the difference between drifting and just turning the car. Integrating the rate of rotation works sort of, but errors accumulate quickly and can throw it off.

    This was a quick and dirty test I did in python, but here's the resulting output done with RaceRender, so considering it's a quick and dirty job done with just GPS and an IMU, It could be worse.... Please excuse the mediocre drifting :) I'm still quite new to it lol.

    I applied a scalar to the estimated drift value, and I think I might be too pesimistic haha. After checking the in-car against a satellite view, i'm like 1/2 of the actual angle.... whoops. Ez fix, but you get the idea....


  • edited August 2023
    @jlvaldez I've given some thought to the drift angle calculation and I haven't figured out how it could work with IMU and a single GPS alone. I think this is done traditionally with an RTK GPS which in this case would probably be two GPS's connected together, one antenna at rear and one at the front.

    I think that I one tried to use two independent GPS's, the data would not correlate enough to allow accurate drift angle calculation.
  • Makes sense if you can't ensure the two gps readings are accurate or synched enough.


    I'm merely integrating the IMU data to estimate the angle

  • edited August 2023
    I'm unsure if that will work. Probably the IMU has so much noise that it won't work with that alone, and you'll at least need magnetometer or something to steady the gyro drift. I think RTK-GPS system would be the best for this.
  • And you'd need 2x RTK-GPS in order to achieve the accuracy needed to determine the angle between front and rear. This is a tricky one!
  • Hello!
    I had a competition last week and the OBD MX+ kept disconnecting during the session. Before leaving the track, everything connected, but then the data was not recorded.
    And another question, when comparing two circles, two arrows follow a trajectory side by side, it used to be, on a fast circle, the arrow went forward and it was clearly visible where you were going faster and slower. Or is it somewhere in the settings?
    iOS OS
  • @Igorka I don't know about the disconnection issue, but the latter might be just a setting. In graph settings you can choose "Comparison X-axis scaling". In "Position" mode the arrows are always side by side, i.e. you will get what you just described I think. "Disabled" might be what you are looking for.
Sign In or Register to comment.