Thread Tools
This thread is privately moderated by erikspencer, who may elect to delete unwanted replies.
Dec 30, 2018, 02:26 AM
Registered User
Thread OP

Racing rates, A tutorial

I thought we'd have a look at some of Velocidrone's top racers who've had quite a bit of success on real tracks as well: SFPV and Holdox.

When looking at their rates, I'd encourage thinking of them in terms of:
  1. What is the slope at mid-stick?
  2. What is the max deg/sec?
  3. How does Yaw compare to Roll or Pitch?
  4. What is the general shape of the curve between mid-stick and max?

Name: sfpv.JPG
Views: 43
Size: 154.6 KB
  • SFPV uses fairly low sensitivity at mid-stick. On Roll/Pitch the slope is 0.8. On Yaw, it's 0.6.
  • At 239 deg/s, his Yaw is quite low.
  • He uses zero Expo (typical), and uses Super to control curve growth. This means the slope of the curve grows faster than linear, in a consistent manner (we'll explain later).

Name: holdox_correct.JPG
Views: 26
Size: 151.0 KB
  • Holdox max deg/sec is fairly high at 524 deg/s.
  • Yaw/Pitch/Roll all have the same max deg/s.
  • Yaw is less sensitive at mid-stick compared to Roll and Pitch.
  • His equivalent slope at mid-stick is 1.44 on Roll/Pitch and 1.03 on Yaw, substantially higher than SFPV's 0.8 and 0.6.
  • He uses zero Super, and uses Expo to control curve shape. Meaning, the curve grows slightly faster than linear of the first half, and growth is more exponential in the latter half.

The last bullet probably causes the most confusion. Basically, SFPV has used the traditional approach of setting a Super Rate value to control curve growth, whereas, Holdox sets the max deg/sec using RC Rate, and then applies Expo to flatten out the middle portion. In terms of the basic shape of the curve this produces, they are not equivalent. To illustrate the difference in shape, the two curves on the plot below start off with the same slope and end at the same max deg/s, but note how the lower curve (per Holdox approach) is quasi-linear, out to about 40% command, whereas the SFPV approach has a more [B}consistent growth[/B] versus command %.
Name: comparison.JPG
Views: 15
Size: 58.3 KB
You may have heard people talk about using Expo to "flatten out" the middle portion. Technically, it affects the whole curve, but the distinguishing feature is the flat growth (linear) in the middle. Adding a bit of Expo to a curve that has Super rate just results in a shape somewhere between these two.

Okay, so now the important part, what can we learn from this comparison?
  • Initial slope controls precision, your ability to make small adjustments. Both racers have more precision (lower initial slope) on Yaw, giving them smoother Yaw movements and greater resolution. You'll quickly feel what is too sensitive or too dead for you in terms of initial slope. The harder choice is Max deg/s and curve shape (more on this later). During turns, there's coordination between Roll and Yaw required; but at high camera angle, not as much Yaw is needed. If you're flying at >50 deg or so, I'd suggest adopting their approach of giving Yaw a lower slope than what you're using on Roll/Pitch.
  • The initial slope you use will depend on the feel of your gimbals, length of your sticks, accuracy of your fingers, how quickly you like the quad to respond, and so on. It may depend on the race as well: smaller gates require greater precision. I'd suggest testing on a track that requires precision versus a track that requires quick turns, and find a value that works for both. You'll likely continue making tweaks over time.
  • The Max deg/s of the two racers differ substantially. Your preference probably depends on how you fly. Some racers like pushing their sticks full-over in turns. In this case, they might use a lower max deg/sec that gives a predictable max-turn rate. Other racers like doing quick snaps to adjust their orientation, so require higher max values.
  • (Holdox-advocate) His approach has greater linearity near mid-stick. The linear-growth is more tamed and allows you to be a bit more fat-fingered and aggressive. You can actually use a large initial slope and have it "feel" less aggressive than a curve based on Super Rate. So, if your fingers flow easily, or you've enjoyed the predictability of linear rate curves in the past, then take a hard look at his approach.
  • (SFPV-advocate) His low rates/slopes certainly require a lot of stick movement. If you're just starting out, be warned you won't get immediate feedback on what your fingers are doing. But with more experience, the lower rates provide greater resolution and control. In terms of setting the curve shape based on Super Rate, this is the traditional approach preferred by most racers. I speculate the reason being is that it sort-of allows you to have the best of two worlds: (1) low initial slope for precision in the 0-15% of stick movement, (2) fast-enough growth that stick throws > 15% don't feel smushy.

In summary, it's a matter of preference, but hopefully, this discussion provides a framework that'll help get you started. If your approach or understanding of rates differs from what I've presented, I'd be happy to hear. Lastly, I've made a web app that might help:
Last edited by erikspencer; Mar 04, 2019 at 12:42 AM.
Sign up now
to remove ads between posts

Quick Reply
Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion ADVANCED Electronics Tutorial (Why LiPo "C" Ratings are Fake) WallacEngineer Electric Power Cars 8 Jan 09, 2019 11:32 AM
Discussion Learn Acro Flying Tutorial Lesson Rate Mode dermott Mini Multirotor Drones 6 Nov 07, 2017 02:53 AM
Discussion XC Racing 1/18 Buggy Build Tutorial Teeder Micro Cars 0 Apr 20, 2016 08:51 PM
Mini-HowTo Internal Resistance & C Rating: Tutorial From Electric Clinic #3 Paul Naton Electric Competition Soaring-F5J/ALES/e-Soaring 5 Mar 05, 2016 09:10 AM
Discussion Blade MQX tutorial on Spektrum DX6i dual rate settings that make it easy to fly! the desolate one Multirotor Drone Talk 0 Feb 20, 2012 01:56 PM