diff --git a/src/scgenerator/noise.py b/src/scgenerator/noise.py
index ec0cd8b..63cf7a0 100644
--- a/src/scgenerator/noise.py
+++ b/src/scgenerator/noise.py
@@ -104,7 +104,8 @@ class NoiseMeasurement:
         dB: bool = True,
         wavelength: float | None = None,
         power: float | None = None,
-        crop: int | float = 1,
+        left: int = 1,
+        right: int = -1,
     ) -> tuple[np.ndarray, np.ndarray]:
         """
         Transforms the PSD in a way that makes it easy to plot
@@ -116,9 +117,12 @@ class NoiseMeasurement:
         wavelength, power : float | None, optional
             if both are provided, will be used to compute the quantum noise limit and the PSD will
             be output relative to that amount
-        crop : int, optional
-            how many low frequency points to drop. The default (1) is to drop only the 0 frequency
+        left : int, optional
+            Index of the first frequency point. The default (1) is to drop only the 0 frequency
             point, as it is outside the plot range when plotting the PSD on a log-log plot.
+        right : int, optional
+            index of the last frequency point, non inclusive (i.e. like slices). The default (-1) is
+            to drop only the Nyquist frequency, as segmentation artificially reduces its value.
 
         Returns
         -------
@@ -131,7 +135,7 @@ class NoiseMeasurement:
         -------
         >>> noise = NoiseMeasurement(*np.load("my_measurement.npy"))
         >>> plt.plot(*noise.plottable(wavelength=800e-9, power=0.3)) # dB above quantum limit
-        >>> plt.xscale("log")
+        >>> plt.xscale("log") # creates log-log plot is dB is already log scale for y axis
         >>> plt.show()
         """
         psd = self.psd
@@ -141,10 +145,7 @@ class NoiseMeasurement:
         if dB:
             psd = math.to_dB(psd, ref=1.0)
 
-        if isinstance(crop, (float, np.floating)):
-            crop = math.argclosest(self.freq, crop)
-
-        return self.freq[crop:], psd[crop:]
+        return self.freq[left:right], psd[left:right]
 
     def sample_spectrum(
         self, nf: int, dt: float | None = None, log_mode: bool = False