diff --git a/Makefile b/Makefile
index bf9fb93..1170055 100644
--- a/Makefile
+++ b/Makefile
@@ -19,8 +19,8 @@ camera.o: camera.c
 scene.o: scene.c
 	gcc $(LINKS) -c scene.c -o scene.o
 
-presentation: presentation.md
-    pandoc -t beamer -s presentation.md -o pres.pdf    
+pres: presentation.md
+	pandoc -t beamer presentation.md -o pres.pdf    
 
 clean:
 	rm *.o blackpink
diff --git a/img/b4.jpg b/img/b4.jpg
new file mode 100644
index 0000000..1ae5728
Binary files /dev/null and b/img/b4.jpg differ
diff --git a/presentation.md b/presentation.md
index f570892..ad7f33c 100644
--- a/presentation.md
+++ b/presentation.md
@@ -65,16 +65,29 @@ At one point we were leaking memory at approximately 150Mb per second.
 
 ---
 
-The first success was 1pm on saturday, where we got a circle drawn to the
-screen. 
+![1pm Saturday](img/b3.png)
 
-![circle](img/b3.png)
+---
+
+![Fail states were often interesting](img/b2.png)
 
 ---
 
-Fail states were often interesting. 
+12 Hours later, we could render 
+
+- circles
+- boxes, and 
+- straight lines 
+
+from one angle, and move them around the scene 
+
+- as long as we got lucky with whatever the 4th dimension was doing. 
 
-![cool errors](img/b2.png)
+Plus some post-processing and lighting.
+
+---
+
+![a green sphere](img/b4.jpg)
 
 ---
 
@@ -84,7 +97,6 @@ video? image? demo?
 
 ---
 
-
 # Takeaways 1
 
 - sticking raymarching vectors to manifolds in 4 dimensions is possible but has