diff --git a/Application/closed_loop.py b/Application/closed_loop.py
index d59c07b02e2e4575ade92be49c4434ce75e4b2f5..2b27ed456f30cd5d90a437b6731eaf4f38b7edb9 100644
--- a/Application/closed_loop.py
+++ b/Application/closed_loop.py
@@ -1,5 +1,4 @@
 import numpy as np
-import cv2 as cv
 import TRex
 import subprocess
 from queue import Queue, Empty
@@ -68,11 +67,11 @@ def update_tracking():
 
         if len(midlines) > i and not np.isinf(midline[0]).any():
             print(midlines, i, midline[0].min(), midline[0].max())
-            cv.circle(image, tuple(midline[0]), 5, (255, 0, 0), -1)
-            for j in range(1, len(midline)):
-                cv.line(image, tuple(midline[j-1, :]), tuple(midline[j]), (255, 255, 255))
+            #cv.circle(image, tuple(midline[0]), 5, (255, 0, 0), -1)
+            #for j in range(1, len(midline)):
+            #    cv.line(image, tuple(midline[j-1, :]), tuple(midline[j]), (255, 255, 255))
 
-        cv.circle(image, pos, 5, color, -1)
+        #cv.circle(image, pos, 5, color, -1)
 
         if key != 1 or i >= len(visual_field):
            continue 
@@ -81,9 +80,9 @@ def update_tracking():
                 continue
             j = np.where(ids == id)[0]
             other = tuple(((positions[j] + centers[j]) * scale_factor)[0].astype(np.int))
-            cv.line(image, pos, other, (255, 255, 255))
+            #cv.line(image, pos, other, (255, 255, 255))
         #print("tracking", frame, key, positions[key][3])
         #cv.imwrite("image.png", image)
-    cv.putText(image, str(frame), (10, 10), cv.FONT_HERSHEY_PLAIN, 0.5, (255, 255, 255))
-    image = cv.cvtColor(image, cv.COLOR_BGR2RGB)
-    TRex.imshow("image", image)
+    #cv.putText(image, str(frame), (10, 10), cv.FONT_HERSHEY_PLAIN, 0.5, (255, 255, 255))
+    #image = cv.cvtColor(image, cv.COLOR_BGR2RGB)
+    #TRex.imshow("image", image)
diff --git a/Application/learn_static.py b/Application/learn_static.py
index 3fc5a8e6271592fb9377561443a38d6609525e7e..6ffa4ce644695ba74ffcc25265586824416da5c2 100644
--- a/Application/learn_static.py
+++ b/Application/learn_static.py
@@ -12,7 +12,6 @@ from keras.models import Sequential
 import keras
 import keras.backend as K
 import numpy as np
-import cv2 as cv
 from keras.preprocessing.image import ImageDataGenerator
 from keras.utils import np_utils
 import TRex
diff --git a/Application/utils.py b/Application/utils.py
index 4680f47b7d00b2b9a24ed9a37a4070bc0a3b4fe0..f2cc3dd982a9e32e6f16c11c3f718e91221b1903 100644
--- a/Application/utils.py
+++ b/Application/utils.py
@@ -23,7 +23,6 @@ def figure_as_image():
     from PIL import Image
     import matplotlib.pyplot as plt
     import numpy as np
-    import cv2 as cv
     import os
     
     plt.gcf().set_tight_layout(True)
@@ -43,7 +42,6 @@ def show_figure(title="plot", output_path="", im = None):
     import TRex
     from PIL import Image
     import numpy as np
-    import cv2 as cv
     import os
 
     if type(im) == type(None):
diff --git a/conda/meta.yaml b/conda/meta.yaml
index d46547cf92c665d9a74d8547571c47a2a544987e..11b3102b2a82778d854eb73933f12d590311e67c 100644
--- a/conda/meta.yaml
+++ b/conda/meta.yaml
@@ -4,7 +4,7 @@ package:
   
 source:
     git_url: https://github.com/mooch443/trex.git
-    git_rev: 55c25dd862ceb5931193cf0d91f783562203d193
+    git_rev: 24f2483b311d0944374f0b817b1cab91a294f94c
     git_depth: 1
 
 build:
@@ -73,8 +73,6 @@ requirements:
     - python [win]
     - ffmpeg ==4.0 [not win]
     - python =3.6 [not win]
-    - py-opencv
-    - opencv
     - glfw [linux]
 
 about:
diff --git a/docs/install.rst b/docs/install.rst
index 3c5384791329f8275a7df00f876412e3ad9dc694..2cfc80bebb8c46f2750ceedf55557476f963e915 100644
--- a/docs/install.rst
+++ b/docs/install.rst
@@ -12,11 +12,11 @@ Installation
 The easy way
 ************
 
-|trex| supports all major platforms. There is an easy way to install |trex| using Anaconda, by creating a new virtual environment (here named ``tracking``, which you can replace)::
+|trex| supports all major platforms. There is an easy way to install |trex| using Anaconda, by creating a new virtual environment (here named ``tracking``, which you can replace) -- the down-side is, that it needs to be compiled with fewer optimzations and features than a manually compiled one due to compatibility and licensing issues::
 
 	conda create -n tracking -c main -c conda-forge -c trexing trex
 
-The conda version does not offer support for Basler cameras. If you need to use |grabs| with machine vision cameras, please consider compiling the software yourself -- it has other advantages, too (such as enabling some Metal features on macOS and getting a squeaky clean, new version)!
+The conda version does not offer support for Basler cameras. If you need to use |grabs| with machine vision cameras, please consider compiling the software yourself -- it has other advantages, too (such as enabling some Metal features on macOS, architecture-specific optimizations and just getting a squeaky clean, new version)!
 
 Compile it yourself
 *******************