5 
SHORT TERM ============= (*needed* for streamlines & tractography) 
SHORT TERM ============= (*needed* for streamlines & tractography) 
6 
======================== 
======================== 
7 


8 

Remove CL from compiler 
9 


10 
[GLK:3] Add sequence types (needed for evals & evecs) 
[GLK:3] Add sequence types (needed for evals & evecs) 
11 
syntax 
syntax 
12 
types: ty '{' INT '}' 
types: ty '{' INT '}' 
25 
SHORTISH TERM ========= (to make using Diderot less annoying to 
SHORTISH TERM ========= (to make using Diderot less annoying to 
26 
======================== program in, and slow to execute) 
======================== program in, and slow to execute) 
27 


28 
valuenumbering optimization [DONE, but needs more testing] 
valuenumbering optimization [DONE] 
29 


30 
Allow ".ddro" file extensions in addition to ".diderot" 
Allow ".ddro" file extensions in addition to ".diderot" 
31 


32 
Be able to output values of type tensor[2,2] and tensor[3,3] 
Be able to output values of type tensor[2,2] and tensor[3,3]; 
33 
(currently only scalars & vectors) 
(currently only scalars & vectors). Want to add some regression tests 
34 

based on this and currently can't 
35 


36 
[GLK:1] Add a clamp function, which takes three arguments; either 
[GLK:1] Add a clamp function, which takes three arguments; either 
37 
three scalars: 
three scalars: 
43 
One question: clamp(x, lo, hi) is the argument order used in OpenCL 
One question: clamp(x, lo, hi) is the argument order used in OpenCL 
44 
and other places, but clamp(lo, hi, x) is much more consistent with 
and other places, but clamp(lo, hi, x) is much more consistent with 
45 
lerp(lo, hi, x), hence GLK's preference 
lerp(lo, hi, x), hence GLK's preference 
46 

[DONE] 
47 


48 
[GLK:2] Proper handling of stabilize method 
[GLK:2] Proper handling of stabilize method 
49 


63 
to index into complete list 
to index into complete list 
64 


65 
[GLK:6] Use of Teem's "hest" commandline parser for getting 
[GLK:6] Use of Teem's "hest" commandline parser for getting 
66 
any input variables that are not defined in the source file 
any "input" variables that are not defined in the source file. 
67 


68 
[GLK:7] ability to declare a field so that probe positions are 
[GLK:7] ability to declare a field so that probe positions are 
69 
*always* "inside"; with various ways of mapping the known image values 
*always* "inside"; with various ways of mapping the known image values 
96 


97 
"initially" supports lists 
"initially" supports lists 
98 


99 
"initially" supports lists of positions output from 
"initially" supports lists of positions output from different 
100 
different initalization Diderot program 
initalization Diderot program (or output from the same program; 
101 

e.g. using output of iso2d.diderot for one isovalue to seed the input 
102 

to another invocation of the same program) 
103 


104 
Communication between strands: they have to be able to learn each 
Communication between strands: they have to be able to learn each 
105 
other's state (at the previous iteration). Early version of this can 
other's state (at the previous iteration). Early version of this can 
210 
field#2(3)[] F = bspln3 ⊛ img; 
field#2(3)[] F = bspln3 ⊛ img; 
211 
or, as a tensor product of kernels, one for each axis, e.g. 
or, as a tensor product of kernels, one for each axis, e.g. 
212 
field#0(3)[] F = (bspln3 ⊗ bspln3 ⊗ tent) ⊛ img; 
field#0(3)[] F = (bspln3 ⊗ bspln3 ⊗ tent) ⊛ img; 
213 
This is especially important for things like timevarying data, or 
This is especially important for things like timevarying fields 
214 
other multidimensional fields where one axis of the domain is very 
and the use of scalespace in field visualization: one axis of the 
215 
different from the rest, and hence must be treated separately when 
must be convolved with a different kernel during probing. 
216 
it comes to convolution. What is very unclear is how, in such cases, 
What is very unclear is how, in such cases, we should notate the 
217 
we should notate the gradient, when we only want to differentiate with 
gradient, when we only want to differentiate with respect to some 
218 
respect to some subset of the axes. One ambitious idea would be: 
subset of the axes. One ambitious idea would be: 
219 
field#0(3)[] Ft = (bspln3 ⊗ bspln3 ⊗ tent) ⊛ img; // 2D timevarying field 
field#0(3)[] Ft = (bspln3 ⊗ bspln3 ⊗ tent) ⊛ img; // 2D timevarying field 
220 
field#0(2)[] F = lambda([x,y], Ft([x,y,42.0])) // restriction to time=42.0 
field#0(2)[] F = lambda([x,y], Ft([x,y,42.0])) // restriction to time=42.0 
221 
vec2 grad = ∇F([x,y]); // 2D gradient 
vec2 grad = ∇F([x,y]); // 2D gradient 
222 


223 

Tensors of order 3 (e.g. gradients of diffusion tensor fields, or 
224 

hessians of vector fields) and order 4 (e.g. Hessians of diffusion 
225 

tensor fields). 
226 


227 
representation of tensor symmetry 
representation of tensor symmetry 
228 
(have to identify the group of index permutations that are symmetries) 
(have to identify the group of index permutations that are symmetries) 
229 


231 


232 
outer works on all tensors 
outer works on all tensors 
233 


234 

Help for debugging Diderot programs: need to be able to uniquely 
235 

identify strands, and for particular strands that are known to behave 
236 

badly, do something like printf or other logging of their computations 
237 

and updates. 
238 


239 

Permit writing dimensionally general code: Have some statement of the 
240 

dimension of the world "W" (or have it be learned from one particular 
241 

field of interest), and then able to write "vec" instead of 
242 

"vec2/vec3", and perhaps "tensor[W,W]" instead of 
243 

"tensor[2,2]/tensor[3,3]" 
244 


245 

Traits: all things things that have boilerplate code (especially 
246 

volume rendering) should be expressed in terms of the unique 
247 

computational core. Different kinds of streamline/tractography 
248 

computation will be another example, as well as particle systems. 
249 


250 
Einstein summation notation 
Einstein summation notation 
251 


252 
"tensor comprehension" (like list comprehension) 
"tensor comprehension" (like list comprehension) 
253 


254 

Fields coming from different sources of data: 
255 

* triangular or tetrahedral meshes over 2D or 3D domains (of the 
256 

source produced by finiteelement codes; these will come with their 
257 

own specialized kinds of reconstruction kernels, called "basis 
258 

functions" in this context) 
259 

* Large point clouds, with some radial basis function around each point, 
260 

which will be tuned by parameters of the point (at least one parameter 
261 

giving some notion of radius) 
262 


263 
====================== 
====================== 
264 
BUGS ================= 
BUGS ================= 
265 
====================== 
====================== 