diff --git a/dda.c b/dda.c
index 16b8c3c..7416461 100644
--- a/dda.c
+++ b/dda.c
@@ -395,14 +395,15 @@ void dda_create(DDA *dda, TARGET *target) {
         dda_join_moves(prev_dda, dda);
         dda->n = dda->start_steps;
         if (dda->n == 0)
-          dda->c = pgm_read_dword(&c0_P[X]);
+          dda->c = pgm_read_dword(&c0_P[dda->fast_axis]);
         else
-          dda->c = ((pgm_read_dword(&c0_P[X]) >> 8) * int_inv_sqrt(dda->n)) >> 5;
+          dda->c = ((pgm_read_dword(&c0_P[dda->fast_axis]) >> 8) *
+                    int_inv_sqrt(dda->n)) >> 5;
         if (dda->c < dda->c_min)
           dda->c = dda->c_min;
       #else
         dda->n = 0;
-        dda->c = pgm_read_dword(&c0_P[X]);
+        dda->c = pgm_read_dword(&c0_P[dda->fast_axis]);
       #endif
 
 		#elif defined ACCELERATION_TEMPORAL
@@ -865,11 +866,12 @@ void dda_clock() {
     }
     if (recalc_speed) {
       if (dda->n == 0)
-        move_c = pgm_read_dword(&c0_P[X]);
+        move_c = pgm_read_dword(&c0_P[dda->fast_axis]);
       else
         // Explicit formula: c0 * (sqrt(n + 1) - sqrt(n)),
         // approximation here: c0 * (1 / (2 * sqrt(n))).
-        move_c = ((pgm_read_dword(&c0_P[X]) >> 8) * int_inv_sqrt(dda->n)) >> 5;
+        move_c = ((pgm_read_dword(&c0_P[dda->fast_axis]) >> 8) *
+                  int_inv_sqrt(dda->n)) >> 5;
 
       // TODO: most likely this whole check is obsolete. It was left as a
       //       safety margin, only. Rampup steps calculation should be accurate