Posted on Mar 17 • Edited on Mar 30

Comparing LEGO SPIKE Prime Programming : Which Is Best for Robotics Competitions? - 2

Does the Programming Environment Affect Precision? A 360° Rotation Test
When using LEGO SPIKE Prime in robotics competitions, does the choice of programming environment impact movement precision?
To investigate this, I conducted an experiment comparing different programming environments.

Tested Programming Environments

I compared the following four environments:

Word Blocks (SPIKE App 3) → Download here
Python (SPIKE App 3) → Download here
Python (Pybricks) → More info
C Language (spike-rt) → GitHub repository

Robot Configuration

For the test, I used a car-type robot with the following setup:
•Left motor: Port A
•Right motor: Port B

Test Method

To compare the environments, I conducted the following test:

Command the robot to rotate 360° using a specified motor angle
Measure the difference between the target and actual rotation angles
Perform 10 trials for each environment and calculate the average error
The same logic was used for all environments

Program Code (for all environments):
Word Blocks (SPIKE App3)

Python (SPIKE App3)

tread = 8 d_tire = 5.6 goal = 360; for i in range(10): motor.reset_relative_position(port.A, 0) start_angle = motor.relative_position(port.A) motor.run(port.A, -300) motor.run(port.B, -300) while tread/d_tire*goal > -motor.relative_position(port.A): pass motor.stop(port.A, stop=motor.BRAKE) motor.stop(port.B, stop=motor.BRAKE) time.sleep_ms(500) print("error:", abs(motor.relative_position(port.A)-start_angle) - tread/d_tire*goal )

Python (Pybricks)

hub = PrimeHub() motorA = Motor(Port.A, Direction.COUNTERCLOCKWISE) motorB = Motor(Port.B, Direction.CLOCKWISE) tread = 8 d_tire = 5.6 goal = 360 for i in range(10): start_angle = motorA.angle() motorA.run(300) motorB.run(-300) while tread/d_tire*goal > motorA.angle() - start_angle: pass motorA.brake() motorB.brake() wait(500) print("error", motorA.angle()-start_angle - tread/d_tire*goal )

C Language (spike-rt)

void Main(intptr_t exinf) { dly_tsk(5000000); uint8_t tread = 8; float d_tire = 5.6; uint32_t goal = 360; motorA = pup_motor_init(PBIO_PORT_ID_A, PUP_DIRECTION_COUNTERCLOCKWISE); motorB = pup_motor_init(PBIO_PORT_ID_B, PUP_DIRECTION_CLOCKWISE); int8_t i; for (i = 0; i < 10; i++){ pup_motor_reset_count(motorA); pup_motor_set_speed(motorA, 300); pup_motor_set_speed(motorB, -300); while(tread/d_tire*goal > pup_motor_get_count(motorA)); pup_motor_brake(motorA); pup_motor_brake(motorB); dly_tsk(500000); float error = pup_motor_get_count(motorA) - (tread / d_tire * goal); syslog(LOG_NOTICE, "error: %d.%05d", (int)error, (int)((error - (int)error) * 100000)); } }

Results: Which Environment Was Most Precise?

Here are the average rotation errors (smaller is better):
1️⃣ 12° - C Language (spike-rt) 🏆
2️⃣ 13° - Python (Pybricks)
2️⃣ 13° - Python (SPIKE App 3)
4️⃣ 14° - Word Blocks (SPIKE App 3)
Additionally, I measured the error fluctuation range (consistency):
1️⃣ 2° - Python (SPIKE App 3) 🏆
2️⃣ 3° - C Language (spike-rt)
2️⃣ 3° - Word Blocks (SPIKE App 3)
4️⃣ 5° - Python (Pybricks)

Key Takeaways:
C Language (spike-rt) had the most accurate rotation (smallest error)
Python (SPIKE App 3) had the most stable results (smallest variation)

Want to Try C Programming on LEGO SPIKE Prime?

If you’re interested in trying C on SPIKE Prime, there are beginner-friendly learning materials available. As of March 2025, a trial version is also accessible—give it a try!