diff --git a/examples/3_wave_optics_optimization.jl b/examples/3_wave_optics_optimization.jl
index f02fa6b..be78cbf 100644
--- a/examples/3_wave_optics_optimization.jl
+++ b/examples/3_wave_optics_optimization.jl
@@ -32,6 +32,26 @@ using ImageShow, ImageIO, PlutoUI, IndexFunArrays, Optim, RadonKA, FileIO, Color
 # ╔═╡ 50bb6567-4fea-4794-8a5d-2c519e065296
 using CUDA
 
+# ╔═╡ 86b8c061-2f0e-4e4d-9d5e-7694e152813a
+using NDTools
+
+# ╔═╡ bf0c719f-4fa7-4f58-b031-e535813acedf
+md"""# Supplemental simulations
+
+Note, in the publication we optimize a volume of 550x550x550 voxels which only runs on a large GPU (NVIDIA A100 with 80GB). Hence, here a smaller version which produces similar but not identical results because of a different discretization of the boat.
+
+```bibtex
+@misc{wechsler2024wave,
+      title={Wave optical model for tomographic volumetric additive manufacturing}, 
+      author={Felix Wechsler and Carlo Gigli and Jorge Madrid-Wolff and Christophe Moser},
+      year={2024},
+      eprint={2402.06283},
+      archivePrefix={arXiv},
+      primaryClass={physics.optics}
+}
+```
+"""
+
 # ╔═╡ ba186bc1-2d3d-4422-92b9-98e2657e2d47
 md"# 0. Load packages
 On the first run, Julia is going to install some packages automatically. So start this notebook and give it some minutes (5-10min) to install all packages. 
@@ -80,17 +100,30 @@ loss = LossThreshold(thresholds=(0.65, 0.75))
 # ╔═╡ 7f812aa1-7fae-40fd-8601-a5d8c19a5e49
 md"# 2. Target"
 
-# ╔═╡ f6f61a7e-7e2f-430b-801b-63c495ec79d6
-begin
-	KK = 64
-	target_3D = box(Float32, (KK, KK, KK), (40, 40, 50)) .- 
-	box(Float32, (KK, KK, KK), (30, 30, 50));
-	target = Float32.(Bool.(target_3D) .|| (rr2(size(target_3D)) .< 18^2))
-	target = target
-end;
+# ╔═╡ 1931aa0f-966e-4ca0-b6d6-2ff6995ff8cb
+md"# 1. Load Benchy"
+
+# ╔═╡ d708c3a1-6e59-4001-a93a-69442466a9d5
+function load_benchy(sz, sz_file, path)
+	target = zeros(Float32, sz)
+	#@show size(load(joinpath(path, string("slice_", string(1, pad=3) ,".png"))))
+	
+	for i in 0:sz_file[1]-1
+		target[:, :, 40 + i+1] .= select_region(Gray.(load(joinpath(path, string("boat_", string(i, pad=3) ,".png")))), new_size=(sz))
+	end
+
+	#target2 = zeros(Float32, sz)
+	#WaveOpticsPropagation.set_center!(target2, target
+	target2 = select_region(target, new_size=sz)
+	target2 = permutedims(target2, (3,1,2))[end:-1:begin, :, :]
+	return togoc(target2)
+end
+
+# ╔═╡ 70375fb2-a3f9-482d-b637-7b5867b1e2d0
+target = permutedims(load_benchy((180, 180, 180), (100, 100, 100), "/home/felix/Downloads/benchy/files/output_100/"), (1,2,3));
 
 # ╔═╡ 0594a1b3-cac0-4829-a801-78e2420022eb
-md"z slide value $(@bind slice PlutoUI.Slider(axes(target_3D, 3), show_value=true, default=0.5))"
+md"z slide value $(@bind slice PlutoUI.Slider(axes(target, 3), show_value=true, default=0.5))"
 
 # ╔═╡ cc95eb12-e948-44ba-8165-edf6886efb37
 simshow(Array(target[:, :, slice]))
@@ -135,7 +168,7 @@ res
 md"Threshold value=$(@bind thresh4 PlutoUI.Slider(0:0.01:1, show_value=true, default=0.7))"
 
 # ╔═╡ 605ba331-a480-492f-98aa-402155d33ebf
-md"z slider value $(@bind slice2 PlutoUI.Slider(axes(target_3D, 3), show_value=true, default=0.5))"
+md"z slider value $(@bind slice2 PlutoUI.Slider(axes(target, 3), show_value=true, default=0.5))"
 
 # ╔═╡ 6e67afcc-597d-4c18-a351-e24c5b732f2d
 [simshow(Array(printed[:, :, slice2]), set_one=false) simshow(ones((size(target, 1), 5))) simshow(thresh4 .< Array(printed[:, :, slice2])) simshow(ones((size(target, 1), 5))) simshow(target[:, :, slice2])]
@@ -150,6 +183,7 @@ md"angle $(@bind angle PlutoUI.Slider(axes(patterns, 2), show_value=true, defaul
 simshow(Array(patterns[:,angle,:]), cmap=:turbo, set_one=true)
 
 # ╔═╡ Cell order:
+# ╟─bf0c719f-4fa7-4f58-b031-e535813acedf
 # ╠═ba186bc1-2d3d-4422-92b9-98e2657e2d47
 # ╠═874ffaea-c7ae-11ee-11c6-a7ddbc5bf852
 # ╠═5d5d3113-e8d6-40a9-82d6-1d4f52e10d47
@@ -164,7 +198,10 @@ simshow(Array(patterns[:,angle,:]), cmap=:turbo, set_one=true)
 # ╟─0cae8621-9b65-4506-b604-72719293d64d
 # ╠═ed6b2393-0ebb-4a24-b872-7424723e1f98
 # ╟─7f812aa1-7fae-40fd-8601-a5d8c19a5e49
-# ╠═f6f61a7e-7e2f-430b-801b-63c495ec79d6
+# ╟─1931aa0f-966e-4ca0-b6d6-2ff6995ff8cb
+# ╠═86b8c061-2f0e-4e4d-9d5e-7694e152813a
+# ╠═d708c3a1-6e59-4001-a93a-69442466a9d5
+# ╠═70375fb2-a3f9-482d-b637-7b5867b1e2d0
 # ╟─0594a1b3-cac0-4829-a801-78e2420022eb
 # ╠═cc95eb12-e948-44ba-8165-edf6886efb37
 # ╟─ceb4233f-efaa-4e78-922d-92acf6625ff3